Merge pull request 'jsonrpc_err_in_rent' (#41) from jsonrpc_err_in_rent into gitea_feats

Reviewed-on: #41

README.rst

@@ -1,162 +1,161 @@
 piker
 -----
-trading gear for hackers.
+trading gear for hackers
 
 |gh_actions|
 
 .. |gh_actions| image:: https://img.shields.io/endpoint.svg?url=https%3A%2F%2Factions-badge.atrox.dev%2Fpikers%2Fpiker%2Fbadge&style=popout-square
     :target: https://actions-badge.atrox.dev/piker/pikers/goto
 
-``piker`` is a broker agnostic, next-gen FOSS toolset for real-time
-computational trading targeted at `hardcore Linux users <comp_trader>`_ .
+``piker`` is a broker agnostic, next-gen FOSS toolset and runtime for
+real-time computational trading targeted at `hardcore Linux users
+<comp_trader>`_ .
 
-we use as much bleeding edge tech as possible including (but not limited to):
+we use much bleeding edge tech including (but not limited to):
 
 - latest python for glue_
-- trio_ & tractor_ for our distributed, multi-core, real-time streaming
-  `structured concurrency`_ runtime B)
-- Qt_ for pristine high performance UIs
-- pyqtgraph_ for real-time charting
-- ``polars`` ``numpy`` and ``numba`` for `fast numerics`_
-- `apache arrow and parquet`_ for time series history management
-  persistence and sharing
-- (prototyped) techtonicdb_ for L2 book storage
+- uv_ for packaging and distribution
+- trio_ & tractor_ for our distributed `structured concurrency`_ runtime
+- Qt_ for pristine low latency UIs
+- pyqtgraph_ (which we've extended) for real-time charting and graphics
+- ``polars`` ``numpy`` and ``numba`` for redic `fast numerics`_
+- `apache arrow and parquet`_ for time-series storage
 
-.. |travis| image:: https://img.shields.io/travis/pikers/piker/master.svg
-    :target: https://travis-ci.org/pikers/piker
+potential projects we might integrate with soon,
+
+- (already prototyped in ) techtonicdb_ for L2 book storage
+
+.. _comp_trader: https://jfaleiro.wordpress.com/2019/10/09/computational-trader/
+.. _glue: https://numpy.org/doc/stable/user/c-info.python-as-glue.html#using-python-as-glue
+.. _uv: https://docs.astral.sh/uv/
 .. _trio: https://github.com/python-trio/trio
 .. _tractor: https://github.com/goodboy/tractor
 .. _structured concurrency: https://trio.discourse.group/
-.. _marketstore: https://github.com/alpacahq/marketstore
-.. _techtonicdb: https://github.com/0b01/tectonicdb
 .. _Qt: https://www.qt.io/
 .. _pyqtgraph: https://github.com/pyqtgraph/pyqtgraph
-.. _glue: https://numpy.org/doc/stable/user/c-info.python-as-glue.html#using-python-as-glue
 .. _apache arrow and parquet: https://arrow.apache.org/faq/
 .. _fast numerics: https://zerowithdot.com/python-numpy-and-pandas-performance/
-.. _comp_trader: https://jfaleiro.wordpress.com/2019/10/09/computational-trader/
+.. _techtonicdb: https://github.com/0b01/tectonicdb
 
 
-focus and features:
-*******************
-- 100% federated: your code, your hardware, your data feeds, your broker fills.
-- zero web: low latency, native software that doesn't try to re-invent the OS
-- maximal **privacy**: prevent brokers and mms from knowing your
-  planz; smack their spreads with dark volume.
-- zero clutter: modal, context oriented UIs that echew minimalism, reduce
-  thought noise and encourage un-emotion.
-- first class parallelism: built from the ground up on next-gen structured concurrency
-  primitives.
-- traders first: broker/exchange/asset-class agnostic
-- systems grounded: real-time financial signal processing that will
-  make any queuing or DSP eng juice their shorts.
-- non-tina UX: sleek, powerful keyboard driven interaction with expected use in tiling wms
-- data collaboration: every process and protocol is multi-host scalable.
-- fight club ready: zero interest in adoption by suits; no corporate friendly license, ever.
+focus and feats:
+****************
+fitting with these tenets, we're always open to new
+framework/lib/service interop suggestions and ideas!
 
-fitting with these tenets, we're always open to new framework suggestions and ideas.
+- **100% federated**:
+  your code, your hardware, your data feeds, your broker fills.
 
-building the best looking, most reliable, keyboard friendly trading
-platform is the dream; join the cause.
+- **zero web**:
+  low latency as a prime objective, native UIs and modern IPC
+  protocols without trying to re-invent the "OS-as-an-app"..
+
+- **maximal privacy**:
+  prevent brokers and mms from knowing your planz; smack their
+  spreads with dark volume from a VPN tunnel.
+
+- **zero clutter**:
+  modal, context oriented UIs that echew minimalism, reduce thought
+  noise and encourage un-emotion.
+
+- **first class parallelism**:
+  built from the ground up on a next-gen structured concurrency
+  supervision sys.
+
+- **traders first**:
+  broker/exchange/venue/asset-class/money-sys agnostic
+
+- **systems grounded**:
+  real-time financial signal processing (fsp) that will make any
+  queuing or DSP eng juice their shorts.
+
+- **non-tina UX**:
+  sleek, powerful keyboard driven interaction with expected use in
+  tiling wms (or maybe even a DDE).
+
+- **data collab at scale**:
+  every actor-process and protocol is multi-host aware.
+
+- **fight club ready**:
+  zero interest in adoption by suits; no corporate friendly license,
+  ever.
+
+building the hottest looking, fastest, most reliable, keyboard
+friendly FOSS trading platform is the dream; join the cause.
 
 
-sane install with `poetry`
-**************************
-TODO!
+a sane install with `uv`
+************************
+bc why install with `python` when you can faster with `rust` ::
 
-
-rigorous install on ``nixos`` using ``poetry2nix``
-**************************************************
-TODO!
+    uv lock
 
 
 hacky install on nixos
 **********************
-`NixOS` is our core devs' distro of choice for which we offer
+``NixOS`` is our core devs' distro of choice for which we offer
 a stringently defined development shell envoirment that can be loaded with::
 
-    nix-shell develop.nix
-
-this will setup the required python environment to run piker, make sure to
-run::
-
-    pip install -r requirements.txt -e .
-
-once after loading the shell
+    nix-shell default.nix
 
 
-install wild-west style via `pip`
-*********************************
-``piker`` is currently under heavy pre-alpha development and as such
-should be cloned from this repo and hacked on directly.
+start a chart
+*************
+run a realtime OHLCV chart stand-alone::
 
-for a development install::
+    piker -l info chart btcusdt.spot.binance xmrusdt.spot.kraken
 
-    git clone git@github.com:pikers/piker.git
-    cd piker
-    virtualenv env
-    source ./env/bin/activate
-    pip install -r requirements.txt -e .
+this runs a chart UI (with 1m sampled OHLCV) and shows 2 spot markets from 2 diff cexes
+overlayed on the same graph. Use of `piker` without first starting
+a daemon (`pikerd` - see below) means there is an implicit spawning of the
+multi-actor-runtime (implemented as a `tractor` app).
+
+For additional subsystem feats available through our chart UI see the
+various sub-readmes:
+
+- order control using a mouse-n-keyboard UX B)
+- cross venue market-pair (what most call "symbol") search, select, overlay Bo
+- financial-signal-processing (`piker.fsp`) write-n-reload to sub-chart BO
+- src-asset derivatives scan for anal, like the infamous "max pain" XO
 
 
-check out our charts
-********************
-bet you weren't expecting this from the foss::
-
-    piker -l info -b kraken -b binance chart btcusdt.binance --pdb
-
-
-this runs the main chart (currently with 1m sampled OHLC) in in debug
-mode and you can practice paper trading using the following
-micro-manual:
-
-``order_mode`` (
-    edge triggered activation by any of the following keys,
-    ``mouse-click`` on y-level to submit at that price
-    ):
-
-    - ``f``/ ``ctl-f`` to stage buy
-    - ``d``/ ``ctl-d`` to stage sell
-    - ``a`` to stage alert
-
-
-``search_mode`` (
-    ``ctl-l`` or ``ctl-space`` to open,
-    ``ctl-c`` or ``ctl-space`` to close
-    ) :
-
-    - begin typing to have symbol search automatically lookup
-      symbols from all loaded backend (broker) providers
-    - arrow keys and mouse click to navigate selection
-    - vi-like ``ctl-[hjkl]`` for navigation
-
-
-you can also configure your position allocation limits from the
-sidepane.
-
-
-run in distributed mode
-***********************
-start the service manager and data feed daemon in the background and
-connect to it::
+spawn a daemon standalone
+*************************
+we call the root actor-process the ``pikerd``. it can be (and is
+recommended normally to be) started separately from the ``piker
+chart`` program::
 
     pikerd -l info --pdb
 
+the daemon does nothing until a ``piker``-client (like ``piker
+chart``) connects and requests some particular sub-system. for
+a connecting chart ``pikerd`` will spawn and manage at least,
 
-connect your chart::
+- a data-feed daemon: ``datad`` which does all the work of comms with
+  the backend provider (in this case the ``binance`` cex).
+- a paper-trading engine instance, ``paperboi.binance``, (if no live
+  account has been configured) which allows for auto/manual order
+  control against the live quote stream.
 
-    piker -l info -b kraken -b binance chart xmrusdt.binance --pdb
+*using* an actor-service (aka micro-daemon) manager which dynamically
+supervises various sub-subsystems-as-services throughout the ``piker``
+runtime-stack.
 
+now you can (implicitly) connect your chart::
 
-enjoy persistent real-time data feeds tied to daemon lifetime. the next
-time you spawn a chart it will load much faster since the data feed has
-been cached and is now always running live in the background until you
-kill ``pikerd``.
+    piker chart btcusdt.spot.binance
+
+since ``pikerd`` was started separately you can now enjoy a persistent
+real-time data stream tied to the daemon-tree's lifetime. i.e. the next
+time you spawn a chart it will obviously not only load much faster
+(since the underlying ``datad.binance`` is left running with its
+in-memory IPC data structures) but also the data-feed and any order
+mgmt states should be persistent until you finally cancel ``pikerd``.
 
 
 if anyone asks you what this project is about
 *********************************************
-you don't talk about it.
+you don't talk about it; just use it.
 
 
 how do i get involved?
@@ -166,6 +165,15 @@ enter the matrix.
 
 how come there ain't that many docs
 ***********************************
-suck it up, learn the code; no one is trying to sell you on anything.
-also, we need lotsa help so if you want to start somewhere and can't
-necessarily write serious code, this might be the place for you!
+i mean we want/need them but building the core right has been higher
+prio then marketting (and likely will stay that way Bp).
+
+soo, suck it up bc,
+
+- no one is trying to sell you on anything
+- learning the code base is prolly way more valuable
+- the UI/UXs are intended to be "intuitive" for any hacker..
+
+we obviously need tonz help so if you want to start somewhere and
+can't necessarily write "advanced" concurrent python/rust code, this
+helping document literally anything might be the place for you!

config/brokers.toml

@@ -2,13 +2,15 @@
 # ---- CEXY ----
 ################
 [binance]
+accounts.paper = 'paper'
+
 accounts.usdtm = 'futes'
-futes.use_testnet = true
+futes.use_testnet = false
 futes.api_key = ''
 futes.api_secret = ''
 
 accounts.spot = 'spot'
-spot.use_testnet = true
+spot.use_testnet = false
 spot.api_key = ''
 spot.api_secret = ''
 

default.nix

@@ -0,0 +1,134 @@
+with (import <nixpkgs> {});
+let
+  glibStorePath = lib.getLib glib;
+  zlibStorePath = lib.getLib zlib;
+  zstdStorePath = lib.getLib zstd;
+  dbusStorePath = lib.getLib dbus;
+  libGLStorePath = lib.getLib libGL;
+  freetypeStorePath = lib.getLib freetype;
+  qt6baseStorePath = lib.getLib qt6.qtbase;
+  fontconfigStorePath = lib.getLib fontconfig;
+  libxkbcommonStorePath = lib.getLib libxkbcommon;
+  xcbutilcursorStorePath = lib.getLib xcb-util-cursor;
+
+  qtpyStorePath = lib.getLib python312Packages.qtpy;
+  pyqt6StorePath = lib.getLib python312Packages.pyqt6;
+  pyqt6SipStorePath = lib.getLib python312Packages.pyqt6-sip;
+  rapidfuzzStorePath = lib.getLib python312Packages.rapidfuzz;
+  qdarkstyleStorePath = lib.getLib python312Packages.qdarkstyle;
+
+  xorgLibX11StorePath = lib.getLib xorg.libX11;
+  xorgLibxcbStorePath = lib.getLib xorg.libxcb;
+  xorgxcbutilwmStorePath = lib.getLib xorg.xcbutilwm;
+  xorgxcbutilimageStorePath = lib.getLib xorg.xcbutilimage;
+  xorgxcbutilerrorsStorePath = lib.getLib xorg.xcbutilerrors;
+  xorgxcbutilkeysymsStorePath = lib.getLib xorg.xcbutilkeysyms;
+  xorgxcbutilrenderutilStorePath = lib.getLib xorg.xcbutilrenderutil;
+in
+stdenv.mkDerivation {
+  name = "piker-qt6-uv";
+  buildInputs = [
+    # System requirements.
+    glib
+    zlib
+    dbus
+    zstd
+    libGL
+    freetype
+    qt6.qtbase
+    libgcc.lib
+    fontconfig
+    libxkbcommon
+
+    # Xorg requirements
+    xcb-util-cursor
+    xorg.libxcb
+    xorg.libX11
+    xorg.xcbutilwm
+    xorg.xcbutilimage
+    xorg.xcbutilerrors
+    xorg.xcbutilkeysyms
+    xorg.xcbutilrenderutil
+
+    # Python requirements.
+    python312Full
+    python312Packages.uv
+    python312Packages.qdarkstyle
+    python312Packages.rapidfuzz
+    python312Packages.pyqt6
+    python312Packages.qtpy
+  ];
+  src = null;
+  shellHook = ''
+    set -e
+
+    # Set the Qt plugin path
+    # export QT_DEBUG_PLUGINS=1
+
+    QTBASE_PATH="${qt6baseStorePath}/lib"
+    QT_PLUGIN_PATH="$QTBASE_PATH/qt-6/plugins"
+    QT_QPA_PLATFORM_PLUGIN_PATH="$QT_PLUGIN_PATH/platforms"
+
+    LIB_GCC_PATH="${libgcc.lib}/lib"
+    GLIB_PATH="${glibStorePath}/lib"
+    ZSTD_PATH="${zstdStorePath}/lib"
+    ZLIB_PATH="${zlibStorePath}/lib"
+    DBUS_PATH="${dbusStorePath}/lib"
+    LIBGL_PATH="${libGLStorePath}/lib"
+    FREETYPE_PATH="${freetypeStorePath}/lib"
+    FONTCONFIG_PATH="${fontconfigStorePath}/lib"
+    LIB_XKB_COMMON_PATH="${libxkbcommonStorePath}/lib"
+
+    XCB_UTIL_CURSOR_PATH="${xcbutilcursorStorePath}/lib"
+    XORG_LIB_X11_PATH="${xorgLibX11StorePath}/lib"
+    XORG_LIB_XCB_PATH="${xorgLibxcbStorePath}/lib"
+    XORG_XCB_UTIL_IMAGE_PATH="${xorgxcbutilimageStorePath}/lib"
+    XORG_XCB_UTIL_WM_PATH="${xorgxcbutilwmStorePath}/lib"
+    XORG_XCB_UTIL_RENDER_UTIL_PATH="${xorgxcbutilrenderutilStorePath}/lib"
+    XORG_XCB_UTIL_KEYSYMS_PATH="${xorgxcbutilkeysymsStorePath}/lib"
+    XORG_XCB_UTIL_ERRORS_PATH="${xorgxcbutilerrorsStorePath}/lib"
+
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$QTBASE_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$QT_PLUGIN_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$QT_QPA_PLATFORM_PLUGIN_PATH"
+
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$LIB_GCC_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$DBUS_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$GLIB_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$ZLIB_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$ZSTD_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$LIBGL_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$FONTCONFIG_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$FREETYPE_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$LIB_XKB_COMMON_PATH"
+
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$XCB_UTIL_CURSOR_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$XORG_LIB_X11_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$XORG_LIB_XCB_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$XORG_XCB_UTIL_IMAGE_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$XORG_XCB_UTIL_WM_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$XORG_XCB_UTIL_RENDER_UTIL_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$XORG_XCB_UTIL_KEYSYMS_PATH"
+    LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$XORG_XCB_UTIL_ERRORS_PATH"
+
+    export LD_LIBRARY_PATH
+
+    RPDFUZZ_PATH="${rapidfuzzStorePath}/lib/python3.12/site-packages"
+    QDRKSTYLE_PATH="${qdarkstyleStorePath}/lib/python3.12/site-packages"
+    QTPY_PATH="${qtpyStorePath}/lib/python3.12/site-packages"
+    PYQT6_PATH="${pyqt6StorePath}/lib/python3.12/site-packages"
+    PYQT6_SIP_PATH="${pyqt6SipStorePath}/lib/python3.12/site-packages"
+
+    PATCH="$PATCH:$RPDFUZZ_PATH"
+    PATCH="$PATCH:$QDRKSTYLE_PATH"
+    PATCH="$PATCH:$QTPY_PATH"
+    PATCH="$PATCH:$PYQT6_PATH"
+    PATCH="$PATCH:$PYQT6_SIP_PATH"
+
+    export PATCH
+
+    # Install deps
+    uv lock
+
+  '';
+}

develop.nix

@@ -1,28 +1,34 @@
 with (import <nixpkgs> {});
-with python310Packages;
+
 stdenv.mkDerivation {
-  name = "pip-env";
+  name = "poetry-env";
   buildInputs = [
     # System requirements.
     readline
 
     # TODO: hacky non-poetry install stuff we need to get rid of!!
-    virtualenv
-    setuptools
-    pip
-
-    # obviously, and see below for hacked linking
-    pyqt5
+    poetry
+    # virtualenv
+    # setuptools
+    # pip
 
     # Python requirements (enough to get a virtualenv going).
-    python310Full
+    python311Full
+
+    # obviously, and see below for hacked linking
+    python311Packages.pyqt5
+    python311Packages.pyqt5_sip
+    # python311Packages.qtpy
 
     # numerics deps
-    python310Packages.python-Levenshtein
-    python310Packages.fastparquet
-    python310Packages.polars
+    python311Packages.levenshtein
+    python311Packages.fastparquet
+    python311Packages.polars
 
   ];
+  # environment.sessionVariables = {
+  #   LD_LIBRARY_PATH = "${pkgs.stdenv.cc.cc.lib}/lib";
+  # };
   src = null;
   shellHook = ''
     # Allow the use of wheels.
@@ -30,13 +36,12 @@ stdenv.mkDerivation {
 
     # Augment the dynamic linker path
     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:${R}/lib/R/lib:${readline}/lib
-
     export QT_QPA_PLATFORM_PLUGIN_PATH="${qt5.qtbase.bin}/lib/qt-${qt5.qtbase.version}/plugins";
 
-    if [ ! -d "venv" ]; then
-        virtualenv venv
+    if [ ! -d ".venv" ]; then
+        poetry install --with uis
     fi
 
-    source venv/bin/activate
+    poetry shell
   '';
 }

dockering/ib/docker-compose.yml

@@ -19,8 +19,9 @@ services:
 
     # other image tags available:
     # https://github.com/waytrade/ib-gateway-docker#supported-tags
-    # image: waytrade/ib-gateway:981.3j
-    image: waytrade/ib-gateway:1012.2i
+    # image: waytrade/ib-gateway:1012.2i
+    image: ghcr.io/gnzsnz/ib-gateway:latest
+
     restart: 'no'  # restart on boot whenev there's a crash or user clicsk
     network_mode: 'host'
 

dockering/ib/ibc.ini

@@ -117,9 +117,57 @@ SecondFactorDevice=
 
 # If you use the IBKR Mobile app for second factor authentication,
 # and you fail to complete the process before the time limit imposed
-# by IBKR, you can use this setting to tell IBC to exit: arrangements
-# can then be made to automatically restart IBC in order to initiate
-# the login sequence afresh. Otherwise, manual intervention at TWS's
+# by IBKR, this setting tells IBC whether to automatically restart
+# the login sequence, giving you another opportunity to complete
+# second factor authentication. 
+#
+# Permitted values are 'yes' and 'no'.
+#
+# If this setting is not present or has no value, then the value
+# of the deprecated ExitAfterSecondFactorAuthenticationTimeout is
+# used instead. If this also has no value, then this setting defaults
+# to 'no'.
+#
+# NB: you must be using IBC v3.14.0 or later to use this setting:
+# earlier versions ignore it.
+
+ReloginAfterSecondFactorAuthenticationTimeout=
+
+
+# This setting is only relevant if
+# ReloginAfterSecondFactorAuthenticationTimeout is set to 'yes',
+# or if ExitAfterSecondFactorAuthenticationTimeout is set to 'yes'.
+#
+# It controls how long (in seconds) IBC waits for login to complete
+# after the user acknowledges the second factor authentication
+# alert at the IBKR Mobile app. If login has not completed after
+# this time, IBC terminates.
+# The default value is 60.
+
+SecondFactorAuthenticationExitInterval=
+
+
+# This setting specifies the timeout for second factor authentication
+# imposed by IB. The value is in seconds. You should not change this
+# setting unless you have reason to believe that IB has changed the
+# timeout. The default value is 180.
+
+SecondFactorAuthenticationTimeout=180
+
+
+# DEPRECATED SETTING
+# ------------------
+#
+# ExitAfterSecondFactorAuthenticationTimeout - THIS SETTING WILL BE
+# REMOVED IN A FUTURE RELEASE. For IBC version 3.14.0 and later, see
+# the notes for ReloginAfterSecondFactorAuthenticationTimeout above.
+#
+# For IBC versions earlier than 3.14.0: If you use the IBKR Mobile
+# app for second factor authentication, and you fail to complete the
+# process before the time limit imposed by IBKR, you can use this
+# setting to tell IBC to exit: arrangements can then be made to
+# automatically restart IBC in order to initiate the login sequence
+# afresh. Otherwise, manual intervention at TWS's
 # Second Factor Authentication dialog is needed to complete the
 # login.
 #
@@ -132,29 +180,18 @@ SecondFactorDevice=
 ExitAfterSecondFactorAuthenticationTimeout=no
 
 
-# This setting is only relevant if
-# ExitAfterSecondFactorAuthenticationTimeout is set to 'yes'.
-#
-# It controls how long (in seconds) IBC waits for login to complete
-# after the user acknowledges the second factor authentication
-# alert at the IBKR Mobile app. If login has not completed after
-# this time, IBC terminates.
-# The default value is 40.
-
-SecondFactorAuthenticationExitInterval=
-
-
 # Trading Mode
 # ------------
 #
-# TWS 955 introduced a new Trading Mode combo box on its login
-# dialog. This indicates whether the live account or the paper
-# trading account corresponding to the supplied credentials is
-# to be used. The allowed values are 'live' (the default) and
-# 'paper'. For earlier versions of TWS this setting has no
-# effect.
+# This indicates whether the live account or the paper trading
+# account corresponding to the supplied credentials is to be used.
+# The allowed values are 'live' (the default) and 'paper'.
+#
+# If this is set to 'live', then the credentials for the live
+# account must be supplied. If it is set to 'paper', then either
+# the live or the paper-trading credentials may be supplied.
 
-TradingMode=
+TradingMode=paper
 
 
 # Paper-trading Account Warning
@@ -188,7 +225,7 @@ AcceptNonBrokerageAccountWarning=yes
 #
 # The default value is 60.
 
-LoginDialogDisplayTimeout=20
+LoginDialogDisplayTimeout=60
 
 
 
@@ -217,7 +254,15 @@ LoginDialogDisplayTimeout=20
 # but they are acceptable.
 #
 # The default is the current working directory when IBC is
-# started.
+# started, unless the TWS_SETTINGS_PATH setting in the relevant
+# start script is set.
+#
+# If both this setting and TWS_SETTINGS_PATH are set, then this
+# setting takes priority. Note that if they have different values,
+# auto-restart will not work.
+#
+# NB: this setting is now DEPRECATED. You should use the
+# TWS_SETTINGS_PATH setting in the relevant start script.
 
 IbDir=/root/Jts
 
@@ -284,15 +329,32 @@ ExistingSessionDetectedAction=primary
 # Override TWS API Port Number
 # ----------------------------
 #
-# If OverrideTwsApiPort is set to an integer, IBC changes the
-# 'Socket port' in TWS's API configuration to that number shortly
-# after startup. Leaving the setting blank will make no change to
-# the current setting. This setting is only intended for use in
-# certain specialized situations where the port number needs to
+# If OverrideTwsApiPort is set to an integer, IBC changes the 
+# 'Socket port' in TWS's API configuration to that number shortly 
+# after startup (but note that for the FIX Gateway, this setting is
+# actually stored in jts.ini rather than the Gateway's settings
+# file). Leaving the setting blank will make no change to 
+# the current setting. This setting is only intended for use in 
+# certain specialized situations where the port number needs to 
+# be set dynamically at run-time, and for the FIX Gateway: most
+# non-FIX users will never need it, so don't use it unless you know
+# you need it.
+
+OverrideTwsApiPort=4000
+
+
+# Override TWS Master Client ID
+# -----------------------------
+#
+# If OverrideTwsMasterClientID is set to an integer, IBC changes the
+# 'Master Client ID' value in TWS's API configuration to that 
+# value shortly after startup. Leaving the setting blank will make
+# no change to the current setting. This setting is only intended 
+# for use in certain specialized situations where the value needs to
 # be set dynamically at run-time: most users will never need it,
 # so don't use it unless you know you need it.
 
-; OverrideTwsApiPort=4002
+OverrideTwsMasterClientID=
 
 
 # Read-only Login
@@ -302,11 +364,13 @@ ExistingSessionDetectedAction=primary
 # account security programme, the user will not be asked to perform
 # the second factor authentication action, and login to TWS will
 # occur automatically in read-only mode: in this mode, placing or
-# managing orders is not allowed. If set to 'no', and the user is
-# enrolled in IB's account security programme, the user must perform
-# the relevant second factor authentication action to complete the
-# login. 
-
+# managing orders is not allowed. 
+#
+# If set to 'no', and the user is enrolled in IB's account security
+# programme, the second factor authentication process is handled
+# according to the Second Factor Authentication Settings described
+# elsewhere in this file.
+#
 # If the user is not enrolled in IB's account security programme,
 # this setting is ignored. The default is 'no'.
 
@@ -326,7 +390,44 @@ ReadOnlyLogin=no
 # set the relevant checkbox (this only needs to be done once) and
 # not provide a value for this setting.
 
-ReadOnlyApi=no
+ReadOnlyApi=
+
+
+# API Precautions
+# ---------------
+# 
+# These settings relate to the corresponding 'Precautions' checkboxes in the
+# API section of the Global Configuration dialog.
+#
+# For all of these, the accepted values are:
+# - 'yes' sets the checkbox
+# - 'no' clears the checkbox
+# - if not set, the existing TWS/Gateway configuration is unchanged
+#
+# NB: thess settings are really only supplied for the benefit of new TWS
+# or Gateway instances that are being automatically installed and
+# started without user intervention, or where user settings are not preserved
+# between sessions (eg some Docker containers). Where a user is involved, they
+# should use the Global Configuration to set the relevant checkboxes and not
+# provide values for these settings.
+
+BypassOrderPrecautions=
+
+BypassBondWarning=
+
+BypassNegativeYieldToWorstConfirmation=
+
+BypassCalledBondWarning=
+
+BypassSameActionPairTradeWarning=
+
+BypassPriceBasedVolatilityRiskWarning=
+
+BypassUSStocksMarketDataInSharesWarning=
+
+BypassRedirectOrderWarning=
+
+BypassNoOverfillProtectionPrecaution=
 
 
 # Market data size for US stocks - lots or shares
@@ -381,54 +482,145 @@ AcceptBidAskLastSizeDisplayUpdateNotification=accept
 SendMarketDataInLotsForUSstocks=
 
 
+# Trusted API Client IPs
+# ----------------------
+#
+# NB: THIS SETTING IS ONLY RELEVANT FOR THE GATEWAY, AND ONLY WHEN FIX=yes.
+# In all other cases it is ignored.
+#
+# This is a list of IP addresses separated by commas. API clients with IP
+# addresses in this list are able to connect to the API without Gateway
+# generating the 'Incoming connection' popup.
+#
+# Note that 127.0.0.1 is always permitted to connect, so do not include it
+# in this setting.
+
+TrustedTwsApiClientIPs=
+
+
+# Reset Order ID Sequence
+# -----------------------
+#
+# The setting resets the order id sequence for orders submitted via the API, so
+# that the next invocation of the `NextValidId` API callback will return the
+# value 1. The reset occurs when TWS starts.
+#
+# Note that order ids are reset for all API clients, except those that have
+# outstanding (ie incomplete) orders: their order id sequence carries on as
+# before.
+#
+# Valid values are 'yes', 'true', 'false' and 'no'. The default is 'no'.
+
+ResetOrderIdsAtStart=
+
+
+# This setting specifies IBC's action when TWS displays the dialog asking for
+# confirmation of a request to reset the API order id sequence.
+#
+# Note that the Gateway never displays this dialog, so this setting is ignored
+# for a Gateway session.
+#
+# Valid values consist of two strings separated by a solidus '/'. The first
+# value specifies the action to take when the order id reset request resulted
+# from setting ResetOrderIdsAtStart=yes. The second specifies the action to
+# take when the order id reset request is a result of the user clicking the
+# 'Reset API order ID sequence' button in the API configuration. Each value
+# must be one of the following:
+#
+#    'confirm' 
+#        order ids will be reset
+#
+#    'reject' 
+#        order ids will not be reset
+#
+#    'ignore' 
+#        IBC will ignore the dialog. The user must take action.
+#
+#    The default setting is ignore/ignore
+
+# Examples:
+#
+#    'confirm/reject' - confirm order id reset only if ResetOrderIdsAtStart=yes
+#                       and reject any user-initiated requests
+#
+#    'ignore/confirm' - user must decide what to do if ResetOrderIdsAtStart=yes
+#                       and confirm user-initiated requests
+#
+#    'reject/ignore'  - reject order id reset if  ResetOrderIdsAtStart=yes but
+#                       allow user to handle user-initiated requests 
+
+ConfirmOrderIdReset=
+
+
 
 # =============================================================================
-# 4.   TWS Auto-Closedown
+# 4.   TWS Auto-Logoff and Auto-Restart
 # =============================================================================
 #
-# IMPORTANT NOTE: Starting with TWS 974, this setting no longer
-# works properly, because IB have changed the way TWS handles its
-# autologoff mechanism. 
+# TWS and Gateway insist on being restarted every day. Two alternative
+# automatic options are offered: 
 #
-# You should now configure the TWS autologoff time to something
-# convenient for you, and restart IBC each day.
+#    - Auto-Logoff: at a specified time, TWS shuts down tidily, without
+#      restarting.
 #
-# Alternatively, discontinue use of IBC and use the auto-relogin
-# mechanism within TWS 974 and later versions (note that the 
-# auto-relogin mechanism provided by IB is not available if you
-# use IBC).
+#    - Auto-Restart: at a specified time, TWS shuts down and then restarts
+#      without the user having to re-autheticate.
+#
+# The normal way to configure the time at which this happens is via the Lock
+# and Exit section of the Configuration dialog. Once this time has been
+# configured in this way, the setting persists until the user changes it again.
+#
+# However, there are situations where there is no user available to do this
+# configuration, or where there is no persistent storage (for example some
+# Docker images). In such cases, the auto-restart or auto-logoff time can be
+# set whenever IBC starts with the settings below.
+#
+# The value, if specified, must be a time in HH:MM AM/PM format, for example
+# 08:00 AM or 10:00 PM. Note that there must be a single space between the
+# two parts of this value; also that midnight is "12:00 AM" and midday is
+# "12:00 PM".
+#
+# If no value is specified for either setting, the currently configured
+# settings will apply. If a value is supplied for one setting, the other
+# setting is cleared. If values are supplied for both settings, only the
+# auto-restart time is set, and the auto-logoff time is cleared.
+#
+# Note that for a normal TWS/Gateway installation with persistent storage
+# (for example on a desktop computer) the value will be persisted as if the
+# user had set it via the configuration dialog.
+#
+# If you choose to auto-restart, you should take note of the considerations
+# described at the link below. Note that where this information mentions
+# 'manual authentication', restarting IBC will do the job (IBKR does not
+# recognise the existence of IBC in its docuemntation).
+#
+#  https://www.interactivebrokers.com/en/software/tws/twsguide.htm#usersguidebook/configuretws/auto_restart_info.htm
+#
+# If you use the "RESTART" command via the IBC command server, and IBC is
+# running any version of the Gateway (or a version of TWS earlier than 1018),
+# note that this will set the Auto-Restart time in Gateway/TWS's configuration
+# dialog to the time at which the restart actually happens (which may be up to
+# a minute after the RESTART command is issued). To prevent future auto-
+# restarts at this time, you must make sure you have set AutoLogoffTime or
+# AutoRestartTime to your desired value before running IBC. NB: this does not
+# apply to TWS from version 1018 onwards.
 
-# Set to yes or no (lower case).
-#
-#   yes   means allow TWS to shut down automatically at its
-# 	  specified shutdown time, which is set via the TWS
-#	  configuration menu.
-#
-#   no    means TWS never shuts down automatically.
-#
-# NB: IB recommends that you do not keep TWS running
-# continuously. If you set this setting to 'no', you may
-# experience incorrect TWS operation.
-#
-# NB: the default for this setting is 'no'. Since this will
-# only work properly with TWS versions earlier than 974, you
-# should explicitly set this to 'yes' for version 974 and later.
-
-IbAutoClosedown=yes
+AutoLogoffTime=
 
+AutoRestartTime=
 
 
 # =============================================================================
 # 5.   TWS Tidy Closedown Time
 # =============================================================================
 #
-# NB: starting with TWS 974 this is no longer a useful option
-# because both TWS and Gateway now have the same auto-logoff
-# mechanism, and IBC can no longer avoid this.
+# Specifies a time at which TWS will close down tidily, with no restart.
 #
-# Note that giving this setting a value does not change TWS's
-# auto-logoff in any way: any setting will be additional to the
-# TWS auto-logoff.
+# There is little reason to use this setting. It is similar to AutoLogoffTime,
+# but can include a day-of-the-week, whereas AutoLogoffTime and AutoRestartTime
+# apply every day. So for example you could use ClosedownAt in conjunction with
+# AutoRestartTime to shut down TWS on Friday evenings after the markets
+# close, without it running on Saturday as well.
 #
 # To tell IBC to tidily close TWS at a specified time every
 # day, set this value to <hh:mm>, for example:
@@ -487,7 +679,7 @@ AcceptIncomingConnectionAction=reject
 #   no    means the dialog remains on display and must be
 #         handled by the user.
 
-AllowBlindTrading=yes
+AllowBlindTrading=no
 
 
 # Save Settings on a Schedule
@@ -530,6 +722,26 @@ AllowBlindTrading=yes
 SaveTwsSettingsAt=
 
 
+# Confirm Crypto Currency Orders Automatically
+# --------------------------------------------
+#
+# When you place an order for a cryptocurrency contract, a dialog is displayed
+# asking you to confirm that you want to place the order, and notifying you
+# that you are placing an order to trade cryptocurrency with Paxos, a New York
+# limited trust company, and not at Interactive Brokers.
+#
+#   transmit    means that the order will be placed automatically, and the
+#               dialog will then be closed
+#
+#   cancel      means that the order will not be placed, and the dialog will
+#               then be closed
+#
+#   manual      means that IBC will take no action and the user must deal
+#               with the dialog
+
+ConfirmCryptoCurrencyOrders=transmit
+
+
 
 # =============================================================================
 # 7.   Settings Specific to Indian Versions of TWS
@@ -566,13 +778,17 @@ DismissNSEComplianceNotice=yes
 #
 # The port number that IBC listens on for commands
 # such as "STOP". DO NOT set this to the port number
-# used for TWS API connections. There is no good reason
-# to change this setting unless the port is used by
-# some other application (typically another instance of
-# IBC). The default value is 0, which tells IBC not to
-# start the command server
+# used for TWS API connections.
+#
+# The convention is to use 7462 for this port,
+# but it must be set to a different value from any other
+# IBC instance that might run at the same time.
+#
+# The default value is 0, which tells IBC not to start
+# the command server
 
 #CommandServerPort=7462
+CommandServerPort=0
 
 
 # Permitted Command Sources
@@ -583,19 +799,19 @@ DismissNSEComplianceNotice=yes
 # IBC.  Commands can always be sent from the
 # same host as IBC is running on.
 
-ControlFrom=127.0.0.1
+ControlFrom=
 
 
 # Address for Receiving Commands
 # ------------------------------
 #
 # Specifies the IP address on which the Command Server
-# is so listen. For a multi-homed host, this can be used
+# is to listen. For a multi-homed host, this can be used
 # to specify that connection requests are only to be
 # accepted on the specified address. The default is to
 # accept connection requests on all local addresses.
 
-BindAddress=127.0.0.1
+BindAddress=
 
 
 # Command Prompt
@@ -621,7 +837,7 @@ CommandPrompt=
 # information is sent. The default is that such information
 # is not sent.
 
-SuppressInfoMessages=no
+SuppressInfoMessages=yes
 
 
 
@@ -651,10 +867,10 @@ SuppressInfoMessages=no
 # The LogStructureScope setting indicates which windows are
 # eligible for structure logging:
 #
-#    - if set to 'known', only windows that IBC recognizes
-#      are eligible - these are windows that IBC has some
-#      interest in monitoring, usually to take some action
-#      on the user's behalf;
+#    - (default value) if set to 'known', only windows that
+#      IBC recognizes are eligible - these are windows that
+#      IBC has some interest in monitoring, usually to take
+#      some action on the user's behalf;
 #
 #    - if set to 'unknown', only windows that IBC does not
 #      recognize are eligible. Most windows displayed by
@@ -667,9 +883,8 @@ SuppressInfoMessages=no
 #    - if set to 'all', then every window displayed by TWS
 #      is eligible.
 #
-# The default value is 'known'.
 
-LogStructureScope=all
+LogStructureScope=known
 
 
 # When to Log Window Structure
@@ -682,13 +897,15 @@ LogStructureScope=all
 #       structure of an eligible window the first time it
 #       is encountered;
 #
+#     - if set to 'openclose', the structure is logged every
+#       time an eligible window is opened or closed;
+#
 #    - if set to 'activate', the structure is logged every
 #      time an eligible window is made active;
 #
-#    - if set to 'never' or 'no' or 'false', structure
-#      information is never logged.
+#    - (default value) if set to 'never' or 'no' or 'false',
+#      structure information is never logged.
 #
-# The default value is 'never'.
 
 LogStructureWhen=never
 
@@ -708,4 +925,3 @@ LogStructureWhen=never
 #LogComponents=
 
 
-

flake.lock

@@ -0,0 +1,138 @@
+{
+  "nodes": {
+    "flake-utils": {
+      "inputs": {
+        "systems": "systems"
+      },
+      "locked": {
+        "lastModified": 1689068808,
+        "narHash": "sha256-6ixXo3wt24N/melDWjq70UuHQLxGV8jZvooRanIHXw0=",
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "rev": "919d646de7be200f3bf08cb76ae1f09402b6f9b4",
+        "type": "github"
+      },
+      "original": {
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "type": "github"
+      }
+    },
+    "flake-utils_2": {
+      "inputs": {
+        "systems": "systems_2"
+      },
+      "locked": {
+        "lastModified": 1689068808,
+        "narHash": "sha256-6ixXo3wt24N/melDWjq70UuHQLxGV8jZvooRanIHXw0=",
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "rev": "919d646de7be200f3bf08cb76ae1f09402b6f9b4",
+        "type": "github"
+      },
+      "original": {
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "type": "github"
+      }
+    },
+    "nix-github-actions": {
+      "inputs": {
+        "nixpkgs": [
+          "poetry2nix",
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1688870561,
+        "narHash": "sha256-4UYkifnPEw1nAzqqPOTL2MvWtm3sNGw1UTYTalkTcGY=",
+        "owner": "nix-community",
+        "repo": "nix-github-actions",
+        "rev": "165b1650b753316aa7f1787f3005a8d2da0f5301",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-community",
+        "repo": "nix-github-actions",
+        "type": "github"
+      }
+    },
+    "nixpkgs": {
+      "locked": {
+        "lastModified": 1692174805,
+        "narHash": "sha256-xmNPFDi/AUMIxwgOH/IVom55Dks34u1g7sFKKebxUm0=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "caac0eb6bdcad0b32cb2522e03e4002c8975c62e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "NixOS",
+        "ref": "nixos-unstable",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
+    "poetry2nix": {
+      "inputs": {
+        "flake-utils": "flake-utils_2",
+        "nix-github-actions": "nix-github-actions",
+        "nixpkgs": [
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1692048894,
+        "narHash": "sha256-cDw03rso2V4CDc3Mll0cHN+ztzysAvdI8pJ7ybbz714=",
+        "ref": "refs/heads/pyqt6",
+        "rev": "b059ad4c3051f45d6c912e17747aae37a9ec1544",
+        "revCount": 2276,
+        "type": "git",
+        "url": "file:///home/lord_fomo/repos/poetry2nix"
+      },
+      "original": {
+        "type": "git",
+        "url": "file:///home/lord_fomo/repos/poetry2nix"
+      }
+    },
+    "root": {
+      "inputs": {
+        "flake-utils": "flake-utils",
+        "nixpkgs": "nixpkgs",
+        "poetry2nix": "poetry2nix"
+      }
+    },
+    "systems": {
+      "locked": {
+        "lastModified": 1681028828,
+        "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
+        "owner": "nix-systems",
+        "repo": "default",
+        "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-systems",
+        "repo": "default",
+        "type": "github"
+      }
+    },
+    "systems_2": {
+      "locked": {
+        "lastModified": 1681028828,
+        "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
+        "owner": "nix-systems",
+        "repo": "default",
+        "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-systems",
+        "repo": "default",
+        "type": "github"
+      }
+    }
+  },
+  "root": "root",
+  "version": 7
+}

flake.nix

@@ -0,0 +1,180 @@
+# NOTE: to convert to a poetry2nix env like this here are the
+# steps:
+# - install poetry in your system nix config
+# - convert the repo to use poetry using `poetry init`:
+#   https://python-poetry.org/docs/basic-usage/#initialising-a-pre-existing-project
+# - then manually ensuring all deps are converted over:
+# - add this file to the repo and commit it
+# - 
+
+# GROKin tips:
+# - CLI eps are (ostensibly) added via an `entry_points.txt`:
+#   - https://packaging.python.org/en/latest/specifications/entry-points/#file-format
+#   - https://github.com/nix-community/poetry2nix/blob/master/editable.nix#L49
+{
+  description = "piker: trading gear for hackers (pkged with poetry2nix)";
+
+  inputs.flake-utils.url = "github:numtide/flake-utils";
+  inputs.nixpkgs.url = "github:NixOS/nixpkgs/nixos-unstable";
+
+  # see https://github.com/nix-community/poetry2nix/tree/master#api
+  inputs.poetry2nix = {
+    # url = "github:nix-community/poetry2nix";
+    # url = "github:K900/poetry2nix/qt5-explicit-deps";
+    url = "/home/lord_fomo/repos/poetry2nix";
+
+    inputs.nixpkgs.follows = "nixpkgs";
+  };
+
+  outputs = {
+    self,
+    nixpkgs,
+    flake-utils,
+    poetry2nix,
+  }:
+    # TODO: build cross-OS and use the `${system}` var thingy..
+    flake-utils.lib.eachDefaultSystem (system:
+      let
+        # use PWD as sources
+        projectDir = ./.;
+        pyproject = ./pyproject.toml;
+        poetrylock = ./poetry.lock;
+
+        # TODO: port to 3.11 and support both versions?
+        python = "python3.10";
+
+        # for more functions and examples.
+        # inherit
+        # (poetry2nix.legacyPackages.${system})
+        # mkPoetryApplication;
+        # pkgs = nixpkgs.legacyPackages.${system};
+
+        pkgs = nixpkgs.legacyPackages.x86_64-linux;
+        lib = pkgs.lib;
+        p2npkgs = poetry2nix.legacyPackages.x86_64-linux;
+
+        # define all pkg overrides per dep, see edgecases.md:
+        # https://github.com/nix-community/poetry2nix/blob/master/docs/edgecases.md
+        # TODO: add these into the json file:
+        # https://github.com/nix-community/poetry2nix/blob/master/overrides/build-systems.json
+        pypkgs-build-requirements = {
+          asyncvnc = [ "setuptools" ];
+          eventkit = [ "setuptools" ];
+          ib-insync = [ "setuptools" "flake8" ];
+          msgspec = [ "setuptools"];
+          pdbp = [ "setuptools" ];
+          pyqt6-sip = [ "setuptools" ];
+          tabcompleter = [ "setuptools" ];
+          tractor = [ "setuptools" ];
+          tricycle = [ "setuptools" ];
+          trio-typing = [ "setuptools" ];
+          trio-util = [ "setuptools" ];
+          xonsh = [ "setuptools" ];
+        };
+
+        # auto-generate override entries
+        p2n-overrides = p2npkgs.defaultPoetryOverrides.extend (self: super:
+          builtins.mapAttrs (package: build-requirements:
+            (builtins.getAttr package super).overridePythonAttrs (old: {
+              buildInputs = (
+                old.buildInputs or [ ]
+              ) ++ (
+                builtins.map (
+                  pkg: if builtins.isString pkg then builtins.getAttr pkg super else pkg
+                  ) build-requirements
+              );
+            })
+          ) pypkgs-build-requirements
+        );
+
+        # override some ahead-of-time compiled extensions
+        # to be built with their wheels.
+        ahot_overrides = p2n-overrides.extend(
+          final: prev: {
+
+            # llvmlite = prev.llvmlite.override {
+            #   preferWheel = false;
+            # };
+
+            # TODO: get this workin with p2n and nixpkgs..
+            # pyqt6 = prev.pyqt6.override {
+            #   preferWheel = true;
+            # };
+
+            # NOTE: this DOESN'T work atm but after a fix
+            # to poetry2nix, it will and actually this line
+            # won't be needed - thanks @k900:
+            # https://github.com/nix-community/poetry2nix/pull/1257
+            pyqt5 = prev.pyqt5.override {
+              # withWebkit = false;
+              preferWheel = true;
+            };
+
+            # see PR from @k900:
+            # https://github.com/nix-community/poetry2nix/pull/1257
+            # pyqt5-qt5 = prev.pyqt5-qt5.override {
+            #   withWebkit = false;
+            #   preferWheel = true;
+            # };
+
+            # TODO: patch in an override for polars to build
+            # from src! See the details likely needed from
+            # the cryptography entry:
+            # https://github.com/nix-community/poetry2nix/blob/master/overrides/default.nix#L426-L435
+            polars = prev.polars.override {
+              preferWheel = true;
+            };
+          }
+      );
+
+      # WHY!? -> output-attrs that `nix develop` scans for:
+      # https://nixos.org/manual/nix/stable/command-ref/new-cli/nix3-develop.html#flake-output-attributes
+      in
+      rec {
+        packages = {
+          # piker = poetry2nix.legacyPackages.x86_64-linux.mkPoetryEditablePackage {
+          #   editablePackageSources = { piker = ./piker; };
+
+          piker = p2npkgs.mkPoetryApplication {
+            projectDir = projectDir;
+
+            # SEE ABOVE for auto-genned input set, override
+            # buncha deps with extras.. like `setuptools` mostly.
+            # TODO: maybe propose a patch to p2n to show that you
+            # can even do this in the edgecases docs?
+            overrides = ahot_overrides;
+
+            # XXX: won't work on llvmlite..
+            # preferWheels = true;
+          };
+        };
+
+        # devShells.default = pkgs.mkShell {
+        #   projectDir = projectDir;
+        #   python = "python3.10";
+        #   overrides = ahot_overrides;
+        #   inputsFrom = [ self.packages.x86_64-linux.piker ];
+        #   packages = packages;
+        #   # packages = [ poetry2nix.packages.${system}.poetry ];
+        # };
+
+        # TODO: grok the difference here..
+        # - avoid re-cloning git repos on every develop entry..
+        # - ideally allow hacking on the src code of some deps
+        #   (tractor, pyqtgraph, tomlkit, etc.) WITHOUT having to
+        #   re-install them every time a change is made.
+        # - boot a usable xonsh inside the poetry virtualenv when
+        #   defined via a custom entry point?
+        devShells.default = p2npkgs.mkPoetryEnv {
+        # env = p2npkgs.mkPoetryEnv {
+            projectDir = projectDir;
+            python = pkgs.python310;
+            overrides = ahot_overrides;
+            editablePackageSources = packages;
+              # piker = "./";
+              # tractor = "../tractor/";
+            # };  # wut?
+        };
+      }
+    );  # end of .outputs scope
+}

piker/accounting/README.rst

@@ -0,0 +1,16 @@
+.accounting
+-----------
+A subsystem for transaction processing, storage and historical
+measurement.
+
+
+.pnl
+----
+BEP, the break even price: the price at which liquidating
+a remaining position results in a zero PnL since the position was
+"opened" in the destination asset.
+
+PPU: price-per-unit: the "average cost" (in cumulative mean terms)
+of the "entry" transactions which "make a position larger"; taking
+a profit relative to this price means that you will "make more
+profit then made prior" since the position was opened.

piker/accounting/__init__.py

@@ -21,17 +21,21 @@ for tendiez.
 '''
 from ..log import get_logger
 
-from ._ledger import (
+from .calc import (
     iter_by_dt,
+)
+from ._ledger import (
     Transaction,
     TransactionLedger,
     open_trade_ledger,
 )
 from ._pos import (
-    load_pps_from_ledger,
+    Account,
+    load_account,
+    load_account_from_ledger,
     open_pps,
+    open_account,
     Position,
-    PpTable,
 )
 from ._mktinfo import (
     Asset,
@@ -47,22 +51,25 @@ from ._allocate import (
     Allocator,
 )
 
+
 log = get_logger(__name__)
 
 __all__ = [
+    'Account',
     'Allocator',
     'Asset',
     'MktPair',
     'Position',
-    'PpTable',
     'Symbol',
     'Transaction',
     'TransactionLedger',
     'dec_digits',
     'digits_to_dec',
     'iter_by_dt',
-    'load_pps_from_ledger',
+    'load_account',
+    'load_account_from_ledger',
     'mk_allocator',
+    'open_account',
     'open_pps',
     'open_trade_ledger',
     'unpack_fqme',
@@ -82,7 +89,7 @@ def get_likely_pair(
 
     '''
     try:
-        src_name_start = bs_mktid.rindex(src)
+        src_name_start: str = bs_mktid.rindex(src)
     except (
         ValueError,   # substr not found
     ):
@@ -93,25 +100,8 @@ def get_likely_pair(
         # log.warning(
         #     f'No src fiat {src} found in {bs_mktid}?'
         # )
-        return
+        return None
 
-    likely_dst = bs_mktid[:src_name_start]
+    likely_dst: str = bs_mktid[:src_name_start]
     if likely_dst == dst:
         return bs_mktid
-
-
-if __name__ == '__main__':
-    import sys
-    from pprint import pformat
-
-    args = sys.argv
-    assert len(args) > 1, 'Specifiy account(s) from `brokers.toml`'
-    args = args[1:]
-    for acctid in args:
-        broker, name = acctid.split('.')
-        trans, updated_pps = load_pps_from_ledger(broker, name)
-        print(
-            f'Processing transactions into pps for {broker}:{acctid}\n'
-            f'{pformat(trans)}\n\n'
-            f'{pformat(updated_pps)}'
-        )

piker/accounting/_allocate.py

@@ -25,7 +25,7 @@ from bidict import bidict
 
 from ._pos import Position
 from . import MktPair
-from ..data.types import Struct
+from piker.types import Struct
 
 
 _size_units = bidict({
@@ -118,9 +118,9 @@ class Allocator(Struct):
         ld: int = mkt.size_tick_digits
 
         size_unit = self.size_unit
-        live_size = live_pp.size
+        live_size = live_pp.cumsize
         abs_live_size = abs(live_size)
-        abs_startup_size = abs(startup_pp.size)
+        abs_startup_size = abs(startup_pp.cumsize)
 
         u_per_slot, currency_per_slot = self.step_sizes()
 
@@ -213,8 +213,6 @@ class Allocator(Struct):
             slots_used = self.slots_used(
                 Position(
                     mkt=mkt,
-                    size=order_size,
-                    ppu=price,
                     bs_mktid=mkt.bs_mktid,
                 )
             )
@@ -241,7 +239,7 @@ class Allocator(Struct):
         Calc and return the number of slots used by this ``Position``.
 
         '''
-        abs_pp_size = abs(pp.size)
+        abs_pp_size = abs(pp.cumsize)
 
         if self.size_unit == 'currency':
             # live_currency_size = size or (abs_pp_size * pp.ppu)

piker/accounting/_ledger.py

@@ -21,65 +21,77 @@ Trade and transaction ledger processing.
 from __future__ import annotations
 from collections import UserDict
 from contextlib import contextmanager as cm
+from functools import partial
 from pathlib import Path
+from pprint import pformat
+from types import ModuleType
 from typing import (
     Any,
     Callable,
-    Iterator,
-    Union,
-    Generator
+    Generator,
+    Literal,
+    TYPE_CHECKING,
 )
 
 from pendulum import (
-    datetime,
     DateTime,
-    from_timestamp,
-    parse,
 )
 import tomli_w  # for fast ledger writing
 
-from .. import config
-from ..data.types import Struct
+from piker.types import Struct
+from piker import config
 from ..log import get_logger
-from ._mktinfo import (
-    Symbol,  # legacy
-    MktPair,
-    Asset,
+from .calc import (
+    iter_by_dt,
 )
 
+if TYPE_CHECKING:
+    from ..data._symcache import (
+        SymbologyCache,
+    )
+
 log = get_logger(__name__)
 
 
+TxnType = Literal[
+    'clear',
+    'transfer',
+
+    # TODO: see https://github.com/pikers/piker/issues/510
+    # 'split',
+    # 'rename',
+    # 'resize',
+    # 'removal',
+]
+
+
 class Transaction(Struct, frozen=True):
 
-    # TODO: unify this with the `MktPair`,
-    # once we have that as a required field,
-    # we don't really need the fqme any more..
+    # NOTE: this is a unified acronym also used in our `MktPair`
+    # and can stand for any of a
+    # "fully qualified <blank> endpoint":
+    # - "market" in the case of financial trades
+    #   (btcusdt.spot.binance).
+    # - "merkel (tree)" aka a blockchain system "wallet tranfers"
+    #   (btc.blockchain)
+    # - "money" for tradtitional (digital databases)
+    #   *bank accounts* (usd.swift, eur.sepa)
     fqme: str
 
-    tid: Union[str, int]  # unique transaction id
+    tid: str | int  # unique transaction id
     size: float
     price: float
     cost: float  # commisions or other additional costs
-    dt: datetime
+    dt: DateTime
+
+    # the "event type" in terms of "market events" see above and
+    # https://github.com/pikers/piker/issues/510
+    etype: TxnType = 'clear'
 
     # TODO: we can drop this right since we
     # can instead expect the backend to provide this
     # via the `MktPair`?
-    expiry: datetime | None = None
-
-    # TODO: drop the Symbol type, construct using
-    # t.sys (the transaction system)
-
-    # the underlying "transaction system", normally one of a ``MktPair``
-    # (a description of a tradable double auction) or a ledger-recorded
-    # ("ledger" in any sense as long as you can record transfers) of any
-    # sort) ``Asset``.
-    sym: MktPair | Asset | Symbol | None = None
-
-    @property
-    def sys(self) -> Symbol:
-        return self.sym
+    expiry: DateTime | None = None
 
     # (optional) key-id defined by the broker-service backend which
     # ensures the instrument-symbol market key for this record is unique
@@ -88,15 +100,16 @@ class Transaction(Struct, frozen=True):
     # service.
     bs_mktid: str | int | None = None
 
-    def to_dict(self) -> dict:
-        dct = super().to_dict()
-
-        # TODO: switch to sys!
-        dct.pop('sym')
+    def to_dict(
+        self,
+        **kwargs,
+    ) -> dict:
+        dct: dict[str, Any] = super().to_dict(**kwargs)
 
         # ensure we use a pendulum formatted
         # ISO style str here!@
         dct['dt'] = str(self.dt)
+
         return dct
 
 
@@ -108,17 +121,45 @@ class TransactionLedger(UserDict):
     outside.
 
     '''
+    # NOTE: see `open_trade_ledger()` for defaults, this should
+    # never be constructed manually!
     def __init__(
         self,
         ledger_dict: dict,
         file_path: Path,
+        account: str,
+        mod: ModuleType,  # broker mod
         tx_sort: Callable,
+        symcache: SymbologyCache,
 
     ) -> None:
-        self.file_path = file_path
-        self.tx_sort = tx_sort
+        self.account: str = account
+        self.file_path: Path = file_path
+        self.mod: ModuleType = mod
+        self.tx_sort: Callable = tx_sort
+
+        self._symcache: SymbologyCache = symcache
+
+        # any added txns we keep in that form for meta-data
+        # gathering purposes
+        self._txns: dict[str, Transaction] = {}
+
         super().__init__(ledger_dict)
 
+    def __repr__(self) -> str:
+        return (
+            f'TransactionLedger: {len(self)}\n'
+            f'{pformat(list(self.data))}'
+        )
+
+    @property
+    def symcache(self) -> SymbologyCache:
+        '''
+        Read-only ref to backend's ``SymbologyCache``.
+
+        '''
+        return self._symcache
+
     def update_from_t(
         self,
         t: Transaction,
@@ -129,14 +170,14 @@ class TransactionLedger(UserDict):
 
         '''
         self.data[t.tid] = t.to_dict()
+        self._txns[t.tid] = t
 
-    def iter_trans(
+    def iter_txns(
         self,
-        mkt_by_fqme: dict[str, MktPair],
-        broker: str = 'paper',
+        symcache: SymbologyCache | None = None,
 
     ) -> Generator[
-        tuple[str, Transaction],
+        Transaction,
         None,
         None,
     ]:
@@ -145,129 +186,125 @@ class TransactionLedger(UserDict):
         form via generator.
 
         '''
-        if broker != 'paper':
-            raise NotImplementedError('Per broker support not dun yet!')
+        symcache = symcache or self._symcache
 
-            # TODO: lookup some standard normalizer
-            # func in the backend?
-            # from ..brokers import get_brokermod
-            # mod = get_brokermod(broker)
-            # trans_dict = mod.norm_trade_records(self.data)
-
-            # NOTE: instead i propose the normalizer is
-            # a one shot routine (that can be lru cached)
-            # and instead call it for each entry incrementally:
-            # normer = mod.norm_trade_record(txdict)
-
-        # TODO: use tx_sort here yah?
-        for txdict in self.tx_sort(self.data.values()):
-        # for tid, txdict in self.data.items():
-            # special field handling for datetimes
-            # to ensure pendulum is used!
-            tid: str = txdict['tid']
-            fqme: str = txdict.get('fqme') or txdict['fqsn']
-            dt: DateTime = parse(txdict['dt'])
-            expiry: str | None = txdict.get('expiry')
-
-            if not (mkt := mkt_by_fqme.get(fqme)):
-                # we can't build a trans if we don't have
-                # the ``.sys: MktPair`` info, so skip.
-                continue
-
-            tx = Transaction(
-                fqme=fqme,
-                tid=txdict['tid'],
-                dt=dt,
-                price=txdict['price'],
-                size=txdict['size'],
-                cost=txdict.get('cost', 0),
-                bs_mktid=txdict['bs_mktid'],
-
-                # TODO: change to .sys!
-                sym=mkt,
-                expiry=parse(expiry) if expiry else None,
+        if self.account == 'paper':
+            from piker.clearing import _paper_engine
+            norm_trade: Callable = partial(
+                _paper_engine.norm_trade,
+                brokermod=self.mod,
             )
-            yield tid, tx
 
-    def to_trans(
+        else:
+            norm_trade: Callable = self.mod.norm_trade
+
+        # datetime-sort and pack into txs
+        for tid, txdict in self.tx_sort(self.data.items()):
+            txn: Transaction = norm_trade(
+                tid,
+                txdict,
+                pairs=symcache.pairs,
+                symcache=symcache,
+            )
+            yield txn
+
+    def to_txns(
         self,
-        **kwargs,
+        symcache: SymbologyCache | None = None,
 
     ) -> dict[str, Transaction]:
         '''
-        Return entire output from ``.iter_trans()`` in a ``dict``.
+        Return entire output from ``.iter_txns()`` in a ``dict``.
 
         '''
-        return dict(self.iter_trans(**kwargs))
+        txns: dict[str, Transaction] = {}
+        for t in self.iter_txns(symcache=symcache):
 
-    def write_config(
-        self,
+            if not t:
+                log.warning(f'{self.mod.name}:{self.account} TXN is -> {t}')
+                continue
 
-    ) -> None:
+            txns[t.tid] = t
+
+        return txns
+
+    def write_config(self) -> None:
         '''
-        Render the self.data ledger dict to it's TOML file form.
+        Render the self.data ledger dict to its TOML file form.
+
+        ALWAYS order datetime sorted!
 
         '''
-        cpy = self.data.copy()
+        is_paper: bool = self.account == 'paper'
+
+        symcache: SymbologyCache = self._symcache
         towrite: dict[str, Any] = {}
-        for tid, trans in cpy.items():
-
-            # drop key for non-expiring assets
-            txdict = towrite[tid] = self.data[tid]
+        for tid, txdict in self.tx_sort(self.data.copy()):
+            # write blank-str expiry for non-expiring assets
             if (
                 'expiry' in txdict
                 and txdict['expiry'] is None
             ):
-                txdict.pop('expiry')
+                txdict['expiry'] = ''
 
-            # re-write old acro-key
-            fqme = txdict.get('fqsn')
-            if fqme:
+            # (maybe) re-write old acro-key
+            if (
+                is_paper
+                # if symcache is empty/not supported (yet), don't
+                # bother xD
+                and symcache.mktmaps
+            ):
+                fqme: str = txdict.pop('fqsn', None) or txdict['fqme']
+                bs_mktid: str | None = txdict.get('bs_mktid')
+
+                if (
+
+                    fqme not in symcache.mktmaps
+                    or (
+                        # also try to see if this is maybe a paper
+                        # engine ledger in which case the bs_mktid
+                        # should be the fqme as well!
+                        bs_mktid
+                        and fqme != bs_mktid
+                    )
+                ):
+                    # always take any (paper) bs_mktid if defined and
+                    # in the backend's cache key set.
+                    if bs_mktid in symcache.mktmaps:
+                        fqme: str = bs_mktid
+                    else:
+                        best_fqme: str = list(symcache.search(fqme))[0]
+                        log.warning(
+                            f'Could not find FQME: {fqme} in qualified set?\n'
+                            f'Qualifying and expanding {fqme} -> {best_fqme}'
+                        )
+                        fqme = best_fqme
+
+                if (
+                    bs_mktid
+                    and bs_mktid != fqme
+                ):
+                    # in paper account case always make sure both the
+                    # fqme and bs_mktid are fully qualified..
+                    txdict['bs_mktid'] = fqme
+
+                # in paper ledgers always write the latest
+                # symbology key field: an FQME.
                 txdict['fqme'] = fqme
 
+            towrite[tid] = txdict
+
         with self.file_path.open(mode='wb') as fp:
             tomli_w.dump(towrite, fp)
 
 
-def iter_by_dt(
-    records: dict[str, dict[str, Any]] | list[dict],
-
-    # NOTE: parsers are looked up in the insert order
-    # so if you know that the record stats show some field
-    # is more common then others, stick it at the top B)
-    parsers: dict[tuple[str], Callable] = {
-        'dt': None,  # parity case
-        'datetime': parse,  # datetime-str
-        'time': from_timestamp,  # float epoch
-    },
-    key: Callable | None = None,
-
-) -> Iterator[tuple[str, dict]]:
-    '''
-    Iterate entries of a ``records: dict`` table sorted by entry recorded
-    datetime presumably set at the ``'dt'`` field in each entry.
-
-    '''
-    def dyn_parse_to_dt(txdict: dict[str, Any]) -> DateTime:
-        k, v, parser = next(
-            (k, txdict[k], parsers[k]) for k in parsers if k in txdict
-        )
-        return parser(v) if parser else v
-
-    if isinstance(records, dict):
-        records = records.values()
-
-    for entry in sorted(
-        records,
-        key=key or dyn_parse_to_dt,
-    ):
-        yield entry
-
-
 def load_ledger(
     brokername: str,
     acctid: str,
 
+    # for testing or manual load from file
+    dirpath: Path | None = None,
+
 ) -> tuple[dict, Path]:
     '''
     Load a ledger (TOML) file from user's config directory:
@@ -282,7 +319,11 @@ def load_ledger(
     except ModuleNotFoundError:
         import tomli as tomllib
 
-    ldir: Path = config._config_dir / 'accounting' / 'ledgers'
+    ldir: Path = (
+        dirpath
+        or
+        config._config_dir / 'accounting' / 'ledgers'
+    )
     if not ldir.is_dir():
         ldir.mkdir()
 
@@ -308,8 +349,15 @@ def open_trade_ledger(
     broker: str,
     account: str,
 
+    allow_from_sync_code: bool = False,
+    symcache: SymbologyCache | None = None,
+
     # default is to sort by detected datetime-ish field
     tx_sort: Callable = iter_by_dt,
+    rewrite: bool = False,
+
+    # for testing or manual load from file
+    _fp: Path | None = None,
 
 ) -> Generator[TransactionLedger, None, None]:
     '''
@@ -321,18 +369,52 @@ def open_trade_ledger(
     name as defined in the user's ``brokers.toml`` config.
 
     '''
-    ledger_dict, fpath = load_ledger(broker, account)
+    from ..brokers import get_brokermod
+    mod: ModuleType = get_brokermod(broker)
+
+    ledger_dict, fpath = load_ledger(
+        broker,
+        account,
+        dirpath=_fp,
+    )
     cpy = ledger_dict.copy()
+
+    # XXX NOTE: if not provided presume we are being called from
+    # sync code and need to maybe run `trio` to generate..
+    if symcache is None:
+
+        # XXX: be mega pendantic and ensure the caller knows what
+        # they're doing!
+        if not allow_from_sync_code:
+            raise RuntimeError(
+                'You MUST set `allow_from_sync_code=True` when '
+                'calling `open_trade_ledger()` from sync code! '
+                'If you are calling from async code you MUST '
+                'instead pass a `symcache: SymbologyCache`!'
+            )
+
+        from ..data._symcache import (
+            get_symcache,
+        )
+        symcache: SymbologyCache = get_symcache(broker)
+
+    assert symcache
+
     ledger = TransactionLedger(
         ledger_dict=cpy,
         file_path=fpath,
-        tx_sort=tx_sort,
+        account=account,
+        mod=mod,
+        symcache=symcache,
+        tx_sort=getattr(mod, 'tx_sort', tx_sort),
     )
     try:
         yield ledger
     finally:
-        if ledger.data != ledger_dict:
-
+        if (
+            ledger.data != ledger_dict
+            or rewrite
+        ):
             # TODO: show diff output?
             # https://stackoverflow.com/questions/12956957/print-diff-of-python-dictionaries
             log.info(f'Updating ledger for {fpath}:\n')

piker/accounting/_mktinfo.py

@@ -36,7 +36,7 @@ from typing import (
     Literal,
 )
 
-from ..data.types import Struct
+from piker.types import Struct
 
 
 # TODO: make these literals..
@@ -130,8 +130,29 @@ class Asset(Struct, frozen=True):
     # should not be explicitly required in our generic API.
     info: dict | None = None
 
-    # TODO?
-    # _to_dict_skip = {'info'}
+    # `None` is not toml-compat so drop info
+    # if no extra data added..
+    def to_dict(
+        self,
+        **kwargs,
+    ) -> dict:
+        dct = super().to_dict(**kwargs)
+        if (info := dct.pop('info', None)):
+            dct['info'] = info
+
+        assert dct['tx_tick']
+        return dct
+
+    @classmethod
+    def from_msg(
+        cls,
+        msg: dict[str, Any],
+    ) -> Asset:
+        return cls(
+            tx_tick=Decimal(str(msg.pop('tx_tick'))),
+            info=msg.pop('info', None),
+            **msg,
+        )
 
     def __str__(self) -> str:
         return self.name
@@ -288,12 +309,40 @@ class MktPair(Struct, frozen=True):
     # strike price, call or put, swap type, exercise model, etc.
     contract_info: list[str] | None = None
 
+    # TODO: rename to sectype since all of these can
+    # be considered "securities"?
     _atype: str = ''
 
+    # allow explicit disable of the src part of the market
+    # pair name -> useful for legacy markets like qqq.nasdaq.ib
+    _fqme_without_src: bool = False
+
     # NOTE: when cast to `str` return fqme
     def __str__(self) -> str:
         return self.fqme
 
+    def to_dict(
+        self,
+        **kwargs,
+    ) -> dict:
+        d = super().to_dict(**kwargs)
+        d['src'] = self.src.to_dict(**kwargs)
+
+        if not isinstance(self.dst, str):
+            d['dst'] = self.dst.to_dict(**kwargs)
+        else:
+            d['dst'] = str(self.dst)
+
+        d['price_tick'] = str(self.price_tick)
+        d['size_tick'] = str(self.size_tick)
+
+        if self.contract_info is None:
+            d.pop('contract_info')
+
+        # d.pop('_fqme_without_src')
+
+        return d
+
     @classmethod
     def from_msg(
         cls,
@@ -304,36 +353,32 @@ class MktPair(Struct, frozen=True):
         Constructor for a received msg-dict normally received over IPC.
 
         '''
-        dst_asset_msg = msg.pop('dst')
-        src_asset_msg = msg.pop('src')
-
-        if isinstance(dst_asset_msg, str):
-            src: str = str(src_asset_msg)
-            assert isinstance(src, str)
-            return cls.from_fqme(
-                dst_asset_msg,
-                src=src,
-                **msg,
-            )
-
+        if not isinstance(
+            dst_asset_msg := msg.pop('dst'),
+            str,
+        ):
+            dst: Asset = Asset.from_msg(dst_asset_msg)  # .copy()
         else:
-            # NOTE: we call `.copy()` here to ensure
-            # type casting!
-            dst = Asset(**dst_asset_msg).copy()
-            if not isinstance(src_asset_msg, str):
-                src = Asset(**src_asset_msg).copy()
-            else:
-                src = str(src_asset_msg)
+            dst: str = dst_asset_msg
 
+        src_asset_msg: dict = msg.pop('src')
+        src: Asset = Asset.from_msg(src_asset_msg)  # .copy()
+
+        # XXX NOTE: ``msgspec`` can encode `Decimal` but it doesn't
+        # decide to it by default since we aren't spec-cing these
+        # msgs as structs proper to get them to decode implictily
+        # (yet) as per,
+        # - https://github.com/pikers/piker/pull/354
+        # - https://github.com/goodboy/tractor/pull/311
+        # SO we have to ensure we do a struct type
+        # case (which `.copy()` does) to ensure we get the right
+        # type!
         return cls(
             dst=dst,
             src=src,
+            price_tick=Decimal(msg.pop('price_tick')),
+            size_tick=Decimal(msg.pop('size_tick')),
             **msg,
-        # XXX NOTE: ``msgspec`` can encode `Decimal`
-        # but it doesn't decide to it by default since
-        # we aren't spec-cing these msgs as structs, SO
-        # we have to ensure we do a struct type case (which `.copy()`
-        # does) to ensure we get the right type!
         ).copy()
 
     @property
@@ -361,7 +406,20 @@ class MktPair(Struct, frozen=True):
         ):
             _fqme = f'{fqme}.{broker}'
 
-        broker, mkt_ep_key, venue, suffix = unpack_fqme(_fqme)
+        broker, mkt_ep_key, venue, expiry = unpack_fqme(_fqme)
+
+        kven: str = kwargs.pop('venue', venue)
+        if venue:
+            assert venue == kven
+        else:
+            venue = kven
+
+        exp: str = kwargs.pop('expiry', expiry)
+        if expiry:
+            assert exp == expiry
+        else:
+            expiry = exp
+
         dst: Asset = Asset.guess_from_mkt_ep_key(
             mkt_ep_key,
             atype=kwargs.get('_atype'),
@@ -373,14 +431,15 @@ class MktPair(Struct, frozen=True):
         # which we expect to be filled in by some
         # backend client with access to that data-info.
         return cls(
-            # XXX: not resolved to ``Asset`` :(
             dst=dst,
+            # XXX: not resolved to ``Asset`` :(
+            #src=src,
 
             broker=broker,
             venue=venue,
             # XXX NOTE: we presume this token
             # if the expiry for now!
-            expiry=suffix,
+            expiry=expiry,
 
             price_tick=price_tick,
             size_tick=size_tick,
@@ -486,7 +545,7 @@ class MktPair(Struct, frozen=True):
         '''
         key: str = (
             self.pair(delim_char=delim_char)
-            if not without_src
+            if not (without_src or self._fqme_without_src)
             else str(self.dst)
         )
 
@@ -555,7 +614,7 @@ class MktPair(Struct, frozen=True):
         if isinstance(self.dst, Asset):
             return str(self.dst.atype)
 
-        return 'unknown'
+        return 'UNKNOWN'
 
     @property
     def price_tick_digits(self) -> int:

piker/accounting/calc.py

@@ -0,0 +1,698 @@
+# piker: trading gear for hackers
+# Copyright (C) Tyler Goodlet (in stewardship for pikers)
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+'''
+Calculation routines for balance and position tracking such that
+you know when you're losing money (if possible) XD
+
+'''
+from __future__ import annotations
+from collections.abc import ValuesView
+from contextlib import contextmanager as cm
+from math import copysign
+from typing import (
+    Any,
+    Callable,
+    Iterator,
+    TYPE_CHECKING,
+)
+
+import polars as pl
+from pendulum import (
+    DateTime,
+    from_timestamp,
+    parse,
+)
+
+if TYPE_CHECKING:
+    from ._ledger import (
+        Transaction,
+        TransactionLedger,
+    )
+
+
+def ppu(
+    clears: Iterator[Transaction],
+
+    # include transaction cost in breakeven price
+    # and presume the worst case of the same cost
+    # to exit this transaction (even though in reality
+    # it will be dynamic based on exit stratetgy).
+    cost_scalar: float = 2,
+
+    # return the ledger of clears as a (now dt sorted) dict with
+    # new position fields inserted alongside each entry.
+    as_ledger: bool = False,
+
+) -> float | list[(str, dict)]:
+    '''
+    Compute the "price-per-unit" price for the given non-zero sized
+    rolling position.
+
+    The recurrence relation which computes this (exponential) mean
+    per new clear which **increases** the accumulative postiion size
+    is:
+
+    ppu[-1] = (
+        ppu[-2] * accum_size[-2]
+        +
+        ppu[-1] * size
+    ) / accum_size[-1]
+
+    where `cost_basis` for the current step is simply the price
+    * size of the most recent clearing transaction.
+
+    -----
+    TODO: get the BEP computed and working similarly!
+    -----
+    the equivalent "break even price" or bep at each new clear
+    event step conversely only changes when an "position exiting
+    clear" which **decreases** the cumulative dst asset size:
+
+    bep[-1] = ppu[-1] - (cum_pnl[-1] / cumsize[-1])
+
+    '''
+    asize_h: list[float] = []  # historical accumulative size
+    ppu_h: list[float] = []  # historical price-per-unit
+    # ledger: dict[str, dict] = {}
+    ledger: list[dict] = []
+
+    t: Transaction
+    for t in clears:
+        clear_size: float = t.size
+        clear_price: str | float = t.price
+        is_clear: bool = not isinstance(clear_price, str)
+
+        last_accum_size = asize_h[-1] if asize_h else 0
+        accum_size: float = last_accum_size + clear_size
+        accum_sign = copysign(1, accum_size)
+        sign_change: bool = False
+
+        # on transfers we normally write some non-valid
+        # price since withdrawal to another account/wallet
+        # has nothing to do with inter-asset-market prices.
+        # TODO: this should be better handled via a `type: 'tx'`
+        # field as per existing issue surrounding all this:
+        # https://github.com/pikers/piker/issues/510
+        if isinstance(clear_price, str):
+            # TODO: we can't necessarily have this commit to
+            # the overall pos size since we also need to
+            # include other positions contributions to this
+            # balance or we might end up with a -ve balance for
+            # the position..
+            continue
+
+        # test if the pp somehow went "passed" a net zero size state
+        # resulting in a change of the "sign" of the size (+ve for
+        # long, -ve for short).
+        sign_change = (
+            copysign(1, last_accum_size) + accum_sign == 0
+            and last_accum_size != 0
+        )
+
+        # since we passed the net-zero-size state the new size
+        # after sum should be the remaining size the new
+        # "direction" (aka, long vs. short) for this clear.
+        if sign_change:
+            clear_size: float = accum_size
+            abs_diff: float = abs(accum_size)
+            asize_h.append(0)
+            ppu_h.append(0)
+
+        else:
+            # old size minus the new size gives us size diff with
+            # +ve -> increase in pp size
+            # -ve -> decrease in pp size
+            abs_diff = abs(accum_size) - abs(last_accum_size)
+
+        # XXX: LIFO breakeven price update. only an increaze in size
+        # of the position contributes the breakeven price,
+        # a decrease does not (i.e. the position is being made
+        # smaller).
+        # abs_clear_size = abs(clear_size)
+        abs_new_size: float | int = abs(accum_size)
+
+        if (
+            abs_diff > 0
+            and is_clear
+        ):
+            cost_basis = (
+                # cost basis for this clear
+                clear_price * abs(clear_size)
+                +
+                # transaction cost
+                accum_sign * cost_scalar * t.cost
+            )
+
+            if asize_h:
+                size_last: float = abs(asize_h[-1])
+                cb_last: float = ppu_h[-1] * size_last
+                ppu: float = (cost_basis + cb_last) / abs_new_size
+
+            else:
+                ppu: float = cost_basis / abs_new_size
+
+        else:
+            # TODO: for PPU we should probably handle txs out
+            # (aka withdrawals) similarly by simply not having
+            # them contrib to the running PPU calc and only
+            # when the next entry clear comes in (which will
+            # then have a higher weighting on the PPU).
+
+            # on "exit" clears from a given direction,
+            # only the size changes not the price-per-unit
+            # need to be updated since the ppu remains constant
+            # and gets weighted by the new size.
+            ppu: float = ppu_h[-1] if ppu_h else 0  # set to previous value
+
+        # extend with new rolling metric for this step
+        ppu_h.append(ppu)
+        asize_h.append(accum_size)
+
+        # ledger[t.tid] = {
+            # 'txn': t,
+        # ledger[t.tid] = t.to_dict() | {
+        ledger.append((
+            t.tid,
+            t.to_dict() | {
+                'ppu': ppu,
+                'cumsize': accum_size,
+                'sign_change': sign_change,
+
+                # TODO: cum_pnl, bep
+            }
+        ))
+
+    final_ppu = ppu_h[-1] if ppu_h else 0
+    # TODO: once we have etypes in all ledger entries..
+    # handle any split info entered (for now) manually by user
+    # if self.split_ratio is not None:
+    #     final_ppu /= self.split_ratio
+
+    if as_ledger:
+        return ledger
+
+    else:
+        return final_ppu
+
+
+def iter_by_dt(
+    records: (
+        dict[str, dict[str, Any]]
+        | ValuesView[dict]  # eg. `Position._events.values()`
+        | list[dict]
+        | list[Transaction]  # XXX preferred!
+    ),
+
+    # NOTE: parsers are looked up in the insert order
+    # so if you know that the record stats show some field
+    # is more common then others, stick it at the top B)
+    parsers: dict[str, Callable | None] = {
+        'dt': parse,  # parity case
+        'datetime': parse,  # datetime-str
+        'time': from_timestamp,  # float epoch
+    },
+    key: Callable | None = None,
+
+) -> Iterator[tuple[str, dict]]:
+    '''
+    Iterate entries of a transaction table sorted by entry recorded
+    datetime presumably set at the ``'dt'`` field in each entry.
+
+    '''
+    if isinstance(records, dict):
+        records: list[tuple[str, dict]] = list(records.items())
+
+    def dyn_parse_to_dt(
+        tx: tuple[str, dict[str, Any]] | Transaction,
+    ) -> DateTime:
+
+        # handle `.items()` inputs
+        if isinstance(tx, tuple):
+            tx = tx[1]
+
+        # dict or tx object?
+        isdict: bool = isinstance(tx, dict)
+
+        # get best parser for this record..
+        for k in parsers:
+            if (
+                isdict and k in tx
+                 or getattr(tx, k, None)
+            ):
+                v = tx[k] if isdict else tx.dt
+                assert v is not None, f'No valid value for `{k}`!?'
+
+                # only call parser on the value if not None from
+                # the `parsers` table above (when NOT using
+                # `.get()`), otherwise pass through the value and
+                # sort on it directly
+                if (
+                    not isinstance(v, DateTime)
+                    and (parser := parsers.get(k))
+                ):
+                    return parser(v)
+                else:
+                    return v
+
+        else:
+            # XXX: should never get here..
+            breakpoint()
+
+    entry: tuple[str, dict] | Transaction
+    for entry in sorted(
+        records,
+        key=key or dyn_parse_to_dt,
+    ):
+        # NOTE the type sig above; either pairs or txns B)
+        yield entry
+
+
+# TODO: probably just move this into the test suite or
+# keep it here for use from as such?
+# def ensure_state(self) -> None:
+#     '''
+#     Audit either the `.cumsize` and `.ppu` local instance vars against
+#     the clears table calculations and return the calc-ed values if
+#     they differ and log warnings to console.
+
+#     '''
+#     # clears: list[dict] = self._clears
+
+#     # self.first_clear_dt = min(clears, key=lambda e: e['dt'])['dt']
+#     last_clear: dict = clears[-1]
+#     csize: float = self.calc_size()
+#     accum: float = last_clear['accum_size']
+
+#     if not self.expired():
+#         if (
+#             csize != accum
+#             and csize != round(accum * (self.split_ratio or 1))
+#         ):
+#             raise ValueError(f'Size mismatch: {csize}')
+#     else:
+#         assert csize == 0, 'Contract is expired but non-zero size?'
+
+#     if self.cumsize != csize:
+#         log.warning(
+#             'Position state mismatch:\n'
+#             f'{self.cumsize} => {csize}'
+#         )
+#         self.cumsize = csize
+
+#     cppu: float = self.calc_ppu()
+#     ppu: float = last_clear['ppu']
+#     if (
+#         cppu != ppu
+#         and self.split_ratio is not None
+
+#         # handle any split info entered (for now) manually by user
+#         and cppu != (ppu / self.split_ratio)
+#     ):
+#         raise ValueError(f'PPU mismatch: {cppu}')
+
+#     if self.ppu != cppu:
+#         log.warning(
+#             'Position state mismatch:\n'
+#             f'{self.ppu} => {cppu}'
+#         )
+#         self.ppu = cppu
+
+
+@cm
+def open_ledger_dfs(
+
+    brokername: str,
+    acctname: str,
+
+    ledger: TransactionLedger | None = None,
+
+    **kwargs,
+
+) -> tuple[
+    dict[str, pl.DataFrame],
+    TransactionLedger,
+]:
+    '''
+    Open a ledger of trade records (presumably from some broker
+    backend), normalize the records into `Transactions` via the
+    backend's declared endpoint, cast to a `polars.DataFrame` which
+    can update the ledger on exit.
+
+    '''
+    from piker.toolz import open_crash_handler
+    with open_crash_handler():
+        if not ledger:
+            import time
+            from ._ledger import open_trade_ledger
+
+            now = time.time()
+
+            with open_trade_ledger(
+                    brokername,
+                    acctname,
+                    rewrite=True,
+                    allow_from_sync_code=True,
+
+                    # proxied through from caller
+                    **kwargs,
+
+            ) as ledger:
+                if not ledger:
+                    raise ValueError(f'No ledger for {acctname}@{brokername} exists?')
+
+                print(f'LEDGER LOAD TIME: {time.time() - now}')
+
+        yield ledger_to_dfs(ledger), ledger
+
+
+def ledger_to_dfs(
+    ledger: TransactionLedger,
+
+) -> dict[str, pl.DataFrame]:
+
+    txns: dict[str, Transaction] = ledger.to_txns()
+
+    # ldf = pl.DataFrame(
+    #     list(txn.to_dict() for txn in txns.values()),
+    ldf = pl.from_dicts(
+        list(txn.to_dict() for txn in txns.values()),
+
+        # only for ordering the cols
+        schema=[
+            ('fqme', str),
+            ('tid', str),
+            ('bs_mktid', str),
+            ('expiry', str),
+            ('etype', str),
+            ('dt', str),
+            ('size', pl.Float64),
+            ('price', pl.Float64),
+            ('cost', pl.Float64),
+        ],
+    ).sort(  # chronological order
+        'dt'
+    ).with_columns([
+        pl.col('dt').str.to_datetime(),
+        # pl.col('expiry').str.to_datetime(),
+        # pl.col('expiry').dt.date(),
+    ])
+
+    # filter out to the columns matching values filter passed
+    # as input.
+    # if filter_by_ids:
+    #     for col, vals in filter_by_ids.items():
+    #         str_vals = set(map(str, vals))
+    #         pred: pl.Expr = pl.col(col).eq(str_vals.pop())
+    #         for val in str_vals:
+    #             pred |= pl.col(col).eq(val)
+
+    #     fdf = df.filter(pred)
+
+    # TODO: originally i had tried just using a plain ol' groupby
+    # + agg here but the issue was re-inserting to the src frame.
+    # however, learning more about `polars` seems like maybe we can
+    # use `.over()`?
+    # https://pola-rs.github.io/polars/py-polars/html/reference/expressions/api/polars.Expr.over.html#polars.Expr.over
+    # => CURRENTLY we break up into a frame per mkt / fqme
+    dfs: dict[str, pl.DataFrame] = ldf.partition_by(
+        'bs_mktid',
+        as_dict=True,
+    )
+
+    # TODO: not sure if this is even possible but..
+    # - it'd be more ideal to use `ppt = df.groupby('fqme').agg([`
+    # - ppu and bep calcs!
+    for key in dfs:
+
+        # covert to lazy form (since apparently we might need it
+        # eventually ...)
+        df: pl.DataFrame = dfs[key]
+
+        ldf: pl.LazyFrame = df.lazy()
+
+        df = dfs[key] = ldf.with_columns([
+
+            pl.cumsum('size').alias('cumsize'),
+
+            # amount of source asset "sent" (via buy txns in
+            # the market) to acquire the dst asset, PER txn.
+            # when this value is -ve (i.e. a sell operation) then
+            # the amount sent is actually "returned".
+            (
+                (pl.col('price') * pl.col('size'))
+                +
+                (pl.col('cost')) # * pl.col('size').sign())
+            ).alias('dst_bot'),
+
+        ]).with_columns([
+
+            # rolling balance in src asset units
+            (pl.col('dst_bot').cumsum() * -1).alias('src_balance'),
+
+            # "position operation type" in terms of increasing the
+            # amount in the dst asset (entering) or decreasing the
+            # amount in the dst asset (exiting).
+            pl.when(
+                pl.col('size').sign() == pl.col('cumsize').sign()
+
+            ).then(
+                pl.lit('enter')  # see above, but is just price * size per txn
+
+            ).otherwise(
+                pl.when(pl.col('cumsize') == 0)
+                .then(pl.lit('exit_to_zero'))
+                .otherwise(pl.lit('exit'))
+            ).alias('descr'),
+
+            (pl.col('cumsize').sign() == pl.col('size').sign())
+            .alias('is_enter'),
+
+        ]).with_columns([
+
+            # pl.lit(0, dtype=pl.Utf8).alias('virt_cost'),
+            pl.lit(0, dtype=pl.Float64).alias('applied_cost'),
+            pl.lit(0, dtype=pl.Float64).alias('pos_ppu'),
+            pl.lit(0, dtype=pl.Float64).alias('per_txn_pnl'),
+            pl.lit(0, dtype=pl.Float64).alias('cum_pos_pnl'),
+            pl.lit(0, dtype=pl.Float64).alias('pos_bep'),
+            pl.lit(0, dtype=pl.Float64).alias('cum_ledger_pnl'),
+            pl.lit(None, dtype=pl.Float64).alias('ledger_bep'),
+
+            # TODO: instead of the iterative loop below i guess we
+            # could try using embedded lists to track which txns
+            # are part of which ppu / bep calcs? Not sure this will
+            # look any better nor be any more performant though xD
+            # pl.lit([[0]], dtype=pl.List(pl.Float64)).alias('list'),
+
+        # choose fields to emit for accounting puposes
+        ]).select([
+            pl.exclude([
+                'tid',
+                # 'dt',
+                'expiry',
+                'bs_mktid',
+                'etype',
+                # 'is_enter',
+            ]),
+        ]).collect()
+
+        # compute recurrence relations for ppu and bep
+        last_ppu: float = 0
+        last_cumsize: float = 0
+        last_ledger_pnl: float = 0
+        last_pos_pnl: float = 0
+        virt_costs: list[float, float] = [0., 0.]
+
+        # imperatively compute the PPU (price per unit) and BEP
+        # (break even price) iteratively over the ledger, oriented
+        # around each position state: a state of split balances in
+        # > 1 asset.
+        for i, row in enumerate(df.iter_rows(named=True)):
+
+            cumsize: float = row['cumsize']
+            is_enter: bool = row['is_enter']
+            price: float = row['price']
+            size: float = row['size']
+
+            # the profit is ALWAYS decreased, aka made a "loss"
+            # by the constant fee charged by the txn provider!
+            # see below in final PnL calculation and row element
+            # set.
+            txn_cost: float = row['cost']
+            pnl: float = 0
+
+            # ALWAYS reset per-position cum PnL
+            if last_cumsize == 0:
+                last_pos_pnl: float = 0
+
+            # a "position size INCREASING" or ENTER transaction
+            # which "makes larger", in src asset unit terms, the
+            # trade's side-size of the destination asset:
+            # - "buying" (more) units of the dst asset
+            # - "selling" (more short) units of the dst asset
+            if is_enter:
+
+                # Naively include transaction cost in breakeven
+                # price and presume the worst case of the
+                # exact-same-cost-to-exit this transaction's worth
+                # of size even though in reality it will be dynamic
+                # based on exit strategy, price, liquidity, etc..
+                virt_cost: float = txn_cost
+
+                # cpu: float = cost / size
+                # cummean of the cost-per-unit used for modelling
+                # a projected future exit cost which we immediately
+                # include in the costs incorporated to BEP on enters
+                last_cum_costs_size, last_cpu = virt_costs
+                cum_costs_size: float = last_cum_costs_size + abs(size)
+                cumcpu = (
+                    (last_cpu * last_cum_costs_size)
+                    +
+                    txn_cost
+                ) / cum_costs_size
+                virt_costs = [cum_costs_size, cumcpu]
+
+                txn_cost = txn_cost + virt_cost
+                # df[i, 'virt_cost'] = f'{-virt_cost} FROM {cumcpu}@{cum_costs_size}'
+
+                # a cumulative mean of the price-per-unit acquired
+                # in the destination asset:
+                # https://en.wikipedia.org/wiki/Moving_average#Cumulative_average
+                # You could also think of this measure more
+                # generally as an exponential mean with `alpha
+                # = 1/N` where `N` is the current number of txns
+                # included in the "position" defining set:
+                # https://en.wikipedia.org/wiki/Exponential_smoothing
+                ppu: float = (
+                    (
+                        (last_ppu * last_cumsize)
+                        +
+                        (price * size)
+                    ) /
+                    cumsize
+                )
+
+            # a "position size DECREASING" or EXIT transaction
+            # which "makes smaller" the trade's side-size of the
+            # destination asset:
+            # - selling previously bought units of the dst asset
+            #   (aka 'closing' a long position).
+            # - buying previously borrowed and sold (short) units
+            #   of the dst asset (aka 'covering'/'closing' a short
+            #   position).
+            else:
+                # only changes on position size increasing txns
+                ppu: float = last_ppu
+
+                # UNWIND IMPLIED COSTS FROM ENTRIES
+                # => Reverse the virtual/modelled (2x predicted) txn
+                # cost that was included in the least-recently
+                # entered txn that is still part of the current CSi
+                # set.
+                # => we look up the cost-per-unit cumsum and apply
+                # if over the current txn size (by multiplication)
+                # and then reverse that previusly applied cost on
+                # the txn_cost for this record.
+                #
+                # NOTE: current "model" is just to previously assumed 2x
+                # the txn cost for a matching enter-txn's
+                # cost-per-unit; we then immediately reverse this
+                # prediction and apply the real cost received here.
+                last_cum_costs_size, last_cpu = virt_costs
+                prev_virt_cost: float = last_cpu * abs(size)
+                txn_cost: float = txn_cost - prev_virt_cost  # +ve thus a "reversal"
+                cum_costs_size: float = last_cum_costs_size - abs(size)
+                virt_costs = [cum_costs_size, last_cpu]
+
+                # df[i, 'virt_cost'] = (
+                #     f'{-prev_virt_cost} FROM {last_cpu}@{cum_costs_size}'
+                # )
+
+                # the per-txn profit or loss (PnL) given we are
+                # (partially) "closing"/"exiting" the position via
+                # this txn.
+                pnl: float = (last_ppu - price) * size
+
+            # always subtract txn cost from total txn pnl
+            txn_pnl: float = pnl - txn_cost
+
+            # cumulative PnLs per txn
+            last_ledger_pnl = (
+                last_ledger_pnl + txn_pnl
+            )
+            last_pos_pnl = df[i, 'cum_pos_pnl'] = (
+                last_pos_pnl + txn_pnl
+            )
+
+            if cumsize == 0:
+                last_ppu = ppu = 0
+
+            # compute the BEP: "break even price", a value that
+            # determines at what price the remaining cumsize can be
+            # liquidated such that the net-PnL on the current
+            # position will result in ZERO gain or loss from open
+            # to close including all txn costs B)
+            if (
+                abs(cumsize) > 0  # non-exit-to-zero position txn
+            ):
+                cumsize_sign: float = copysign(1, cumsize)
+                ledger_bep: float = (
+                    (
+                        (ppu * cumsize)
+                        -
+                        (last_ledger_pnl * cumsize_sign)
+                    ) / cumsize
+                )
+
+                # NOTE: when we "enter more" dst asset units (aka
+                # increase position state) AFTER having exited some
+                # units (aka decreasing the pos size some) the bep
+                # needs to be RECOMPUTED based on new ppu such that
+                # liquidation of the cumsize at the bep price
+                # results in a zero-pnl for the existing position
+                # (since the last one).
+                # for position lifetime BEP we never can have
+                # a valid value once the position is "closed"
+                # / full exitted Bo
+                pos_bep: float = (
+                    (
+                        (ppu * cumsize)
+                        -
+                        (last_pos_pnl * cumsize_sign)
+                    ) / cumsize
+                )
+
+            # inject DF row with all values
+            df[i, 'pos_ppu'] = ppu
+            df[i, 'per_txn_pnl'] = txn_pnl
+            df[i, 'applied_cost'] = -txn_cost
+            df[i, 'cum_pos_pnl'] = last_pos_pnl
+            df[i, 'pos_bep'] = pos_bep
+            df[i, 'cum_ledger_pnl'] = last_ledger_pnl
+            df[i, 'ledger_bep'] = ledger_bep
+
+            # keep backrefs to suffice reccurence relation
+            last_ppu: float = ppu
+            last_cumsize: float = cumsize
+
+    # TODO?: pass back the current `Position` object loaded from
+    # the account as well? Would provide incentive to do all
+    # this ledger loading inside a new async open_account().
+    # bs_mktid: str = df[0]['bs_mktid']
+    # pos: Position = acnt.pps[bs_mktid]
+
+    return dfs

piker/accounting/cli.py

@@ -19,6 +19,7 @@ CLI front end for trades ledger and position tracking management.
 
 '''
 from __future__ import annotations
+from pprint import pformat
 
 
 from rich.console import Console
@@ -37,13 +38,11 @@ from ..calc import humanize
 from ..brokers._daemon import broker_init
 from ._ledger import (
     load_ledger,
+    TransactionLedger,
     # open_trade_ledger,
-    # TransactionLedger,
 )
-from ._pos import (
-    PpTable,
-    load_pps_from_ledger,
-    # load_account,
+from .calc import (
+    open_ledger_dfs,
 )
 
 
@@ -240,54 +239,73 @@ def sync(
 def disect(
     # "fully_qualified_account_name"
     fqan: str,
-    bs_mktid: str,  # for ib
-    pdb: bool = False,
+    fqme: str,  # for ib
 
+    # TODO: in tractor we should really have
+    # a debug_mode ctx for wrapping any kind of code no?
+    pdb: bool = False,
+    bs_mktid: str = typer.Option(
+        None,
+        "-bid",
+    ),
     loglevel: str = typer.Option(
         'error',
         "-l",
     ),
 ):
+    from piker.log import get_console_log
+    from piker.toolz import open_crash_handler
+    get_console_log(loglevel)
+
     pair: tuple[str, str]
     if not (pair := unpack_fqan(fqan)):
         raise ValueError('{fqan} malformed!?')
 
     brokername, account = pair
 
-    # ledger: TransactionLedger
-    records: dict[str, dict]
-    table: PpTable
-    records, table = load_pps_from_ledger(
-        brokername,
-        account,
-        filter_by_ids={bs_mktid},
-    )
-    df = pl.DataFrame(
-        list(records.values()),
-        # schema=[
-        #     ('tid', str),
-        #     ('fqme', str),
-        #     ('dt', str),
-        #     ('size', pl.Float64),
-        #     ('price', pl.Float64),
-        #     ('cost', pl.Float64),
-        #     ('expiry', str),
-        #     ('bs_mktid', str),
-        # ],
-    ).select([
-        pl.col('fqme'),
-        pl.col('dt').str.to_datetime(),
-        # pl.col('expiry').dt.datetime(),
-        pl.col('size'),
-        pl.col('price'),
-    ])
+    # ledger dfs groupby-partitioned by fqme
+    dfs: dict[str, pl.DataFrame]
+    # actual ledger instance
+    ldgr: TransactionLedger
 
-    assert not df.is_empty()
-    breakpoint()
-    # tractor.pause_from_sync()
-    # with open_trade_ledger(
-    #     brokername,
-    #     account,
-    # ) as ledger:
-    #     for tid, rec in ledger.items():
-    #         bs_mktid: str = rec['bs_mktid']
+    pl.Config.set_tbl_cols(-1)
+    pl.Config.set_tbl_rows(-1)
+    with (
+        open_crash_handler(),
+        open_ledger_dfs(
+            brokername,
+            account,
+        ) as (dfs, ldgr),
+    ):
+
+        # look up specific frame for fqme-selected asset
+        if (df := dfs.get(fqme)) is None:
+            mktids2fqmes: dict[str, list[str]] = {}
+            for bs_mktid in dfs:
+                df: pl.DataFrame = dfs[bs_mktid]
+                fqmes: pl.Series[str] = df['fqme']
+                uniques: list[str] = fqmes.unique()
+                mktids2fqmes[bs_mktid] = set(uniques)
+                if fqme in uniques:
+                    break
+            print(
+                f'No specific ledger for fqme={fqme} could be found in\n'
+                f'{pformat(mktids2fqmes)}?\n'
+                f'Maybe the `{brokername}` backend uses something '
+                'else for its `bs_mktid` then the `fqme`?\n'
+                'Scanning for matches in unique fqmes per frame..\n'
+            )
+
+        # :pray:
+        assert not df.is_empty()
+
+        # muck around in pdbp REPL
+        breakpoint()
+
+        # TODO: we REALLY need a better console REPL for this
+        # kinda thing..
+        # - `xonsh` is an obvious option (and it looks amazin) but
+        # we need to figure out how to embed it better then just:
+        # from xonsh.main import main
+        # main(argv=[])
+        # which will not actually inject the `df` to globals?

piker/brokers/__init__.py

@@ -50,7 +50,7 @@ __brokers__: list[str] = [
     'binance',
     'ib',
     'kraken',
-    'kucoin'
+    'kucoin',
 
     # broken but used to work
     # 'questrade',
@@ -71,7 +71,7 @@ def get_brokermod(brokername: str) -> ModuleType:
     Return the imported broker module by name.
 
     '''
-    module = import_module('.' + brokername, 'piker.brokers')
+    module: ModuleType = import_module('.' + brokername, 'piker.brokers')
     # we only allow monkeying because it's for internal keying
     module.name = module.__name__.split('.')[-1]
     return module
@@ -106,6 +106,6 @@ async def open_cached_client(
     ) as (cache_hit, client):
 
         if cache_hit:
-            log.info(f'Reusing existing {client}')
+            log.runtime(f'Reusing existing {client}')
 
         yield client

piker/brokers/_daemon.py

@@ -179,9 +179,6 @@ def broker_init(
         subpath: str = f'{modpath}.{submodname}'
         enabled.append(subpath)
 
-        # TODO XXX: DO WE NEED THIS?
-        # enabled.append('piker.data.feed')
-
     return (
         brokermod,
         start_actor_kwargs,  # to `ActorNursery.start_actor()`

piker/brokers/_util.py

@@ -18,10 +18,11 @@
 Handy cross-broker utils.
 
 """
+from __future__ import annotations
 from functools import partial
 
 import json
-import asks
+import httpx
 import logging
 
 from ..log import (
@@ -50,6 +51,7 @@ class SymbolNotFound(BrokerError):
     "Symbol not found by broker search"
 
 
+# TODO: these should probably be moved to `.tsp/.data`?
 class NoData(BrokerError):
     '''
     Symbol data not permitted or no data
@@ -59,14 +61,15 @@ class NoData(BrokerError):
     def __init__(
         self,
         *args,
-        frame_size: int = 1000,
+        info: dict|None = None,
 
     ) -> None:
         super().__init__(*args)
+        self.info: dict|None = info
 
         # when raised, machinery can check if the backend
         # set a "frame size" for doing datetime calcs.
-        self.frame_size: int = 1000
+        # self.frame_size: int = 1000
 
 
 class DataUnavailable(BrokerError):
@@ -88,16 +91,18 @@ class DataThrottle(BrokerError):
 
 
 def resproc(
-    resp: asks.response_objects.Response,
+    resp: httpx.Response,
     log: logging.Logger,
     return_json: bool = True,
     log_resp: bool = False,
 
-) -> asks.response_objects.Response:
-    """Process response and return its json content.
+) -> httpx.Response:
+    '''
+    Process response and return its json content.
 
     Raise the appropriate error on non-200 OK responses.
-    """
+
+    '''
     if not resp.status_code == 200:
         raise BrokerError(resp.body)
     try:

piker/brokers/binance/__init__.py

@@ -32,12 +32,19 @@ from .feed import (
 )
 from .broker import (
     open_trade_dialog,
+    get_cost,
+)
+from .venues import (
+    SpotPair,
+    FutesPair,
 )
-
 
 __all__ = [
     'get_client',
     'get_mkt_info',
+    'get_cost',
+    'SpotPair',
+    'FutesPair',
     'open_trade_dialog',
     'open_history_client',
     'open_symbol_search',

piker/brokers/binance/api.py

@@ -1,8 +1,8 @@
 # piker: trading gear for hackers
 # Copyright (C)
-#   Guillermo Rodriguez (aka ze jefe)
-#   Tyler Goodlet
-#   (in stewardship for pikers)
+#  Guillermo Rodriguez (aka ze jefe)
+#  Tyler Goodlet
+#  (in stewardship for pikers)
 
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU Affero General Public License as published by
@@ -25,6 +25,7 @@ from __future__ import annotations
 from collections import ChainMap
 from contextlib import (
     asynccontextmanager as acm,
+    AsyncExitStack,
 )
 from datetime import datetime
 from pprint import pformat
@@ -41,8 +42,7 @@ import trio
 from pendulum import (
     now,
 )
-import asks
-from fuzzywuzzy import process as fuzzy
+import httpx
 import numpy as np
 
 from piker import config
@@ -52,9 +52,13 @@ from piker.clearing._messages import (
 from piker.accounting import (
     Asset,
     digits_to_dec,
+    MktPair,
+)
+from piker.types import Struct
+from piker.data import (
+    def_iohlcv_fields,
+    match_from_pairs,
 )
-from piker.data.types import Struct
-from piker.data import def_iohlcv_fields
 from piker.brokers import (
     resproc,
     SymbolNotFound,
@@ -64,7 +68,6 @@ from .venues import (
     PAIRTYPES,
     Pair,
     MarketType,
-
     _spot_url,
     _futes_url,
     _testnet_futes_url,
@@ -74,16 +77,18 @@ from .venues import (
 log = get_logger('piker.brokers.binance')
 
 
-def get_config() -> dict:
-
+def get_config() -> dict[str, Any]:
     conf: dict
     path: Path
-    conf, path = config.load()
-
-    section = conf.get('binance')
-
+    conf, path = config.load(
+        conf_name='brokers',
+        touch_if_dne=True,
+    )
+    section: dict = conf.get('binance')
     if not section:
-        log.warning(f'No config section found for binance in {path}')
+        log.warning(
+            f'No config section found for binance in {path}'
+        )
         return {}
 
     return section
@@ -139,7 +144,7 @@ def binance_timestamp(
 
 class Client:
     '''
-    Async ReST API client using ``trio`` + ``asks`` B)
+    Async ReST API client using `trio` + `httpx` B)
 
     Supports all of the spot, margin and futures endpoints depending
     on method.
@@ -148,10 +153,17 @@ class Client:
     def __init__(
         self,
 
+        venue_sessions: dict[
+            str,  # venue key
+            tuple[httpx.AsyncClient, str]  # session, eps path
+        ],
+        conf: dict[str, Any],
         # TODO: change this to `Client.[mkt_]venue: MarketType`?
         mkt_mode: MarketType = 'spot',
 
     ) -> None:
+        self.conf = conf
+
         # build out pair info tables for each market type
         # and wrap in a chain-map view for search / query.
         self._spot_pairs: dict[str, Pair] = {}  # spot info table
@@ -178,44 +190,13 @@ class Client:
         # market symbols for use by search. See `.exch_info()`.
         self._pairs: ChainMap[str, Pair] = ChainMap()
 
-        # spot EPs sesh
-        self._sesh = asks.Session(connections=4)
-        self._sesh.base_location: str = _spot_url
-        # spot testnet
-        self._test_sesh: asks.Session = asks.Session(connections=4)
-        self._test_sesh.base_location: str = _testnet_spot_url
-
-        # margin and extended spot endpoints session.
-        self._sapi_sesh = asks.Session(connections=4)
-        self._sapi_sesh.base_location: str = _spot_url
-
-        # futes EPs sesh
-        self._fapi_sesh = asks.Session(connections=4)
-        self._fapi_sesh.base_location: str = _futes_url
-        # futes testnet
-        self._test_fapi_sesh: asks.Session = asks.Session(connections=4)
-        self._test_fapi_sesh.base_location: str = _testnet_futes_url
-
         # global client "venue selection" mode.
         # set this when you want to switch venues and not have to
         # specify the venue for the next request.
         self.mkt_mode: MarketType = mkt_mode
 
-        # per 8
-        self.venue_sesh: dict[
-            str,  # venue key
-            tuple[asks.Session, str]  # session, eps path
-        ] = {
-            'spot': (self._sesh, '/api/v3/'),
-            'spot_testnet': (self._test_sesh, '/fapi/v1/'),
-
-            'margin': (self._sapi_sesh, '/sapi/v1/'),
-
-            'usdtm_futes': (self._fapi_sesh, '/fapi/v1/'),
-            'usdtm_futes_testnet': (self._test_fapi_sesh, '/fapi/v1/'),
-
-            # 'futes_coin': self._dapi,  # TODO
-        }
+        # per-mkt-venue API client table
+        self.venue_sesh = venue_sessions
 
         # lookup for going from `.mkt_mode: str` to the config
         # subsection `key: str`
@@ -230,40 +211,6 @@ class Client:
             'futes': ['usdtm_futes'],
         }
 
-        # for creating API keys see,
-        # https://www.binance.com/en/support/faq/how-to-create-api-keys-on-binance-360002502072
-        self.conf: dict = get_config()
-
-        for key, subconf in self.conf.items():
-            if api_key := subconf.get('api_key', ''):
-                venue_keys: list[str] = self.confkey2venuekeys[key]
-
-                venue_key: str
-                sesh: asks.Session
-                for venue_key in venue_keys:
-                    sesh, _ = self.venue_sesh[venue_key]
-
-                    api_key_header: dict = {
-                        # taken from official:
-                        # https://github.com/binance/binance-futures-connector-python/blob/main/binance/api.py#L47
-                        "Content-Type": "application/json;charset=utf-8",
-
-                        # TODO: prolly should just always query and copy
-                        # in the real latest ver?
-                        "User-Agent": "binance-connector/6.1.6smbz6",
-                        "X-MBX-APIKEY": api_key,
-                    }
-                    sesh.headers.update(api_key_header)
-
-                    # if `.use_tesnet = true` in the config then
-                    # also add headers for the testnet session which
-                    # will be used for all order control
-                    if subconf.get('use_testnet', False):
-                        testnet_sesh, _ = self.venue_sesh[
-                            venue_key + '_testnet'
-                        ]
-                        testnet_sesh.headers.update(api_key_header)
-
     def _mk_sig(
         self,
         data: dict,
@@ -282,7 +229,6 @@ class Client:
                 'to define the creds for auth-ed endpoints!?'
             )
 
-
         # XXX: Info on security and authentification
         # https://binance-docs.github.io/apidocs/#endpoint-security-type
         if not (api_secret := subconf.get('api_secret')):
@@ -311,7 +257,7 @@ class Client:
         params: dict,
 
         method: str = 'get',
-        venue: str | None = None,  # if None use `.mkt_mode` state
+        venue: str|None = None,  # if None use `.mkt_mode` state
         signed: bool = False,
         allow_testnet: bool = False,
 
@@ -322,8 +268,9 @@ class Client:
         - /fapi/v3/ USD-M FUTURES, or
         - /api/v3/ SPOT/MARGIN
 
-        account/market endpoint request depending on either passed in `venue: str`
-        or the current setting `.mkt_mode: str` setting, default `'spot'`.
+        account/market endpoint request depending on either passed in
+        `venue: str` or the current setting `.mkt_mode: str` setting,
+        default `'spot'`.
 
 
         Docs per venue API:
@@ -352,9 +299,6 @@ class Client:
                 venue=venue_key,
             )
 
-        sesh: asks.Session
-        path: str
-
         # Check if we're configured to route order requests to the
         # venue equivalent's testnet.
         use_testnet: bool = False
@@ -379,11 +323,12 @@ class Client:
             # ctl machinery B)
             venue_key += '_testnet'
 
-        sesh, path = self.venue_sesh[venue_key]
-
-        meth: Callable = getattr(sesh, method)
+        client: httpx.AsyncClient
+        path: str
+        client, path = self.venue_sesh[venue_key]
+        meth: Callable = getattr(client, method)
         resp = await meth(
-            path=path + endpoint,
+            url=path + endpoint,
             params=params,
             timeout=float('inf'),
         )
@@ -396,7 +341,6 @@ class Client:
     ) -> None:
         # lookup internal mkt-specific pair table to update
         pair_table: dict[str, Pair] = self._venue2pairs[venue]
-        asset_table: dict[str, Asset] = self._venue2assets[venue]
 
         # make API request(s)
         resp = await self._api(
@@ -408,6 +352,7 @@ class Client:
             venue=venue,
             allow_testnet=False,  # XXX: never use testnet for symbol lookups
         )
+
         mkt_pairs = resp['symbols']
         if not mkt_pairs:
             raise SymbolNotFound(f'No market pairs found!?:\n{resp}')
@@ -425,28 +370,60 @@ class Client:
                 item['filters'] = filters
 
             pair_type: Type = PAIRTYPES[venue]
-            pair: Pair = pair_type(**item)
+            try:
+                pair: Pair = pair_type(**item)
+            except Exception as e:
+                e.add_note(
+                    "\nDon't panic, prolly stupid binance changed their symbology schema again..\n"
+                    'Check out their API docs here:\n\n'
+                    'https://binance-docs.github.io/apidocs/spot/en/#exchange-information'
+                )
+                raise
             pair_table[pair.symbol.upper()] = pair
 
             # update an additional top-level-cross-venue-table
             # `._pairs: ChainMap` for search B0
             pairs_view_subtable[pair.bs_fqme] = pair
 
+            # XXX WOW: TURNS OUT THIS ISN'T TRUE !?
+            # > (populate `Asset` table for spot mkts only since it
+            # > should be a superset of any other venues such as
+            # > futes or margin)
             if venue == 'spot':
-                if (name := pair.quoteAsset) not in asset_table:
-                    asset_table[name] = Asset(
-                        name=name,
-                        atype='crypto_currency',
-                        tx_tick=digits_to_dec(pair.quoteAssetPrecision),
-                    )
+                dst_sectype: str = 'crypto_currency'
 
-                if (name := pair.baseAsset) not in asset_table:
-                    asset_table[name] = Asset(
-                        name=name,
-                        atype='crypto_currency',
-                        tx_tick=digits_to_dec(pair.baseAssetPrecision),
-                    )
+            elif venue in {'usdtm_futes'}:
+                dst_sectype: str = 'future'
+                if pair.contractType == 'PERPETUAL':
+                    dst_sectype: str = 'perpetual_future'
 
+            spot_asset_table: dict[str, Asset] = self._venue2assets['spot']
+            ven_asset_table: dict[str, Asset] = self._venue2assets[venue]
+
+            if (
+                (name := pair.quoteAsset) not in spot_asset_table
+            ):
+                spot_asset_table[pair.bs_src_asset] = Asset(
+                    name=name,
+                    atype='crypto_currency',
+                    tx_tick=digits_to_dec(pair.quoteAssetPrecision),
+                )
+
+            if (
+                (name := pair.baseAsset) not in ven_asset_table
+            ):
+                if venue != 'spot':
+                    assert dst_sectype != 'crypto_currency'
+
+                ven_asset_table[pair.bs_dst_asset] = Asset(
+                    name=name,
+                    atype=dst_sectype,
+                    tx_tick=digits_to_dec(pair.baseAssetPrecision),
+                )
+
+        # log.warning(
+        #     f'Assets not YET found in spot set: `{pformat(dne)}`!?'
+        # )
         # NOTE: make merged view of all market-type pairs but
         # use market specific `Pair.bs_fqme` for keys!
         # this allows searching for market pairs with different
@@ -458,16 +435,29 @@ class Client:
         if venue == 'spot':
             return
 
-        assets: list[dict] = resp.get('assets', ())
-        for entry in assets:
-            name: str = entry['asset']
-            asset_table[name] = self._venue2assets['spot'].get(name)
+        # TODO: maybe use this assets response for non-spot venues?
+        # -> issue is we do the exch_info queries conc, so we can't
+        # guarantee order for inter-table lookups..
+        # if venue ep delivers an explicit set of assets copy just
+        # ensure they are also already listed in the spot equivs.
+        # assets: list[dict] = resp.get('assets', ())
+        # for entry in assets:
+        #     name: str = entry['asset']
+        #     spot_asset_table: dict[str, Asset] = self._venue2assets['spot']
+        #     if name not in spot_asset_table:
+        #         log.warning(
+        #             f'COULDNT FIND ASSET {name}\n{entry}\n'
+        #             f'ADDING AS FUTES ONLY!?'
+        #         )
+        #     asset_table: dict[str, Asset] = self._venue2assets[venue]
+        #     asset_table[name] = spot_asset_table.get(name)
 
     async def exch_info(
         self,
         sym: str | None = None,
 
         venue: MarketType | None = None,
+        expiry: str | None = None,
 
     ) -> dict[str, Pair] | Pair:
         '''
@@ -483,9 +473,20 @@ class Client:
 
         '''
         pair_table: dict[str, Pair] = self._venue2pairs[
-            venue or self.mkt_mode
+            venue
+            or
+            self.mkt_mode
         ]
-        if cached_pair := pair_table.get(sym):
+        if (
+            expiry
+            and 'perp' not in expiry.lower()
+        ):
+            sym: str = f'{sym}_{expiry}'
+
+        if (
+            sym
+            and (cached_pair := pair_table.get(sym))
+        ):
             return cached_pair
 
         venues: list[str] = ['spot', 'usdtm_futes']
@@ -493,14 +494,52 @@ class Client:
             venues: list[str] = [venue]
 
         # batch per-venue download of all exchange infos
-        async with trio.open_nursery() as rn:
+        async with trio.open_nursery() as tn:
             for ven in venues:
-                rn.start_soon(
+                tn.start_soon(
                     self._cache_pairs,
                     ven,
                 )
 
-        return pair_table[sym] if sym else self._pairs
+        if sym:
+            return pair_table[sym]
+        else:
+            return self._pairs
+
+    async def get_assets(
+        self,
+        venue: str | None = None,
+
+    ) -> dict[str, Asset]:
+        if (
+            venue
+            and venue != 'spot'
+        ):
+            venues = [venue]
+        else:
+            venues = ['usdtm_futes']
+
+        ass_table: dict[str, Asset] = self._venue2assets['spot']
+
+        # merge in futes contracts with a sectype suffix
+        for venue in venues:
+            ass_table |= self._venue2assets[venue]
+
+        return ass_table
+
+
+    async def get_mkt_pairs(self) -> dict[str, Pair]:
+        '''
+        Flatten the multi-venue (chain) map of market pairs
+        to a fqme indexed table for data layer caching.
+
+        '''
+        flat: dict[str, Pair] = {}
+        for venmap in self._pairs.maps:
+            for bs_fqme, pair in venmap.items():
+                flat[pair.bs_fqme] = pair
+
+        return flat
 
     # TODO: unused except by `brokers.core.search_symbols()`?
     async def search_symbols(
@@ -510,20 +549,32 @@ class Client:
 
     ) -> dict[str, Any]:
 
-        fq_pairs: dict = await self.exch_info()
+        fq_pairs: dict[str, Pair] = await self.exch_info()
 
-        matches = fuzzy.extractBests(
-            pattern,
-            fq_pairs,
+        # TODO: cache this list like we were in
+        # `open_symbol_search()`?
+        # keys: list[str] = list(fq_pairs)
+
+        return match_from_pairs(
+            pairs=fq_pairs,
+            query=pattern.upper(),
             score_cutoff=50,
         )
-        # repack in dict form
-        return {item[0]['symbol']: item[0]
-                for item in matches}
+
+    def pair2venuekey(
+        self,
+        pair: Pair,
+    ) -> str:
+        return {
+            'USDTM': 'usdtm_futes',
+            'SPOT': 'spot',
+            # 'COINM': 'coin_futes',
+            # ^-TODO-^ bc someone might want it..?
+        }[pair.venue]
 
     async def bars(
         self,
-        symbol: str,
+        mkt: MktPair,
 
         start_dt: datetime | None = None,
         end_dt: datetime | None = None,
@@ -553,16 +604,20 @@ class Client:
         start_time = binance_timestamp(start_dt)
         end_time = binance_timestamp(end_dt)
 
+        bs_pair: Pair = self._pairs[mkt.bs_fqme.upper()]
+
         # https://binance-docs.github.io/apidocs/spot/en/#kline-candlestick-data
         bars = await self._api(
             'klines',
             params={
-                'symbol': symbol.upper(),
+                # NOTE: always query using their native symbology!
+                'symbol': mkt.bs_mktid.upper(),
                 'interval': '1m',
                 'startTime': start_time,
                 'endTime': end_time,
                 'limit': limit
             },
+            venue=self.pair2venuekey(bs_pair),
             allow_testnet=False,
         )
         new_bars: list[tuple] = []
@@ -879,17 +934,148 @@ class Client:
         await self.close_listen_key(key)
 
 
+_venue_urls: dict[str, str] = {
+    'spot': (
+        _spot_url,
+        '/api/v3/',
+    ),
+    'spot_testnet': (
+        _testnet_spot_url,
+        '/fapi/v1/'
+    ),
+    # margin and extended spot endpoints session.
+    # TODO: did this ever get implemented fully?
+    # 'margin': (
+    #     _spot_url,
+    #     '/sapi/v1/'
+    # ),
+
+    'usdtm_futes': (
+        _futes_url,
+        '/fapi/v1/',
+    ),
+
+    'usdtm_futes_testnet': (
+        _testnet_futes_url,
+        '/fapi/v1/',
+    ),
+
+    # TODO: for anyone who actually needs it ;P
+    # 'coin_futes': ()
+}
+
+
+def init_api_keys(
+    client: Client,
+    conf: dict[str, Any],
+) -> None:
+    '''
+    Set up per-venue API keys each http client according to the user's
+    `brokers.conf`.
+
+    For ex, to use spot-testnet and live usdt futures APIs:
+
+    ```toml
+        [binance]
+        # spot test net
+        spot.use_testnet = true
+        spot.api_key = '<spot_api_key_from_binance_account>'
+        spot.api_secret = '<spot_api_key_password>'
+
+        # futes live
+        futes.use_testnet = false
+        accounts.usdtm = 'futes'
+        futes.api_key = '<futes_api_key_from_binance>'
+        futes.api_secret = '<futes_api_key_password>''
+
+        # if uncommented will use the built-in paper engine and not
+        # connect to `binance` API servers for order ctl.
+        # accounts.paper = 'paper'
+    ```
+
+    '''
+    for key, subconf in conf.items():
+        if api_key := subconf.get('api_key', ''):
+            venue_keys: list[str] = client.confkey2venuekeys[key]
+
+            venue_key: str
+            client: httpx.AsyncClient
+            for venue_key in venue_keys:
+                client, _ = client.venue_sesh[venue_key]
+
+                api_key_header: dict = {
+                    # taken from official:
+                    # https://github.com/binance/binance-futures-connector-python/blob/main/binance/api.py#L47
+                    "Content-Type": "application/json;charset=utf-8",
+
+                    # TODO: prolly should just always query and copy
+                    # in the real latest ver?
+                    "User-Agent": "binance-connector/6.1.6smbz6",
+                    "X-MBX-APIKEY": api_key,
+                }
+                client.headers.update(api_key_header)
+
+                # if `.use_tesnet = true` in the config then
+                # also add headers for the testnet session which
+                # will be used for all order control
+                if subconf.get('use_testnet', False):
+                    testnet_sesh, _ = client.venue_sesh[
+                        venue_key + '_testnet'
+                    ]
+                    testnet_sesh.headers.update(api_key_header)
+
+
 @acm
-async def get_client() -> Client:
+async def get_client(
+    mkt_mode: MarketType = 'spot',
+) -> Client:
+    '''
+    Construct an single `piker` client which composes multiple underlying venue
+    specific API clients both for live and test networks.
 
-    client = Client()
-    await client.exch_info()
-    log.info(
-        f'{client} in {client.mkt_mode} mode: caching exchange infos..\n'
-        'Cached multi-market pairs:\n'
-        f'spot: {len(client._spot_pairs)}\n'
-        f'usdtm_futes: {len(client._ufutes_pairs)}\n'
-        f'Total: {len(client._pairs)}\n'
-    )
+    '''
+    venue_sessions: dict[
+        str,  # venue key
+        tuple[httpx.AsyncClient, str]  # session, eps path
+    ] = {}
+    async with AsyncExitStack() as client_stack:
+        for name, (base_url, path) in _venue_urls.items():
+            api: httpx.AsyncClient = await client_stack.enter_async_context(
+                httpx.AsyncClient(
+                    base_url=base_url,
+                    # headers={},
 
-    yield client
+                    # TODO: is there a way to numerate this?
+                    # https://www.python-httpx.org/advanced/clients/#why-use-a-client
+                    # connections=4
+                )
+            )
+            venue_sessions[name] = (
+                api,
+                path,
+            )
+
+        conf: dict[str, Any] = get_config()
+        # for creating API keys see,
+        # https://www.binance.com/en/support/faq/how-to-create-api-keys-on-binance-360002502072
+        client = Client(
+            venue_sessions=venue_sessions,
+            conf=conf,
+            mkt_mode=mkt_mode,
+        )
+        init_api_keys(
+            client=client,
+            conf=conf,
+        )
+        fq_pairs: dict[str, Pair] = await client.exch_info()
+        assert fq_pairs
+        log.info(
+            f'Loaded multi-venue `Client` in mkt_mode={client.mkt_mode!r}\n\n'
+            f'Symbology Summary:\n'
+            f'------ - ------\n'
+            f'spot: {len(client._spot_pairs)}\n'
+            f'usdtm_futes: {len(client._ufutes_pairs)}\n'
+            '------ - ------\n'
+            f'total: {len(client._pairs)}\n'
+        )
+        yield client

piker/brokers/binance/broker.py

@@ -36,7 +36,6 @@ import trio
 
 from piker.accounting import (
     Asset,
-    # MktPair,
 )
 from piker.brokers._util import (
     get_logger,
@@ -49,7 +48,9 @@ from piker.brokers import (
     open_cached_client,
     BrokerError,
 )
-from piker.clearing import OrderDialogs
+from piker.clearing import (
+    OrderDialogs,
+)
 from piker.clearing._messages import (
     BrokerdOrder,
     BrokerdOrderAck,
@@ -71,6 +72,33 @@ from .api import Client
 log = get_logger('piker.brokers.binance')
 
 
+# Fee schedule template, mostly for paper engine fees modelling.
+# https://www.binance.com/en/support/faq/what-are-market-makers-and-takers-360007720071
+def get_cost(
+    price: float,
+    size: float,
+    is_taker: bool = False,
+
+) -> float:
+
+    # https://www.binance.com/en/fee/trading
+    cb: float = price * size
+    match is_taker:
+        case True:
+            return cb * 0.001000
+
+        case False if cb < 1e6:
+            return cb * 0.001000
+
+        case False if 1e6 >= cb < 5e6:
+            return cb * 0.000900
+
+        # NOTE: there's more but are you really going
+        # to have a cb bigger then this per trade?
+        case False if cb >= 5e6:
+            return cb * 0.000800
+
+
 async def handle_order_requests(
     ems_order_stream: tractor.MsgStream,
     client: Client,
@@ -232,16 +260,24 @@ async def open_trade_dialog(
     account_name: str = 'usdtm'
     use_testnet: bool = False
 
+    # TODO: if/when we add .accounting support we need to
+    # do a open_symcache() call.. though maybe we can hide
+    # this in a new async version of open_account()?
     async with open_cached_client('binance') as client:
-        subconf: dict = client.conf[venue_name]
-        use_testnet = subconf.get('use_testnet', False)
+        subconf: dict|None = client.conf.get(venue_name)
 
         # XXX: if no futes.api_key or spot.api_key has been set we
         # always fall back to the paper engine!
-        if not subconf.get('api_key'):
+        if (
+            not subconf
+            or
+            not subconf.get('api_key')
+        ):
             await ctx.started('paper')
             return
 
+        use_testnet: bool = subconf.get('use_testnet', False)
+
     async with (
         open_cached_client('binance') as client,
     ):
@@ -321,7 +357,7 @@ async def open_trade_dialog(
 
                         if balance > 0:
                             balances[spot_asset] = (balance, last_update_t)
-                            # await tractor.breakpoint()
+                            # await tractor.pause()
 
                 # @position response:
                 # {'positions': [{'entryPrice': '0.0',
@@ -400,7 +436,7 @@ async def open_trade_dialog(
             #   and comparison with binance's own position calcs.
             # - load pps and accounts using accounting apis, write
             #   the ledger and account files
-            #   - table: PpTable
+            #   - table: Account
             #   - ledger: TransactionLedger
 
             async with (

piker/brokers/binance/feed.py

@@ -24,8 +24,11 @@ from contextlib import (
     aclosing,
 )
 from datetime import datetime
-from functools import partial
+from functools import (
+    partial,
+)
 import itertools
+from pprint import pformat
 from typing import (
     Any,
     AsyncGenerator,
@@ -39,12 +42,12 @@ from trio_typing import TaskStatus
 from pendulum import (
     from_timestamp,
 )
-from fuzzywuzzy import process as fuzzy
 import numpy as np
 import tractor
 
 from piker.brokers import (
     open_cached_client,
+    NoData,
 )
 from piker._cacheables import (
     async_lifo_cache,
@@ -54,9 +57,8 @@ from piker.accounting import (
     DerivTypes,
     MktPair,
     unpack_fqme,
-    digits_to_dec,
 )
-from piker.data.types import Struct
+from piker.types import Struct
 from piker.data.validate import FeedInit
 from piker.data._web_bs import (
     open_autorecon_ws,
@@ -108,6 +110,7 @@ class AggTrade(Struct, frozen=True):
 
 async def stream_messages(
     ws: NoBsWs,
+
 ) -> AsyncGenerator[NoBsWs, dict]:
 
     # TODO: match syntax here!
@@ -218,6 +221,8 @@ def make_sub(pairs: list[str], sub_name: str, uid: int) -> dict[str, str]:
     }
 
 
+# TODO, why aren't frame resp `log.info()`s showing in upstream
+# code?!
 @acm
 async def open_history_client(
     mkt: MktPair,
@@ -250,24 +255,30 @@ async def open_history_client(
             else:
                 client.mkt_mode = 'spot'
 
-            # NOTE: always query using their native symbology!
-            mktid: str = mkt.bs_mktid
-            array = await client.bars(
-                mktid,
+            array: np.ndarray = await client.bars(
+                mkt=mkt,
                 start_dt=start_dt,
                 end_dt=end_dt,
             )
+            if array.size == 0:
+                raise NoData(
+                    f'No frame for {start_dt} -> {end_dt}\n'
+                )
+
             times = array['time']
-            if (
-                end_dt is None
-            ):
-                inow = round(time.time())
+            if not times.any():
+                raise ValueError(
+                    'Bad frame with null-times?\n\n'
+                    f'{times}'
+                )
+
+            if end_dt is None:
+                inow: int = round(time.time())
                 if (inow - times[-1]) > 60:
-                    await tractor.breakpoint()
+                    await tractor.pause()
 
             start_dt = from_timestamp(times[0])
             end_dt = from_timestamp(times[-1])
-
             return array, start_dt, end_dt
 
         yield get_ohlc, {'erlangs': 3, 'rate': 3}
@@ -277,69 +288,113 @@ async def open_history_client(
 async def get_mkt_info(
     fqme: str,
 
-) -> tuple[MktPair, Pair]:
+) -> tuple[MktPair, Pair] | None:
 
     # uppercase since kraken bs_mktid is always upper
-    if 'binance' not in fqme:
+    if 'binance' not in fqme.lower():
         fqme += '.binance'
 
-    bs_fqme, _, broker = fqme.rpartition('.')
+    mkt_mode: str = ''
     broker, mkt_ep, venue, expiry = unpack_fqme(fqme)
 
-    # NOTE: see the `FutesPair.bs_fqme: str` implementation
-    # to understand the reverse market info lookup below.
-    mkt_mode = venue = venue.lower() or 'spot'
-    _atype: str = ''
+    # NOTE: we always upper case all tokens to be consistent with
+    # binance's symbology style for pairs, like `BTCUSDT`, but in
+    # theory we could also just keep things lower case; as long as
+    # we're consistent and the symcache matches whatever this func
+    # returns, always!
+    expiry: str = expiry.upper()
+    venue: str = venue.upper()
+    venue_lower: str = venue.lower()
+
+    # XXX TODO: we should change the usdtm_futes name to just
+    # usdm_futes (dropping the tether part) since it turns out that
+    # there are indeed USD-tokens OTHER THEN tether being used as
+    # the margin assets.. it's going to require a wholesale
+    # (variable/key) rename as well as file name adjustments to any
+    # existing tsdb set..
+    if 'usd' in venue_lower:
+        mkt_mode: str = 'usdtm_futes'
+
+    # NO IDEA what these contracts (some kinda DEX-ish futes?) are
+    # but we're masking them for now..
+    elif (
+        'defi' in venue_lower
+
+        # TODO: handle coinm futes which have a margin asset that
+        # is some crypto token!
+        # https://binance-docs.github.io/apidocs/delivery/en/#exchange-information
+        or 'btc' in venue_lower
+    ):
+        return None
+
+    else:
+        # NOTE: see the `FutesPair.bs_fqme: str` implementation
+        # to understand the reverse market info lookup below.
+        mkt_mode = venue_lower or 'spot'
+
     if (
         venue
-        and 'spot' not in venue.lower()
+        and 'spot' not in venue_lower
 
         # XXX: catch all in case user doesn't know which
         # venue they want (usdtm vs. coinm) and we can choose
         # a default (via config?) once we support coin-m APIs.
-        or 'perp' in bs_fqme.lower()
+        or 'perp' in venue_lower
     ):
-        mkt_mode: str = f'{venue.lower()}_futes'
-        if 'perp' in expiry:
-            _atype = 'perpetual_future'
-
-        else:
-            _atype = 'future'
+        if not mkt_mode:
+            mkt_mode: str = f'{venue_lower}_futes'
 
     async with open_cached_client(
         'binance',
     ) as client:
 
-        # switch mode depending on input pattern parsing
+        assets: dict[str, Asset] = await client.get_assets()
+        pair_str: str = mkt_ep.upper()
+
+        # switch venue-mode depending on input pattern parsing
+        # since we want to use a particular endpoint (set) for
+        # pair info lookup!
         client.mkt_mode = mkt_mode
 
-        pair_str: str = mkt_ep.upper()
-        pair: Pair = await client.exch_info(pair_str)
+        pair: Pair = await client.exch_info(
+            pair_str,
+            venue=mkt_mode,  # explicit
+            expiry=expiry,
+        )
 
         if 'futes' in mkt_mode:
             assert isinstance(pair, FutesPair)
 
+        dst: Asset | None = assets.get(pair.bs_dst_asset)
+        if (
+            not dst
+            # TODO: a known asset DNE list?
+            # and pair.baseAsset == 'DEFI'
+        ):
+            log.warning(
+                f'UNKNOWN {venue} asset {pair.baseAsset} from,\n'
+                f'{pformat(pair.to_dict())}'
+            )
+
+            # XXX UNKNOWN missing "asset", though no idea why?
+            # maybe it's only avail in the margin venue(s): /dapi/ ?
+            return None
+
         mkt = MktPair(
-            dst=Asset(
-                name=pair.baseAsset,
-                atype='crypto',
-                tx_tick=digits_to_dec(pair.baseAssetPrecision),
-            ),
-            src=Asset(
-                name=pair.quoteAsset,
-                atype='crypto',
-                tx_tick=digits_to_dec(pair.quoteAssetPrecision),
-            ),
+            dst=dst,
+            src=assets[pair.bs_src_asset],
             price_tick=pair.price_tick,
             size_tick=pair.size_tick,
             bs_mktid=pair.symbol,
             expiry=expiry,
             venue=venue,
             broker='binance',
-            _atype=_atype,
+
+            # NOTE: sectype is always taken from dst, see
+            # `MktPair.type_key` and `Client._cache_pairs()`
+            # _atype=sectype,
         )
-        both = mkt, pair
-        return both
+        return mkt, pair
 
 
 @acm
@@ -410,6 +465,8 @@ async def stream_quotes(
     ):
         init_msgs: list[FeedInit] = []
         for sym in symbols:
+            mkt: MktPair
+            pair: Pair
             mkt, pair = await get_mkt_info(sym)
 
             # build out init msgs according to latest spec
@@ -458,7 +515,6 @@ async def stream_quotes(
 
             # start streaming
             async for typ, quote in msg_gen:
-
                 # period = time.time() - last
                 # hz = 1/period if period else float('inf')
                 # if hz > 60:
@@ -472,10 +528,11 @@ async def open_symbol_search(
     ctx: tractor.Context,
 ) -> Client:
 
+    # NOTE: symbology tables are loaded as part of client
+    # startup in ``.api.get_client()`` and in this case
+    # are stored as `Client._pairs`.
     async with open_cached_client('binance') as client:
 
-        # load all symbols locally for fast search
-        fqpairs_cache = await client.exch_info()
         # TODO: maybe we should deliver the cache
         # so that client's can always do a local-lookup-first
         # style try and then update async as (new) match results
@@ -486,14 +543,15 @@ async def open_symbol_search(
 
             pattern: str
             async for pattern in stream:
-                matches = fuzzy.extractBests(
+                # NOTE: pattern fuzzy-matching is done within
+                # the methd impl.
+                pairs: dict[str, Pair] = await client.search_symbols(
                     pattern,
-                    fqpairs_cache,
-                    score_cutoff=50,
                 )
 
-                # repack in dict form
-                await stream.send({
-                    item[0].bs_fqme: item[0]
-                    for item in matches
-                })
+                # repack in fqme-keyed table
+                byfqme: dict[str, Pair] = {}
+                for pair in pairs.values():
+                    byfqme[pair.bs_fqme] = pair
+
+                await stream.send(byfqme)

piker/brokers/binance/venues.py

@@ -1,8 +1,5 @@
 # piker: trading gear for hackers
-# Copyright (C)
-#   Guillermo Rodriguez (aka ze jefe)
-#   Tyler Goodlet
-#   (in stewardship for pikers)
+# Copyright (C) Tyler Goodlet (in stewardship for pikers)
 
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU Affero General Public License as published by
@@ -29,7 +26,7 @@ from decimal import Decimal
 
 from msgspec import field
 
-from piker.data.types import Struct
+from piker.types import Struct
 
 
 # API endpoint paths by venue / sub-API
@@ -65,7 +62,7 @@ MarketType = Literal[
     'spot',
     # 'margin',
     'usdtm_futes',
-    # 'coin_futes',
+    # 'coinm_futes',
 ]
 
 
@@ -87,6 +84,7 @@ def get_api_eps(venue: MarketType) -> tuple[str, str]:
 
 
 class Pair(Struct, frozen=True, kw_only=True):
+
     symbol: str
     status: str
     orderTypes: list[str]
@@ -120,6 +118,10 @@ class Pair(Struct, frozen=True, kw_only=True):
     def bs_fqme(self) -> str:
         return self.symbol
 
+    @property
+    def bs_mktid(self) -> str:
+        return f'{self.symbol}.{self.venue}'
+
 
 class SpotPair(Pair, frozen=True):
 
@@ -135,15 +137,31 @@ class SpotPair(Pair, frozen=True):
     quoteOrderQtyMarketAllowed: bool
     isSpotTradingAllowed: bool
     isMarginTradingAllowed: bool
+    otoAllowed: bool
 
     defaultSelfTradePreventionMode: str
     allowedSelfTradePreventionModes: list[str]
     permissions: list[str]
+    permissionSets: list[list[str]]
+
+    # NOTE: see `.data._symcache.SymbologyCache.load()` for why
+    ns_path: str = 'piker.brokers.binance:SpotPair'
+
+    @property
+    def venue(self) -> str:
+        return 'SPOT'
 
     @property
     def bs_fqme(self) -> str:
         return f'{self.symbol}.SPOT'
 
+    @property
+    def bs_src_asset(self) -> str:
+        return f'{self.quoteAsset}'
+
+    @property
+    def bs_dst_asset(self) -> str:
+        return f'{self.baseAsset}'
 
 
 class FutesPair(Pair):
@@ -163,12 +181,14 @@ class FutesPair(Pair):
     quoteAsset: str  # 'USDT',
     quotePrecision: int  # 8,
     requiredMarginPercent: float  # '5.0000',
-    settlePlan: int  # 0,
     timeInForce: list[str]  # ['GTC', 'IOC', 'FOK', 'GTX'],
     triggerProtect: float  # '0.0500',
     underlyingSubType: list[str]  # ['PoW'],
     underlyingType: str  # 'COIN'
 
+    # NOTE: see `.data._symcache.SymbologyCache.load()` for why
+    ns_path: str = 'piker.brokers.binance:FutesPair'
+
     # NOTE: for compat with spot pairs and `MktPair.src: Asset`
     # processing..
     @property
@@ -176,32 +196,101 @@ class FutesPair(Pair):
         return self.quotePrecision
 
     @property
-    def bs_fqme(self) -> str:
+    def expiry(self) -> str:
+        symbol: str = self.symbol
+        contype: str = self.contractType
+        match contype:
+            case (
+                'CURRENT_QUARTER'
+                | 'CURRENT_QUARTER DELIVERING'
+                | 'NEXT_QUARTER'  # su madre binance..
+            ):
+                pair, _, expiry = symbol.partition('_')
+                assert pair == self.pair  # sanity
+                return f'{expiry}'
+
+            case 'PERPETUAL':
+                return 'PERP'
+
+            case '':
+                subtype: list[str] = self.underlyingSubType
+                if not subtype:
+                    if self.status == 'PENDING_TRADING':
+                        return 'PENDING'
+
+                match subtype:
+                    case ['DEFI']:
+                        return 'PERP'
+
+        # wow, just wow you binance guys suck..
+        if self.status == 'PENDING_TRADING':
+            return 'PENDING'
+
+        # XXX: yeah no clue then..
+        raise ValueError(
+            f'Bad .expiry token match: {contype} for {symbol}'
+        )
+
+    @property
+    def venue(self) -> str:
         symbol: str = self.symbol
         ctype: str = self.contractType
         margin: str = self.marginAsset
 
         match ctype:
             case 'PERPETUAL':
-                return f'{symbol}.{margin}M.PERP'
+                return f'{margin}M'
 
-            case 'CURRENT_QUARTER':
-                pair, _, expiry = symbol.partition('_')
-                return f'{pair}.{margin}M.{expiry}'
+            case (
+                'CURRENT_QUARTER'
+                | 'CURRENT_QUARTER DELIVERING'
+                | 'NEXT_QUARTER'  # su madre binance..
+            ):
+                _, _, expiry = symbol.partition('_')
+                return f'{margin}M'
 
             case '':
                 subtype: list[str] = self.underlyingSubType
                 if not subtype:
                     if self.status == 'PENDING_TRADING':
-                        return f'{symbol}.{margin}M.PENDING'
+                        return f'{margin}M'
 
-                match subtype[0]:
-                    case 'DEFI':
-                        return f'{symbol}.{subtype}.PERP'
+                match subtype:
+                    case (
+                        ['DEFI']
+                        | ['USDC']
+                    ):
+                        return f'{subtype[0]}'
 
         # XXX: yeah no clue then..
-        return f'{symbol}.WTF.PWNED.BBQ'
+        raise ValueError(
+            f'Bad .venue token match: {ctype}'
+        )
 
+    @property
+    def bs_fqme(self) -> str:
+        symbol: str = self.symbol
+        ctype: str = self.contractType
+        venue: str = self.venue
+        pair: str = self.pair
+
+        match ctype:
+            case (
+                'CURRENT_QUARTER'
+                | 'NEXT_QUARTER'  # su madre binance..
+            ):
+                pair, _, expiry = symbol.partition('_')
+                assert pair == self.pair
+
+        return f'{pair}.{venue}.{self.expiry}'
+
+    @property
+    def bs_src_asset(self) -> str:
+        return f'{self.quoteAsset}'
+
+    @property
+    def bs_dst_asset(self) -> str:
+        return f'{self.baseAsset}.{self.venue}'
 
 
 PAIRTYPES: dict[MarketType, Pair] = {

piker/brokers/cli.py

@@ -454,8 +454,18 @@ def mkt_info(
 
 @cli.command()
 @click.argument('pattern', required=True)
+# TODO: move this to top level click/typer context for all subs
+@click.option(
+    '--pdb',
+    is_flag=True,
+    help='Enable tractor debug mode',
+)
 @click.pass_obj
-def search(config, pattern):
+def search(
+    config: dict,
+    pattern: str,
+    pdb: bool,
+):
     '''
     Search for symbols from broker backend(s).
 
@@ -468,23 +478,26 @@ def search(config, pattern):
 
         async with maybe_open_pikerd(
             loglevel=config['loglevel'],
+            debug_mode=pdb,
         ):
             return await func()
 
-    quotes = trio.run(
-        main,
-        partial(
-            core.symbol_search,
-            brokermods,
-            pattern,
-        ),
-    )
+    from piker.toolz import open_crash_handler
+    with open_crash_handler():
+        quotes = trio.run(
+            main,
+            partial(
+                core.symbol_search,
+                brokermods,
+                pattern,
+            ),
+        )
 
-    if not quotes:
-        log.error(f"No matches could be found for {pattern}?")
-        return
+        if not quotes:
+            log.error(f"No matches could be found for {pattern}?")
+            return
 
-    click.echo(colorize_json(quotes))
+        click.echo(colorize_json(quotes))
 
 
 @cli.command()
@@ -493,9 +506,11 @@ def search(config, pattern):
 @click.option('--delete', '-d', flag_value=True, help='Delete section')
 @click.pass_obj
 def brokercfg(config, section, value, delete):
-    """If invoked with no arguments, open an editor to edit broker configs file
-    or get / update an individual section.
-    """
+    '''
+    If invoked with no arguments, open an editor to edit broker
+    configs file or get / update an individual section.
+
+    '''
     from .. import config
 
     if section:

piker/brokers/core.py

@@ -95,15 +95,15 @@ async def option_chain(
             return await client.option_chains(contracts)
 
 
-async def contracts(
-    brokermod: ModuleType,
-    symbol: str,
-) -> Dict[str, Dict[str, Dict[str, Any]]]:
-    """Return option contracts (all expiries) for ``symbol``.
-    """
-    async with brokermod.get_client() as client:
-        # return await client.get_all_contracts([symbol])
-        return await client.get_all_contracts([symbol])
+# async def contracts(
+#     brokermod: ModuleType,
+#     symbol: str,
+# ) -> Dict[str, Dict[str, Dict[str, Any]]]:
+#     """Return option contracts (all expiries) for ``symbol``.
+#     """
+#     async with brokermod.get_client() as client:
+#         # return await client.get_all_contracts([symbol])
+#         return await client.get_all_contracts([symbol])
 
 
 async def bars(
@@ -117,21 +117,6 @@ async def bars(
         return await client.bars(symbol, **kwargs)
 
 
-async def mkt_info(
-    brokermod: ModuleType,
-    fqme: str,
-    **kwargs,
-
-) -> MktPair:
-    '''
-    Return MktPair info from broker including src and dst assets.
-
-    '''
-    return await brokermod.get_mkt_info(
-        fqme.replace(brokermod.name, '')
-    )
-
-
 async def search_w_brokerd(name: str, pattern: str) -> dict:
 
     async with open_cached_client(name) as client:
@@ -160,7 +145,11 @@ async def symbol_search(
 
         async with maybe_spawn_brokerd(
             mod.name,
-            infect_asyncio=getattr(mod, '_infect_asyncio', False),
+            infect_asyncio=getattr(
+                mod,
+                '_infect_asyncio',
+                False,
+            ),
         ) as portal:
 
             results.append((
@@ -178,3 +167,20 @@ async def symbol_search(
             n.start_soon(search_backend, mod.name)
 
     return results
+
+
+async def mkt_info(
+    brokermod: ModuleType,
+    fqme: str,
+    **kwargs,
+
+) -> MktPair:
+    '''
+    Return MktPair info from broker including src and dst assets.
+
+    '''
+    async with open_cached_client(brokermod.name) as client:
+        assert client
+        return await brokermod.get_mkt_info(
+            fqme.replace(brokermod.name, '')
+        )

piker/brokers/deribit/api.py

@@ -34,7 +34,7 @@ from typing import (
 import pendulum
 import trio
 from trio_typing import TaskStatus
-from fuzzywuzzy import process as fuzzy
+from rapidfuzz import process as fuzzy
 import numpy as np
 from tractor.trionics import (
     broadcast_receiver,
@@ -52,8 +52,11 @@ from cryptofeed.defines import (
 )
 from cryptofeed.symbols import Symbol
 
-from piker.data.types import Struct
-from piker.data import def_iohlcv_fields
+from piker.data import (
+    def_iohlcv_fields,
+    match_from_pairs,
+    Struct,
+)
 from piker.data._web_bs import (
     open_jsonrpc_session
 )
@@ -79,7 +82,7 @@ _testnet_ws_url = 'wss://test.deribit.com/ws/api/v2'
 class JSONRPCResult(Struct):
     jsonrpc: str = '2.0'
     id: int
-    result: Optional[dict] = None
+    result: Optional[list[dict]] = None
     error: Optional[dict] = None
     usIn: int
     usOut: int
@@ -289,24 +292,29 @@ class Client:
         currency: str = 'btc',  # BTC, ETH, SOL, USDC
         kind: str = 'option',
         expired: bool = False
-    ) -> dict[str, Any]:
-        """Get symbol info for the exchange.
 
-        """
+    ) -> dict[str, dict]:
+        '''
+        Get symbol infos.
+
+        '''
         if self._pairs:
             return self._pairs
 
         # will retrieve all symbols by default
-        params = {
+        params: dict[str, str] = {
             'currency': currency.upper(),
             'kind': kind,
             'expired': str(expired).lower()
         }
 
-        resp = await self.json_rpc('public/get_instruments', params)
-        results = resp.result
-
-        instruments = {
+        resp: JSONRPCResult = await self.json_rpc(
+            'public/get_instruments',
+            params,
+        )
+        # convert to symbol-keyed table
+        results: list[dict] | None = resp.result
+        instruments: dict[str, dict] = {
             item['instrument_name'].lower(): item
             for item in results
         }
@@ -319,6 +327,7 @@ class Client:
     async def cache_symbols(
         self,
     ) -> dict:
+
         if not self._pairs:
             self._pairs = await self.symbol_info()
 
@@ -329,17 +338,23 @@ class Client:
         pattern: str,
         limit: int = 30,
     ) -> dict[str, Any]:
-        data = await self.symbol_info()
+        '''
+        Fuzzy search symbology set for pairs matching `pattern`.
 
-        matches = fuzzy.extractBests(
-            pattern,
-            data,
+        '''
+        pairs: dict[str, Any] = await self.symbol_info()
+        matches: dict[str, Pair] = match_from_pairs(
+            pairs=pairs,
+            query=pattern.upper(),
             score_cutoff=35,
             limit=limit
         )
-        # repack in dict form
-        return {item[0]['instrument_name'].lower(): item[0]
-                for item in matches}
+
+       # repack in name-keyed table
+        return {
+            pair['instrument_name'].lower(): pair
+            for pair in matches.values()
+        }
 
     async def bars(
         self,

piker/brokers/deribit/feed.py

@@ -26,7 +26,7 @@ import time
 import trio
 from trio_typing import TaskStatus
 import pendulum
-from fuzzywuzzy import process as fuzzy
+from rapidfuzz import process as fuzzy
 import numpy as np
 import tractor
 

piker/brokers/ib/__init__.py

@@ -30,23 +30,33 @@ from .api import (
 )
 from .feed import (
     open_history_client,
-    open_symbol_search,
     stream_quotes,
 )
 from .broker import (
     open_trade_dialog,
 )
 from .ledger import (
+    norm_trade,
     norm_trade_records,
+    tx_sort,
+)
+from .symbols import (
+    get_mkt_info,
+    open_symbol_search,
+    _search_conf,
 )
 
 __all__ = [
     'get_client',
+    'get_mkt_info',
+    'norm_trade',
     'norm_trade_records',
     'open_trade_dialog',
     'open_history_client',
     'open_symbol_search',
     'stream_quotes',
+    '_search_conf',
+    'tx_sort',
 ]
 
 _brokerd_mods: list[str] = [
@@ -56,6 +66,7 @@ _brokerd_mods: list[str] = [
 
 _datad_mods: list[str] = [
     'feed',
+    'symbols',
 ]
 
 
@@ -75,3 +86,8 @@ _spawn_kwargs = {
 # know if ``brokerd`` should be spawned with
 # ``tractor``'s aio mode.
 _infect_asyncio: bool = True
+
+# XXX NOTE: for now we disable symcache with this backend since
+# there is no clearly simple nor practical way to download "all
+# symbology info" for all supported venues..
+_no_symcache: bool = True

piker/brokers/ib/_flex_reports.py

@@ -159,7 +159,11 @@ def load_flex_trades(
     for acctid in trades_by_account:
         trades_by_id = trades_by_account[acctid]
 
-        with open_trade_ledger('ib', acctid) as ledger_dict:
+        with open_trade_ledger(
+            'ib',
+            acctid,
+            allow_from_sync_code=True,
+        ) as ledger_dict:
             tid_delta = set(trades_by_id) - set(ledger_dict)
             log.info(
                 'New trades detected\n'

piker/brokers/ib/_util.py

@@ -88,16 +88,25 @@ async def data_reset_hack(
     api_port: str = str(ib_client.client.port)
     vnc_host: str
     vnc_port: int
-    vnc_host, vnc_port = client.conf['vnc_addrs'].get(
-        api_port,
-        ('localhost', 3003)
-    )
+    vnc_sockaddr: tuple[str] | None = client.conf.get('vnc_addrs')
 
     no_setup_msg:str = (
-        f'No data reset hack test setup for {vnc_host}!\n'
-        'See setup @\n'
+        f'No data reset hack test setup for {vnc_sockaddr}!\n'
+        'See config setup tips @\n'
         'https://github.com/pikers/piker/tree/master/piker/brokers/ib'
     )
+
+    if not vnc_sockaddr:
+        log.warning(
+            no_setup_msg
+            +
+            'REQUIRES A `vnc_addrs: array` ENTRY'
+        )
+
+    vnc_host, vnc_port = vnc_sockaddr.get(
+        api_port,
+        ('localhost', 3003)
+    )
     global _reset_tech
 
     match _reset_tech:
@@ -250,7 +259,7 @@ def i3ipc_xdotool_manual_click_hack() -> None:
                         timeout=timeout,
                     )
 
-    # re-activate and focus original window
+        # re-activate and focus original window
         subprocess.call([
             'xdotool',
             'windowactivate', '--sync', str(orig_win_id),

piker/brokers/ib/feed.py

@@ -21,33 +21,40 @@ from __future__ import annotations
 import asyncio
 from contextlib import (
     asynccontextmanager as acm,
-    nullcontext,
 )
-from decimal import Decimal
 from dataclasses import asdict
 from datetime import datetime
 from functools import partial
+from pprint import pformat
 from math import isnan
 import time
 from typing import (
     Any,
     Callable,
-    Awaitable,
+    TYPE_CHECKING,
 )
 
 from async_generator import aclosing
-from fuzzywuzzy import process as fuzzy
 import ib_insync as ibis
 import numpy as np
-import pendulum
+from pendulum import (
+    now,
+    from_timestamp,
+    # DateTime,
+    Duration,
+    duration as mk_duration,
+)
 import tractor
 import trio
 from trio_typing import TaskStatus
 
-from .._util import (
+from piker.accounting import (
+    MktPair,
+)
+from piker.data.validate import FeedInit
+from piker.brokers._util import (
     NoData,
     DataUnavailable,
-    SymbolNotFound,
 )
 from .api import (
     # _adhoc_futes_set,
@@ -63,14 +70,10 @@ from .api import (
     RequestError,
 )
 from ._util import data_reset_hack
-from piker._cacheables import (
-    async_lifo_cache,
-)
-from piker.accounting import (
-    Asset,
-    MktPair,
-)
-from piker.data.validate import FeedInit
+from .symbols import get_mkt_info
+
+if TYPE_CHECKING:
+    from trio._core._run import Task
 
 
 # XXX NOTE: See available types table docs:
@@ -140,32 +143,21 @@ async def open_history_client(
     #   memory.
 
     # IB's internal symbology does not expect the "source asset" in
-    # the "symbol name", what we call the "market name". This is
+    # the "symbol name", what we call the "market pair name". This is
     # common in most legacy market brokers since it's presumed that
     # given a certain stock exchange, listed assets are traded
-    # "from" a particular source fiat, normally something like USD.
-    if (
-        mkt.src
-        and mkt.src.atype == 'fiat'
-    ):
-        fqme_kwargs: dict[str, Any] = {}
+    # "from" a particular source fiat, normally something like USD
+    # on the given venue-provider, eg. nasdaq, nyse, etc.
+    fqme_kwargs: dict[str, Any] = {}
+    if mkt.dst.atype != 'fiat':
+        fqme_kwargs  = {
+            'without_src': True,  # default is True
+            'delim_char': '',  # bc they would normally use a frickin `.` smh
+        }
 
-        if mkt.dst.atype == 'forex':
-
-            # XXX: for now we do need the src token kept in since
-            fqme_kwargs  = {
-                'without_src': False,  # default is True
-                'delim_char': '',  # bc they would normally use a frickin `.` smh
-            }
-
-        fqme: str = mkt.get_bs_fqme(**(fqme_kwargs))
-
-    else:
-        fqme = mkt.bs_fqme
+    fqme: str = mkt.get_bs_fqme(**(fqme_kwargs))
 
     async with open_data_client() as proxy:
-
-
         max_timeout: float = 2.
         mean: float = 0
         count: int = 0
@@ -173,12 +165,13 @@ async def open_history_client(
         head_dt: None | datetime = None
         if (
             # fx cons seem to not provide this endpoint?
+            # TODO: guard against all contract types which don't
+            # support it?
             'idealpro' not in fqme
         ):
-            try:
-                head_dt = await proxy.get_head_time(fqme=fqme)
-            except RequestError:
-                head_dt = None
+            head_dt: datetime | None = await proxy.maybe_get_head_time(
+                fqme=fqme
+            )
 
         async def get_hist(
             timeframe: float,
@@ -186,8 +179,15 @@ async def open_history_client(
             start_dt: datetime | None = None,
 
         ) -> tuple[np.ndarray, str]:
+
             nonlocal max_timeout, mean, count
 
+            if (
+                start_dt
+                and start_dt.timestamp() == 0
+            ):
+                await tractor.pause()
+
             query_start = time.time()
             out, timedout = await get_bars(
                 proxy,
@@ -208,24 +208,48 @@ async def open_history_client(
                     f'mean: {mean}'
                 )
 
-            if (
-                out is None
-            ):
-                # could be trying to retreive bars over weekend
+            # could be trying to retreive bars over weekend
+            if out is None:
                 log.error(f"Can't grab bars starting at {end_dt}!?!?")
-                raise NoData(
-                    f'{end_dt}',
-                    # frame_size=2000,
-                )
+                if (
+                    end_dt
+                    and head_dt
+                    and end_dt <= head_dt
+                ):
+                    raise DataUnavailable(
+                        f'First timestamp is {head_dt}\n'
+                        f'But {end_dt} was requested..'
+                    )
 
-            if (
-                end_dt
-                and head_dt
-                and end_dt <= head_dt
-            ):
-                raise DataUnavailable(f'First timestamp is {head_dt}')
+                else:
+                    raise NoData(
+                        info={
+                            'fqme': fqme,
+                            'head_dt': head_dt,
+                            'start_dt': start_dt,
+                            'end_dt': end_dt,
+                            'timedout': timedout,
+                        },
+                    )
 
-            bars, bars_array, first_dt, last_dt = out
+            # also see return type for `get_bars()`
+            bars: ibis.objects.BarDataList
+            bars_array: np.ndarray
+            first_dt: datetime
+            last_dt: datetime
+            (
+                bars,
+                bars_array,
+                first_dt,
+                last_dt,
+            ) = out
+
+            # TODO: audit the sampling period here as well?
+            # timestep should always be at least as large as the
+            # period step.
+            # tdiff: np.ndarray = np.diff(bars_array['time'])
+            # if (tdiff < timeframe).any():
+            #     await tractor.pause()
 
             # volume cleaning since there's -ve entries,
             # wood luv to know what crookery that is..
@@ -239,7 +263,18 @@ async def open_history_client(
         # quite sure why.. needs some tinkering and probably
         # a lookthrough of the ``ib_insync`` machinery, for eg. maybe
         # we have to do the batch queries on the `asyncio` side?
-        yield get_hist, {'erlangs': 1, 'rate': 3}
+        yield (
+            get_hist,
+            {
+                'erlangs': 1,  # max conc reqs
+                'rate': 3,  # max req rate
+                'frame_types': {  # expected frame sizes
+                    1: mk_duration(seconds=2e3),
+                    60: mk_duration(days=2),
+                }
+
+            },
+        )
 
 
 _pacing: str = (
@@ -325,7 +360,7 @@ async def wait_on_data_reset(
     return False
 
 
-_data_resetter_task: trio.Task | None = None
+_data_resetter_task: Task | None = None
 _failed_resets: int = 0
 
 
@@ -351,7 +386,15 @@ async def get_bars(
 
     task_status: TaskStatus[trio.CancelScope] = trio.TASK_STATUS_IGNORED,
 
-) -> (dict, np.ndarray):
+) -> tuple[
+    tuple[  # result tuple
+        ibis.objects.BarDataList,
+        np.ndarray,
+        datetime,
+        datetime,
+    ] | None,
+    bool,  # timed out hint
+]:
     '''
     Retrieve historical data from a ``trio``-side task using
     a ``MethoProxy``.
@@ -376,7 +419,11 @@ async def get_bars(
 
         while _failed_resets < max_failed_resets:
             try:
-                out = await proxy.bars(
+                (
+                    bars,
+                    bars_array,
+                    dt_duration,
+                ) = await proxy.bars(
                     fqme=fqme,
                     end_dt=end_dt,
                     sample_period_s=timeframe,
@@ -387,44 +434,58 @@ async def get_bars(
                     # current impl) to detect a cancel case.
                     # timeout=timeout,
                 )
-                if out is None:
-                    raise NoData(f'{end_dt}')
-
-                bars, bars_array, dt_duration = out
+                # usually either a request during a venue closure
+                # or into a large (weekend) closure gap.
+                if not bars:
+                    # no data returned?
+                    log.warning(
+                        'History frame is blank?\n'
+                        f'start_dt: {start_dt}\n'
+                        f'end_dt: {end_dt}\n'
+                        f'duration: {dt_duration}\n'
+                    )
+                    # NOTE: REQUIRED to pass back value..
+                    result = None
+                    return None
 
                 # not enough bars signal, likely due to venue
                 # operational gaps.
-                too_little: bool = False
-                if (
-                    end_dt
-                    and (
-                        not bars
-                        or (too_little :=
-                            start_dt
-                            and (len(bars) * timeframe)
-                                < dt_duration.in_seconds()
-                        )
-                    )
-                ):
-                    if (
-                        end_dt
-                        or too_little
-                    ):
+                if end_dt:
+                    dur_s: float = len(bars) * timeframe
+                    bars_dur = Duration(seconds=dur_s)
+                    dt_dur_s: float = dt_duration.in_seconds()
+                    if dur_s < dt_dur_s:
                         log.warning(
-                            f'History is blank for {dt_duration} from {end_dt}'
+                            'History frame is shorter then expected?\n'
+                            f'start_dt: {start_dt}\n'
+                            f'end_dt: {end_dt}\n'
+                            f'duration: {dt_dur_s}\n'
+                            f'frame duration seconds: {dur_s}\n'
+                            f'dur diff: {dt_duration - bars_dur}\n'
                         )
-                        end_dt -= dt_duration
-                        continue
+                        # NOTE: we used to try to get a minimal
+                        # set of bars by recursing but this ran
+                        # into possible infinite query loops
+                        # when logic in the `Client.bars()` dt
+                        # diffing went bad. So instead for now
+                        # we just return the
+                        # shorter-then-expected history with
+                        # a warning.
+                        # TODO: in the future it prolly makes
+                        # the most send to do venue operating
+                        # hours lookup and
+                        # timestamp-in-operating-range set
+                        # checking to know for sure if we can
+                        # safely and quickly ignore non-uniform history
+                        # frame timestamp gaps..
+                        # end_dt -= dt_duration
+                        # continue
+                        # await tractor.pause()
 
-                    raise NoData(f'{end_dt}')
-
-                if bars_array is None:
-                    raise SymbolNotFound(fqme)
-
-                first_dt = pendulum.from_timestamp(
+                first_dt = from_timestamp(
                     bars[0].date.timestamp())
 
-                last_dt = pendulum.from_timestamp(
+                last_dt = from_timestamp(
                     bars[-1].date.timestamp())
 
                 time = bars_array['time']
@@ -437,15 +498,22 @@ async def get_bars(
                 if data_cs:
                     data_cs.cancel()
 
-                result = (bars, bars_array, first_dt, last_dt)
+                # NOTE: setting this is critical!
+                result = (
+                    bars,  # ib native
+                    bars_array,  # numpy
+                    first_dt,
+                    last_dt,
+                )
 
                 # signal data reset loop parent task
                 result_ready.set()
 
+                # NOTE: this isn't getting collected anywhere!
                 return result
 
             except RequestError as err:
-                msg = err.message
+                msg: str = err.message
 
                 if 'No market data permissions for' in msg:
                     # TODO: signalling for no permissions searches
@@ -467,7 +535,7 @@ async def get_bars(
                     if end_dt is not None:
                         end_dt = end_dt.subtract(days=1)
                     elif end_dt is None:
-                        end_dt = pendulum.now().subtract(days=1)
+                        end_dt = now().subtract(days=1)
 
                     log.warning(
                         f'NO DATA found ending @ {end_dt}\n'
@@ -483,21 +551,29 @@ async def get_bars(
                     nodatas_count += 1
                     continue
 
-                elif 'API historical data query cancelled' in err.message:
+                elif (
+                    'API historical data query cancelled'
+                    in
+                    err.message
+                ):
                     log.warning(
                         'Query cancelled by IB (:eyeroll:):\n'
                         f'{err.message}'
                     )
                     continue
+
                 elif (
                     'Trading TWS session is connected from a different IP'
-                        in err.message
+                    in
+                    err.message
                 ):
                     log.warning("ignoring ip address warning")
                     continue
 
                 # XXX: more or less same as above timeout case
-                elif _pacing in msg:
+                elif (
+                    _pacing in msg
+                ):
                     log.warning(
                         'History throttle rate reached!\n'
                         'Resetting farms with `ctrl-alt-f` hack\n'
@@ -549,9 +625,10 @@ async def get_bars(
                 # don't double invoke the reset hack if another
                 # requester task already has it covered.
                 continue
+
             else:
                 _data_resetter_task = trio.lowlevel.current_task()
-                unset_resetter = True
+                unset_resetter: bool = True
 
             # spawn new data reset task
             data_cs, reset_done = await nurse.start(
@@ -564,31 +641,16 @@ async def get_bars(
             # sync wait on reset to complete
             await reset_done.wait()
 
-    _data_resetter_task = None if unset_resetter else _data_resetter_task
+    _data_resetter_task = (
+        None
+        if unset_resetter
+        else _data_resetter_task
+    )
     assert result
-    return result, data_cs is not None
-
-
-# re-mapping to piker asset type names
-# https://github.com/erdewit/ib_insync/blob/master/ib_insync/contract.py#L113
-_asset_type_map = {
-    'STK': 'stock',
-    'OPT': 'option',
-    'FUT': 'future',
-    'CONTFUT': 'continuous_future',
-    'CASH': 'forex',
-    'IND': 'index',
-    'CFD': 'cfd',
-    'BOND': 'bond',
-    'CMDTY': 'commodity',
-    'FOP': 'futures_option',
-    'FUND': 'mutual_fund',
-    'WAR': 'warrant',
-    'IOPT': 'warran',
-    'BAG': 'bag',
-    'CRYPTO': 'crypto',  # bc it's diff then fiat?
-    # 'NEWS': 'news',
-}
+    return (
+        result,
+        data_cs is not None,
+    )
 
 
 _quote_streams: dict[str, trio.abc.ReceiveStream] = {}
@@ -609,8 +671,8 @@ async def _setup_quote_stream(
         # making them mostly useless and explains why the scanner
         # is always slow XD
         # '293',  # Trade count for day
-        '294',  # Trade rate / minute
-        '295',  # Vlm rate / minute
+        # '294',  # Trade rate / minute
+        # '295',  # Vlm rate / minute
     ),
     contract: Contract | None = None,
 
@@ -794,96 +856,6 @@ def normalize(
     return data
 
 
-@async_lifo_cache()
-async def get_mkt_info(
-    fqme: str,
-
-    proxy: MethodProxy | None = None,
-
-) -> tuple[MktPair, ibis.ContractDetails]:
-
-    # XXX: we don't need to split off any fqme broker part?
-    # bs_fqme, _, broker = fqme.partition('.')
-
-    proxy: MethodProxy
-    if proxy is not None:
-        client_ctx = nullcontext(proxy)
-    else:
-        client_ctx = open_data_client
-
-    async with client_ctx as proxy:
-        try:
-            (
-                con,  # Contract
-                details,  # ContractDetails
-            ) = await proxy.get_sym_details(symbol=fqme)
-        except ConnectionError:
-            log.exception(f'Proxy is ded {proxy._aio_ns}')
-            raise
-
-    # TODO: more consistent field translation
-    atype = _asset_type_map[con.secType]
-
-    if atype == 'commodity':
-        venue: str = 'cmdty'
-    else:
-        venue = con.primaryExchange or con.exchange
-
-    price_tick: Decimal = Decimal(str(details.minTick))
-
-    if atype == 'stock':
-        # XXX: GRRRR they don't support fractional share sizes for
-        # stocks from the API?!
-        # if con.secType == 'STK':
-        size_tick = Decimal('1')
-    else:
-        size_tick: Decimal = Decimal(str(details.minSize).rstrip('0'))
-        # |-> TODO: there is also the Contract.sizeIncrement, bt wtf is it?
-
-    # NOTE: this is duplicate from the .broker.norm_trade_records()
-    # routine, we should factor all this parsing somewhere..
-    expiry_str = str(con.lastTradeDateOrContractMonth)
-    # if expiry:
-    #     expiry_str: str = str(pendulum.parse(
-    #         str(expiry).strip(' ')
-    #     ))
-
-    # TODO: currently we can't pass the fiat src asset because
-    # then we'll get a `MNQUSD` request for history data..
-    # we need to figure out how we're going to handle this (later?)
-    # but likely we want all backends to eventually handle
-    # ``dst/src.venue.`` style !?
-    src: str | Asset = ''
-    if atype == 'forex':
-        src = Asset(
-            name=str(con.currency),
-            atype='fiat',
-            tx_tick=Decimal('0.01'),  # right?
-        )
-
-    mkt = MktPair(
-        dst=Asset(
-            name=con.symbol.lower(),
-            atype=atype,
-            tx_tick=size_tick,
-        ),
-        src=src,
-
-        price_tick=price_tick,
-        size_tick=size_tick,
-
-        bs_mktid=str(con.conId),
-        venue=str(venue),
-        expiry=expiry_str,
-        broker='ib',
-
-        # TODO: options contract info as str?
-        # contract_info=<optionsdetails>
-    )
-
-    return mkt, details
-
-
 async def stream_quotes(
 
     send_chan: trio.abc.SendChannel,
@@ -911,7 +883,10 @@ async def stream_quotes(
     proxy: MethodProxy
     mkt: MktPair
     details: ibis.ContractDetails
-    async with open_data_client() as proxy:
+    async with (
+        open_data_client() as proxy,
+        # trio.open_nursery() as tn,
+    ):
         mkt, details = await get_mkt_info(
             sym,
             proxy=proxy,  # passed to avoid implicit client load
@@ -920,7 +895,7 @@ async def stream_quotes(
         init_msg = FeedInit(mkt_info=mkt)
 
         if mkt.dst.atype in {
-            'forex',
+            'fiat',
             'index',
             'commodity',
         }:
@@ -931,27 +906,50 @@ async def stream_quotes(
         init_msgs.append(init_msg)
 
         con: Contract = details.contract
-        first_ticker: Ticker = await proxy.get_quote(contract=con)
-        first_quote: dict = normalize(first_ticker)
-
-        log.warning(f'FIRST QUOTE: {first_quote}')
-
-        # TODO: we should instead spawn a task that waits on a feed to start
-        # and let it wait indefinitely..instead of this hard coded stuff.
+        first_ticker: Ticker | None = None
         with trio.move_on_after(1):
-            first_ticker = await proxy.get_quote(contract=con)
+            first_ticker: Ticker = await proxy.get_quote(
+                contract=con,
+                raise_on_timeout=False,
+            )
 
-        # it might be outside regular trading hours so see if we can at
-        # least grab history.
+        if first_ticker:
+            first_quote: dict = normalize(first_ticker)
+
+            # TODO: we need a stack-oriented log levels filters for
+            # this!
+            # log.info(message, filter={'stack': 'live_feed'}) ?
+            log.runtime(
+                'Rxed init quote:\n\n'
+                f'{pformat(first_quote)}\n'
+            )
+
+        # NOTE: it might be outside regular trading hours for
+        # assets with "standard venue operating hours" so we
+        # only "pretend the feed is live" when the dst asset
+        # type is NOT within the NON-NORMAL-venue set: aka not
+        # commodities, forex or crypto currencies which CAN
+        # always return a NaN on a snap quote request during
+        # normal venue hours. In the case of a closed venue
+        # (equitiies, futes, bonds etc.) we at least try to
+        # grab the OHLC history.
         if (
-            isnan(first_ticker.last)  # last quote price value is nan
+            first_ticker
+            and
+            isnan(first_ticker.last)
+            # SO, if the last quote price value is NaN we ONLY
+            # "pretend to do" `feed_is_live.set()` if it's a known
+            # dst asset venue with a lot of closed operating hours.
             and mkt.dst.atype not in {
                 'commodity',
-                'forex',
+                'fiat',
                 'crypto',
             }
         ):
-            task_status.started((init_msgs, first_quote))
+            task_status.started((
+                init_msgs,
+                first_quote,
+            ))
 
             # it's not really live but this will unblock
             # the brokerd feed task to tell the ui to update?
@@ -961,6 +959,28 @@ async def stream_quotes(
             await trio.sleep_forever()
             return  # we never expect feed to come up?
 
+        # TODO: we should instead spawn a task that waits on a feed
+        # to start and let it wait indefinitely..instead of this
+        # hard coded stuff.
+        # async def wait_for_first_quote():
+        #     with trio.CancelScope() as cs:
+
+        # XXX: MUST acquire a ticker + first quote before starting
+        # the live quotes loop!
+        # with trio.move_on_after(1):
+        first_ticker = await proxy.get_quote(
+            contract=con,
+            raise_on_timeout=True,
+        )
+        first_quote: dict = normalize(first_ticker)
+
+        # TODO: we need a stack-oriented log levels filters for
+        # this!
+        # log.info(message, filter={'stack': 'live_feed'}) ?
+        log.runtime(
+            'Rxed init quote:\n'
+            f'{pformat(first_quote)}'
+        )
         cs: trio.CancelScope | None = None
         startup: bool = True
         while (
@@ -981,8 +1001,11 @@ async def stream_quotes(
 
                     # only on first entry at feed boot up
                     if startup:
-                        startup = False
-                        task_status.started((init_msgs, first_quote))
+                        startup: bool = False
+                        task_status.started((
+                            init_msgs,
+                            first_quote,
+                        ))
 
                     # start a stream restarter task which monitors the
                     # data feed event.
@@ -1006,7 +1029,7 @@ async def stream_quotes(
 
                             # generally speaking these feeds don't
                             # include vlm data.
-                            atype = mkt.dst.atype
+                            atype: str = mkt.dst.atype
                             log.info(
                                 f'No-vlm {mkt.fqme}@{atype}, skipping quote poll'
                             )
@@ -1042,7 +1065,8 @@ async def stream_quotes(
                             quote = normalize(ticker)
                             log.debug(f"First ticker received {quote}")
 
-                        # tell caller quotes are now coming in live
+                        # tell data-layer spawner-caller that live
+                        # quotes are now streaming.
                         feed_is_live.set()
 
                         # last = time.time()
@@ -1054,141 +1078,3 @@ async def stream_quotes(
                             # ugh, clear ticks since we've consumed them
                             ticker.ticks = []
                             # last = time.time()
-
-
-@tractor.context
-async def open_symbol_search(
-    ctx: tractor.Context,
-
-) -> None:
-
-    # TODO: load user defined symbol set locally for fast search?
-    await ctx.started({})
-
-    async with (
-        open_client_proxies() as (proxies, _),
-        open_data_client() as data_proxy,
-    ):
-        async with ctx.open_stream() as stream:
-
-            # select a non-history client for symbol search to lighten
-            # the load in the main data node.
-            proxy = data_proxy
-            for name, proxy in proxies.items():
-                if proxy is data_proxy:
-                    continue
-                break
-
-            ib_client = proxy._aio_ns.ib
-            log.info(f'Using {ib_client} for symbol search')
-
-            last = time.time()
-            async for pattern in stream:
-                log.info(f'received {pattern}')
-                now = time.time()
-
-                # this causes tractor hang...
-                # assert 0
-
-                assert pattern, 'IB can not accept blank search pattern'
-
-                # throttle search requests to no faster then 1Hz
-                diff = now - last
-                if diff < 1.0:
-                    log.debug('throttle sleeping')
-                    await trio.sleep(diff)
-                    try:
-                        pattern = stream.receive_nowait()
-                    except trio.WouldBlock:
-                        pass
-
-                if (
-                    not pattern
-                    or pattern.isspace()
-
-                    # XXX: not sure if this is a bad assumption but it
-                    # seems to make search snappier?
-                    or len(pattern) < 1
-                ):
-                    log.warning('empty pattern received, skipping..')
-
-                    # TODO: *BUG* if nothing is returned here the client
-                    # side will cache a null set result and not showing
-                    # anything to the use on re-searches when this query
-                    # timed out. We probably need a special "timeout" msg
-                    # or something...
-
-                    # XXX: this unblocks the far end search task which may
-                    # hold up a multi-search nursery block
-                    await stream.send({})
-
-                    continue
-
-                log.info(f'searching for {pattern}')
-
-                last = time.time()
-
-                # async batch search using api stocks endpoint and module
-                # defined adhoc symbol set.
-                stock_results = []
-
-                async def stash_results(target: Awaitable[list]):
-                    try:
-                        results = await target
-                    except tractor.trionics.Lagged:
-                        print("IB SYM-SEARCH OVERRUN?!?")
-                        return
-
-                    stock_results.extend(results)
-
-                for i in range(10):
-                    with trio.move_on_after(3) as cs:
-                        async with trio.open_nursery() as sn:
-                            sn.start_soon(
-                                stash_results,
-                                proxy.search_symbols(
-                                    pattern=pattern,
-                                    upto=5,
-                                ),
-                            )
-
-                            # trigger async request
-                            await trio.sleep(0)
-
-                    if cs.cancelled_caught:
-                        log.warning(
-                            f'Search timeout? {proxy._aio_ns.ib.client}'
-                        )
-                        continue
-                    else:
-                        break
-
-                    # # match against our ad-hoc set immediately
-                    # adhoc_matches = fuzzy.extractBests(
-                    #     pattern,
-                    #     list(_adhoc_futes_set),
-                    #     score_cutoff=90,
-                    # )
-                    # log.info(f'fuzzy matched adhocs: {adhoc_matches}')
-                    # adhoc_match_results = {}
-                    # if adhoc_matches:
-                    #     # TODO: do we need to pull contract details?
-                    #     adhoc_match_results = {i[0]: {} for i in
-                    #     adhoc_matches}
-
-                log.debug(f'fuzzy matching stocks {stock_results}')
-                stock_matches = fuzzy.extractBests(
-                    pattern,
-                    stock_results,
-                    score_cutoff=50,
-                )
-
-                # matches = adhoc_match_results | {
-                matches = {
-                    item[0]: {} for item in stock_matches
-                }
-                # TODO: we used to deliver contract details
-                # {item[2]: item[0] for item in stock_matches}
-
-                log.debug(f"sending matches: {matches.keys()}")
-                await stream.send(matches)

piker/brokers/ib/ledger.py

@@ -18,159 +18,382 @@
 Trade transaction accounting and normalization.
 
 '''
+from __future__ import annotations
 from bisect import insort
+from dataclasses import asdict
 from decimal import Decimal
+from functools import partial
 from pprint import pformat
 from typing import (
     Any,
+    Callable,
+    TYPE_CHECKING,
 )
 
 from bidict import bidict
-import pendulum
+from pendulum import (
+    DateTime,
+    parse,
+    from_timestamp,
+)
+from ib_insync import (
+    Contract,
+    Commodity,
+    Fill,
+    Execution,
+    CommissionReport,
+)
 
+from piker.types import Struct
+from piker.data import (
+    SymbologyCache,
+)
 from piker.accounting import (
+    Asset,
     dec_digits,
     digits_to_dec,
     Transaction,
     MktPair,
+    iter_by_dt,
 )
 from ._flex_reports import parse_flex_dt
 from ._util import log
 
+if TYPE_CHECKING:
+    from .api import (
+        Client,
+        MethodProxy,
+    )
 
-def norm_trade_records(
-    ledger: dict[str, Any],
 
-) -> dict[str, Transaction]:
-    '''
-    Normalize a flex report or API retrieved executions
-    ledger into our standard record format.
+tx_sort: Callable = partial(
+    iter_by_dt,
+    parsers={
+        'dateTime': parse_flex_dt,
+        'datetime': parse,
 
-    '''
-    records: list[Transaction] = []
+        # XXX: for some some fucking 2022 and
+        # back options records.. f@#$ me..
+        'date': parse,
+    }
+)
 
-    for tid, record in ledger.items():
-        conid = record.get('conId') or record['conid']
-        comms = record.get('commission')
-        if comms is None:
-            comms = -1*record['ibCommission']
 
-        price = record.get('price') or record['tradePrice']
+def norm_trade(
+    tid: str,
+    record: dict[str, Any],
 
-        # the api doesn't do the -/+ on the quantity for you but flex
-        # records do.. are you fucking serious ib...!?
-        size = record.get('quantity') or record['shares'] * {
-            'BOT': 1,
-            'SLD': -1,
-        }[record['side']]
+    # this is the dict that was returned from
+    # `Client.get_mkt_pairs()` and when running offline ledger
+    # processing from `.accounting`, this will be the table loaded
+    # into `SymbologyCache.pairs`.
+    pairs: dict[str, Struct],
+    symcache: SymbologyCache | None = None,
 
-        exch = record['exchange']
-        lexch = record.get('listingExchange')
+) -> Transaction | None:
 
-        # NOTE: remove null values since `tomlkit` can't serialize
-        # them to file.
-        dnc = record.pop('deltaNeutralContract', False)
-        if dnc is not None:
-            record['deltaNeutralContract'] = dnc
+    conid: int = str(record.get('conId') or record['conid'])
+    bs_mktid: str = str(conid)
 
-        suffix = lexch or exch
-        symbol = record['symbol']
+    # NOTE: sometimes weird records (like BTTX?)
+    # have no field for this?
+    comms: float = -1 * (
+        record.get('commission')
+        or record.get('ibCommission')
+        or 0
+    )
+    if not comms:
+        log.warning(
+            'No commissions found for record?\n'
+            f'{pformat(record)}\n'
+        )
 
-        # likely an opts contract record from a flex report..
-        # TODO: no idea how to parse ^ the strike part from flex..
-        # (00010000 any, or 00007500 tsla, ..)
-        # we probably must do the contract lookup for this?
-        if '   ' in symbol or '--' in exch:
-            underlying, _, tail = symbol.partition('   ')
-            suffix = exch = 'opt'
-            expiry = tail[:6]
-            # otype = tail[6]
-            # strike = tail[7:]
+    price: float = (
+        record.get('price')
+        or record.get('tradePrice')
+    )
+    if price is None:
+        log.warning(
+            'No `price` field found in record?\n'
+            'Skipping normalization..\n'
+            f'{pformat(record)}\n'
+        )
+        return None
 
-            print(f'skipping opts contract {symbol}')
-            continue
+    # the api doesn't do the -/+ on the quantity for you but flex
+    # records do.. are you fucking serious ib...!?
+    size: float|int = (
+        record.get('quantity')
+        or record['shares']
+    ) * {
+        'BOT': 1,
+        'SLD': -1,
+    }[record['side']]
 
-        # timestamping is way different in API records
-        dtstr = record.get('datetime')
-        date = record.get('date')
-        flex_dtstr = record.get('dateTime')
+    symbol: str = record['symbol']
+    exch: str = (
+        record.get('listingExchange')
+        or record.get('primaryExchange')
+        or record['exchange']
+    )
 
-        if dtstr or date:
-            dt = pendulum.parse(dtstr or date)
+    # NOTE: remove null values since `tomlkit` can't serialize
+    # them to file.
+    if dnc := record.pop('deltaNeutralContract', None):
+        record['deltaNeutralContract'] = dnc
 
-        elif flex_dtstr:
-            # probably a flex record with a wonky non-std timestamp..
-            dt = parse_flex_dt(record['dateTime'])
+    # likely an opts contract record from a flex report..
+    # TODO: no idea how to parse ^ the strike part from flex..
+    # (00010000 any, or 00007500 tsla, ..)
+    # we probably must do the contract lookup for this?
+    if (
+        '   ' in symbol
+        or '--' in exch
+    ):
+        underlying, _, tail = symbol.partition('   ')
+        exch: str = 'opt'
+        expiry: str = tail[:6]
+        # otype = tail[6]
+        # strike = tail[7:]
 
-        # special handling of symbol extraction from
-        # flex records using some ad-hoc schema parsing.
-        asset_type: str = record.get(
-            'assetCategory'
-        ) or record.get('secType', 'STK')
+        log.warning(
+            f'Skipping option contract -> NO SUPPORT YET!\n'
+            f'{symbol}\n'
+        )
+        return None
 
-        # TODO: XXX: WOA this is kinda hacky.. probably
-        # should figure out the correct future pair key more
-        # explicitly and consistently?
-        if asset_type == 'FUT':
-            # (flex) ledger entries don't have any simple 3-char key?
-            symbol = record['symbol'][:3]
-            asset_type: str = 'future'
+    # timestamping is way different in API records
+    dtstr: str = record.get('datetime')
+    date: str = record.get('date')
+    flex_dtstr: str = record.get('dateTime')
 
-        elif asset_type == 'STK':
-            asset_type: str = 'stock'
+    if dtstr or date:
+        dt: DateTime = parse(dtstr or date)
 
-        # try to build out piker fqme from record.
-        expiry = (
+    elif flex_dtstr:
+        # probably a flex record with a wonky non-std timestamp..
+        dt: DateTime = parse_flex_dt(record['dateTime'])
+
+    # special handling of symbol extraction from
+    # flex records using some ad-hoc schema parsing.
+    asset_type: str = (
+        record.get('assetCategory')
+        or record.get('secType')
+        or 'STK'
+    )
+
+    if (expiry := (
             record.get('lastTradeDateOrContractMonth')
             or record.get('expiry')
         )
+    ):
+        expiry: str = str(expiry).strip(' ')
+        # NOTE: we directly use the (simple and usually short)
+        # date-string expiry token when packing the `MktPair`
+        # since we want the fqme to contain *that* token.
+        # It might make sense later to instead parse and then
+        # render different output str format(s) for this same
+        # purpose depending on asset-type-market down the road.
+        # Eg. for derivs we use the short token only for fqme
+        # but use the isoformat('T') for transactions and
+        # account file position entries?
+        # dt_str: str = pendulum.parse(expiry).isoformat('T')
 
-        if expiry:
-            expiry = str(expiry).strip(' ')
-            suffix = f'{exch}.{expiry}'
-            expiry = pendulum.parse(expiry)
+    # XXX: pretty much all legacy market assets have a fiat
+    # currency (denomination) determined by their venue.
+    currency: str = record['currency']
+    src = Asset(
+        name=currency.lower(),
+        atype='fiat',
+        tx_tick=Decimal('0.01'),
+    )
 
-        # src: str = record['currency']
-        price_tick: Decimal = digits_to_dec(dec_digits(price))
+    match asset_type:
+        case 'FUT':
+            # XXX (flex) ledger entries don't necessarily have any
+            # simple 3-char key.. sometimes the .symbol is some
+            # weird internal key that we probably don't want in the
+            # .fqme => we should probably just wrap `Contract` to
+            # this like we do other crypto$ backends XD
 
-        pair = MktPair.from_fqme(
-            fqme=f'{symbol}.{suffix}.ib',
-            bs_mktid=str(conid),
-            _atype=str(asset_type),  # XXX: can't serlialize `tomlkit.String`
+            # NOTE: at least older FLEX records should have
+            # this field.. no idea about API entries..
+            local_symbol: str | None = record.get('localSymbol')
+            underlying_key: str = record.get('underlyingSymbol')
+            descr: str | None = record.get('description')
 
-            price_tick=price_tick,
-            # NOTE: for "legacy" assets, volume is normally discreet, not
-            # a float, but we keep a digit in case the suitz decide
-            # to get crazy and change it; we'll be kinda ready
-            # schema-wise..
-            size_tick='1',
+            if (
+                not (
+                    local_symbol
+                    and symbol in local_symbol
+                )
+                and (
+                        descr
+                        and symbol not in descr
+                    )
+            ):
+                con_key, exp_str = descr.split(' ')
+                symbol: str = underlying_key or con_key
+
+            dst = Asset(
+                name=symbol.lower(),
+                atype='future',
+                tx_tick=Decimal('1'),
+            )
+
+        case 'STK':
+            dst = Asset(
+                name=symbol.lower(),
+                atype='stock',
+                tx_tick=Decimal('1'),
+            )
+
+        case 'CASH':
+            if currency not in symbol:
+                # likely a dict-casted `Forex` contract which
+                # has .symbol as the dst and .currency as the
+                # src.
+                name: str = symbol.lower()
+            else:
+                # likely a flex-report record which puts
+                # EUR.USD as the symbol field and just USD in
+                # the currency field.
+                name: str = symbol.lower().replace(f'.{src.name}', '')
+
+            dst = Asset(
+                name=name,
+                atype='fiat',
+                tx_tick=Decimal('0.01'),
+            )
+
+        case 'OPT':
+            dst = Asset(
+                name=symbol.lower(),
+                atype='option',
+                tx_tick=Decimal('1'),
+
+                # TODO: we should probably always cast to the
+                # `Contract` instance then dict-serialize that for
+                # the `.info` field!
+                # info=asdict(Option()),
+            )
+
+        case 'CMDTY':
+            from .symbols import _adhoc_symbol_map
+            con_kwargs, _ = _adhoc_symbol_map[symbol.upper()]
+            dst = Asset(
+                name=symbol.lower(),
+                atype='commodity',
+                tx_tick=Decimal('1'),
+                info=asdict(Commodity(**con_kwargs)),
+            )
+
+    # try to build out piker fqme from record.
+    # src: str = record['currency']
+    price_tick: Decimal = digits_to_dec(dec_digits(price))
+
+    # NOTE: can't serlialize `tomlkit.String` so cast to native
+    atype: str = str(dst.atype)
+
+    # if not (mkt := symcache.mktmaps.get(bs_mktid)):
+    mkt = MktPair(
+        bs_mktid=bs_mktid,
+        dst=dst,
+
+        price_tick=price_tick,
+        # NOTE: for "legacy" assets, volume is normally discreet, not
+        # a float, but we keep a digit in case the suitz decide
+        # to get crazy and change it; we'll be kinda ready
+        # schema-wise..
+        size_tick=Decimal('1'),
+
+        src=src,  # XXX: normally always a fiat
+
+        _atype=atype,
+
+        venue=exch,
+        expiry=expiry,
+        broker='ib',
+
+        _fqme_without_src=(atype != 'fiat'),
+    )
+
+    fqme: str = mkt.fqme
+
+    # XXX: if passed in, we fill out the symcache ad-hoc in order
+    # to make downstream accounting work..
+    if symcache is not None:
+        orig_mkt: MktPair | None  = symcache.mktmaps.get(bs_mktid)
+        if (
+            orig_mkt
+            and orig_mkt.fqme != mkt.fqme
+        ):
+            log.warning(
+            # print(
+                f'Contracts with common `conId`: {bs_mktid} mismatch..\n'
+                f'{orig_mkt.fqme} -> {mkt.fqme}\n'
+                # 'with DIFF:\n'
+                # f'{mkt - orig_mkt}'
+            )
+
+        symcache.mktmaps[bs_mktid] = mkt
+        symcache.mktmaps[fqme] = mkt
+        symcache.assets[src.name] = src
+        symcache.assets[dst.name] = dst
+
+    # NOTE: for flex records the normal fields for defining an fqme
+    # sometimes won't be available so we rely on two approaches for
+    # the "reverse lookup" of piker style fqme keys:
+    # - when dealing with API trade records received from
+    #   `IB.trades()` we do a contract lookup at he time of processing
+    # - when dealing with flex records, it is assumed the record
+    #   is at least a day old and thus the TWS position reporting system
+    #   should already have entries if the pps are still open, in
+    #   which case, we can pull the fqme from that table (see
+    #   `trades_dialogue()` above).
+    return Transaction(
+        fqme=fqme,
+        tid=tid,
+        size=size,
+        price=price,
+        cost=comms,
+        dt=dt,
+        expiry=expiry,
+        bs_mktid=str(conid),
+    )
+
+
+
+def norm_trade_records(
+    ledger: dict[str, Any],
+    symcache: SymbologyCache | None = None,
+
+) -> dict[str, Transaction]:
+    '''
+    Normalize (xml) flex-report or (recent) API trade records into
+    our ledger format with parsing for `MktPair` and `Asset`
+    extraction to fill in the `Transaction.sys: MktPair` field.
+
+    '''
+    records: list[Transaction] = []
+    for tid, record in ledger.items():
+
+        txn = norm_trade(
+            tid,
+            record,
+
+            # NOTE: currently no symcache support
+            pairs={},
+            symcache=symcache,
         )
 
-        fqme = pair.fqme
+        if txn is None:
+            continue
 
-        # NOTE: for flex records the normal fields for defining an fqme
-        # sometimes won't be available so we rely on two approaches for
-        # the "reverse lookup" of piker style fqme keys:
-        # - when dealing with API trade records received from
-        #   `IB.trades()` we do a contract lookup at he time of processing
-        # - when dealing with flex records, it is assumed the record
-        #   is at least a day old and thus the TWS position reporting system
-        #   should already have entries if the pps are still open, in
-        #   which case, we can pull the fqme from that table (see
-        #   `trades_dialogue()` above).
+        # inject txns sorted by datetime
         insort(
             records,
-            Transaction(
-                fqme=fqme,
-                sym=pair,
-                tid=tid,
-                size=size,
-                price=price,
-                cost=comms,
-                dt=dt,
-                expiry=expiry,
-                bs_mktid=str(conid),
-            ),
+            txn,
             key=lambda t: t.dt
         )
 
@@ -179,50 +402,49 @@ def norm_trade_records(
 
 def api_trades_to_ledger_entries(
     accounts: bidict[str, str],
+    fills: list[Fill],
 
-    # TODO: maybe we should just be passing through the
-    # ``ib_insync.order.Trade`` instance directly here
-    # instead of pre-casting to dicts?
-    trade_entries: list[dict],
-
-) -> dict:
+) -> dict[str, dict]:
     '''
-    Convert API execution objects entry objects into ``dict`` form,
-    pretty much straight up without modification except add
-    a `pydatetime` field from the parsed timestamp.
+    Convert API execution objects entry objects into
+    flattened-``dict`` form, pretty much straight up without
+    modification except add a `pydatetime` field from the parsed
+    timestamp so that on write
 
     '''
-    trades_by_account = {}
-    for t in trade_entries:
-        # NOTE: example of schema we pull from the API client.
-        # {
-        #     'commissionReport': CommissionReport(...
-        #     'contract': {...
-        #     'execution': Execution(...
-        #     'time': 1654801166.0
-        # }
+    trades_by_account: dict[str, dict] = {}
+    for fill in fills:
 
-        # flatten all sub-dicts and values into one top level entry.
-        entry = {}
-        for section, val in t.items():
-            match section:
+        # NOTE: for the schema, see the defn for `Fill` which is
+        # a `NamedTuple` subtype
+        fdict: dict = fill._asdict()
+
+        # flatten all (sub-)objects and convert to dicts.
+        # with values packed into one top level entry.
+        val: CommissionReport | Execution | Contract
+        txn_dict: dict[str, Any] = {}
+        for attr_name, val in fdict.items():
+            match attr_name:
+                # value is a `@dataclass` subtype
                 case 'contract' | 'execution' | 'commissionReport':
-                    # sub-dict cases
-                    entry.update(val)
+                    txn_dict.update(asdict(val))
 
                 case 'time':
                     # ib has wack ns timestamps, or is that us?
                     continue
 
+                # TODO: we can remove this case right since there's
+                # only 4 fields on a `Fill`?
                 case _:
-                    entry[section] = val
+                    txn_dict[attr_name] = val
 
-        tid = str(entry['execId'])
-        dt = pendulum.from_timestamp(entry['time'])
-        # TODO: why isn't this showing seconds in the str?
-        entry['pydatetime'] = dt
-        entry['datetime'] = str(dt)
-        acctid = accounts[entry['acctNumber']]
+        tid = str(txn_dict['execId'])
+        dt = from_timestamp(txn_dict['time'])
+        txn_dict['datetime'] = str(dt)
+        acctid = accounts[txn_dict['acctNumber']]
+
+        # NOTE: only inserted (then later popped) for sorting below!
+        txn_dict['pydatetime'] = dt
 
         if not tid:
             # this is likely some kind of internal adjustment
@@ -233,13 +455,18 @@ def api_trades_to_ledger_entries(
             #   the user from the accounts window in TWS where they can
             #   manually set the avg price and size:
             #   https://api.ibkr.com/lib/cstools/faq/web1/index.html#/tag/DTWS_ADJ_AVG_COST
-            log.warning(f'Skipping ID-less ledger entry:\n{pformat(entry)}')
+            log.warning(
+                'Skipping ID-less ledger txn_dict:\n'
+                f'{pformat(txn_dict)}'
+            )
             continue
 
         trades_by_account.setdefault(
             acctid, {}
-        )[tid] = entry
+        )[tid] = txn_dict
 
+    # TODO: maybe we should just bisect.insort() into a list of
+    # tuples and then return a dict of that?
     # sort entries in output by python based datetime
     for acctid in trades_by_account:
         trades_by_account[acctid] = dict(sorted(
@@ -248,3 +475,55 @@ def api_trades_to_ledger_entries(
         ))
 
     return trades_by_account
+
+
+async def update_ledger_from_api_trades(
+    fills: list[Fill],
+    client: Client | MethodProxy,
+    accounts_def_inv: bidict[str, str],
+
+    # NOTE: provided for ad-hoc insertions "as transactions are
+    # processed" -> see `norm_trade()` signature requirements.
+    symcache: SymbologyCache | None = None,
+
+) -> tuple[
+    dict[str, Transaction],
+    dict[str, dict],
+]:
+    # XXX; ERRGGG..
+    # pack in the "primary/listing exchange" value from a
+    # contract lookup since it seems this isn't available by
+    # default from the `.fills()` method endpoint...
+    fill: Fill
+    for fill in fills:
+        con: Contract = fill.contract
+        conid: str = con.conId
+        pexch: str | None = con.primaryExchange
+
+        if not pexch:
+            cons = await client.get_con(conid=conid)
+            if cons:
+                con = cons[0]
+                pexch = con.primaryExchange or con.exchange
+            else:
+                # for futes it seems like the primary is always empty?
+                pexch: str = con.exchange
+
+        # pack in the ``Contract.secType``
+        # entry['asset_type'] = condict['secType']
+
+    entries: dict[str, dict] = api_trades_to_ledger_entries(
+        accounts_def_inv,
+        fills,
+    )
+    # normalize recent session's trades to the `Transaction` type
+    trans_by_acct: dict[str, dict[str, Transaction]] = {}
+
+    for acctid, trades_by_id in entries.items():
+        # normalize to transaction form
+        trans_by_acct[acctid] = norm_trade_records(
+            trades_by_id,
+            symcache=symcache,
+        )
+
+    return trans_by_acct, entries

piker/brokers/ib/symbols.py

@@ -0,0 +1,615 @@
+# piker: trading gear for hackers
+# Copyright (C) Tyler Goodlet (in stewardship for pikers)
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+'''
+Symbology search and normalization.
+
+'''
+from __future__ import annotations
+from contextlib import (
+    nullcontext,
+)
+from decimal import Decimal
+import time
+from typing import (
+    Awaitable,
+    TYPE_CHECKING,
+)
+
+from rapidfuzz import process as fuzzy
+import ib_insync as ibis
+import tractor
+import trio
+
+from piker.accounting import (
+    Asset,
+    MktPair,
+    unpack_fqme,
+)
+from piker._cacheables import (
+    async_lifo_cache,
+)
+
+from ._util import (
+    log,
+)
+
+if TYPE_CHECKING:
+    from .api import (
+        MethodProxy,
+        Client,
+    )
+
+_futes_venues = (
+    'GLOBEX',
+    'NYMEX',
+    'CME',
+    'CMECRYPTO',
+    'COMEX',
+    # 'CMDTY',  # special name case..
+    'CBOT',  # (treasury) yield futures
+)
+
+_adhoc_cmdty_set = {
+    # metals
+    # https://misc.interactivebrokers.com/cstools/contract_info/v3.10/index.php?action=Conid%20Info&wlId=IB&conid=69067924
+    'xauusd.cmdty',  # london gold spot ^
+    'xagusd.cmdty',  # silver spot
+}
+
+# NOTE: if you aren't seeing one of these symbol's futues contracts
+# show up, it's likely the `.<venue>` part is wrong!
+_adhoc_futes_set = {
+
+    # equities
+    'nq.cme',
+    'mnq.cme',  # micro
+
+    'es.cme',
+    'mes.cme',  # micro
+
+    # cypto$
+    'brr.cme',
+    'mbt.cme',  # micro
+    'ethusdrr.cme',
+
+    # agriculture
+    'he.comex',  # lean hogs
+    'le.comex',  # live cattle (geezers)
+    'gf.comex',  # feeder cattle (younguns)
+
+    # raw
+    'lb.comex',  # random len lumber
+
+    'gc.comex',
+    'mgc.comex',  # micro
+
+    # oil & gas
+    'cl.nymex',
+
+    'ni.comex',  # silver futes
+    'qi.comex',  # mini-silver futes
+
+    # treasury yields
+    # etfs by duration:
+    # SHY -> IEI -> IEF -> TLT
+    'zt.cbot',  # 2y
+    'z3n.cbot',  # 3y
+    'zf.cbot',  # 5y
+    'zn.cbot',  # 10y
+    'zb.cbot',  # 30y
+
+    # (micros of above)
+    '2yy.cbot',
+    '5yy.cbot',
+    '10y.cbot',
+    '30y.cbot',
+}
+
+
+# taken from list here:
+# https://www.interactivebrokers.com/en/trading/products-spot-currencies.php
+_adhoc_fiat_set = set((
+    'USD, AED, AUD, CAD,'
+    'CHF, CNH, CZK, DKK,'
+    'EUR, GBP, HKD, HUF,'
+    'ILS, JPY, MXN, NOK,'
+    'NZD, PLN, RUB, SAR,'
+    'SEK, SGD, TRY, ZAR'
+    ).split(' ,')
+)
+
+# manually discovered tick discrepancies,
+# onl god knows how or why they'd cuck these up..
+_adhoc_mkt_infos: dict[int | str, dict] = {
+    'vtgn.nasdaq': {'price_tick': Decimal('0.01')},
+}
+
+
+# map of symbols to contract ids
+_adhoc_symbol_map = {
+    # https://misc.interactivebrokers.com/cstools/contract_info/v3.10/index.php?action=Conid%20Info&wlId=IB&conid=69067924
+
+    # NOTE: some cmdtys/metals don't have trade data like gold/usd:
+    # https://groups.io/g/twsapi/message/44174
+    'XAUUSD': ({'conId': 69067924}, {'whatToShow': 'MIDPOINT'}),
+}
+for qsn in _adhoc_futes_set:
+    sym, venue = qsn.split('.')
+    assert venue.upper() in _futes_venues, f'{venue}'
+    _adhoc_symbol_map[sym.upper()] = (
+        {'exchange': venue},
+        {},
+    )
+
+
+# exchanges we don't support at the moment due to not knowing
+# how to do symbol-contract lookup correctly likely due
+# to not having the data feeds subscribed.
+_exch_skip_list = {
+
+    'ASX',  # aussie stocks
+    'MEXI',  # mexican stocks
+
+    # no idea
+    'NSE',
+    'VALUE',
+    'FUNDSERV',
+    'SWB2',
+    'PSE',
+    'PHLX',
+}
+
+# optional search config the backend can register for
+# it's symbol search handling (in this case we avoid
+# accepting patterns before the kb has settled more then
+# a quarter second).
+_search_conf = {
+    'pause_period': 6 / 16,
+}
+
+
+@tractor.context
+async def open_symbol_search(ctx: tractor.Context) -> None:
+    '''
+    Symbology search brokerd-endpoint.
+
+    '''
+    from .api import open_client_proxies
+    from .feed import open_data_client
+
+    # TODO: load user defined symbol set locally for fast search?
+    await ctx.started({})
+
+    async with (
+        open_client_proxies() as (proxies, _),
+        open_data_client() as data_proxy,
+    ):
+        async with ctx.open_stream() as stream:
+
+            # select a non-history client for symbol search to lighten
+            # the load in the main data node.
+            proxy = data_proxy
+            for name, proxy in proxies.items():
+                if proxy is data_proxy:
+                    continue
+                break
+
+            ib_client = proxy._aio_ns.ib
+            log.info(
+                f'Using API client for symbol-search\n'
+                f'{ib_client}\n'
+            )
+
+            last = time.time()
+            async for pattern in stream:
+                log.info(f'received {pattern}')
+                now: float = time.time()
+
+                # this causes tractor hang...
+                # assert 0
+
+                assert pattern, 'IB can not accept blank search pattern'
+
+                # throttle search requests to no faster then 1Hz
+                diff = now - last
+                if diff < 1.0:
+                    log.debug('throttle sleeping')
+                    await trio.sleep(diff)
+                    try:
+                        pattern = stream.receive_nowait()
+                    except trio.WouldBlock:
+                        pass
+
+                if (
+                    not pattern
+                    or pattern.isspace()
+
+                    # XXX: not sure if this is a bad assumption but it
+                    # seems to make search snappier?
+                    or len(pattern) < 1
+                ):
+                    log.warning('empty pattern received, skipping..')
+
+                    # TODO: *BUG* if nothing is returned here the client
+                    # side will cache a null set result and not showing
+                    # anything to the use on re-searches when this query
+                    # timed out. We probably need a special "timeout" msg
+                    # or something...
+
+                    # XXX: this unblocks the far end search task which may
+                    # hold up a multi-search nursery block
+                    await stream.send({})
+
+                    continue
+
+                log.info(f'searching for {pattern}')
+
+                last = time.time()
+
+                # async batch search using api stocks endpoint and module
+                # defined adhoc symbol set.
+                stock_results = []
+
+                async def extend_results(
+                    target: Awaitable[list]
+                ) -> None:
+                    try:
+                        results = await target
+                    except tractor.trionics.Lagged:
+                        print("IB SYM-SEARCH OVERRUN?!?")
+                        return
+
+                    stock_results.extend(results)
+
+                for _ in range(10):
+                    with trio.move_on_after(3) as cs:
+                        async with trio.open_nursery() as sn:
+                            sn.start_soon(
+                                extend_results,
+                                proxy.search_symbols(
+                                    pattern=pattern,
+                                    upto=5,
+                                ),
+                            )
+
+                            # trigger async request
+                            await trio.sleep(0)
+
+                    if cs.cancelled_caught:
+                        log.warning(
+                            f'Search timeout? {proxy._aio_ns.ib.client}'
+                        )
+                        continue
+                    elif stock_results:
+                        break
+                    # else:
+                    # await tractor.pause()
+
+                    # # match against our ad-hoc set immediately
+                    # adhoc_matches = fuzzy.extract(
+                    #     pattern,
+                    #     list(_adhoc_futes_set),
+                    #     score_cutoff=90,
+                    # )
+                    # log.info(f'fuzzy matched adhocs: {adhoc_matches}')
+                    # adhoc_match_results = {}
+                    # if adhoc_matches:
+                    #     # TODO: do we need to pull contract details?
+                    #     adhoc_match_results = {i[0]: {} for i in
+                    #     adhoc_matches}
+
+                log.debug(f'fuzzy matching stocks {stock_results}')
+                stock_matches = fuzzy.extract(
+                    pattern,
+                    stock_results,
+                    score_cutoff=50,
+                )
+
+                # matches = adhoc_match_results | {
+                matches = {
+                    item[0]: {} for item in stock_matches
+                }
+                # TODO: we used to deliver contract details
+                # {item[2]: item[0] for item in stock_matches}
+
+                log.debug(f"sending matches: {matches.keys()}")
+                await stream.send(matches)
+
+
+# re-mapping to piker asset type names
+# https://github.com/erdewit/ib_insync/blob/master/ib_insync/contract.py#L113
+_asset_type_map = {
+    'STK': 'stock',
+    'OPT': 'option',
+    'FUT': 'future',
+    'CONTFUT': 'continuous_future',
+    'CASH': 'fiat',
+    'IND': 'index',
+    'CFD': 'cfd',
+    'BOND': 'bond',
+    'CMDTY': 'commodity',
+    'FOP': 'futures_option',
+    'FUND': 'mutual_fund',
+    'WAR': 'warrant',
+    'IOPT': 'warran',
+    'BAG': 'bag',
+    'CRYPTO': 'crypto',  # bc it's diff then fiat?
+    # 'NEWS': 'news',
+}
+
+
+def parse_patt2fqme(
+    # client: Client,
+    pattern: str,
+
+) -> tuple[str, str, str, str]:
+
+    # TODO: we can't use this currently because
+    # ``wrapper.starTicker()`` currently cashes ticker instances
+    # which means getting a singel quote will potentially look up
+    # a quote for a ticker that it already streaming and thus run
+    # into state clobbering (eg. list: Ticker.ticks). It probably
+    # makes sense to try this once we get the pub-sub working on
+    # individual symbols...
+
+    # XXX UPDATE: we can probably do the tick/trades scraping
+    # inside our eventkit handler instead to bypass this entirely?
+
+    currency = ''
+
+    # fqme parsing stage
+    # ------------------
+    if '.ib' in pattern:
+        _, symbol, venue, expiry = unpack_fqme(pattern)
+
+    else:
+        symbol = pattern
+        expiry = ''
+
+        # # another hack for forex pairs lul.
+        # if (
+        #     '.idealpro' in symbol
+        #     # or '/' in symbol
+        # ):
+        #     exch: str = 'IDEALPRO'
+        #     symbol = symbol.removesuffix('.idealpro')
+        #     if '/' in symbol:
+        #         symbol, currency = symbol.split('/')
+
+        # else:
+        # TODO: yes, a cache..
+        # try:
+        #     # give the cache a go
+        #     return client._contracts[symbol]
+        # except KeyError:
+        #     log.debug(f'Looking up contract for {symbol}')
+        expiry: str = ''
+        if symbol.count('.') > 1:
+            symbol, _, expiry = symbol.rpartition('.')
+
+        # use heuristics to figure out contract "type"
+        symbol, venue = symbol.upper().rsplit('.', maxsplit=1)
+
+    return symbol, currency, venue, expiry
+
+
+def con2fqme(
+    con: ibis.Contract,
+    _cache: dict[int, (str, bool)] = {}
+
+) -> tuple[str, bool]:
+    '''
+    Convert contracts to fqme-style strings to be used both in
+    symbol-search matching and as feed tokens passed to the front
+    end data deed layer.
+
+    Previously seen contracts are cached by id.
+
+    '''
+    # should be real volume for this contract by default
+    calc_price: bool = False
+    if con.conId:
+        try:
+            # TODO: LOL so apparently IB just changes the contract
+            # ID (int) on a whim.. so we probably need to use an
+            # FQME style key after all...
+            return _cache[con.conId]
+        except KeyError:
+            pass
+
+    suffix: str = con.primaryExchange or con.exchange
+    symbol: str = con.symbol
+    expiry: str = con.lastTradeDateOrContractMonth or ''
+
+    match con:
+        case ibis.Option():
+            # TODO: option symbol parsing and sane display:
+            symbol = con.localSymbol.replace(' ', '')
+
+        case (
+            ibis.Commodity()
+            # search API endpoint returns std con box..
+            | ibis.Contract(secType='CMDTY')
+        ):
+            # commodities and forex don't have an exchange name and
+            # no real volume so we have to calculate the price
+            suffix = con.secType
+
+            # no real volume on this tract
+            calc_price = True
+
+        case ibis.Forex() | ibis.Contract(secType='CASH'):
+            dst, src = con.localSymbol.split('.')
+            symbol = ''.join([dst, src])
+            suffix = con.exchange or 'idealpro'
+
+            # no real volume on forex feeds..
+            calc_price = True
+
+    if not suffix:
+        entry = _adhoc_symbol_map.get(
+            con.symbol or con.localSymbol
+        )
+        if entry:
+            meta, kwargs = entry
+            cid = meta.get('conId')
+            if cid:
+                assert con.conId == meta['conId']
+            suffix = meta['exchange']
+
+    # append a `.<suffix>` to the returned symbol
+    # key for derivatives that normally is the expiry
+    # date key.
+    if expiry:
+        suffix += f'.{expiry}'
+
+    fqme_key = symbol.lower()
+    if suffix:
+        fqme_key = '.'.join((fqme_key, suffix)).lower()
+
+    _cache[con.conId] = fqme_key, calc_price
+    return fqme_key, calc_price
+
+
+@async_lifo_cache()
+async def get_mkt_info(
+    fqme: str,
+
+    proxy: MethodProxy | None = None,
+
+) -> tuple[MktPair, ibis.ContractDetails]:
+
+    if '.ib' not in fqme:
+        fqme += '.ib'
+    broker, pair, venue, expiry = unpack_fqme(fqme)
+
+    proxy: MethodProxy
+    if proxy is not None:
+        client_ctx = nullcontext(proxy)
+    else:
+        from .feed import (
+            open_data_client,
+        )
+        client_ctx = open_data_client
+
+    async with client_ctx as proxy:
+        try:
+            (
+                con,  # Contract
+                details,  # ContractDetails
+            ) = await proxy.get_sym_details(fqme=fqme)
+        except ConnectionError:
+            log.exception(f'Proxy is ded {proxy._aio_ns}')
+            raise
+
+    # TODO: more consistent field translation
+    atype = _asset_type_map[con.secType]
+
+    if atype == 'commodity':
+        venue: str = 'cmdty'
+    else:
+        venue = con.primaryExchange or con.exchange
+
+    price_tick: Decimal = Decimal(str(details.minTick))
+    ib_min_tick_gt_2: Decimal = Decimal('0.01')
+    if (
+        price_tick < ib_min_tick_gt_2
+    ):
+        # TODO: we need to add some kinda dynamic rounding sys
+        # to our MktPair i guess?
+        # not sure where the logic should sit, but likely inside
+        # the `.clearing._ems` i suppose...
+        log.warning(
+            'IB seems to disallow a min price tick < 0.01 '
+            'when the price is > 2.0..?\n'
+            f'Decreasing min tick precision for {fqme} to 0.01'
+        )
+        # price_tick = ib_min_tick
+        # await tractor.pause()
+
+    if atype == 'stock':
+        # XXX: GRRRR they don't support fractional share sizes for
+        # stocks from the API?!
+        # if con.secType == 'STK':
+        size_tick = Decimal('1')
+    else:
+        size_tick: Decimal = Decimal(
+            str(details.minSize).rstrip('0')
+        )
+        # |-> TODO: there is also the Contract.sizeIncrement, bt wtf is it?
+
+    # NOTE: this is duplicate from the .broker.norm_trade_records()
+    # routine, we should factor all this parsing somewhere..
+    expiry_str = str(con.lastTradeDateOrContractMonth)
+    # if expiry:
+    #     expiry_str: str = str(pendulum.parse(
+    #         str(expiry).strip(' ')
+    #     ))
+
+    # TODO: currently we can't pass the fiat src asset because
+    # then we'll get a `MNQUSD` request for history data..
+    # we need to figure out how we're going to handle this (later?)
+    # but likely we want all backends to eventually handle
+    # ``dst/src.venue.`` style !?
+    src = Asset(
+        name=str(con.currency).lower(),
+        atype='fiat',
+        tx_tick=Decimal('0.01'),  # right?
+    )
+    dst = Asset(
+        name=con.symbol.lower(),
+        atype=atype,
+        tx_tick=size_tick,
+    )
+
+    mkt = MktPair(
+        src=src,
+        dst=dst,
+
+        price_tick=price_tick,
+        size_tick=size_tick,
+
+        bs_mktid=str(con.conId),
+        venue=str(venue),
+        expiry=expiry_str,
+        broker='ib',
+
+        # TODO: options contract info as str?
+        # contract_info=<optionsdetails>
+        _fqme_without_src=(atype != 'fiat'),
+    )
+
+    # just.. wow.
+    if entry := _adhoc_mkt_infos.get(mkt.bs_fqme):
+        log.warning(f'Frickin {mkt.fqme} has an adhoc {entry}..')
+        new = mkt.to_dict()
+        new['price_tick'] = entry['price_tick']
+        new['src'] = src
+        new['dst'] = dst
+        mkt = MktPair(**new)
+
+    # if possible register the bs_mktid to the just-built
+    # mkt so that it can be retreived by order mode tasks later.
+    # TODO NOTE: this is going to be problematic if/when we split
+    # out the datatd vs. brokerd actors since the mktmap lookup
+    # table will now be inaccessible..
+    if proxy is not None:
+        client: Client = proxy._aio_ns
+        client._contracts[mkt.bs_fqme] = con
+        client._cons2mkts[con] = mkt
+
+    return mkt, details

piker/brokers/kraken/__init__.py

@@ -19,23 +19,36 @@ Kraken backend.
 
 Sub-modules within break into the core functionalities:
 
-- ``broker.py`` part for orders / trading endpoints
-- ``feed.py`` for real-time data feed endpoints
-- ``api.py`` for the core API machinery which is ``trio``-ized
-  wrapping around ``ib_insync``.
+- .api: for the core API machinery which generally
+        a ``asks``/``trio-websocket`` implemented ``Client``.
+- .broker: part for orders / trading endpoints.
+- .feed: for real-time and historical data query endpoints.
+- .ledger: for transaction processing as it pertains to accounting.
+- .symbols: for market (name) search and symbology meta-defs.
 
 '''
+from .symbols import (
+    Pair,  # for symcache
+    open_symbol_search,
+    # required by `.accounting`, `.data`
+    get_mkt_info,
+)
+# required by `.brokers`
 from .api import (
     get_client,
 )
 from .feed import (
-    get_mkt_info,
-    open_history_client,
-    open_symbol_search,
+    # required by `.data`
     stream_quotes,
+    open_history_client,
 )
 from .broker import (
+    # required by `.clearing`
     open_trade_dialog,
+)
+from .ledger import (
+    # required by `.accounting`
+    norm_trade,
     norm_trade_records,
 )
 
@@ -43,17 +56,20 @@ from .broker import (
 __all__ = [
     'get_client',
     'get_mkt_info',
+    'Pair',
     'open_trade_dialog',
     'open_history_client',
     'open_symbol_search',
     'stream_quotes',
     'norm_trade_records',
+    'norm_trade',
 ]
 
 
 # tractor RPC enable arg
 __enable_modules__: list[str] = [
     'api',
-    'feed',
     'broker',
+    'feed',
+    'symbols',
 ]

piker/brokers/kraken/api.py

@@ -15,12 +15,11 @@
 # along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
 '''
-Kraken web API wrapping.
+Core (web) API client
 
 '''
 from contextlib import asynccontextmanager as acm
 from datetime import datetime
-from decimal import Decimal
 import itertools
 from typing import (
     Any,
@@ -28,10 +27,8 @@ from typing import (
 )
 import time
 
-from bidict import bidict
+import httpx
 import pendulum
-import asks
-from fuzzywuzzy import process as fuzzy
 import numpy as np
 import urllib.parse
 import hashlib
@@ -40,11 +37,14 @@ import base64
 import trio
 
 from piker import config
-from piker.data.types import Struct
-from piker.data import def_iohlcv_fields
+from piker.data import (
+    def_iohlcv_fields,
+    match_from_pairs,
+)
 from piker.accounting._mktinfo import (
     Asset,
     digits_to_dec,
+    dec_digits,
 )
 from piker.brokers._util import (
     resproc,
@@ -54,11 +54,17 @@ from piker.brokers._util import (
 )
 from piker.accounting import Transaction
 from piker.log import get_logger
+from .symbols import Pair
 
 log = get_logger('piker.brokers.kraken')
 
 # <uri>/<version>/
 _url = 'https://api.kraken.com/0'
+
+_headers: dict[str, str] = {
+    'User-Agent': 'krakenex/2.1.0 (+https://github.com/veox/python3-krakenex)'
+}
+
 # TODO: this is the only backend providing this right?
 # in which case we should drop it from the defaults and
 # instead make a custom fields descr in this module!
@@ -69,12 +75,18 @@ _symbol_info_translation: dict[str, str] = {
 
 
 def get_config() -> dict[str, Any]:
+    '''
+    Load our section from `piker/brokers.toml`.
 
-    conf, path = config.load()
-    section = conf.get('kraken')
-
-    if section is None:
-        log.warning(f'No config section found for kraken in {path}')
+    '''
+    conf, path = config.load(
+        conf_name='brokers',
+        touch_if_dne=True,
+    )
+    if (section := conf.get('kraken')) is None:
+        log.warning(
+            f'No config section found for kraken in {path}'
+        )
         return {}
 
     return section
@@ -105,96 +117,51 @@ class InvalidKey(ValueError):
     '''
 
 
-# https://www.kraken.com/features/api#get-tradable-pairs
-class Pair(Struct):
-    altname: str  # alternate pair name
-    wsname: str  # WebSocket pair name (if available)
-    aclass_base: str  # asset class of base component
-    base: str  # asset id of base component
-    aclass_quote: str  # asset class of quote component
-    quote: str  # asset id of quote component
-    lot: str  # volume lot size
-
-    cost_decimals: int
-    costmin: float
-    pair_decimals: int  # scaling decimal places for pair
-    lot_decimals: int  # scaling decimal places for volume
-
-    # amount to multiply lot volume by to get currency volume
-    lot_multiplier: float
-
-    # array of leverage amounts available when buying
-    leverage_buy: list[int]
-    # array of leverage amounts available when selling
-    leverage_sell: list[int]
-
-    # fee schedule array in [volume, percent fee] tuples
-    fees: list[tuple[int, float]]
-
-    # maker fee schedule array in [volume, percent fee] tuples (if on
-    # maker/taker)
-    fees_maker: list[tuple[int, float]]
-
-    fee_volume_currency: str  # volume discount currency
-    margin_call: str  # margin call level
-    margin_stop: str  # stop-out/liquidation margin level
-    ordermin: float  # minimum order volume for pair
-    tick_size: float  # min price step size
-    status: str
-
-    short_position_limit: float = 0
-    long_position_limit: float = float('inf')
-
-    @property
-    def price_tick(self) -> Decimal:
-        return digits_to_dec(self.pair_decimals)
-
-    @property
-    def size_tick(self) -> Decimal:
-        return digits_to_dec(self.lot_decimals)
-
-    @property
-    def bs_fqme(self) -> str:
-        return f'{self.symbol}.SPOT'
-
-
 class Client:
 
-    # symbol mapping from all names to the altname
-    _ntable: dict[str, str] = {}
+    # assets and mkt pairs are key-ed by kraken's ReST response
+    # symbol-bs_mktids (we call them "X-keys" like fricking
+    # "XXMRZEUR"). these keys used directly since ledger endpoints
+    # return transaction sets keyed with the same set!
+    _Assets: dict[str, Asset] = {}
+    _AssetPairs: dict[str, Pair] = {}
 
-    # 2-way map of symbol names to their "alt names" ffs XD
-    _altnames: bidict[str, str] = bidict()
+    # offer lookup tables for all .altname and .wsname
+    # to the equivalent .xname so that various symbol-schemas
+    # can be mapped to `Pair`s in the tables above.
+    _altnames: dict[str, str] = {}
+    _wsnames: dict[str, str] = {}
 
+    # key-ed by `Pair.bs_fqme: str`, and thus used for search
+    # allowing for lookup using piker's own FQME symbology sys.
     _pairs: dict[str, Pair] = {}
+    _assets: dict[str, Asset] = {}
 
     def __init__(
         self,
         config: dict[str, str],
+        httpx_client: httpx.AsyncClient,
+
         name: str = '',
         api_key: str = '',
         secret: str = ''
     ) -> None:
-        self._sesh = asks.Session(connections=4)
-        self._sesh.base_location = _url
-        self._sesh.headers.update({
-            'User-Agent':
-                'krakenex/2.1.0 (+https://github.com/veox/python3-krakenex)'
-        })
+
+        self._sesh: httpx.AsyncClient = httpx_client
+
         self._name = name
         self._api_key = api_key
         self._secret = secret
 
         self.conf: dict[str, str] = config
-        self.assets: dict[str, Asset] = {}
 
     @property
     def pairs(self) -> dict[str, Pair]:
+
         if self._pairs is None:
             raise RuntimeError(
-                "Make sure to run `cache_symbols()` on startup!"
+                "Client didn't run `.get_mkt_pairs()` on startup?!"
             )
-            # retreive and cache all symbols
 
         return self._pairs
 
@@ -203,10 +170,9 @@ class Client:
         method: str,
         data: dict,
     ) -> dict[str, Any]:
-        resp = await self._sesh.post(
-            path=f'/public/{method}',
+        resp: httpx.Response = await self._sesh.post(
+            url=f'/public/{method}',
             json=data,
-            timeout=float('inf')
         )
         return resproc(resp, log)
 
@@ -217,18 +183,18 @@ class Client:
         uri_path: str
     ) -> dict[str, Any]:
         headers = {
-            'Content-Type':
-                'application/x-www-form-urlencoded',
-            'API-Key':
-                self._api_key,
-            'API-Sign':
-                get_kraken_signature(uri_path, data, self._secret)
+            'Content-Type': 'application/x-www-form-urlencoded',
+            'API-Key': self._api_key,
+            'API-Sign': get_kraken_signature(
+                uri_path,
+                data,
+                self._secret,
+            ),
         }
-        resp = await self._sesh.post(
-            path=f'/private/{method}',
+        resp: httpx.Response = await self._sesh.post(
+            url=f'/private/{method}',
             data=data,
             headers=headers,
-            timeout=float('inf')
         )
         return resproc(resp, log)
 
@@ -254,17 +220,29 @@ class Client:
             'Balance',
             {},
         )
-        by_bsmktid = resp['result']
+        by_bsmktid: dict[str, dict] = resp['result']
 
-        # TODO: we need to pull out the "asset" decimals
-        # data and return a `decimal.Decimal` instead here!
-        # using the underlying Asset
-        return {
-            self._altnames[sym].lower(): float(bal)
-            for sym, bal in by_bsmktid.items()
-        }
+        balances: dict = {}
+        for xname, bal in by_bsmktid.items():
+            asset: Asset = self._Assets[xname]
 
-    async def get_assets(self) -> dict[str, Asset]:
+            # TODO: which KEY should we use? it's used to index
+            # the `Account.pps: dict` ..
+            key: str = asset.name.lower()
+            # TODO: should we just return a `Decimal` here
+            # or is the rounded version ok?
+            balances[key] = round(
+                float(bal),
+                ndigits=dec_digits(asset.tx_tick)
+            )
+
+        return balances
+
+    async def get_assets(
+        self,
+        reload: bool = False,
+
+    ) -> dict[str, Asset]:
         '''
         Load and cache all asset infos and pack into
         our native ``Asset`` struct.
@@ -282,21 +260,37 @@ class Client:
             }
 
         '''
-        resp = await self._public('Assets', {})
-        assets = resp['result']
+        if (
+            not self._assets
+            or reload
+        ):
+            resp = await self._public('Assets', {})
+            assets: dict[str, dict] = resp['result']
 
-        for bs_mktid, info in assets.items():
-            altname = self._altnames[bs_mktid] = info['altname']
-            aclass: str = info['aclass']
+            for bs_mktid, info in assets.items():
 
-            self.assets[bs_mktid] = Asset(
-                name=altname.lower(),
-                atype=f'crypto_{aclass}',
-                tx_tick=digits_to_dec(info['decimals']),
-                info=info,
-            )
+                altname: str = info['altname']
+                aclass: str = info['aclass']
+                asset = Asset(
+                    name=altname,
+                    atype=f'crypto_{aclass}',
+                    tx_tick=digits_to_dec(info['decimals']),
+                    info=info,
+                )
+                # NOTE: yes we keep 2 sets since kraken insists on
+                # keeping 3 frickin sets bc apparently they have
+                # no sane data engineers whol all like different
+                # keys for their fricking symbology sets..
+                self._Assets[bs_mktid] = asset
+                self._assets[altname.lower()] = asset
+                self._assets[altname] = asset
 
-        return self.assets
+        # we return the "most native" set merged with our preferred
+        # naming (which i guess is the "altname" one) since that's
+        # what the symcache loader will be storing, and we need the
+        # keys that are easiest to match against in any trade
+        # records.
+        return self._Assets | self._assets
 
     async def get_trades(
         self,
@@ -377,23 +371,25 @@ class Client:
         #     'amount': '0.00300726', 'fee': '0.00001000', 'time':
         #     1658347714, 'status': 'Success'}]}
 
+        if xfers:
+            import tractor
+            await tractor.pp()
+
         trans: dict[str, Transaction] = {}
         for entry in xfers:
-
             # look up the normalized name and asset info
-            asset_key = entry['asset']
-            asset = self.assets[asset_key]
-            asset_key = self._altnames[asset_key].lower()
+            asset_key: str = entry['asset']
+            asset: Asset = self._Assets[asset_key]
+            asset_key: str = asset.name.lower()
 
             # XXX: this is in the asset units (likely) so it isn't
             # quite the same as a commisions cost necessarily..)
+            # TODO: also round this based on `Pair` cost precision info?
             cost = float(entry['fee'])
-
-            fqme = asset_key + '.kraken'
+            # fqme: str = asset_key + '.kraken'
 
             tx = Transaction(
-                fqme=fqme,
-                sym=asset,
+                fqme=asset_key,  # this must map to an entry in .assets!
                 tid=entry['txid'],
                 dt=pendulum.from_timestamp(entry['time']),
                 bs_mktid=f'{asset_key}{src_asset}',
@@ -408,6 +404,11 @@ class Client:
 
                 # XXX: see note above
                 cost=cost,
+
+                # not a trade but a withdrawal or deposit on the
+                # asset (chain) system.
+                etype='transfer',
+
             )
             trans[tx.tid] = tx
 
@@ -458,7 +459,7 @@ class Client:
         # txid is a transaction id given by kraken
         return await self.endpoint('CancelOrder', {"txid": reqid})
 
-    async def pair_info(
+    async def asset_pairs(
         self,
         pair_patt: str | None = None,
 
@@ -470,64 +471,76 @@ class Client:
         https://docs.kraken.com/rest/#tag/Market-Data/operation/getTradableAssetPairs
 
         '''
-        # get all pairs by default, or filter
-        # to whatever pattern is provided as input.
-        pairs: dict[str, str] | None = None
-        if pair_patt is not None:
-            pairs = {'pair': pair_patt}
+        if not self._AssetPairs:
+            # get all pairs by default, or filter
+            # to whatever pattern is provided as input.
+            req_pairs: dict[str, str] | None = None
+            if pair_patt is not None:
+                req_pairs = {'pair': pair_patt}
 
-        resp = await self._public(
-            'AssetPairs',
-            pairs,
-        )
-        err = resp['error']
-        if err:
-            raise SymbolNotFound(pair_patt)
+            resp = await self._public(
+                'AssetPairs',
+                req_pairs,
+            )
+            err = resp['error']
+            if err:
+                raise SymbolNotFound(pair_patt)
 
-        pairs: dict[str, Pair] = {
+            # NOTE: we try to key pairs by our custom defined
+            # `.bs_fqme` field since we want to offer search over
+            # this pattern set, callers should fill out lookup
+            # tables for kraken's bs_mktid keys to map to these
+            # keys!
+            # XXX: FURTHER kraken's data eng team decided to offer
+            # 3 frickin market-pair-symbol key sets depending on
+            # which frickin API is being used.
+            # Example for the trading pair 'LTC<EUR'
+            # - the "X-key" from rest eps 'XLTCZEUR'
+            # - the "websocket key" from ws msgs is 'LTC/EUR'
+            # - the "altname key" also delivered in pair info is 'LTCEUR'
+            for xkey, data in resp['result'].items():
 
-            key: Pair(**data)
-            for key, data in resp['result'].items()
-        }
-        # always cache so we can possibly do faster lookup
-        self._pairs.update(pairs)
+                # NOTE: always cache in pairs tables for faster lookup
+                pair = Pair(xname=xkey, **data)
+
+                # register the above `Pair` structs for all
+                # key-sets/monikers: a set of 4 (frickin) tables
+                # acting as a combined surjection of all possible
+                # (and stupid) kraken names to their `Pair` obj.
+                self._AssetPairs[xkey] = pair
+                self._pairs[pair.bs_fqme] = pair
+                self._altnames[pair.altname] = pair
+                self._wsnames[pair.wsname] = pair
 
         if pair_patt is not None:
-            return next(iter(pairs.items()))[1]
+            return next(iter(self._pairs.items()))[1]
 
-        return pairs
+        return self._AssetPairs
 
-    async def cache_symbols(self) -> dict:
+    async def get_mkt_pairs(
+        self,
+        reload: bool = False,
+    ) -> dict:
         '''
-        Load all market pair info build and cache it for downstream use.
+        Load all market pair info build and cache it for downstream
+        use.
 
-        A ``._ntable: dict[str, str]`` is available for mapping the
-        websocket pair name-keys and their http endpoint API (smh)
-        equivalents to the "alternative name" which is generally the one
-        we actually want to use XD
+        Multiple pair info lookup tables (like ``._altnames:
+        dict[str, str]``) are created for looking up the
+        piker-native `Pair`-struct from any input of the three
+        (yes, it's that idiotic..) available symbol/pair-key-sets
+        that kraken frickin offers depending on the API including
+        the .altname, .wsname and the weird ass default set they
+        return in ReST responses .xname..
 
         '''
-        if not self._pairs:
-            pairs = await self.pair_info()
-            assert self._pairs == pairs
+        if (
+            not self._pairs
+            or reload
+        ):
+            await self.asset_pairs()
 
-            # table of all ws and rest keys to their alt-name values.
-            ntable: dict[str, str] = {}
-
-            for rest_key in list(pairs.keys()):
-
-                pair: Pair = pairs[rest_key]
-                altname = pair.altname
-                wsname = pair.wsname
-                ntable[altname] = ntable[rest_key] = ntable[wsname] = altname
-
-                # register the pair under all monikers, a giant flat
-                # surjection of all possible names to each info obj.
-                self._pairs[altname] = self._pairs[wsname] = pair
-
-            self._ntable.update(ntable)
-
-        return self._pairs
+        return self._AssetPairs
 
     async def search_symbols(
         self,
@@ -543,16 +556,20 @@ class Client:
 
         '''
         if not len(self._pairs):
-            await self.cache_symbols()
-            assert self._pairs, '`Client.cache_symbols()` was never called!?'
+            await self.get_mkt_pairs()
+            assert self._pairs, '`Client.get_mkt_pairs()` was never called!?'
 
-        matches = fuzzy.extractBests(
-            pattern,
-            self._pairs,
+        matches: dict[str, Pair] = match_from_pairs(
+            pairs=self._pairs,
+            query=pattern.upper(),
             score_cutoff=50,
         )
-        # repack in dict form
-        return {item[0].altname: item[0] for item in matches}
+
+        # repack in .altname-keyed output table
+        return {
+            pair.altname: pair
+            for pair in matches.values()
+        }
 
     async def bars(
         self,
@@ -632,10 +649,10 @@ class Client:
                 raise BrokerError(errmsg)
 
     @classmethod
-    def normalize_symbol(
+    def to_bs_fqme(
         cls,
-        ticker: str
-    ) -> tuple[str, Pair]:
+        pair_str: str
+    ) -> str:
         '''
         Normalize symbol names to to a 3x3 pair from the global
         definition map which we build out from the data retreived from
@@ -643,7 +660,7 @@ class Client:
 
         '''
         try:
-            return cls._ntable[ticker]
+            return cls._altnames[pair_str.upper()].bs_fqme
         except KeyError as ke:
             raise SymbolNotFound(f'kraken has no {ke.args[0]}')
 
@@ -651,21 +668,36 @@ class Client:
 @acm
 async def get_client() -> Client:
 
-    conf = get_config()
-    if conf:
-        client = Client(
-            conf,
-            name=conf['key_descr'],
-            api_key=conf['api_key'],
-            secret=conf['secret']
-        )
-    else:
-        client = Client({})
+    conf: dict[str, Any] = get_config()
+    async with httpx.AsyncClient(
+        base_url=_url,
+        headers=_headers,
 
-    # at startup, load all symbols, and asset info in
-    # batch requests.
-    async with trio.open_nursery() as nurse:
-        nurse.start_soon(client.get_assets)
-        await client.cache_symbols()
+        # TODO: is there a way to numerate this?
+        # https://www.python-httpx.org/advanced/clients/#why-use-a-client
+        # connections=4
+    ) as trio_client:
+        if conf:
+            client = Client(
+                conf,
+                httpx_client=trio_client,
 
-    yield client
+                # TODO: don't break these up and just do internal
+                # conf lookups instead..
+                name=conf['key_descr'],
+                api_key=conf['api_key'],
+                secret=conf['secret']
+            )
+        else:
+            client = Client(
+                conf={},
+                httpx_client=trio_client,
+            )
+
+        # at startup, load all symbols, and asset info in
+        # batch requests.
+        async with trio.open_nursery() as nurse:
+            nurse.start_soon(client.get_assets)
+            await client.get_mkt_pairs()
+
+        yield client

piker/brokers/kraken/broker.py

@@ -24,7 +24,6 @@ from contextlib import (
 )
 from functools import partial
 from itertools import count
-import math
 from pprint import pformat
 import time
 from typing import (
@@ -35,21 +34,16 @@ from typing import (
 )
 
 from bidict import bidict
-import pendulum
 import trio
 import tractor
 
 from piker.accounting import (
     Position,
-    PpTable,
+    Account,
     Transaction,
     TransactionLedger,
     open_trade_ledger,
-    open_pps,
-    get_likely_pair,
-)
-from piker.accounting._mktinfo import (
-    MktPair,
+    open_account,
 )
 from piker.clearing import(
     OrderDialogs,
@@ -65,18 +59,24 @@ from piker.clearing._messages import (
     BrokerdPosition,
     BrokerdStatus,
 )
+from piker.brokers import (
+    open_cached_client,
+)
+from piker.data import open_symcache
 from .api import (
     log,
     Client,
     BrokerError,
-    get_client,
 )
 from .feed import (
-    get_mkt_info,
     open_autorecon_ws,
     NoBsWs,
     stream_messages,
 )
+from .ledger import (
+    norm_trade_records,
+    verify_balances,
+)
 
 MsgUnion = Union[
     BrokerdCancel,
@@ -371,7 +371,8 @@ async def subscribe(
 
 
 def trades2pps(
-    table: PpTable,
+    acnt: Account,
+    ledger: TransactionLedger,
     acctid: str,
     new_trans: dict[str, Transaction] = {},
 
@@ -379,13 +380,14 @@ def trades2pps(
 
 ) -> list[BrokerdPosition]:
     if new_trans:
-        updated = table.update_from_trans(
+        updated = acnt.update_from_ledger(
             new_trans,
+            symcache=ledger.symcache,
         )
         log.info(f'Updated pps:\n{pformat(updated)}')
 
-    pp_entries, closed_pp_objs = table.dump_active()
-    pp_objs: dict[Union[str, int], Position] = table.pps
+    pp_entries, closed_pp_objs = acnt.dump_active()
+    pp_objs: dict[Union[str, int], Position] = acnt.pps
 
     pps: dict[int, Position]
     position_msgs: list[dict] = []
@@ -399,13 +401,13 @@ def trades2pps(
                 # backend suffix prefixed but when
                 # reading accounts from ledgers we
                 # don't need it and/or it's prefixed
-                # in the section table.. we should
+                # in the section acnt.. we should
                 # just strip this from the message
                 # right since `.broker` is already
                 # included?
                 account='kraken.' + acctid,
                 symbol=p.mkt.fqme,
-                size=p.size,
+                size=p.cumsize,
                 avg_price=p.ppu,
                 currency='',
             )
@@ -416,7 +418,7 @@ def trades2pps(
         # as little as possible. we need to either do
         # these writes in another actor, or try out `trio`'s
         # async file IO api?
-        table.write_config()
+        acnt.write_config()
 
     return position_msgs
 
@@ -427,7 +429,12 @@ async def open_trade_dialog(
 
 ) -> AsyncIterator[dict[str, Any]]:
 
-    async with get_client() as client:
+    async with (
+        # TODO: maybe bind these together and deliver
+        # a tuple from `.open_cached_client()`?
+        open_cached_client('kraken') as client,
+        open_symcache('kraken') as symcache,
+    ):
         # make ems flip to paper mode when no creds setup in
         # `brokers.toml` B0
         if not client._api_key:
@@ -457,8 +464,8 @@ async def open_trade_dialog(
         # - delete the *ABSOLUTE LAST* entry from account's corresponding
         #   trade ledgers file (NOTE this MUST be the last record
         #   delivered from the api ledger),
-        # - open you ``pps.toml`` and find that same tid and delete it
-        #   from the pp's clears table,
+        # - open you ``account.kraken.spot.toml`` and find that
+        #   same tid and delete it from the pos's clears table,
         # - set this flag to `True`
         #
         # You should see an update come in after the order mode
@@ -469,172 +476,85 @@ async def open_trade_dialog(
         # update things correctly.
         simulate_pp_update: bool = False
 
-        table: PpTable
+        acnt: Account
         ledger: TransactionLedger
         with (
-            open_pps(
+            open_account(
                 'kraken',
                 acctid,
                 write_on_exit=True,
-            ) as table,
+            ) as acnt,
 
             open_trade_ledger(
                 'kraken',
                 acctid,
+                symcache=symcache,
             ) as ledger,
         ):
-            # transaction-ify the ledger entries
-            ledger_trans = await norm_trade_records(ledger)
+            # TODO: loading ledger entries should all be done
+            # within a newly implemented `async with open_account()
+            # as acnt` where `Account.ledger: TransactionLedger`
+            # can be used to explicitily update and write the
+            # offline TOML files!
+            # ------ - ------
+            # MOL the init sequence is:
+            # - get `Account` (with presumed pre-loaded ledger done
+            #   beind the scenes as part of ctx enter).
+            # - pull new trades from API, update the ledger with
+            #   normalized to `Transaction` entries of those
+            #   records, presumably (and implicitly) update the
+            #   acnt state including expiries, positions,
+            #   transfers..), and finally of course existing
+            #   per-asset balances.
+            # - validate all pos and balances ensuring there's
+            #   no seemingly noticeable discrepancies?
 
-            if not table.pps:
-                # NOTE: we can't use this since it first needs
-                # broker: str input support!
-                # table.update_from_trans(ledger.to_trans())
-                table.update_from_trans(ledger_trans)
-                table.write_config()
+            # LOAD and transaction-ify the EXISTING LEDGER
+            ledger_trans: dict[str, Transaction] = await norm_trade_records(
+                ledger,
+                client,
+                api_name_set='xname',
+            )
+
+            if not acnt.pps:
+                acnt.update_from_ledger(
+                    ledger_trans,
+                    symcache=ledger.symcache,
+                )
+                acnt.write_config()
 
             # TODO: eventually probably only load
             # as far back as it seems is not deliverd in the
             # most recent 50 trades and assume that by ordering we
-            # already have those records in the ledger.
-            tids2trades = await client.get_trades()
+            # already have those records in the ledger?
+            tids2trades: dict[str, dict] = await client.get_trades()
             ledger.update(tids2trades)
             if tids2trades:
                 ledger.write_config()
 
-            api_trans = await norm_trade_records(tids2trades)
+            api_trans: dict[str, Transaction] = await norm_trade_records(
+                tids2trades,
+                client,
+                api_name_set='xname',
+            )
 
             # retrieve kraken reported balances
             # and do diff with ledger to determine
             # what amount of trades-transactions need
             # to be reloaded.
-            balances = await client.get_balances()
+            balances: dict[str, float] = await client.get_balances()
 
-            for dst, size in balances.items():
+            verify_balances(
+                acnt,
+                src_fiat,
+                balances,
+                client,
+                ledger,
+                ledger_trans,
+                api_trans,
+            )
 
-                # we don't care about tracking positions
-                # in the user's source fiat currency.
-                if (
-                    dst == src_fiat
-                    or not any(
-                        dst in bs_mktid for bs_mktid in table.pps
-                    )
-                ):
-                    log.warning(
-                        f'Skipping balance `{dst}`:{size} for position calcs!'
-                    )
-                    continue
-
-                def has_pp(
-                    dst: str,
-                    size: float,
-
-                ) -> Position | None:
-
-                    src2dst: dict[str, str] = {}
-
-                    for bs_mktid in table.pps:
-                        likely_pair = get_likely_pair(
-                            src_fiat,
-                            dst,
-                            bs_mktid,
-                        )
-                        if likely_pair:
-                            src2dst[src_fiat] = dst
-
-                    for src, dst in src2dst.items():
-                        pair = f'{dst}{src_fiat}'
-                        pp = table.pps.get(pair)
-                        if (
-                            pp
-                            and math.isclose(pp.size, size)
-                        ):
-                            return pp
-
-                        elif (
-                            size == 0
-                            and pp.size
-                        ):
-                            log.warning(
-                                f'`kraken` account says you have  a ZERO '
-                                f'balance for {bs_mktid}:{pair}\n'
-                                f'but piker seems to think `{pp.size}`\n'
-                                'This is likely a discrepancy in piker '
-                                'accounting if the above number is'
-                                "large,' though it's likely to due lack"
-                                "f tracking xfers fees.."
-                            )
-                            return pp
-
-                    return None  # signal no entry
-
-                pos = has_pp(dst, size)
-                if not pos:
-
-                    # we have a balance for which there is no pp
-                    # entry? so we have to likely update from the
-                    # ledger.
-                    updated = table.update_from_trans(ledger_trans)
-                    log.info(f'Updated pps from ledger:\n{pformat(updated)}')
-                    pos = has_pp(dst, size)
-
-                    if (
-                        not pos
-                        and not simulate_pp_update
-                    ):
-                        # try reloading from API
-                        table.update_from_trans(api_trans)
-                        pos = has_pp(dst, size)
-                        if not pos:
-
-                            # get transfers to make sense of abs balances.
-                            # NOTE: we do this after ledger and API
-                            # loading since we might not have an entry
-                            # in the ``pps.toml`` for the necessary pair
-                            # yet and thus this likely pair grabber will
-                            # likely fail.
-                            for bs_mktid in table.pps:
-                                likely_pair = get_likely_pair(
-                                    src_fiat,
-                                    dst,
-                                    bs_mktid,
-                                )
-                                if likely_pair:
-                                    break
-                            else:
-                                raise ValueError(
-                                    'Could not find a position pair in '
-                                    'ledger for likely widthdrawal '
-                                    f'candidate: {dst}'
-                                )
-
-                            if likely_pair:
-                                # this was likely pp that had a withdrawal
-                                # from the dst asset out of the account.
-
-                                xfer_trans = await client.get_xfers(
-                                    dst,
-                                    # TODO: not all src assets are
-                                    # 3 chars long...
-                                    src_asset=likely_pair[3:],
-                                )
-                                if xfer_trans:
-                                    updated = table.update_from_trans(
-                                        xfer_trans,
-                                        cost_scalar=1,
-                                    )
-                                    log.info(
-                                        f'Updated {dst} from transfers:\n'
-                                        f'{pformat(updated)}'
-                                    )
-
-                        if has_pp(dst, size):
-                            raise ValueError(
-                                'Could not reproduce balance:\n'
-                                f'dst: {dst}, {size}\n'
-                            )
-
-            # only for simulate-testing a "new fill" since
+            # XXX NOTE: only for simulate-testing a "new fill" since
             # otherwise we have to actually conduct a live clear.
             if simulate_pp_update:
                 tid = list(tids2trades)[0]
@@ -643,25 +563,27 @@ async def open_trade_dialog(
                 reqids2txids[0] = last_trade_dict['ordertxid']
 
             ppmsgs: list[BrokerdPosition] = trades2pps(
-                table,
+                acnt,
+                ledger,
                 acctid,
             )
+            # sync with EMS delivering pps and accounts
             await ctx.started((ppmsgs, [acc_name]))
 
             # TODO: ideally this blocks the this task
             # as little as possible. we need to either do
             # these writes in another actor, or try out `trio`'s
             # async file IO api?
-            table.write_config()
+            acnt.write_config()
 
             # Get websocket token for authenticated data stream
             # Assert that a token was actually received.
             resp = await client.endpoint('GetWebSocketsToken', {})
-            err = resp.get('error')
-            if err:
+            if err := resp.get('error'):
                 raise BrokerError(err)
 
-            token = resp['result']['token']
+            # resp token for ws init
+            token: str = resp['result']['token']
 
             ws: NoBsWs
             async with (
@@ -690,32 +612,35 @@ async def open_trade_dialog(
 
                 # enter relay loop
                 await handle_order_updates(
-                    ws,
-                    stream,
-                    ems_stream,
-                    apiflows,
-                    ids,
-                    reqids2txids,
-                    table,
-                    api_trans,
-                    acctid,
-                    acc_name,
-                    token,
+                    client=client,
+                    ws=ws,
+                    ws_stream=stream,
+                    ems_stream=ems_stream,
+                    apiflows=apiflows,
+                    ids=ids,
+                    reqids2txids=reqids2txids,
+                    acnt=acnt,
+                    ledger=ledger,
+                    acctid=acctid,
+                    acc_name=acc_name,
+                    token=token,
                 )
 
 
 async def handle_order_updates(
+    client: Client,  # only for pairs table needed in ledger proc
     ws: NoBsWs,
     ws_stream: AsyncIterator,
     ems_stream: tractor.MsgStream,
     apiflows: OrderDialogs,
     ids: bidict[str, int],
     reqids2txids: bidict[int, str],
-    table: PpTable,
+    acnt: Account,
 
     # transaction records which will be updated
     # on new trade clearing events (aka order "fills")
-    ledger_trans: dict[str, Transaction],
+    ledger: TransactionLedger,
+    # ledger_trans: dict[str, Transaction],
     acctid: str,
     acc_name: str,
     token: str,
@@ -733,7 +658,7 @@ async def handle_order_updates(
 
             # TODO: turns out you get the fill events from the
             # `openOrders` before you get this, so it might be better
-            # to do all fill/status/pp updates in that sub and just use
+            # to do all fill/status/pos updates in that sub and just use
             # this one for ledger syncs?
 
             # For eg. we could take the "last 50 trades" and do a diff
@@ -775,7 +700,8 @@ async def handle_order_updates(
                     # if tid not in ledger_trans
                 }
                 for tid, trade in trades.items():
-                    assert tid not in ledger_trans
+                    # assert tid not in ledger_trans
+                    assert tid not in ledger
                     txid = trade['ordertxid']
                     reqid = trade.get('userref')
 
@@ -818,12 +744,22 @@ async def handle_order_updates(
                     )
                     await ems_stream.send(status_msg)
 
-                new_trans = await norm_trade_records(trades)
-                ppmsgs = trades2pps(
-                    table,
-                    acctid,
-                    new_trans,
+                new_trans = await norm_trade_records(
+                    trades,
+                    client,
+                    api_name_set='wsname',
                 )
+                ppmsgs: list[BrokerdPosition] = trades2pps(
+                    acnt=acnt,
+                    ledger=ledger,
+                    acctid=acctid,
+                    new_trans=new_trans,
+                )
+                # ppmsgs = trades2pps(
+                #     acnt,
+                #     acctid,
+                #     new_trans,
+                # )
                 for pp_msg in ppmsgs:
                     await ems_stream.send(pp_msg)
 
@@ -1183,36 +1119,3 @@ async def handle_order_updates(
                         })
             case _:
                 log.warning(f'Unhandled trades update msg: {msg}')
-
-
-async def norm_trade_records(
-    ledger: dict[str, Any],
-
-) -> dict[str, Transaction]:
-
-    records: dict[str, Transaction] = {}
-
-    for tid, record in ledger.items():
-
-        size = float(record.get('vol')) * {
-            'buy': 1,
-            'sell': -1,
-        }[record['type']]
-
-        # we normalize to kraken's `altname` always..
-        bs_mktid: str = Client.normalize_symbol(record['pair'])
-        fqme = f'{bs_mktid.lower()}.kraken'
-        mkt: MktPair = (await get_mkt_info(fqme))[0]
-
-        records[tid] = Transaction(
-            fqme=fqme,
-            sym=mkt,
-            tid=tid,
-            size=size,
-            price=float(record['price']),
-            cost=float(record['fee']),
-            dt=pendulum.from_timestamp(float(record['time'])),
-            bs_mktid=bs_mktid,
-        )
-
-    return records

piker/brokers/kraken/feed.py

@@ -30,38 +30,29 @@ from typing import (
 )
 import time
 
-from fuzzywuzzy import process as fuzzy
 import numpy as np
 import pendulum
 from trio_typing import TaskStatus
-import tractor
 import trio
 
 from piker.accounting._mktinfo import (
-    Asset,
     MktPair,
-    unpack_fqme,
 )
 from piker.brokers import (
     open_cached_client,
-    SymbolNotFound,
-)
-from piker._cacheables import (
-    async_lifo_cache,
 )
 from piker.brokers._util import (
     BrokerError,
     DataThrottle,
     DataUnavailable,
 )
-from piker.data.types import Struct
+from piker.types import Struct
 from piker.data.validate import FeedInit
 from piker.data._web_bs import open_autorecon_ws, NoBsWs
 from .api import (
     log,
-    Client,
-    Pair,
 )
+from .symbols import get_mkt_info
 
 
 class OHLC(Struct, frozen=True):
@@ -267,62 +258,6 @@ async def open_history_client(
         yield get_ohlc, {'erlangs': 1, 'rate': 1}
 
 
-@async_lifo_cache()
-async def get_mkt_info(
-    fqme: str,
-
-) -> tuple[MktPair, Pair]:
-    '''
-    Query for and return a `MktPair` and backend-native `Pair` (or
-    wtv else) info.
-
-    If more then one fqme is provided return a ``dict`` of native
-    key-strs to `MktPair`s.
-
-    '''
-    venue: str = 'spot'
-    expiry: str = ''
-    if '.kraken' in fqme:
-        broker, pair, venue, expiry = unpack_fqme(fqme)
-        venue: str = venue or 'spot'
-
-    if venue != 'spot':
-        raise SymbolNotFound(
-            'kraken only supports spot markets right now!\n'
-            f'{fqme}\n'
-        )
-
-    async with open_cached_client('kraken') as client:
-
-        # uppercase since kraken bs_mktid is always upper
-        bs_fqme, _, broker = fqme.partition('.')
-        pair_str: str = bs_fqme.upper()
-        bs_mktid: str = Client.normalize_symbol(pair_str)
-        pair: Pair = await client.pair_info(pair_str)
-
-        assets = client.assets
-        dst_asset: Asset = assets[pair.base]
-        src_asset: Asset = assets[pair.quote]
-
-        mkt = MktPair(
-            dst=dst_asset,
-            src=src_asset,
-
-            price_tick=pair.price_tick,
-            size_tick=pair.size_tick,
-            bs_mktid=bs_mktid,
-
-            expiry=expiry,
-            venue=venue or 'spot',
-
-            # TODO: futes
-            # _atype=_atype,
-
-            broker='kraken',
-        )
-        return mkt, pair
-
-
 async def stream_quotes(
 
     send_chan: trio.abc.SendChannel,
@@ -478,30 +413,3 @@ async def stream_quotes(
                         log.warning(f'Unknown WSS message: {typ}, {quote}')
 
                 await send_chan.send({topic: quote})
-
-
-@tractor.context
-async def open_symbol_search(
-    ctx: tractor.Context,
-
-) -> Client:
-    async with open_cached_client('kraken') as client:
-
-        # load all symbols locally for fast search
-        cache = await client.cache_symbols()
-        await ctx.started(cache)
-
-        async with ctx.open_stream() as stream:
-
-            async for pattern in stream:
-
-                matches = fuzzy.extractBests(
-                    pattern,
-                    cache,
-                    score_cutoff=50,
-                )
-                # repack in dict form
-                await stream.send({
-                    pair[0].altname: pair[0]
-                    for pair in matches
-                })

piker/brokers/kraken/ledger.py

@@ -0,0 +1,269 @@
+# piker: trading gear for hackers
+# Copyright (C) Tyler Goodlet (in stewardship for pikers)
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+'''
+Trade transaction accounting and normalization.
+
+'''
+import math
+from pprint import pformat
+from typing import (
+    Any,
+)
+
+import pendulum
+
+from piker.accounting import (
+    Transaction,
+    Position,
+    Account,
+    get_likely_pair,
+    TransactionLedger,
+    # MktPair,
+)
+from piker.types import Struct
+from piker.data import (
+    SymbologyCache,
+)
+from .api import (
+    log,
+    Client,
+    Pair,
+)
+# from .feed import get_mkt_info
+
+
+def norm_trade(
+    tid: str,
+    record: dict[str, Any],
+
+    # this is the dict that was returned from
+    # `Client.get_mkt_pairs()` and when running offline ledger
+    # processing from `.accounting`, this will be the table loaded
+    # into `SymbologyCache.pairs`.
+    pairs: dict[str, Struct],
+    symcache: SymbologyCache | None = None,
+
+) -> Transaction:
+
+    size: float = float(record.get('vol')) * {
+        'buy': 1,
+        'sell': -1,
+    }[record['type']]
+
+    # NOTE: this value may be either the websocket OR the rest schema
+    # so we need to detect the key format and then choose the
+    # correct symbol lookup table to evetually get a ``Pair``..
+    # See internals of `Client.asset_pairs()` for deats!
+    src_pair_key: str = record['pair']
+
+    # XXX: kraken's data engineering is soo bad they require THREE
+    # different pair schemas (more or less seemingly tied to
+    # transport-APIs)..LITERALLY they return different market id
+    # pairs in the ledger endpoints vs. the websocket event subs..
+    # lookup pair using appropriately provided tabled depending
+    # on API-key-schema..
+    pair: Pair = pairs[src_pair_key]
+    fqme: str = pair.bs_fqme.lower() + '.kraken'
+
+    return Transaction(
+        fqme=fqme,
+        tid=tid,
+        size=size,
+        price=float(record['price']),
+        cost=float(record['fee']),
+        dt=pendulum.from_timestamp(float(record['time'])),
+        bs_mktid=pair.bs_mktid,
+    )
+
+
+async def norm_trade_records(
+    ledger: dict[str, Any],
+    client: Client,
+    api_name_set: str = 'xname',
+
+) -> dict[str, Transaction]:
+    '''
+    Loop through an input ``dict`` of trade records
+    and convert them to ``Transactions``.
+
+    '''
+    records: dict[str, Transaction] = {}
+    for tid, record in ledger.items():
+
+        # manual_fqme: str = f'{bs_mktid.lower()}.kraken'
+        # mkt: MktPair = (await get_mkt_info(manual_fqme))[0]
+        # fqme: str = mkt.fqme
+        # assert fqme == manual_fqme
+        pairs: dict[str, Pair] = {
+            'xname': client._AssetPairs,
+            'wsname': client._wsnames,
+            'altname': client._altnames,
+        }[api_name_set]
+
+        records[tid] = norm_trade(
+            tid,
+            record,
+            pairs=pairs,
+        )
+
+    return records
+
+
+def has_pp(
+    acnt: Account,
+    src_fiat: str,
+    dst: str,
+    size: float,
+
+) -> Position | None:
+
+    src2dst: dict[str, str] = {}
+    for bs_mktid in acnt.pps:
+        likely_pair = get_likely_pair(
+            src_fiat,
+            dst,
+            bs_mktid,
+        )
+        if likely_pair:
+            src2dst[src_fiat] = dst
+
+    for src, dst in src2dst.items():
+        pair: str = f'{dst}{src_fiat}'
+        pos: Position = acnt.pps.get(pair)
+        if (
+            pos
+            and math.isclose(pos.size, size)
+        ):
+            return pos
+
+        elif (
+            size == 0
+            and pos.size
+        ):
+            log.warning(
+                f'`kraken` account says you have  a ZERO '
+                f'balance for {bs_mktid}:{pair}\n'
+                f'but piker seems to think `{pos.size}`\n'
+                'This is likely a discrepancy in piker '
+                'accounting if the above number is'
+                "large,' though it's likely to due lack"
+                "f tracking xfers fees.."
+            )
+            return pos
+
+    return None  # indicate no entry found
+
+
+# TODO: factor most of this "account updating from txns" into the
+# the `Account` impl so has to provide for hiding the mostly
+# cross-provider updates from txn sets
+async def verify_balances(
+    acnt: Account,
+    src_fiat: str,
+    balances: dict[str, float],
+    client: Client,
+    ledger: TransactionLedger,
+    ledger_trans: dict[str, Transaction],  # from toml
+    api_trans: dict[str, Transaction],  # from API
+
+    simulate_pp_update: bool = False,
+
+) -> None:
+    for dst, size in balances.items():
+
+        # we don't care about tracking positions
+        # in the user's source fiat currency.
+        if (
+            dst == src_fiat
+            or not any(
+                dst in bs_mktid for bs_mktid in acnt.pps
+            )
+        ):
+            log.warning(
+                f'Skipping balance `{dst}`:{size} for position calcs!'
+            )
+            continue
+
+        # we have a balance for which there is no pos entry
+        # - we have to likely update from the ledger?
+        if not has_pp(acnt, src_fiat, dst, size):
+            updated = acnt.update_from_ledger(
+                ledger_trans,
+                symcache=ledger.symcache,
+            )
+            log.info(f'Updated pps from ledger:\n{pformat(updated)}')
+
+            # FIRST try reloading from API records
+            if (
+                not has_pp(acnt, src_fiat, dst, size)
+                and not simulate_pp_update
+            ):
+                acnt.update_from_ledger(
+                    api_trans,
+                    symcache=ledger.symcache,
+                )
+
+                # get transfers to make sense of abs
+                # balances.
+                # NOTE: we do this after ledger and API
+                # loading since we might not have an
+                # entry in the
+                # ``account.kraken.spot.toml`` for the
+                # necessary pair yet and thus this
+                # likely pair grabber will likely fail.
+                if not has_pp(acnt, src_fiat, dst, size):
+                    for bs_mktid in acnt.pps:
+                        likely_pair: str | None = get_likely_pair(
+                            src_fiat,
+                            dst,
+                            bs_mktid,
+                        )
+                        if likely_pair:
+                            break
+                    else:
+                        raise ValueError(
+                            'Could not find a position pair in '
+                            'ledger for likely widthdrawal '
+                            f'candidate: {dst}'
+                        )
+
+                    # this was likely pos that had a withdrawal
+                    # from the dst asset out of the account.
+                    if likely_pair:
+                        xfer_trans = await client.get_xfers(
+                            dst,
+
+                            # TODO: not all src assets are
+                            # 3 chars long...
+                            src_asset=likely_pair[3:],
+                        )
+                        if xfer_trans:
+                            updated = acnt.update_from_ledger(
+                                xfer_trans,
+                                cost_scalar=1,
+                                symcache=ledger.symcache,
+                            )
+                            log.info(
+                                f'Updated {dst} from transfers:\n'
+                                f'{pformat(updated)}'
+                            )
+
+                if has_pp(acnt, src_fiat, dst, size):
+                    raise ValueError(
+                        'Could not reproduce balance:\n'
+                        f'dst: {dst}, {size}\n'
+                    )

piker/brokers/kraken/symbols.py

@@ -0,0 +1,206 @@
+# piker: trading gear for hackers
+# Copyright (C) Tyler Goodlet (in stewardship for pikers)
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+'''
+Symbology defs and search.
+
+'''
+from decimal import Decimal
+
+import tractor
+from rapidfuzz import process as fuzzy
+
+from piker._cacheables import (
+    async_lifo_cache,
+)
+from piker.accounting._mktinfo import (
+    digits_to_dec,
+)
+from piker.brokers import (
+    open_cached_client,
+    SymbolNotFound,
+)
+from piker.types import Struct
+from piker.accounting._mktinfo import (
+    Asset,
+    MktPair,
+    unpack_fqme,
+)
+
+
+# https://www.kraken.com/features/api#get-tradable-pairs
+class Pair(Struct):
+    xname: str  # idiotic bs_mktid equiv i guess?
+    altname: str  # alternate pair name
+    wsname: str  # WebSocket pair name (if available)
+    aclass_base: str  # asset class of base component
+    base: str  # asset id of base component
+    aclass_quote: str  # asset class of quote component
+    quote: str  # asset id of quote component
+    lot: str  # volume lot size
+
+    cost_decimals: int
+    costmin: float
+    pair_decimals: int  # scaling decimal places for pair
+    lot_decimals: int  # scaling decimal places for volume
+
+    # amount to multiply lot volume by to get currency volume
+    lot_multiplier: float
+
+    # array of leverage amounts available when buying
+    leverage_buy: list[int]
+    # array of leverage amounts available when selling
+    leverage_sell: list[int]
+
+    # fee schedule array in [volume, percent fee] tuples
+    fees: list[tuple[int, float]]
+
+    # maker fee schedule array in [volume, percent fee] tuples (if on
+    # maker/taker)
+    fees_maker: list[tuple[int, float]]
+
+    fee_volume_currency: str  # volume discount currency
+    margin_call: str  # margin call level
+    margin_stop: str  # stop-out/liquidation margin level
+    ordermin: float  # minimum order volume for pair
+    tick_size: float  # min price step size
+    status: str
+
+    short_position_limit: float = 0
+    long_position_limit: float = float('inf')
+
+    # TODO: should we make this a literal NamespacePath ref?
+    ns_path: str = 'piker.brokers.kraken:Pair'
+
+    @property
+    def bs_mktid(self) -> str:
+        '''
+        Kraken seems to index it's market symbol sets in
+        transaction ledgers using the key returned from rest
+        queries.. so use that since apparently they can't
+        make up their minds on a better key set XD
+
+        '''
+        return self.xname
+
+    @property
+    def price_tick(self) -> Decimal:
+        return digits_to_dec(self.pair_decimals)
+
+    @property
+    def size_tick(self) -> Decimal:
+        return digits_to_dec(self.lot_decimals)
+
+    @property
+    def bs_dst_asset(self) -> str:
+        dst, _ = self.wsname.split('/')
+        return dst
+
+    @property
+    def bs_src_asset(self) -> str:
+        _, src = self.wsname.split('/')
+        return src
+
+    @property
+    def bs_fqme(self) -> str:
+        '''
+        Basically the `.altname` but with special '.' handling and
+        `.SPOT` suffix appending (for future multi-venue support).
+
+        '''
+        dst, src = self.wsname.split('/')
+        # XXX: omg for stupid shite like ETH2.S/ETH..
+        dst = dst.replace('.', '-')
+        return f'{dst}{src}.SPOT'
+
+
+@tractor.context
+async def open_symbol_search(ctx: tractor.Context) -> None:
+    async with open_cached_client('kraken') as client:
+
+        # load all symbols locally for fast search
+        cache = await client.get_mkt_pairs()
+        await ctx.started(cache)
+
+        async with ctx.open_stream() as stream:
+            async for pattern in stream:
+                await stream.send(
+                    await client.search_symbols(pattern)
+                )
+
+
+@async_lifo_cache()
+async def get_mkt_info(
+    fqme: str,
+
+) -> tuple[MktPair, Pair]:
+    '''
+    Query for and return a `MktPair` and backend-native `Pair` (or
+    wtv else) info.
+
+    If more then one fqme is provided return a ``dict`` of native
+    key-strs to `MktPair`s.
+
+    '''
+    venue: str = 'spot'
+    expiry: str = ''
+    if '.kraken' not in fqme:
+        fqme += '.kraken'
+
+    broker, pair, venue, expiry = unpack_fqme(fqme)
+    venue: str = venue or 'spot'
+
+    if venue.lower() != 'spot':
+        raise SymbolNotFound(
+            'kraken only supports spot markets right now!\n'
+            f'{fqme}\n'
+        )
+
+    async with open_cached_client('kraken') as client:
+
+        # uppercase since kraken bs_mktid is always upper
+        # bs_fqme, _, broker = fqme.partition('.')
+        # pair_str: str = bs_fqme.upper()
+        pair_str: str = f'{pair}.{venue}'
+
+        pair: Pair | None = client._pairs.get(pair_str.upper())
+        if not pair:
+            bs_fqme: str = client.to_bs_fqme(pair_str)
+            pair: Pair = client._pairs[bs_fqme]
+
+        if not (assets := client._assets):
+            assets: dict[str, Asset] = await client.get_assets()
+
+        dst_asset: Asset = assets[pair.bs_dst_asset]
+        src_asset: Asset = assets[pair.bs_src_asset]
+
+        mkt = MktPair(
+            dst=dst_asset,
+            src=src_asset,
+
+            price_tick=pair.price_tick,
+            size_tick=pair.size_tick,
+            bs_mktid=pair.bs_mktid,
+
+            expiry=expiry,
+            venue=venue or 'spot',
+
+            # TODO: futes
+            # _atype=_atype,
+
+            broker='kraken',
+        )
+        return mkt, pair

piker/brokers/kucoin.py

@@ -16,10 +16,9 @@
 # along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
 '''
-Kucoin broker backend
+Kucoin cex API backend.
 
 '''
-
 from contextlib import (
     asynccontextmanager as acm,
     aclosing,
@@ -41,9 +40,8 @@ from typing import (
 import wsproto
 from uuid import uuid4
 
-from fuzzywuzzy import process as fuzzy
 from trio_typing import TaskStatus
-import asks
+import httpx
 from bidict import bidict
 import numpy as np
 import pendulum
@@ -64,8 +62,11 @@ from piker._cacheables import (
 )
 from piker.log import get_logger
 from piker.data.validate import FeedInit
-from piker.data.types import Struct
-from piker.data import def_iohlcv_fields
+from piker.types import Struct  # NOTE, this is already a `tractor.msg.Struct`
+from piker.data import (
+    def_iohlcv_fields,
+    match_from_pairs,
+)
 from piker.data._web_bs import (
     open_autorecon_ws,
     NoBsWs,
@@ -74,6 +75,8 @@ from ._util import DataUnavailable
 
 log = get_logger(__name__)
 
+_no_symcache: bool = True
+
 
 class KucoinMktPair(Struct, frozen=True):
     '''
@@ -86,18 +89,27 @@ class KucoinMktPair(Struct, frozen=True):
 
     @property
     def price_tick(self) -> Decimal:
-        return Decimal(str(self.baseIncrement))
+        return Decimal(str(self.quoteIncrement))
 
     baseMaxSize: float
     baseMinSize: float
 
     @property
     def size_tick(self) -> Decimal:
-        return Decimal(str(self.baseMinSize))
+        return Decimal(str(self.quoteMinSize))
+
+    callauctionFirstStageStartTime: None|float
+    callauctionIsEnabled: bool
+    callauctionPriceCeiling: float|None
+    callauctionPriceFloor: float|None
+    callauctionSecondStageStartTime: float|None
+    callauctionThirdStageStartTime: float|None
 
     enableTrading: bool
+    feeCategory: int
     feeCurrency: str
     isMarginEnabled: bool
+    makerFeeCoefficient: float
     market: str
     minFunds: float
     name: str
@@ -107,7 +119,10 @@ class KucoinMktPair(Struct, frozen=True):
     quoteIncrement: float
     quoteMaxSize: float
     quoteMinSize: float
+    st: bool
     symbol: str  # our bs_mktid, kucoin's internal id
+    takerFeeCoefficient: float
+    tradingStartTime: float|None
 
 
 class AccountTrade(Struct, frozen=True):
@@ -207,8 +222,13 @@ def get_config() -> BrokerConfig | None:
 
 
 class Client:
-    def __init__(self) -> None:
-        self._config: BrokerConfig | None = get_config()
+
+    def __init__(
+        self,
+        httpx_client: httpx.AsyncClient,
+    ) -> None:
+        self._http: httpx.AsyncClient = httpx_client
+        self._config: BrokerConfig|None = get_config()
         self._pairs: dict[str, KucoinMktPair] = {}
         self._fqmes2mktids: bidict[str, str] = bidict()
         self._bars: list[list[float]] = []
@@ -222,18 +242,24 @@ class Client:
 
     ) -> dict[str, str | bytes]:
         '''
-        Generate authenticated request headers
+        Generate authenticated request headers:
+
         https://docs.kucoin.com/#authentication
+        https://www.kucoin.com/docs/basic-info/connection-method/authentication/creating-a-request
+        https://www.kucoin.com/docs/basic-info/connection-method/authentication/signing-a-message
 
         '''
-
         if not self._config:
             raise ValueError(
-                'No config found when trying to send authenticated request')
+                'No config found when trying to send authenticated request'
+            )
 
         str_to_sign = (
             str(int(time.time() * 1000))
-            + action + f'/api/{api}/{endpoint.lstrip("/")}'
+            +
+            action
+            +
+            f'/api/{api}/{endpoint.lstrip("/")}'
         )
 
         signature = base64.b64encode(
@@ -244,6 +270,7 @@ class Client:
             ).digest()
         )
 
+        # TODO: can we cache this between calls?
         passphrase = base64.b64encode(
             hmac.new(
                 self._config.key_secret.encode('utf-8'),
@@ -265,8 +292,10 @@ class Client:
         self,
         action: Literal['POST', 'GET'],
         endpoint: str,
+
         api: str = 'v2',
         headers: dict = {},
+
     ) -> Any:
         '''
         Generic request wrapper for Kucoin API
@@ -279,14 +308,19 @@ class Client:
                 api,
             )
 
-        api_url = f'https://api.kucoin.com/api/{api}/{endpoint}'
-
-        res = await asks.request(action, api_url, headers=headers)
-
-        json = res.json()
-        if 'data' in json:
-            return json['data']
+        req_meth: Callable = getattr(
+            self._http,
+            action.lower(),
+        )
+        res = await req_meth(
+            url=f'/{api}/{endpoint}',
+            headers=headers,
+        )
+        json: dict = res.json()
+        if (data := json.get('data')) is not None:
+            return data
         else:
+            api_url: str = self._http.base_url
             log.error(
                 f'Error making request to {api_url} ->\n'
                 f'{pformat(res)}'
@@ -306,7 +340,7 @@ class Client:
         '''
         token_type = 'private' if private else 'public'
         try:
-            data: dict[str, Any] | None = await self._request(
+            data: dict[str, Any]|None = await self._request(
                 'POST',
                 endpoint=f'bullet-{token_type}',
                 api='v1'
@@ -344,8 +378,8 @@ class Client:
             currencies: dict[str, Currency] = {}
             entries: list[dict] = await self._request(
                 'GET',
-                api='v1',
                 endpoint='currencies',
+                api='v1',
             )
             for entry in entries:
                 curr = Currency(**entry).copy()
@@ -361,20 +395,29 @@ class Client:
         dict[str, KucoinMktPair],
         bidict[str, KucoinMktPair],
     ]:
-        entries = await self._request('GET', 'symbols')
+        entries = await self._request(
+            'GET',
+            endpoint='symbols',
+        )
         log.info(f' {len(entries)} Kucoin market pairs fetched')
 
         pairs: dict[str, KucoinMktPair] = {}
         fqmes2mktids: bidict[str, str] = bidict()
         for item in entries:
-            pair = pairs[item['name']] = KucoinMktPair(**item)
+            try:
+                pair = pairs[item['name']] = KucoinMktPair(**item)
+            except TypeError as te:
+                raise TypeError(
+                    '`KucoinMktPair` and reponse fields do not match ??\n'
+                    f'{KucoinMktPair.fields_diff(item)}\n'
+                ) from te
             fqmes2mktids[
                 item['name'].lower().replace('-', '')
             ] = pair.name
 
         return pairs, fqmes2mktids
 
-    async def cache_pairs(
+    async def get_mkt_pairs(
         self,
         update: bool = False,
 
@@ -402,16 +445,27 @@ class Client:
 
     ) -> dict[str, KucoinMktPair]:
         '''
-        Use fuzzy search to match against all market names.
+        Use fuzzy search engine to match against pairs, deliver
+        matching ones.
 
         '''
-        data = await self.cache_pairs()
+        if not len(self._pairs):
+            await self.get_mkt_pairs()
+            assert self._pairs, '`Client.get_mkt_pairs()` was never called!?'
 
-        matches = fuzzy.extractBests(
-            pattern, data, score_cutoff=35, limit=limit
+        matches: dict[str, KucoinMktPair] = match_from_pairs(
+            pairs=self._pairs,
+            # query=pattern.upper(),
+            query=pattern.upper(),
+            score_cutoff=35,
+            limit=limit,
         )
+
         # repack in dict form
-        return {item[0].name: item[0] for item in matches}
+        return {
+            pair.name: pair
+            for pair in matches.values()
+        }
 
     async def last_trades(self, sym: str) -> list[AccountTrade]:
         trades = await self._request(
@@ -551,13 +605,21 @@ def fqme_to_kucoin_sym(
 
 @acm
 async def get_client() -> AsyncGenerator[Client, None]:
-    client = Client()
+    '''
+    Load an API `Client` preconfigured from user settings
 
-    async with trio.open_nursery() as n:
-        n.start_soon(client.cache_pairs)
-        await client.get_currencies()
+    '''
+    async with (
+        httpx.AsyncClient(
+            base_url='https://api.kucoin.com/api',
+        ) as trio_client,
+    ):
+        client = Client(httpx_client=trio_client)
+        async with trio.open_nursery() as tn:
+            tn.start_soon(client.get_mkt_pairs)
+            await client.get_currencies()
 
-    yield client
+        yield client
 
 
 @tractor.context
@@ -566,7 +628,7 @@ async def open_symbol_search(
 ) -> None:
     async with open_cached_client('kucoin') as client:
         # load all symbols locally for fast search
-        await client.cache_pairs()
+        await client.get_mkt_pairs()
         await ctx.started()
 
         async with ctx.open_stream() as stream:
@@ -593,7 +655,7 @@ async def open_ping_task(
                 await trio.sleep((ping_interval - 1000) / 1000)
                 await ws.send_msg({'id': connect_id, 'type': 'ping'})
 
-        log.info('Starting ping task for kucoin ws connection')
+        log.warning('Starting ping task for kucoin ws connection')
         n.start_soon(ping_server)
 
         yield
@@ -605,16 +667,21 @@ async def open_ping_task(
 async def get_mkt_info(
     fqme: str,
 
-) -> tuple[MktPair, KucoinMktPair]:
+) -> tuple[
+    MktPair,
+    KucoinMktPair,
+]:
     '''
-    Query for and return a `MktPair` and `KucoinMktPair`.
+    Query for and return both a `piker.accounting.MktPair` and
+    `KucoinMktPair` from provided `fqme: str`
+    (fully-qualified-market-endpoint).
 
     '''
     async with open_cached_client('kucoin') as client:
         # split off any fqme broker part
         bs_fqme, _, broker = fqme.partition('.')
 
-        pairs: dict[str, KucoinMktPair] = await client.cache_pairs()
+        pairs: dict[str, KucoinMktPair] = await client.get_mkt_pairs()
 
         try:
             # likely search result key which is already in native mkt symbol form
@@ -682,6 +749,8 @@ async def stream_quotes(
 
         log.info(f'Starting up quote stream(s) for {symbols}')
         for sym_str in symbols:
+            mkt: MktPair
+            pair: KucoinMktPair
             mkt, pair = await get_mkt_info(sym_str)
             init_msgs.append(
                 FeedInit(mkt_info=mkt)
@@ -689,7 +758,11 @@ async def stream_quotes(
 
         ws: NoBsWs
         token, ping_interval = await client._get_ws_token()
-        connect_id = str(uuid4())
+        log.info('API reported ping_interval: {ping_interval}\n')
+
+        connect_id: str = str(uuid4())
+        typ: str
+        quote: dict
         async with (
             open_autorecon_ws(
                 (
@@ -703,20 +776,37 @@ async def stream_quotes(
                 ),
             ) as ws,
             open_ping_task(ws, ping_interval, connect_id),
-            aclosing(stream_messages(ws, sym_str)) as msg_gen,
+            aclosing(
+                iter_normed_quotes(
+                    ws, sym_str
+                )
+            ) as iter_quotes,
         ):
-            typ, quote = await anext(msg_gen)
+            typ, quote = await anext(iter_quotes)
 
-            while typ != 'trade':
-                # take care to not unblock here until we get a real
-                # trade quote
-                typ, quote = await anext(msg_gen)
+            # take care to not unblock here until we get a real
+            # trade quote?
+            # ^TODO, remove this right?
+            # -[ ] what often blocks chart boot/new-feed switching
+            #   since we'ere waiting for a live quote instead of just
+            #   loading history afap..
+            #  |_ XXX, not sure if we require a bit of rework to core
+            #    feed init logic or if backends justg gotta be
+            #    changed up.. feel like there was some causality
+            #    dilema prolly only seen with IB too..
+            # while typ != 'trade':
+            #     typ, quote = await anext(iter_quotes)
 
             task_status.started((init_msgs, quote))
             feed_is_live.set()
 
-            async for typ, msg in msg_gen:
-                await send_chan.send({sym_str: msg})
+            # XXX NOTE, DO NOT include the `.<backend>` suffix!
+            # OW the sampling loop will not broadcast correctly..
+            # since `bus._subscribers.setdefault(bs_fqme, set())`
+            # is used inside `.data.open_feed_bus()` !!!
+            topic: str = mkt.bs_fqme
+            async for typ, quote in iter_quotes:
+                await send_chan.send({topic: quote})
 
 
 @acm
@@ -771,7 +861,7 @@ async def subscribe(
             )
 
 
-async def stream_messages(
+async def iter_normed_quotes(
     ws: NoBsWs,
     sym: str,
 
@@ -802,6 +892,9 @@ async def stream_messages(
 
                 yield 'trade', {
                     'symbol': sym,
+                    # TODO, is 'last' even used elsewhere/a-good
+                    # semantic? can't we just read the ticks with our
+                    # .data.ticktools.frame_ticks()`/
                     'last': trade_data.price,
                     'brokerd_ts': last_trade_ts,
                     'ticks': [
@@ -894,7 +987,7 @@ async def open_history_client(
             if end_dt is None:
                 inow = round(time.time())
 
-                print(
+                log.debug(
                     f'difference in time between load and processing'
                     f'{inow - times[-1]}'
                 )

piker/clearing/README.rst

@@ -0,0 +1,49 @@
+piker.clearing
+______________
+trade execution-n-control subsys for both live and paper trading as
+well as algo-trading manual override/interaction across any backend
+broker and data provider.
+
+avail UIs
+*********
+
+order ctl
+---------
+the `piker.clearing` subsys is exposed mainly though
+the `piker chart` GUI as a "chart trader" style UX and
+is automatically enabled whenever a chart is opened.
+
+.. ^TODO, more prose here!
+
+the "manual" order control features are exposed via the
+`piker.ui.order_mode` API and can pretty much always be
+used (at least) in simulated-trading mode, aka "paper"-mode, and
+the micro-manual is as follows:
+
+``order_mode`` (
+    edge triggered activation by any of the following keys,
+    ``mouse-click`` on y-level to submit at that price
+    ):
+
+    - ``f``/ ``ctl-f`` to stage buy
+    - ``d``/ ``ctl-d`` to stage sell
+    - ``a`` to stage alert
+
+
+``search_mode`` (
+    ``ctl-l`` or ``ctl-space`` to open,
+    ``ctl-c`` or ``ctl-space`` to close
+    ) :
+
+    - begin typing to have symbol search automatically lookup
+      symbols from all loaded backend (broker) providers
+    - arrow keys and mouse click to navigate selection
+    - vi-like ``ctl-[hjkl]`` for navigation
+
+
+position (pp) mgmt
+------------------
+you can also configure your position allocation limits from the
+sidepane.
+
+.. ^TODO, explain and provide tut once more refined!

piker/clearing/__init__.py

@@ -27,13 +27,28 @@ from ._ems import (
     open_brokerd_dialog,
 )
 from ._util import OrderDialogs
+from ._messages import(
+    Order,
+    Status,
+    Cancel,
+
+    # TODO: deprecate these and replace end-2-end with
+    # client-side-dialog set above B)
+    # https://github.com/pikers/piker/issues/514
+    BrokerdPosition
+)
 
 
 __all__ = [
+    'FeeModel',
     'open_ems',
     'OrderClient',
     'open_brokerd_dialog',
     'OrderDialogs',
+    'Order',
+    'Status',
+    'Cancel',
+    'BrokerdPosition'
 
 ]
 

piker/clearing/_client.py

@@ -30,7 +30,7 @@ from tractor.trionics import broadcast_receiver
 from ._util import (
     log,  # sub-sys logger
 )
-from ..data.types import Struct
+from piker.types import Struct
 from ..service import maybe_open_emsd
 from ._messages import (
     Order,

piker/clearing/_ems.py

@@ -27,7 +27,7 @@ from contextlib import asynccontextmanager as acm
 from decimal import Decimal
 from math import isnan
 from pprint import pformat
-import time
+from time import time_ns
 from types import ModuleType
 from typing import (
     AsyncIterator,
@@ -51,12 +51,13 @@ from ..accounting._mktinfo import (
     unpack_fqme,
     dec_digits,
 )
+from piker.types import Struct
 from ..ui._notify import notify_from_ems_status_msg
 from ..data import iterticks
-from ..data.types import Struct
 from ._messages import (
     Order,
     Status,
+    Error,
     BrokerdCancel,
     BrokerdOrder,
     # BrokerdOrderAck,
@@ -255,7 +256,7 @@ async def clear_dark_triggers(
                                 action=action,
                                 oid=oid,
                                 account=account,
-                                time_ns=time.time_ns(),
+                                time_ns=time_ns(),
                                 symbol=bfqme,
                                 price=submit_price,
                                 size=size,
@@ -268,7 +269,7 @@ async def clear_dark_triggers(
                     # fallthrough logic
                     status = Status(
                         oid=oid,  # ems dialog id
-                        time_ns=time.time_ns(),
+                        time_ns=time_ns(),
                         resp=resp,
                         req=cmd,
                         brokerd_msg=brokerd_msg,
@@ -826,8 +827,8 @@ async def translate_and_relay_brokerd_events(
                 # keep pps per account up to date locally in ``emsd`` mem
                 # sym, broker = pos_msg.symbol, pos_msg.broker
 
+                # NOTE: translate to a FQME!
                 relay.positions.setdefault(
-                    # NOTE: translate to a FQSN!
                     (broker, pos_msg.account),
                     {}
                 )[pos_msg.symbol] = pos_msg
@@ -883,7 +884,7 @@ async def translate_and_relay_brokerd_events(
                         BrokerdCancel(
                             oid=oid,
                             reqid=reqid,
-                            time_ns=time.time_ns(),
+                            time_ns=time_ns(),
                             account=status_msg.req.account,
                         )
                     )
@@ -898,38 +899,75 @@ async def translate_and_relay_brokerd_events(
                 continue
 
             # BrokerdError
+            # TODO: figure out how this will interact with EMS clients
+            # for ex. on an error do we react with a dark orders
+            # management response, like cancelling all dark orders?
+            # This looks like a supervision policy for pending orders on
+            # some unexpected failure - something we need to think more
+            # about.  In most default situations, with composed orders
+            # (ex.  brackets), most brokers seem to use a oca policy.
             case {
                 'name': 'error',
                 'oid': oid,  # ems order-dialog id
                 'reqid': reqid,  # brokerd generated order-request id
             }:
-                status_msg = book._active.get(oid)
+                if (
+                    not oid
+                    # try to lookup any order dialog by
+                    # brokerd-side id..
+                    and not (
+                        oid := book._ems2brokerd_ids.inverse.get(reqid)
+                    )
+                ):
+                    log.warning(
+                        f'Rxed unusable error-msg:\n'
+                        f'{brokerd_msg}'
+                    )
+                    continue
+
                 msg = BrokerdError(**brokerd_msg)
-                log.error(fmsg)  # XXX make one when it's blank?
 
-                # TODO: figure out how this will interact with EMS clients
-                # for ex. on an error do we react with a dark orders
-                # management response, like cancelling all dark orders?
-                # This looks like a supervision policy for pending orders on
-                # some unexpected failure - something we need to think more
-                # about.  In most default situations, with composed orders
-                # (ex.  brackets), most brokers seem to use a oca policy.
-
-                # only relay to client side if we have an active
-                # ongoing dialog
-                if status_msg:
+                # NOTE: retreive the last client-side response
+                # OR create an error when we have no last msg /dialog
+                # on record
+                status_msg: Status
+                if not (status_msg := book._active.get(oid)):
+                    status_msg = Error(
+                        time_ns=time_ns(),
+                        oid=oid,
+                        reqid=reqid,
+                        brokerd_msg=msg,
+                    )
+                else:
+                    # only modify last status if we have an active
+                    # ongoing dialog..
                     status_msg.resp = 'error'
                     status_msg.brokerd_msg = msg
-                    book._active[oid] = status_msg
 
-                    await router.client_broadcast(
-                        status_msg.req.symbol,
-                        status_msg,
+                book._active[oid] = status_msg
+
+                log.error(
+                    'Translating brokerd error to status:\n'
+                    f'{fmsg}'
+                    f'{status_msg.to_dict()}'
+                )
+                if req := status_msg.req:
+                    fqme: str = req.symbol
+                else:
+                    bdmsg: Struct = status_msg.brokerd_msg
+                    fqme: str = (
+                        bdmsg.symbol  # might be None
+                        or
+                        bdmsg.broker_details['flow']
+                        # NOTE: what happens in empty case in the
+                        # broadcast below? it's a problem?
+                        .get('symbol', '')
                     )
 
-                else:
-                    log.error(f'Error for unknown order flow:\n{msg}')
-                    continue
+                await router.client_broadcast(
+                    fqme,
+                    status_msg,
+                )
 
             # BrokerdStatus
             case {
@@ -1070,7 +1108,7 @@ async def translate_and_relay_brokerd_events(
                     status_msg.req = order
 
                     assert status_msg.src  # source tag?
-                    oid = str(status_msg.reqid)
+                    oid: str = str(status_msg.reqid)
 
                     # attempt to avoid collisions
                     status_msg.reqid = oid
@@ -1087,38 +1125,28 @@ async def translate_and_relay_brokerd_events(
                         status_msg,
                     )
 
-                # don't fall through
-                continue
-
-            # brokerd error
-            case {
-                'name': 'status',
-                'status': 'error',
-            }:
-                log.error(f'Broker error:\n{fmsg}')
-                # XXX: we presume the brokerd cancels its own order
-                continue
-
             # TOO FAST ``BrokerdStatus`` that arrives
             # before the ``BrokerdAck``.
+            # NOTE XXX: sometimes there is a race with the backend (like
+            # `ib` where the pending status will be relayed *before*
+            # the ack msg, in which case we just ignore the faster
+            # pending msg and wait for our expected ack to arrive
+            # later (i.e. the first block below should enter).
             case {
-                # XXX: sometimes there is a race with the backend (like
-                # `ib` where the pending stauts will be related before
-                # the ack, in which case we just ignore the faster
-                # pending msg and wait for our expected ack to arrive
-                # later (i.e. the first block below should enter).
                 'name': 'status',
                 'status': status,
                 'reqid': reqid,
             }:
-                oid = book._ems2brokerd_ids.inverse.get(reqid)
-                msg = f'Unhandled broker status for dialog {reqid}:\n'
-                if oid:
-                    status_msg = book._active.get(oid)
-                    # status msg may not have been set yet or popped?
+                msg = (
+                    f'Unhandled broker status for dialog {reqid}:\n'
+                    f'{pformat(brokerd_msg)}'
+                )
+                if (
+                    oid := book._ems2brokerd_ids.inverse.get(reqid)
+                ):
                     # NOTE: have seen a key error here on kraken
                     # clearable limits..
-                    if status_msg:
+                    if status_msg := book._active.get(oid):
                         msg += (
                             f'last status msg: {pformat(status_msg)}\n\n'
                             f'this msg:{fmsg}\n'
@@ -1214,7 +1242,7 @@ async def process_client_order_cmds(
                         BrokerdCancel(
                             oid=oid,
                             reqid=reqid,
-                            time_ns=time.time_ns(),
+                            time_ns=time_ns(),
                             account=order.account,
                         )
                     )
@@ -1289,7 +1317,7 @@ async def process_client_order_cmds(
 
                 msg = BrokerdOrder(
                     oid=oid,  # no ib support for oids...
-                    time_ns=time.time_ns(),
+                    time_ns=time_ns(),
 
                     # if this is None, creates a new order
                     # otherwise will modify any existing one
@@ -1307,7 +1335,7 @@ async def process_client_order_cmds(
                         oid=oid,
                         reqid=reqid,
                         resp='pending',
-                        time_ns=time.time_ns(),
+                        time_ns=time_ns(),
                         brokerd_msg=msg,
                         req=req,
                     )
@@ -1424,7 +1452,7 @@ async def process_client_order_cmds(
                 status = Status(
                     resp=resp,
                     oid=oid,
-                    time_ns=time.time_ns(),
+                    time_ns=time_ns(),
                     req=req,
                     src='dark',
                 )

piker/clearing/_messages.py

@@ -18,39 +18,14 @@
 Clearing sub-system message and protocols.
 
 """
-# from collections import (
-#     ChainMap,
-#     deque,
-# )
+from __future__ import annotations
 from typing import (
     Literal,
 )
 
 from msgspec import field
 
-from ..data.types import Struct
-
-
-# TODO: a composite for tracking msg flow on 2-legged
-# dialogs.
-# class Dialog(ChainMap):
-#     '''
-#     Msg collection abstraction to easily track the state changes of
-#     a msg flow in one high level, query-able and immutable construct.
-
-#     The main use case is to query data from a (long-running)
-#     msg-transaction-sequence
-
-
-#     '''
-#     def update(
-#         self,
-#         msg,
-#     ) -> None:
-#         self.maps.insert(0, msg.to_dict())
-
-#     def flatten(self) -> dict:
-#         return dict(self)
+from piker.types import Struct
 
 
 # TODO: ``msgspec`` stuff worth paying attention to:
@@ -163,6 +138,18 @@ class Status(Struct):
     brokerd_msg: dict = {}
 
 
+class Error(Status):
+    resp: str = 'error'
+
+    # TODO: allow re-wrapping from existing (last) status?
+    @classmethod
+    def from_status(
+        cls,
+        msg: Status,
+    ) -> Error:
+        ...
+
+
 # ---------------
 # emsd -> brokerd
 # ---------------
@@ -226,6 +213,7 @@ class BrokerdOrderAck(Struct):
 
     # emsd id originally sent in matching request msg
     oid: str
+    # TODO: do we need this?
     account: str = ''
     name: str = 'ack'
 
@@ -238,13 +226,14 @@ class BrokerdStatus(Struct):
         'open',
         'canceled',
         'pending',
-        'error',
+        # 'error',  # NOTE: use `BrokerdError`
         'closed',
     ]
+    name: str = 'status'
 
+    oid: str = ''
     # TODO: do we need this?
     account: str | None = None,
-    name: str = 'status'
     filled: float = 0.0
     reason: str = ''
     remaining: float = 0.0
@@ -287,20 +276,22 @@ class BrokerdError(Struct):
     This is still a TODO thing since we're not sure how to employ it yet.
 
     '''
-    oid: str
     reason: str
 
     # TODO: drop this right?
     symbol: str | None = None
 
+    oid: str | None = None
     # if no brokerd order request was actually submitted (eg. we errored
     # at the ``pikerd`` layer) then there will be ``reqid`` allocated.
-    reqid: int | str | None = None
+    reqid: str | None = None
 
     name: str = 'error'
     broker_details: dict = {}
 
 
+# TODO: yeah, so we REALLY need to completely deprecate
+# this and use the `.accounting.Position` msg-type instead..
 class BrokerdPosition(Struct):
     '''
     Position update event from brokerd.

piker/clearing/_paper_engine.py

@@ -26,10 +26,12 @@ from contextlib import asynccontextmanager as acm
 from datetime import datetime
 from operator import itemgetter
 import itertools
+from pprint import pformat
 import time
 from typing import (
     Callable,
 )
+from types import ModuleType
 import uuid
 
 from bidict import bidict
@@ -37,22 +39,26 @@ import pendulum
 import trio
 import tractor
 
-from ..brokers import get_brokermod
-from .. import data
-from ..data.types import Struct
-from ..accounting._mktinfo import (
+from piker.brokers import get_brokermod
+from piker.service import find_service
+from piker.accounting import (
+    Account,
     MktPair,
-)
-from ..accounting import (
     Position,
-    PpTable,
     Transaction,
     TransactionLedger,
+    open_account,
     open_trade_ledger,
-    open_pps,
+    unpack_fqme,
 )
-from ..data import iterticks
-from ..accounting import unpack_fqme
+from piker.data import (
+    Feed,
+    SymbologyCache,
+    iterticks,
+    open_feed,
+    open_symcache,
+)
+from piker.types import Struct
 from ._util import (
     log,  # sub-sys logger
     get_console_log,
@@ -77,11 +83,10 @@ class PaperBoi(Struct):
 
     '''
     broker: str
-
     ems_trades_stream: tractor.MsgStream
-
-    ppt: PpTable
+    acnt: Account
     ledger: TransactionLedger
+    fees: Callable
 
     # map of paper "live" orders which be used
     # to simulate fills based on paper engine settings
@@ -263,29 +268,42 @@ class PaperBoi(Struct):
         # we don't actually have any unique backend symbol ourselves
         # other then this thing, our fqme address.
         bs_mktid: str = fqme
+        if fees := self.fees:
+            cost: float = fees(price, size)
+        else:
+            cost: float = 0
+
         t = Transaction(
             fqme=fqme,
-            sym=self._mkts[fqme],
             tid=oid,
             size=size,
             price=price,
-            cost=0,  # TODO: cost model
+            cost=cost,
             dt=pendulum.from_timestamp(fill_time_s),
             bs_mktid=bs_mktid,
         )
 
         # update in-mem ledger and pos table
         self.ledger.update_from_t(t)
-        self.ppt.update_from_trans({oid: t})
+        self.acnt.update_from_ledger(
+            {oid: t},
+            symcache=self.ledger._symcache,
+
+            # XXX when a backend has no symcache support yet we can
+            # simply pass in the gmi() retreived table created
+            # during init :o
+            _mktmap_table=self._mkts,
+        )
 
         # transmit pp msg to ems
-        pp = self.ppt.pps[bs_mktid]
+        pp: Position = self.acnt.pps[bs_mktid]
+
         pp_msg = BrokerdPosition(
             broker=self.broker,
             account='paper',
             symbol=fqme,
 
-            size=pp.size,
+            size=pp.cumsize,
             avg_price=pp.ppu,
 
             # TODO: we need to look up the asset currency from
@@ -296,7 +314,7 @@ class PaperBoi(Struct):
         # write all updates to filesys immediately
         # (adds latency but that works for simulation anyway)
         self.ledger.write_config()
-        self.ppt.write_config()
+        self.acnt.write_config()
 
         await self.ems_trades_stream.send(pp_msg)
 
@@ -325,6 +343,7 @@ async def simulate_fills(
     # this stream may eventually contain multiple symbols
     async for quotes in quote_stream:
         for sym, quote in quotes.items():
+            # print(sym)
             for tick in iterticks(
                 quote,
                 # dark order price filter(s)
@@ -540,139 +559,185 @@ async def open_trade_dialog(
     # enable piker.clearing console log for *this* subactor
     get_console_log(loglevel)
 
-    ppt: PpTable
-    ledger: TransactionLedger
-    with (
-        open_pps(
-            broker,
-            'paper',
-            write_on_exit=True,
-        ) as ppt,
+    symcache: SymbologyCache
+    async with open_symcache(get_brokermod(broker)) as symcache:
 
-        open_trade_ledger(
-            broker,
-            'paper',
-        ) as ledger
-    ):
-        # NOTE: retreive market(pair) info from the backend broker
-        # since ledger entries (in their backend native format) often
-        # don't contain necessary market info per trade record entry..
-        # - if no fqme was passed in, we presume we're running in
-        #   "ledger-sync-only mode" and thus we load mkt info for
-        #   each symbol found in the ledger to a ppt table manually.
+        acnt: Account
+        ledger: TransactionLedger
+        with (
 
-        # TODO: how to process ledger info from backends?
-        # - should we be rolling our own actor-cached version of these
-        # client API refs or using portal IPC to send requests to the
-        # existing brokerd daemon?
-        # - alternatively we can possibly expect and use
-        # a `.broker.norm_trade_records()` ep?
-        brokermod = get_brokermod(broker)
-        gmi = getattr(brokermod, 'get_mkt_info', None)
+            # TODO: probably do the symcache and ledger loading
+            # implicitly behind this? Deliver an account, and ledger
+            # pair or make the ledger an attr of the account?
+            open_account(
+                broker,
+                'paper',
+                write_on_exit=True,
+            ) as acnt,
 
-        # update all transactions with mkt info before
-        # loading any pps
-        mkt_by_fqme: dict[str, MktPair] = {}
-        if fqme:
-            bs_fqme, _, broker = fqme.rpartition('.')
-            mkt, _ = await brokermod.get_mkt_info(bs_fqme)
-            mkt_by_fqme[mkt.fqme] = mkt
+            open_trade_ledger(
+                broker,
+                'paper',
+                symcache=symcache,
+            ) as ledger
+        ):
+            # NOTE: WE MUST retreive market(pair) info from each
+            # backend broker since ledger entries (in their
+            # provider-native format) often don't contain necessary
+            # market info per trade record entry..
+            # FURTHER, if no fqme was passed in, we presume we're
+            # running in "ledger-sync-only mode" and thus we load
+            # mkt info for each symbol found in the ledger to
+            # an acnt table manually.
 
-        # for each sym in the ledger load it's `MktPair` info
-        for tid, txdict in ledger.data.items():
-            l_fqme: str = txdict.get('fqme') or txdict['fqsn']
+            # TODO: how to process ledger info from backends?
+            # - should we be rolling our own actor-cached version of these
+            #   client API refs or using portal IPC to send requests to the
+            #   existing brokerd daemon?
+            # - alternatively we can possibly expect and use
+            #   a `.broker.ledger.norm_trade()` ep?
+            brokermod: ModuleType = get_brokermod(broker)
+            gmi: Callable = getattr(brokermod, 'get_mkt_info', None)
 
+            # update all transactions with mkt info before
+            # loading any pps
+            mkt_by_fqme: dict[str, MktPair] = {}
             if (
-                gmi
-                and l_fqme not in mkt_by_fqme
+                fqme
+                and fqme not in symcache.mktmaps
             ):
-                mkt, pair = await brokermod.get_mkt_info(
-                    l_fqme.rstrip(f'.{broker}'),
+                log.warning(
+                    f'Symcache for {broker} has no `{fqme}` entry?\n'
+                    'Manually requesting mkt map data via `.get_mkt_info()`..'
                 )
-                mkt_by_fqme[l_fqme] = mkt
 
-            # if an ``fqme: str`` input was provided we only
-            # need a ``MktPair`` for that one market, since we're
-            # running in real simulated-clearing mode, not just ledger
-            # syncing.
-            if (
-                fqme is not None
-                and fqme in mkt_by_fqme
-            ):
-                break
+                bs_fqme, _, broker = fqme.rpartition('.')
+                mkt, pair = await gmi(bs_fqme)
+                mkt_by_fqme[mkt.fqme] = mkt
 
-        # update pos table from ledger history and provide a ``MktPair``
-        # lookup for internal position accounting calcs.
-        ppt.update_from_trans(ledger.to_trans(mkt_by_fqme=mkt_by_fqme))
+            # for each sym in the ledger load its `MktPair` info
+            for tid, txdict in ledger.data.items():
+                l_fqme: str = txdict.get('fqme') or txdict['fqsn']
 
-        pp_msgs: list[BrokerdPosition] = []
-        pos: Position
-        token: str  # f'{symbol}.{self.broker}'
-        for token, pos in ppt.pps.items():
-            pp_msgs.append(BrokerdPosition(
-                broker=broker,
-                account='paper',
-                symbol=pos.mkt.fqme,
-                size=pos.size,
-                avg_price=pos.ppu,
+                if (
+                    gmi
+                    and l_fqme not in symcache.mktmaps
+                    and l_fqme not in mkt_by_fqme
+                ):
+                    log.warning(
+                        f'Symcache for {broker} has no `{l_fqme}` entry?\n'
+                        'Manually requesting mkt map data via `.get_mkt_info()`..'
+                    )
+                    mkt, pair = await gmi(
+                        l_fqme.rstrip(f'.{broker}'),
+                    )
+                    mkt_by_fqme[l_fqme] = mkt
+
+                # if an ``fqme: str`` input was provided we only
+                # need a ``MktPair`` for that one market, since we're
+                # running in real simulated-clearing mode, not just ledger
+                # syncing.
+                if (
+                    fqme is not None
+                    and fqme in mkt_by_fqme
+                ):
+                    break
+
+            # update pos table from ledger history and provide a ``MktPair``
+            # lookup for internal position accounting calcs.
+            acnt.update_from_ledger(
+                ledger,
+
+                # NOTE: if the symcache fails on fqme lookup
+                # (either sycache not yet supported or not filled
+                # in) use manually constructed table from calling
+                # the `.get_mkt_info()` provider EP above.
+                _mktmap_table=mkt_by_fqme,
+            )
+
+            pp_msgs: list[BrokerdPosition] = []
+            pos: Position
+            token: str  # f'{symbol}.{self.broker}'
+            for token, pos in acnt.pps.items():
+
+                pp_msgs.append(BrokerdPosition(
+                    broker=broker,
+                    account='paper',
+                    symbol=pos.mkt.fqme,
+                    size=pos.cumsize,
+                    avg_price=pos.ppu,
+                ))
+
+            await ctx.started((
+                pp_msgs,
+                ['paper'],
             ))
 
-        await ctx.started((
-            pp_msgs,
-            ['paper'],
-        ))
+            # write new positions state in case ledger was
+            # newer then that tracked in pps.toml
+            acnt.write_config()
 
-        # write new positions state in case ledger was
-        # newer then that tracked in pps.toml
-        ppt.write_config()
-
-        # exit early since no fqme was passed,
-        # normally this case is just to load
-        # positions "offline".
-        if fqme is None:
-            log.warning(
-                'Paper engine only running in position delivery mode!\n'
-                'NO SIMULATED CLEARING LOOP IS ACTIVE!'
-            )
-            await trio.sleep_forever()
-            return
-
-        async with (
-            data.open_feed(
-                [fqme],
-                loglevel=loglevel,
-            ) as feed,
-        ):
-            # sanity check all the mkt infos
-            for fqme, flume in feed.flumes.items():
-                assert mkt_by_fqme[fqme] == flume.mkt
+            # exit early since no fqme was passed,
+            # normally this case is just to load
+            # positions "offline".
+            if fqme is None:
+                log.warning(
+                    'Paper engine only running in position delivery mode!\n'
+                    'NO SIMULATED CLEARING LOOP IS ACTIVE!'
+                )
+                await trio.sleep_forever()
+                return
 
+            feed: Feed
             async with (
-                ctx.open_stream() as ems_stream,
-                trio.open_nursery() as n,
+                open_feed(
+                    [fqme],
+                    loglevel=loglevel,
+                ) as feed,
             ):
-                client = PaperBoi(
-                    broker=broker,
-                    ems_trades_stream=ems_stream,
-                    ppt=ppt,
-                    ledger=ledger,
-
-                    _buys=_buys,
-                    _sells=_sells,
-                    _reqids=_reqids,
-
-                    _mkts=mkt_by_fqme,
+                # sanity check all the mkt infos
+                for fqme, flume in feed.flumes.items():
+                    mkt: MktPair = symcache.mktmaps.get(fqme) or mkt_by_fqme[fqme]
+                    if mkt != flume.mkt:
+                        diff: tuple = mkt - flume.mkt
+                        log.warning(
+                            'MktPair sig mismatch?\n'
+                            f'{pformat(diff)}'
+                        )
 
+                get_cost: Callable = getattr(
+                    brokermod,
+                    'get_cost',
+                    None,
                 )
 
-                n.start_soon(
-                    handle_order_requests,
-                    client,
-                    ems_stream,
-                )
+                async with (
+                    ctx.open_stream() as ems_stream,
+                    trio.open_nursery() as n,
+                ):
+                    client = PaperBoi(
+                        broker=broker,
+                        ems_trades_stream=ems_stream,
+                        acnt=acnt,
+                        ledger=ledger,
+                        fees=get_cost,
 
-                # paper engine simulator clearing task
-                await simulate_fills(feed.streams[broker], client)
+                        _buys=_buys,
+                        _sells=_sells,
+                        _reqids=_reqids,
+
+                        _mkts=mkt_by_fqme,
+
+                    )
+
+                    n.start_soon(
+                        handle_order_requests,
+                        client,
+                        ems_stream,
+                    )
+
+                    # paper engine simulator clearing task
+                    await simulate_fills(feed.streams[broker], client)
 
 
 @acm
@@ -696,7 +761,7 @@ async def open_paperboi(
     service_name = f'paperboi.{broker}'
 
     async with (
-        tractor.find_actor(service_name) as portal,
+        find_service(service_name) as portal,
         tractor.open_nursery() as an,
     ):
         # NOTE: only spawn if no paperboi already is up since we likely
@@ -719,7 +784,59 @@ async def open_paperboi(
         ) as (ctx, first):
             yield ctx, first
 
-            # tear down connection and any spawned actor on exit
+            # ALWAYS tear down connection AND any newly spawned
+            # paperboi actor on exit!
             await ctx.cancel()
+
             if we_spawned:
                 await portal.cancel_actor()
+
+
+def norm_trade(
+    tid: str,
+    txdict: dict,
+    pairs: dict[str, Struct],
+    symcache: SymbologyCache | None = None,
+
+    brokermod: ModuleType | None = None,
+
+) -> Transaction:
+    from pendulum import (
+        DateTime,
+        parse,
+    )
+
+    # special field handling for datetimes
+    # to ensure pendulum is used!
+    dt: DateTime = parse(txdict['dt'])
+    expiry: str | None = txdict.get('expiry')
+    fqme: str = txdict.get('fqme') or txdict.pop('fqsn')
+
+    price: float = txdict['price']
+    size: float =  txdict['size']
+    cost: float = txdict.get('cost', 0)
+    if (
+        brokermod
+        and (get_cost := getattr(
+            brokermod,
+            'get_cost',
+            False,
+        ))
+    ):
+        cost = get_cost(
+            price,
+            size,
+            is_taker=True,
+        )
+
+    return Transaction(
+        fqme=fqme,
+        tid=txdict['tid'],
+        dt=dt,
+        price=price,
+        size=size,
+        cost=cost,
+        bs_mktid=txdict['bs_mktid'],
+        expiry=parse(expiry) if expiry else None,
+        etype='clear',
+    )

piker/clearing/_util.py

@@ -25,7 +25,7 @@ from ..log import (
     get_logger,
     get_console_log,
 )
-from piker.data.types import Struct
+from piker.types import Struct
 subsys: str = 'piker.clearing'
 
 log = get_logger(subsys)
@@ -36,9 +36,6 @@ get_console_log = partial(
 )
 
 
-# TODO: use this in other backends like kraken which currently has
-# a less formalized version more or less:
-# `apiflows[reqid].maps.append(status_msg.to_dict())`
 class OrderDialogs(Struct):
     '''
     Order control dialog (and thus transaction) tracking via

piker/cli/__init__.py

@@ -1,30 +1,33 @@
 # piker: trading gear for hackers
-# Copyright (C) 2018-present  Tyler Goodlet (in stewardship of pikers)
+# Copyright (C) 2018-present Tyler Goodlet
+# (in stewardship for pikers, everywhere.)
 
-# This program is free software: you can redistribute it and/or modify
-# it under the terms of the GNU Affero General Public License as published by
-# the Free Software Foundation, either version 3 of the License, or
-# (at your option) any later version.
+# This program is free software: you can redistribute it and/or
+# modify it under the terms of the GNU Affero General Public
+# License as published by the Free Software Foundation, either
+# version 3 of the License, or (at your option) any later version.
 
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU Affero General Public License for more details.
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# Affero General Public License for more details.
 
-# You should have received a copy of the GNU Affero General Public License
-# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+# You should have received a copy of the GNU Affero General Public
+# License along with this program.  If not, see
+# <https://www.gnu.org/licenses/>.
 
 '''
 CLI commons.
 
 '''
 import os
-from contextlib import AsyncExitStack
+# from contextlib import AsyncExitStack
 from types import ModuleType
 
 import click
 import trio
 import tractor
+from tractor._multiaddr import parse_maddr
 
 from ..log import (
     get_console_log,
@@ -42,89 +45,175 @@ from .. import config
 log = get_logger('piker.cli')
 
 
+def load_trans_eps(
+    network: dict | None = None,
+    maddrs: list[tuple] | None = None,
+
+) -> dict[str, dict[str, dict]]:
+
+    # transport-oriented endpoint multi-addresses
+    eps: dict[
+        str,  # service name, eg. `pikerd`, `emsd`..
+
+        # libp2p style multi-addresses parsed into prot layers
+        list[dict[str, str | int]]
+    ] = {}
+
+    if (
+        network
+        and not maddrs
+    ):
+        # load network section and (attempt to) connect all endpoints
+        # which are reachable B)
+        for key, maddrs in network.items():
+            match key:
+
+                # TODO: resolve table across multiple discov
+                # prots Bo
+                case 'resolv':
+                    pass
+
+                case 'pikerd':
+                    dname: str = key
+                    for maddr in maddrs:
+                        layers: dict = parse_maddr(maddr)
+                        eps.setdefault(
+                            dname,
+                            [],
+                        ).append(layers)
+
+    elif maddrs:
+        # presume user is manually specifying the root actor ep.
+        eps['pikerd'] = [parse_maddr(maddr)]
+
+    return eps
+
+
 @click.command()
-@click.option('--loglevel', '-l', default='warning', help='Logging level')
-@click.option('--tl', is_flag=True, help='Enable tractor logging')
-@click.option('--pdb', is_flag=True, help='Enable tractor debug mode')
-@click.option('--host', '-h', default=None, help='Host addr to bind')
-@click.option('--port', '-p', default=None, help='Port number to bind')
 @click.option(
-    '--tsdb',
-    is_flag=True,
-    help='Enable local ``marketstore`` instance'
+    '--loglevel',
+    '-l',
+    default='warning',
+    help='Logging level',
 )
 @click.option(
-    '--es',
+    '--tl',
     is_flag=True,
-    help='Enable local ``elasticsearch`` instance'
+    help='Enable tractor-runtime logs',
 )
+@click.option(
+    '--pdb',
+    is_flag=True,
+    help='Enable tractor debug mode',
+)
+@click.option(
+    '--maddr',
+    '-m',
+    default=None,
+    help='Multiaddrs to bind or contact',
+)
+# @click.option(
+#     '--tsdb',
+#     is_flag=True,
+#     help='Enable local ``marketstore`` instance'
+# )
+# @click.option(
+#     '--es',
+#     is_flag=True,
+#     help='Enable local ``elasticsearch`` instance'
+# )
 def pikerd(
+    maddr: list[str] | None,
     loglevel: str,
-    host: str,
-    port: int,
     tl: bool,
     pdb: bool,
-    tsdb: bool,
-    es: bool,
+    # tsdb: bool,
+    # es: bool,
 ):
     '''
     Spawn the piker broker-daemon.
 
     '''
-    log = get_console_log(loglevel, name='cli')
+    from tractor.devx import maybe_open_crash_handler
+    with maybe_open_crash_handler(pdb=pdb):
+        log = get_console_log(loglevel, name='cli')
 
-    if pdb:
-        log.warning((
-            "\n"
-            "!!! YOU HAVE ENABLED DAEMON DEBUG MODE !!!\n"
-            "When a `piker` daemon crashes it will block the "
-            "task-thread until resumed from console!\n"
-            "\n"
-        ))
+        if pdb:
+            log.warning((
+                "\n"
+                "!!! YOU HAVE ENABLED DAEMON DEBUG MODE !!!\n"
+                "When a `piker` daemon crashes it will block the "
+                "task-thread until resumed from console!\n"
+                "\n"
+            ))
 
-    reg_addr: None | tuple[str, int] = None
-    if host or port:
-        reg_addr = (
-            host or _default_registry_host,
-            int(port) or _default_registry_port,
+        # service-actor registry endpoint socket-address set
+        regaddrs: list[tuple[str, int]] = []
+
+        conf, _ = config.load(
+            conf_name='conf',
         )
-
-    from .. import service
-
-    async def main():
-        service_mngr: service.Services
-
-        async with (
-            service.open_pikerd(
-                loglevel=loglevel,
-                debug_mode=pdb,
-                registry_addr=reg_addr,
-
-            ) as service_mngr,  # normally delivers a ``Services`` handle
-
-            AsyncExitStack() as stack,
+        network: dict = conf.get('network')
+        if (
+            network is None
+            and not maddr
         ):
-            if tsdb:
-                dname, conf = await stack.enter_async_context(
-                    service.marketstore.start_ahab_daemon(
-                        service_mngr,
-                        loglevel=loglevel,
-                    )
-                )
-                log.info(f'TSDB `{dname}` up with conf:\n{conf}')
+            regaddrs = [(
+                _default_registry_host,
+                _default_registry_port,
+            )]
 
-            if es:
-                dname, conf = await stack.enter_async_context(
-                    service.elastic.start_ahab_daemon(
-                        service_mngr,
-                        loglevel=loglevel,
-                    )
-                )
-                log.info(f'DB `{dname}` up with conf:\n{conf}')
+        else:
+            eps: dict = load_trans_eps(
+                network,
+                maddr,
+            )
+            for layers in eps['pikerd']:
+                regaddrs.append((
+                    layers['ipv4']['addr'],
+                    layers['tcp']['port'],
+                ))
 
-            await trio.sleep_forever()
+        from .. import service
 
-    trio.run(main)
+        async def main():
+            service_mngr: service.Services
+
+            async with (
+                service.open_pikerd(
+                    registry_addrs=regaddrs,
+                    loglevel=loglevel,
+                    debug_mode=pdb,
+
+                ) as service_mngr,  # normally delivers a ``Services`` handle
+
+                # AsyncExitStack() as stack,
+            ):
+                # TODO: spawn all other sub-actor daemons according to
+                # multiaddress endpoint spec defined by user config
+                assert service_mngr
+
+                # if tsdb:
+                #     dname, conf = await stack.enter_async_context(
+                #         service.marketstore.start_ahab_daemon(
+                #             service_mngr,
+                #             loglevel=loglevel,
+                #         )
+                #     )
+                #     log.info(f'TSDB `{dname}` up with conf:\n{conf}')
+
+                # if es:
+                #     dname, conf = await stack.enter_async_context(
+                #         service.elastic.start_ahab_daemon(
+                #             service_mngr,
+                #             loglevel=loglevel,
+                #         )
+                #     )
+                #     log.info(f'DB `{dname}` up with conf:\n{conf}')
+
+                await trio.sleep_forever()
+
+        trio.run(main)
 
 
 @click.group(context_settings=config._context_defaults)
@@ -137,8 +226,24 @@ def pikerd(
 @click.option('--loglevel', '-l', default='warning', help='Logging level')
 @click.option('--tl', is_flag=True, help='Enable tractor logging')
 @click.option('--configdir', '-c', help='Configuration directory')
-@click.option('--host', '-h', default=None, help='Host addr to bind')
-@click.option('--port', '-p', default=None, help='Port number to bind')
+@click.option(
+    '--pdb',
+    is_flag=True,
+    help='Enable runtime debug mode ',
+)
+@click.option(
+    '--maddr',
+    '-m',
+    default=None,
+    multiple=True,
+    help='Multiaddr to bind',
+)
+@click.option(
+    '--regaddr',
+    '-r',
+    default=None,
+    help='Registrar addr to contact',
+)
 @click.pass_context
 def cli(
     ctx: click.Context,
@@ -146,8 +251,11 @@ def cli(
     loglevel: str,
     tl: bool,
     configdir: str,
-    host: str,
-    port: int,
+    pdb: bool,
+
+    # TODO: make these list[str] with multiple -m maddr0 -m maddr1
+    maddr: list[str],
+    regaddr: str,
 
 ) -> None:
     if configdir is not None:
@@ -168,12 +276,20 @@ def cli(
     }
     assert brokermods
 
-    reg_addr: None | tuple[str, int] = None
-    if host or port:
-        reg_addr = (
-            host or _default_registry_host,
-            int(port) or _default_registry_port,
-        )
+    # TODO: load endpoints from `conf::[network].pikerd`
+    # - pikerd vs. regd, separate registry daemon?
+    # - expose datad vs. brokerd?
+    # - bind emsd with certain perms on public iface?
+    regaddrs: list[tuple[str, int]] = regaddr or [(
+        _default_registry_host,
+        _default_registry_port,
+    )]
+
+    # TODO: factor [network] section parsing out from pikerd
+    # above and call it here as well.
+    # if maddr:
+    #     for addr in maddr:
+    #         layers: dict = parse_maddr(addr)
 
     ctx.obj.update({
         'brokers': brokers,
@@ -183,7 +299,12 @@ def cli(
         'log': get_console_log(loglevel),
         'confdir': config._config_dir,
         'wl_path': config._watchlists_data_path,
-        'registry_addr': reg_addr,
+        'registry_addrs': regaddrs,
+        'pdb': pdb,  # debug mode flag
+
+        # TODO: endpoint parsing, pinging and binding
+        # on no existing server.
+        # 'maddrs': maddr,
     })
 
     # allow enabling same loglevel in ``tractor`` machinery
@@ -230,8 +351,7 @@ def services(config, tl, ports):
 
 
 def _load_clis() -> None:
-    from ..service import marketstore  # noqa
-    from ..service import elastic  # noqa
+    # from ..service import elastic  # noqa
     from ..brokers import cli  # noqa
     from ..ui import cli  # noqa
     from ..watchlists import cli  # noqa

piker/config.py

@@ -104,14 +104,15 @@ def get_app_dir(
     # `tractor`) with the testing dir and check for it whenever we
     # detect `pytest` is being used (which it isn't under normal
     # operation).
-    if "pytest" in sys.modules:
-        import tractor
-        actor = tractor.current_actor(err_on_no_runtime=False)
-        if actor:  # runtime is up
-            rvs = tractor._state._runtime_vars
-            testdirpath = Path(rvs['piker_vars']['piker_test_dir'])
-            assert testdirpath.exists(), 'piker test harness might be borked!?'
-            app_name = str(testdirpath)
+    # if "pytest" in sys.modules:
+    #     import tractor
+    #     actor = tractor.current_actor(err_on_no_runtime=False)
+    #     if actor:  # runtime is up
+    #         rvs = tractor._state._runtime_vars
+    #         import pdbp; pdbp.set_trace()
+    #         testdirpath = Path(rvs['piker_vars']['piker_test_dir'])
+    #         assert testdirpath.exists(), 'piker test harness might be borked!?'
+    #         app_name = str(testdirpath)
 
     if platform.system() == 'Windows':
         key = "APPDATA" if roaming else "LOCALAPPDATA"
@@ -134,14 +135,19 @@ def get_app_dir(
 
 _click_config_dir: Path = Path(get_app_dir('piker'))
 _config_dir: Path = _click_config_dir
-_parent_user: str = os.environ.get('SUDO_USER')
 
-if _parent_user:
+# NOTE: when using `sudo` we attempt to determine the non-root user
+# and still use their normal config dir.
+if (
+    (_parent_user := os.environ.get('SUDO_USER'))
+    and
+    _parent_user != 'root'
+):
     non_root_user_dir = Path(
         os.path.expanduser(f'~{_parent_user}')
     )
     root: str = 'root'
-    _ccds: str = str(_click_config_dir)  # click config dir string
+    _ccds: str = str(_click_config_dir)  # click config dir as string
     i_tail: int = int(_ccds.rfind(root) + len(root))
     _config_dir = (
         non_root_user_dir
@@ -246,7 +252,8 @@ def repodir() -> Path:
 
 
 def load(
-    conf_name: str = 'brokers',  # appended with .toml suffix
+    # NOTE: always appended with .toml suffix
+    conf_name: str = 'conf',
     path: Path | None = None,
 
     decode: Callable[
@@ -357,7 +364,9 @@ def load_accounts(
 
 ) -> bidict[str, str | None]:
 
-    conf, path = load()
+    conf, path = load(
+        conf_name='brokers',
+    )
     accounts = bidict()
     for provider_name, section in conf.items():
         accounts_section = section.get('accounts')
@@ -378,51 +387,3 @@ def load_accounts(
     accounts['paper'] = None
 
     return accounts
-
-
-# XXX: Recursive getting & setting
-
-def get_value(_dict, _section):
-    subs = _section.split('.')
-    if len(subs) > 1:
-        return get_value(
-            _dict[subs[0]],
-            '.'.join(subs[1:]),
-        )
-
-    else:
-        return _dict[_section]
-
-
-def set_value(_dict, _section, val):
-    subs = _section.split('.')
-    if len(subs) > 1:
-        if subs[0] not in _dict:
-            _dict[subs[0]] = {}
-
-        return set_value(
-            _dict[subs[0]],
-            '.'.join(subs[1:]),
-            val
-        )
-
-    else:
-        _dict[_section] = val
-
-
-def del_value(_dict, _section):
-    subs = _section.split('.')
-    if len(subs) > 1:
-        if subs[0] not in _dict:
-            return
-
-        return del_value(
-            _dict[subs[0]],
-            '.'.join(subs[1:])
-        )
-
-    else:
-        if _section not in _dict:
-            return
-
-        del _dict[_section]

piker/data/__init__.py

@@ -39,18 +39,33 @@ from .feed import (
     open_feed,
 )
 from .flows import Flume
+from ._symcache import (
+    SymbologyCache,
+    open_symcache,
+    get_symcache,
+    match_from_pairs,
+)
+from ._sampling import open_sample_stream
+from ..types import Struct
 
 
-__all__ = [
+__all__: list[str] = [
     'Flume',
     'Feed',
     'open_feed',
     'ShmArray',
     'iterticks',
     'maybe_open_shm_array',
+    'match_from_pairs',
     'attach_shm_array',
     'open_shm_array',
     'get_shm_token',
     'def_iohlcv_fields',
     'def_ohlcv_fields',
+    'open_symcache',
+    'open_sample_stream',
+    'get_symcache',
+    'Struct',
+    'SymbologyCache',
+    'types',
 ]

piker/data/_formatters.py

@@ -1,5 +1,5 @@
 # piker: trading gear for hackers
-# Copyright (C) 2018-present  Tyler Goodlet (in stewardship of piker0)
+# Copyright (C) Tyler Goodlet (in stewardship for pikers)
 
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU Affero General Public License as published by
@@ -13,10 +13,10 @@
 
 # You should have received a copy of the GNU Affero General Public License
 # along with this program.  If not, see <https://www.gnu.org/licenses/>.
-"""
+'''
 Pre-(path)-graphics formatted x/y nd/1d rendering subsystem.
 
-"""
+'''
 from __future__ import annotations
 from typing import (
     Optional,
@@ -39,7 +39,12 @@ if TYPE_CHECKING:
     from ._dataviz import (
         Viz,
     )
-    from .._profile import Profiler
+    from piker.toolz import Profiler
+
+# default gap between bars: "bar gap multiplier"
+# - 0.5 is no overlap between OC arms,
+# - 1.0 is full overlap on each neighbor sample
+BGM: float = 0.16
 
 
 class IncrementalFormatter(msgspec.Struct):
@@ -513,6 +518,7 @@ class IncrementalFormatter(msgspec.Struct):
 
 
 class OHLCBarsFmtr(IncrementalFormatter):
+
     x_offset: np.ndarray = np.array([
         -0.5,
         0,
@@ -604,8 +610,9 @@ class OHLCBarsFmtr(IncrementalFormatter):
         vr: tuple[int, int],
 
         start: int = 0,  # XXX: do we need this?
+
         # 0.5 is no overlap between arms, 1.0 is full overlap
-        w: float = 0.16,
+        gap: float = BGM,
 
     ) -> tuple[
         np.ndarray,
@@ -622,7 +629,7 @@ class OHLCBarsFmtr(IncrementalFormatter):
             array[:-1],
             start,
             bar_w=self.index_step_size,
-            bar_gap=w * self.index_step_size,
+            bar_gap=gap * self.index_step_size,
 
             # XXX: don't ask, due to a ``numba`` bug..
             use_time_index=(self.index_field == 'time'),

piker/data/_pathops.py

@@ -17,11 +17,6 @@
 Super fast ``QPainterPath`` generation related operator routines.
 
 """
-from math import (
-    ceil,
-    floor,
-)
-
 import numpy as np
 from numpy.lib import recfunctions as rfn
 from numba import (
@@ -35,11 +30,6 @@ from numba import (
 # TODO: for ``numba`` typing..
 # from ._source import numba_ohlc_dtype
 from ._m4 import ds_m4
-from .._profile import (
-    Profiler,
-    pg_profile_enabled,
-    ms_slower_then,
-)
 
 
 def xy_downsample(
@@ -135,7 +125,7 @@ def path_arrays_from_ohlc(
     half_w: float = bar_w/2
 
     # TODO: report bug for assert @
-    # /home/goodboy/repos/piker/env/lib/python3.8/site-packages/numba/core/typing/builtins.py:991
+    # ../piker/env/lib/python3.8/site-packages/numba/core/typing/builtins.py:991
     for i, q in enumerate(data[start:], start):
 
         open = q['open']
@@ -237,20 +227,20 @@ def trace_hl(
 
     for i in range(hl.size):
         row = hl[i]
-        l, h = row['low'], row['high']
+        lo, hi = row['low'], row['high']
 
-        up_diff = h - last_l
-        down_diff = last_h - l
+        up_diff = hi - last_l
+        down_diff = last_h - lo
 
         if up_diff > down_diff:
-            out[2*i + 1] = h
+            out[2*i + 1] = hi
             out[2*i] = last_l
         else:
-            out[2*i + 1] = l
+            out[2*i + 1] = lo
             out[2*i] = last_h
 
-        last_l = l
-        last_h = h
+        last_l = lo
+        last_h = hi
 
         x[2*i] = int(i) - margin
         x[2*i + 1] = int(i) + margin

piker/data/_sampling.py

@@ -33,6 +33,11 @@ from typing import (
 )
 
 import tractor
+from tractor import (
+    Context,
+    MsgStream,
+    Channel,
+)
 from tractor.trionics import (
     maybe_open_nursery,
 )
@@ -53,7 +58,10 @@ if TYPE_CHECKING:
     from ._sharedmem import (
         ShmArray,
     )
-    from .feed import _FeedsBus
+    from .feed import (
+        _FeedsBus,
+        Sub,
+    )
 
 
 # highest frequency sample step is 1 second by default, though in
@@ -94,7 +102,7 @@ class Sampler:
         float,
         list[
             float,
-            set[tractor.MsgStream]
+            set[MsgStream]
         ],
     ] = defaultdict(
         lambda: [
@@ -258,8 +266,8 @@ class Sampler:
             f'broadcasting {period_s} -> {last_ts}\n'
             # f'consumers: {subs}'
         )
-        borked: set[tractor.MsgStream] = set()
-        sent: set[tractor.MsgStream] = set()
+        borked: set[MsgStream] = set()
+        sent: set[MsgStream] = set()
         while True:
             try:
                 for stream in (subs - sent):
@@ -314,7 +322,7 @@ class Sampler:
 
 @tractor.context
 async def register_with_sampler(
-    ctx: tractor.Context,
+    ctx: Context,
     period_s: float,
     shms_by_period: dict[float, dict] | None = None,
 
@@ -649,12 +657,7 @@ async def sample_and_broadcast(
             # eventually block this producer end of the feed and
             # thus other consumers still attached.
             sub_key: str = broker_symbol.lower()
-            subs: list[
-                tuple[
-                    tractor.MsgStream | trio.MemorySendChannel,
-                    float | None,  # tick throttle in Hz
-                ]
-            ] = bus.get_subs(sub_key)
+            subs: set[Sub] = bus.get_subs(sub_key)
 
             # NOTE: by default the broker backend doesn't append
             # it's own "name" into the fqme schema (but maybe it
@@ -663,34 +666,40 @@ async def sample_and_broadcast(
             fqme: str = f'{broker_symbol}.{brokername}'
             lags: int = 0
 
-            # TODO: speed up this loop in an AOT compiled lang (like
-            # rust or nim or zig) and/or instead of doing a fan out to
-            # TCP sockets here, we add a shm-style tick queue which
-            # readers can pull from instead of placing the burden of
-            # broadcast on solely on this `brokerd` actor. see issues:
+            # XXX TODO XXX: speed up this loop in an AOT compiled
+            # lang (like rust or nim or zig)!
+            # AND/OR instead of doing a fan out to TCP sockets
+            # here, we add a shm-style tick queue which readers can
+            # pull from instead of placing the burden of broadcast
+            # on solely on this `brokerd` actor. see issues:
             # - https://github.com/pikers/piker/issues/98
             # - https://github.com/pikers/piker/issues/107
 
-            for (stream, tick_throttle) in subs.copy():
+            # for (stream, tick_throttle) in subs.copy():
+            for sub in subs.copy():
+                ipc: MsgStream = sub.ipc
+                throttle: float = sub.throttle_rate
                 try:
                     with trio.move_on_after(0.2) as cs:
-                        if tick_throttle:
+                        if throttle:
+                            send_chan: trio.abc.SendChannel = sub.send_chan
+
                             # this is a send mem chan that likely
                             # pushes to the ``uniform_rate_send()`` below.
                             try:
-                                stream.send_nowait(
+                                send_chan.send_nowait(
                                     (fqme, quote)
                                 )
                             except trio.WouldBlock:
                                 overruns[sub_key] += 1
-                                ctx = stream._ctx
-                                chan = ctx.chan
+                                ctx: Context = ipc._ctx
+                                chan: Channel = ctx.chan
 
                                 log.warning(
                                     f'Feed OVERRUN {sub_key}'
                                     '@{bus.brokername} -> \n'
                                     f'feed @ {chan.uid}\n'
-                                    f'throttle = {tick_throttle} Hz'
+                                    f'throttle = {throttle} Hz'
                                 )
 
                                 if overruns[sub_key] > 6:
@@ -707,33 +716,33 @@ async def sample_and_broadcast(
                                             f'{sub_key}:'
                                             f'{ctx.cid}@{chan.uid}'
                                         )
-                                        await stream.aclose()
+                                        await ipc.aclose()
                                         raise trio.BrokenResourceError
                         else:
-                            await stream.send(
+                            await ipc.send(
                                 {fqme: quote}
                             )
 
                     if cs.cancelled_caught:
                         lags += 1
                         if lags > 10:
-                            await tractor.breakpoint()
+                            await tractor.pause()
 
                 except (
                     trio.BrokenResourceError,
                     trio.ClosedResourceError,
                     trio.EndOfChannel,
                 ):
-                    ctx = stream._ctx
-                    chan = ctx.chan
+                    ctx: Context = ipc._ctx
+                    chan: Channel = ctx.chan
                     if ctx:
                         log.warning(
                             'Dropped `brokerd`-quotes-feed connection:\n'
                             f'{broker_symbol}:'
                             f'{ctx.cid}@{chan.uid}'
                         )
-                    if tick_throttle:
-                        assert stream._closed
+                    if sub.throttle_rate:
+                        assert ipc._closed
 
                     # XXX: do we need to deregister here
                     # if it's done in the fee bus code?
@@ -742,7 +751,7 @@ async def sample_and_broadcast(
                     # since there seems to be some kinda race..
                     bus.remove_subs(
                         sub_key,
-                        {(stream, tick_throttle)},
+                        {sub},
                     )
 
 
@@ -750,7 +759,7 @@ async def uniform_rate_send(
 
     rate: float,
     quote_stream: trio.abc.ReceiveChannel,
-    stream: tractor.MsgStream,
+    stream: MsgStream,
 
     task_status: TaskStatus = trio.TASK_STATUS_IGNORED,
 

piker/data/_sharedmem.py

@@ -34,7 +34,7 @@ import tractor
 
 from ._util import log
 from ._source import def_iohlcv_fields
-from .types import Struct
+from piker.types import Struct
 
 
 def cuckoff_mantracker():

piker/data/_symcache.py

@@ -0,0 +1,510 @@
+# piker: trading gear for hackers
+# Copyright (C) Tyler Goodlet (in stewardship for pikers)
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+'''
+Mega-simple symbology cache via TOML files.
+
+Allow backend data providers and/or brokers to stash their
+symbology sets (aka the meta data we normalize into our
+`.accounting.MktPair` type) to the filesystem for faster lookup and
+offline usage.
+
+'''
+from __future__ import annotations
+from contextlib import (
+    asynccontextmanager as acm,
+)
+from pathlib import Path
+from pprint import pformat
+from typing import (
+    Any,
+    Sequence,
+    Hashable,
+    TYPE_CHECKING,
+)
+from types import ModuleType
+
+from rapidfuzz import process as fuzzy
+import tomli_w  # for fast symbol cache writing
+import tractor
+import trio
+try:
+    import tomllib
+except ModuleNotFoundError:
+    import tomli as tomllib
+from msgspec import field
+
+from piker.log import get_logger
+from piker import config
+from piker.types import Struct
+from piker.brokers import (
+    open_cached_client,
+    get_brokermod,
+)
+
+if TYPE_CHECKING:
+    from ..accounting import (
+        Asset,
+        MktPair,
+    )
+
+log = get_logger('data.cache')
+
+
+class SymbologyCache(Struct):
+    '''
+    Asset meta-data cache which holds lookup tables for 3 sets of
+    market-symbology related struct-types required by the
+    `.accounting` and `.data` subsystems.
+
+    '''
+    mod: ModuleType
+    fp: Path
+
+    # all asset-money-systems descriptions as minimally defined by
+    # in `.accounting.Asset`
+    assets: dict[str, Asset] = field(default_factory=dict)
+
+    # backend-system pairs loaded in provider (schema) specific
+    # structs.
+    pairs: dict[str, Struct] = field(default_factory=dict)
+    # serialized namespace path to the backend's pair-info-`Struct`
+    # defn B)
+    pair_ns_path: tractor.msg.NamespacePath | None = None
+
+    # TODO: piker-normalized `.accounting.MktPair` table?
+    # loaded from the `.pairs` and a normalizer
+    # provided by the backend pkg.
+    mktmaps: dict[str, MktPair] = field(default_factory=dict)
+
+    def write_config(self) -> None:
+
+        # put the backend's pair-struct type ref at the top
+        # of file if possible.
+        cachedict: dict[str, Any] = {
+            'pair_ns_path': str(self.pair_ns_path) or '',
+        }
+
+        # serialize all tables as dicts for TOML.
+        for key, table in {
+            'assets': self.assets,
+            'pairs': self.pairs,
+            'mktmaps': self.mktmaps,
+        }.items():
+            if not table:
+                log.warning(
+                    f'Asset cache table for `{key}` is empty?'
+                )
+                continue
+
+            dct = cachedict[key] = {}
+            for key, struct in table.items():
+                dct[key] = struct.to_dict(include_non_members=False)
+
+        try:
+            with self.fp.open(mode='wb') as fp:
+                tomli_w.dump(cachedict, fp)
+        except TypeError:
+            self.fp.unlink()
+            raise
+
+    async def load(self) -> None:
+        '''
+        Explicitly load the "symbology set" for this provider by using
+        2 required `Client` methods:
+
+          - `.get_assets()`: returning a table of `Asset`s
+          - `.get_mkt_pairs()`: returning a table of pair-`Struct`
+            types, custom defined by the particular backend.
+
+        AND, the required `.get_mkt_info()` module-level endpoint
+        which maps `fqme: str` -> `MktPair`s.
+
+        These tables are then used to fill out the `.assets`, `.pairs` and
+        `.mktmaps` tables on this cache instance, respectively.
+
+        '''
+        async with open_cached_client(self.mod.name) as client:
+
+            if get_assets := getattr(client, 'get_assets', None):
+                assets: dict[str, Asset] = await get_assets()
+                for bs_mktid, asset in assets.items():
+                    self.assets[bs_mktid] = asset
+            else:
+                log.warning(
+                    'No symbology cache `Asset` support for `{provider}`..\n'
+                    'Implement `Client.get_assets()`!'
+                )
+
+            if get_mkt_pairs := getattr(client, 'get_mkt_pairs', None):
+
+                pairs: dict[str, Struct] = await get_mkt_pairs()
+                for bs_fqme, pair in pairs.items():
+
+                    # NOTE: every backend defined pair should
+                    # declare it's ns path for roundtrip
+                    # serialization lookup.
+                    if not getattr(pair, 'ns_path', None):
+                        raise TypeError(
+                            f'Pair-struct for {self.mod.name} MUST define a '
+                            '`.ns_path: str`!\n'
+                            f'{pair}'
+                        )
+
+                    entry = await self.mod.get_mkt_info(pair.bs_fqme)
+                    if not entry:
+                        continue
+
+                    mkt: MktPair
+                    pair: Struct
+                    mkt, _pair = entry
+                    assert _pair is pair, (
+                        f'`{self.mod.name}` backend probably has a '
+                        'keying-symmetry problem between the pair-`Struct` '
+                        'returned from `Client.get_mkt_pairs()`and the '
+                        'module level endpoint: `.get_mkt_info()`\n\n'
+                        "Here's the struct diff:\n"
+                        f'{_pair - pair}'
+                    )
+                    # NOTE XXX: this means backends MUST implement
+                    # a `Struct.bs_mktid: str` field to provide
+                    # a native-keyed map to their own symbol
+                    # set(s).
+                    self.pairs[pair.bs_mktid] = pair
+
+                    # NOTE: `MktPair`s are keyed here using piker's
+                    # internal FQME schema so that search,
+                    # accounting and feed init can be accomplished
+                    # a sane, uniform, normalized basis.
+                    self.mktmaps[mkt.fqme] = mkt
+
+                self.pair_ns_path: str = tractor.msg.NamespacePath.from_ref(
+                    pair,
+                )
+
+            else:
+                log.warning(
+                    'No symbology cache `Pair` support for `{provider}`..\n'
+                    'Implement `Client.get_mkt_pairs()`!'
+                )
+
+        return self
+
+    @classmethod
+    def from_dict(
+        cls: type,
+        data: dict,
+        **kwargs,
+    ) -> SymbologyCache:
+
+        # normal init inputs
+        cache = cls(**kwargs)
+
+        # XXX WARNING: this may break if backend namespacing
+        # changes (eg. `Pair` class def is moved to another
+        # module) in which case you can manually update the
+        # `pair_ns_path` in the symcache file and try again.
+        # TODO: probably a verbose error about this?
+        Pair: type = tractor.msg.NamespacePath(
+            str(data['pair_ns_path'])
+        ).load_ref()
+
+        pairtable = data.pop('pairs')
+        for key, pairtable in pairtable.items():
+
+            # allow each serialized pair-dict-table to declare its
+            # specific struct type's path in cases where a backend
+            # supports multiples (normally with different
+            # schemas..) and we are storing them in a flat `.pairs`
+            # table.
+            ThisPair = Pair
+            if this_pair_type := pairtable.get('ns_path'):
+                ThisPair: type = tractor.msg.NamespacePath(
+                    str(this_pair_type)
+                ).load_ref()
+
+            pair: Struct = ThisPair(**pairtable)
+            cache.pairs[key] = pair
+
+        from ..accounting import (
+            Asset,
+            MktPair,
+        )
+
+        # load `dict` -> `Asset`
+        assettable = data.pop('assets')
+        for name, asdict in assettable.items():
+            cache.assets[name] = Asset.from_msg(asdict)
+
+        # load `dict` -> `MktPair`
+        dne: list[str] = []
+        mkttable = data.pop('mktmaps')
+        for fqme, mktdict in mkttable.items():
+
+            mkt = MktPair.from_msg(mktdict)
+            assert mkt.fqme == fqme
+
+            # sanity check asset refs from those (presumably)
+            # loaded asset set above.
+            src: Asset = cache.assets[mkt.src.name]
+            assert src == mkt.src
+            dst: Asset
+            if not (dst := cache.assets.get(mkt.dst.name)):
+                dne.append(mkt.dst.name)
+                continue
+            else:
+                assert dst.name == mkt.dst.name
+
+            cache.mktmaps[fqme] = mkt
+
+        log.warning(
+            f'These `MktPair.dst: Asset`s DNE says `{cache.mod.name}`?\n'
+            f'{pformat(dne)}'
+        )
+        return cache
+
+    @staticmethod
+    async def from_scratch(
+        mod: ModuleType,
+        fp: Path,
+        **kwargs,
+
+    ) -> SymbologyCache:
+        '''
+        Generate (a) new symcache (contents) entirely from scratch
+        including all (TOML) serialized data and file.
+
+        '''
+        log.info(f'GENERATING symbology cache for `{mod.name}`')
+        cache = SymbologyCache(
+            mod=mod,
+            fp=fp,
+            **kwargs,
+        )
+        await cache.load()
+        cache.write_config()
+        return cache
+
+    def search(
+        self,
+        pattern: str,
+        table: str = 'mktmaps'
+
+    ) -> dict[str, Struct]:
+        '''
+        (Fuzzy) search this cache's `.mktmaps` table, which is
+        keyed by FQMEs, for `pattern: str` and return the best
+        matches in a `dict` including the `MktPair` values.
+
+        '''
+        matches = fuzzy.extract(
+            pattern,
+            getattr(self, table),
+            score_cutoff=50,
+        )
+
+        # repack in dict[fqme, MktPair] form
+        return {
+            item[0].fqme: item[0]
+            for item in matches
+        }
+
+
+# actor-process-local in-mem-cache of symcaches (by backend).
+_caches: dict[str, SymbologyCache] = {}
+
+
+def mk_cachefile(
+    provider: str,
+) -> Path:
+    cachedir: Path = config.get_conf_dir() / '_cache'
+    if not cachedir.is_dir():
+        log.info(f'Creating `nativedb` director: {cachedir}')
+        cachedir.mkdir()
+
+    cachefile: Path = cachedir / f'{str(provider)}.symcache.toml'
+    cachefile.touch()
+    return cachefile
+
+
+@acm
+async def open_symcache(
+    mod_or_name: ModuleType | str,
+
+    reload: bool = False,
+    only_from_memcache: bool = False,  # no API req
+    _no_symcache: bool = False,  # no backend support
+
+) -> SymbologyCache:
+
+    if isinstance(mod_or_name, str):
+        mod = get_brokermod(mod_or_name)
+    else:
+        mod: ModuleType = mod_or_name
+
+    provider: str = mod.name
+    cachefile: Path = mk_cachefile(provider)
+
+    # NOTE: certain backends might not support a symbology cache
+    # (easily) and thus we allow for an empty instance to be loaded
+    # and manually filled in at the whim of the caller presuming
+    # the backend pkg-module is annotated appropriately.
+    if (
+        getattr(mod, '_no_symcache', False)
+        or _no_symcache
+    ):
+        yield SymbologyCache(
+            mod=mod,
+            fp=cachefile,
+        )
+        # don't do nuttin
+        return
+
+    # actor-level cache-cache XD
+    global _caches
+    if not reload:
+        try:
+            yield _caches[provider]
+        except KeyError:
+            msg: str = (
+                f'No asset info cache exists yet for `{provider}`'
+            )
+            if only_from_memcache:
+                raise RuntimeError(msg)
+            else:
+                log.warning(msg)
+
+    # if no cache exists or an explicit reload is requested, load
+    # the provider API and call appropriate endpoints to populate
+    # the mkt and asset tables.
+    if (
+        reload
+        or not cachefile.is_file()
+    ):
+        cache = await SymbologyCache.from_scratch(
+            mod=mod,
+            fp=cachefile,
+        )
+
+    else:
+        log.info(
+            f'Loading EXISTING `{mod.name}` symbology cache:\n'
+            f'> {cachefile}'
+        )
+        import time
+        now = time.time()
+        with cachefile.open('rb') as existing_fp:
+            data: dict[str, dict] = tomllib.load(existing_fp)
+            log.runtime(f'SYMCACHE TOML LOAD TIME: {time.time() - now}')
+
+            # if there's an empty file for some reason we need
+            # to do a full reload as well!
+            if not data:
+                cache = await SymbologyCache.from_scratch(
+                    mod=mod,
+                    fp=cachefile,
+                )
+            else:
+                cache = SymbologyCache.from_dict(
+                    data,
+                    mod=mod,
+                    fp=cachefile,
+                )
+
+
+        # TODO: use a real profiling sys..
+        # https://github.com/pikers/piker/issues/337
+        log.info(f'SYMCACHE LOAD TIME: {time.time() - now}')
+
+    yield cache
+
+    # TODO: write only when changes detected? but that should
+    # never happen right except on reload?
+    # cache.write_config()
+
+
+def get_symcache(
+    provider: str,
+    force_reload: bool = False,
+
+) -> SymbologyCache:
+    '''
+    Get any available symbology/assets cache from sync code by
+    (maybe) manually running `trio` to do the work.
+
+    '''
+    # spawn tractor runtime and generate cache
+    # if not existing.
+    async def sched_gen_symcache():
+        async with (
+            # only for runtime's debug mode
+            tractor.open_nursery(debug_mode=True),
+
+            open_symcache(
+                get_brokermod(provider),
+                reload=force_reload,
+            ) as symcache,
+        ):
+            return symcache
+
+    try:
+        symcache: SymbologyCache = trio.run(sched_gen_symcache)
+        assert symcache
+    except BaseException:
+        import pdbp
+        pdbp.xpm()
+
+    return symcache
+
+
+def match_from_pairs(
+    pairs: dict[str, Struct],
+    query: str,
+    score_cutoff: int = 50,
+    **extract_kwargs,
+
+) -> dict[str, Struct]:
+    '''
+    Fuzzy search over a "pairs table" maintained by most backends
+    as part of their symbology-info caching internals.
+
+    Scan the native symbol key set and return best ranked
+    matches back in a new `dict`.
+
+    '''
+
+    # TODO: somehow cache this list (per call) like we were in
+    # `open_symbol_search()`?
+    keys: list[str] = list(pairs)
+    matches: list[tuple[
+        Sequence[Hashable],  # matching input key
+        Any,  # scores
+        Any,
+    ]] = fuzzy.extract(
+        # NOTE: most backends provide keys uppercased
+        query=query,
+        choices=keys,
+        score_cutoff=score_cutoff,
+        **extract_kwargs,
+    )
+
+    # pop and repack pairs in output dict
+    matched_pairs: dict[str, Struct] = {}
+    for item in matches:
+        pair_key: str = item[0]
+        matched_pairs[pair_key] = pairs[pair_key]
+
+    return matched_pairs

piker/data/_timeseries.py

@@ -1,326 +0,0 @@
-# piker: trading gear for hackers
-# Copyright (C) 2018-present  Tyler Goodlet (in stewardship of pikers)
-
-# This program is free software: you can redistribute it and/or modify
-# it under the terms of the GNU Affero General Public License as published by
-# the Free Software Foundation, either version 3 of the License, or
-# (at your option) any later version.
-
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU Affero General Public License for more details.
-
-# You should have received a copy of the GNU Affero General Public License
-# along with this program.  If not, see <https://www.gnu.org/licenses/>.
-
-'''
-Financial time series processing utilities usually
-pertaining to OHLCV style sampled data.
-
-Routines are generally implemented in either ``numpy`` or ``polars`` B)
-
-'''
-from __future__ import annotations
-from typing import Literal
-from math import (
-    ceil,
-    floor,
-)
-
-import numpy as np
-import polars as pl
-
-from ._sharedmem import ShmArray
-from .._profile import (
-    Profiler,
-    pg_profile_enabled,
-    ms_slower_then,
-)
-
-
-def slice_from_time(
-    arr: np.ndarray,
-    start_t: float,
-    stop_t: float,
-    step: float,  # sampler period step-diff
-
-) -> slice:
-    '''
-    Calculate array indices mapped from a time range and return them in
-    a slice.
-
-    Given an input array with an epoch `'time'` series entry, calculate
-    the indices which span the time range and return in a slice. Presume
-    each `'time'` step increment is uniform and when the time stamp
-    series contains gaps (the uniform presumption is untrue) use
-    ``np.searchsorted()`` binary search to look up the appropriate
-    index.
-
-    '''
-    profiler = Profiler(
-        msg='slice_from_time()',
-        disabled=not pg_profile_enabled(),
-        ms_threshold=ms_slower_then,
-    )
-
-    times = arr['time']
-    t_first = floor(times[0])
-    t_last = ceil(times[-1])
-
-    # the greatest index we can return which slices to the
-    # end of the input array.
-    read_i_max = arr.shape[0]
-
-    # compute (presumed) uniform-time-step index offsets
-    i_start_t = floor(start_t)
-    read_i_start = floor(((i_start_t - t_first) // step)) - 1
-
-    i_stop_t = ceil(stop_t)
-
-    # XXX: edge case -> always set stop index to last in array whenever
-    # the input stop time is detected to be greater then the equiv time
-    # stamp at that last entry.
-    if i_stop_t >= t_last:
-        read_i_stop = read_i_max
-    else:
-        read_i_stop = ceil((i_stop_t - t_first) // step) + 1
-
-    # always clip outputs to array support
-    # for read start:
-    # - never allow a start < the 0 index
-    # - never allow an end index > the read array len
-    read_i_start = min(
-        max(0, read_i_start),
-        read_i_max - 1,
-    )
-    read_i_stop = max(
-        0,
-        min(read_i_stop, read_i_max),
-    )
-
-    # check for larger-then-latest calculated index for given start
-    # time, in which case we do a binary search for the correct index.
-    # NOTE: this is usually the result of a time series with time gaps
-    # where it is expected that each index step maps to a uniform step
-    # in the time stamp series.
-    t_iv_start = times[read_i_start]
-    if (
-        t_iv_start > i_start_t
-    ):
-        # do a binary search for the best index mapping to ``start_t``
-        # given we measured an overshoot using the uniform-time-step
-        # calculation from above.
-
-        # TODO: once we start caching these per source-array,
-        # we can just overwrite ``read_i_start`` directly.
-        new_read_i_start = np.searchsorted(
-            times,
-            i_start_t,
-            side='left',
-        )
-
-        # TODO: minimize binary search work as much as possible:
-        # - cache these remap values which compensate for gaps in the
-        #   uniform time step basis where we calc a later start
-        #   index for the given input ``start_t``.
-        # - can we shorten the input search sequence by heuristic?
-        #   up_to_arith_start = index[:read_i_start]
-
-        if (
-            new_read_i_start <= read_i_start
-        ):
-            # t_diff = t_iv_start - start_t
-            # print(
-            #     f"WE'RE CUTTING OUT TIME - STEP:{step}\n"
-            #     f'start_t:{start_t} -> 0index start_t:{t_iv_start}\n'
-            #     f'diff: {t_diff}\n'
-            #     f'REMAPPED START i: {read_i_start} -> {new_read_i_start}\n'
-            # )
-            read_i_start = new_read_i_start
-
-    t_iv_stop = times[read_i_stop - 1]
-    if (
-        t_iv_stop > i_stop_t
-    ):
-        # t_diff = stop_t - t_iv_stop
-        # print(
-        #     f"WE'RE CUTTING OUT TIME - STEP:{step}\n"
-        #     f'calced iv stop:{t_iv_stop} -> stop_t:{stop_t}\n'
-        #     f'diff: {t_diff}\n'
-        #     # f'SHOULD REMAP STOP: {read_i_start} -> {new_read_i_start}\n'
-        # )
-        new_read_i_stop = np.searchsorted(
-            times[read_i_start:],
-            # times,
-            i_stop_t,
-            side='right',
-        )
-
-        if (
-            new_read_i_stop <= read_i_stop
-        ):
-            read_i_stop = read_i_start + new_read_i_stop + 1
-
-    # sanity checks for range size
-    # samples = (i_stop_t - i_start_t) // step
-    # index_diff = read_i_stop - read_i_start + 1
-    # if index_diff > (samples + 3):
-    #     breakpoint()
-
-    # read-relative indexes: gives a slice where `shm.array[read_slc]`
-    # will be the data spanning the input time range `start_t` ->
-    # `stop_t`
-    read_slc = slice(
-        int(read_i_start),
-        int(read_i_stop),
-    )
-
-    profiler(
-        'slicing complete'
-        # f'{start_t} -> {abs_slc.start} | {read_slc.start}\n'
-        # f'{stop_t} -> {abs_slc.stop} | {read_slc.stop}\n'
-    )
-
-    # NOTE: if caller needs absolute buffer indices they can
-    # slice the buffer abs index like so:
-    # index = arr['index']
-    # abs_indx = index[read_slc]
-    # abs_slc = slice(
-    #     int(abs_indx[0]),
-    #     int(abs_indx[-1]),
-    # )
-
-    return read_slc
-
-
-def detect_null_time_gap(
-    shm: ShmArray,
-    imargin: int = 1,
-
-) -> tuple[float, float] | None:
-    '''
-    Detect if there are any zero-epoch stamped rows in
-    the presumed 'time' field-column.
-
-    Filter to the gap and return a surrounding index range.
-
-    NOTE: for now presumes only ONE gap XD
-
-    '''
-    zero_pred: np.ndarray = shm.array['time'] == 0
-    zero_t: np.ndarray = shm.array[zero_pred]
-    if zero_t.size:
-        istart, iend = zero_t['index'][[0, -1]]
-        start, end = shm._array['time'][
-            [istart - imargin, iend + imargin]
-        ]
-        return (
-            istart - imargin,
-            start,
-            end,
-            iend + imargin,
-        )
-
-    return None
-
-
-t_unit: Literal[
-    'days',
-    'hours',
-    'minutes',
-    'seconds',
-    'miliseconds',
-    'microseconds',
-    'nanoseconds',
-]
-
-
-def with_dts(
-    df: pl.DataFrame,
-    time_col: str = 'time',
-) -> pl.DataFrame:
-    '''
-    Insert datetime (casted) columns to a (presumably) OHLC sampled
-    time series with an epoch-time column keyed by ``time_col``.
-
-    '''
-    return df.with_columns([
-        pl.col(time_col).shift(1).suffix('_prev'),
-        pl.col(time_col).diff().alias('s_diff'),
-        pl.from_epoch(pl.col(time_col)).alias('dt'),
-    ]).with_columns([
-        pl.from_epoch(pl.col(f'{time_col}_prev')).alias('dt_prev'),
-        pl.col('dt').diff().alias('dt_diff'),
-    ]) #.with_columns(
-        # pl.col('dt').diff().dt.days().alias('days_dt_diff'),
-    # )
-
-
-def detect_time_gaps(
-    df: pl.DataFrame,
-
-    time_col: str = 'time',
-    # epoch sampling step diff
-    expect_period: float = 60,
-
-    # datetime diff unit and gap value
-    # crypto mkts
-    # gap_dt_unit: t_unit = 'minutes',
-    # gap_thresh: int = 1,
-
-    # legacy stock mkts
-    gap_dt_unit: t_unit = 'days',
-    gap_thresh: int = 2,
-
-) -> pl.DataFrame:
-    '''
-    Filter to OHLC datums which contain sample step gaps.
-
-    For eg. legacy markets which have venue close gaps and/or
-    actual missing data segments.
-
-    '''
-    dt_gap_col: str = f'{gap_dt_unit}_diff'
-    return with_dts(
-        df
-    ).filter(
-        pl.col('s_diff').abs() > expect_period
-    ).with_columns(
-        getattr(
-            pl.col('dt_diff').dt,
-            gap_dt_unit,  # NOTE: must be valid ``Expr.dt.<name>``
-        )().alias(dt_gap_col)
-    ).filter(
-        pl.col(dt_gap_col).abs() > gap_thresh
-    )
-
-
-def detect_price_gaps(
-    df: pl.DataFrame,
-    gt_multiplier: float = 2.,
-    price_fields: list[str] = ['high', 'low'],
-
-) -> pl.DataFrame:
-    '''
-    Detect gaps in clearing price over an OHLC series.
-
-    2 types of gaps generally exist; up gaps and down gaps:
-
-    - UP gap: when any next sample's lo price is strictly greater
-      then the current sample's hi price.
-
-    - DOWN gap: when any next sample's hi price is strictly
-      less then the current samples lo price.
-
-    '''
-    # return df.filter(
-    #     pl.col('high') - ) > expect_period,
-    # ).select([
-    #     pl.dt.datetime(pl.col(time_col).shift(1)).suffix('_previous'),
-    #     pl.all(),
-    # ]).select([
-    #     pl.all(),
-    #     (pl.col(time_col) - pl.col(f'{time_col}_previous')).alias('diff'),
-    # ])
-    ...

piker/data/_web_bs.py

@@ -50,8 +50,8 @@ from trio_websocket._impl import (
     ConnectionTimeout,
 )
 
+from piker.types import Struct
 from ._util import log
-from .types import Struct
 
 
 class NoBsWs:
@@ -234,10 +234,13 @@ async def _reconnect_forever(
             f'{url} trying (RE)CONNECT'
         )
 
-        async with trio.open_nursery() as n:
-            cs = nobsws._cs = n.cancel_scope
-            ws: WebSocketConnection
-            async with open_websocket_url(url) as ws:
+        ws: WebSocketConnection
+        try:
+            async with (
+                trio.open_nursery() as n,
+                open_websocket_url(url) as ws,
+            ):
+                cs = nobsws._cs = n.cancel_scope
                 nobsws._ws = ws
                 log.info(
                     f'{src_mod}\n'
@@ -269,9 +272,11 @@ async def _reconnect_forever(
                 # to let tasks run **inside** the ws open block above.
                 nobsws._connected.set()
                 await trio.sleep_forever()
+        except HandshakeError:
+            log.exception('Retrying connection')
+
+        # ws & nursery block ends
 
-            # ws open block end
-        # nursery block end
         nobsws._connected = trio.Event()
         if cs.cancelled_caught:
             log.cancel(
@@ -284,7 +289,8 @@ async def _reconnect_forever(
                 and not nobsws._connected.is_set()
             )
 
-        # -> from here, move to next reconnect attempt
+        # -> from here, move to next reconnect attempt iteration
+        # in the while loop above Bp
 
     else:
         log.exception(
@@ -353,8 +359,8 @@ async def open_autorecon_ws(
 
 
 '''
-JSONRPC response-request style machinery for transparent multiplexing of msgs
-over a NoBsWs.
+JSONRPC response-request style machinery for transparent multiplexing
+of msgs over a `NoBsWs`.
 
 '''
 
@@ -371,43 +377,82 @@ async def open_jsonrpc_session(
     url: str,
     start_id: int = 0,
     response_type: type = JSONRPCResult,
-    request_type: Optional[type] = None,
-    request_hook: Optional[Callable] = None,
-    error_hook: Optional[Callable] = None,
+    msg_recv_timeout: float = float('inf'),
+    # ^NOTE, since only `deribit` is using this jsonrpc stuff atm
+    # and options mkts are generally "slow moving"..
+    #
+    # FURTHER if we break the underlying ws connection then since we
+    # don't pass a `fixture` to the task that manages `NoBsWs`, i.e.
+    # `_reconnect_forever()`, the jsonrpc "transport pipe" get's
+    # broken and never restored with wtv init sequence is required to
+    # re-establish a working req-resp session.
+
 ) -> Callable[[str, dict], dict]:
+    '''
+    Init a json-RPC-over-websocket connection to the provided `url`.
+
+    A `json_rpc: Callable[[str, dict], dict` is delivered to the
+    caller for sending requests and a bg-`trio.Task` handles
+    processing of response msgs including error reporting/raising in
+    the parent/caller task.
+
+    '''
+    # NOTE, store all request msgs so we can raise errors on the
+    # caller side!
+    req_msgs: dict[int, dict] = {}
 
     async with (
-        trio.open_nursery() as n,
-        open_autorecon_ws(url) as ws
+        trio.open_nursery() as tn,
+        open_autorecon_ws(
+            url=url,
+            msg_recv_timeout=msg_recv_timeout,
+        ) as ws
     ):
-        rpc_id: Iterable = count(start_id)
+        rpc_id: Iterable[int] = count(start_id)
         rpc_results: dict[int, dict] = {}
 
-        async def json_rpc(method: str, params: dict) -> dict:
+        async def json_rpc(
+            method: str,
+            params: dict,
+        ) -> dict:
             '''
             perform a json rpc call and wait for the result, raise exception in
             case of error field present on response
             '''
+            nonlocal req_msgs
+
+            req_id: int = next(rpc_id)
             msg = {
                 'jsonrpc': '2.0',
-                'id': next(rpc_id),
+                'id': req_id,
                 'method': method,
                 'params': params
             }
             _id = msg['id']
 
-            rpc_results[_id] = {
+            result = rpc_results[_id] = {
                 'result': None,
-                'event': trio.Event()
+                'error': None,
+                'event': trio.Event(),  # signal caller resp arrived
             }
+            req_msgs[_id] = msg
 
             await ws.send_msg(msg)
 
+            # wait for reponse before unblocking requester code
             await rpc_results[_id]['event'].wait()
 
-            ret = rpc_results[_id]['result']
+            if (maybe_result := result['result']):
+                ret = maybe_result
+                del rpc_results[_id]
 
-            del rpc_results[_id]
+            else:
+                err = result['error']
+                raise Exception(
+                    f'JSONRPC request failed\n'
+                    f'req: {msg}\n'
+                    f'resp: {err}\n'
+                )
 
             if ret.error is not None:
                 raise Exception(json.dumps(ret.error, indent=4))
@@ -422,6 +467,7 @@ async def open_jsonrpc_session(
             the server side.
 
             '''
+            nonlocal req_msgs
             async for msg in ws:
                 match msg:
                     case {
@@ -445,19 +491,28 @@ async def open_jsonrpc_session(
                         'params': _,
                     }:
                         log.debug(f'Recieved\n{msg}')
-                        if request_hook:
-                            await request_hook(request_type(**msg))
 
                     case {
                         'error': error
                     }:
-                        log.warning(f'Recieved\n{error}')
-                        if error_hook:
-                            await error_hook(response_type(**msg))
+                        # retreive orig request msg, set error
+                        # response in original "result" msg,
+                        # THEN FINALLY set the event to signal caller
+                        # to raise the error in the parent task.
+                        req_id: int = error['id']
+                        req_msg: dict = req_msgs[req_id]
+                        result: dict = rpc_results[req_id]
+                        result['error'] = error
+                        result['event'].set()
+                        log.error(
+                            f'JSONRPC request failed\n'
+                            f'req: {req_msg}\n'
+                            f'resp: {error}\n'
+                        )
 
                     case _:
                         log.warning(f'Unhandled JSON-RPC msg!?\n{msg}')
 
-        n.start_soon(recv_task)
+        tn.start_soon(recv_task)
         yield json_rpc
-        n.cancel_scope.cancel()
+        tn.cancel_scope.cancel()

piker/data/feed.py

@@ -28,6 +28,7 @@ module.
 from __future__ import annotations
 from collections import (
     defaultdict,
+    abc,
 )
 from contextlib import asynccontextmanager as acm
 from functools import partial
@@ -36,7 +37,6 @@ from types import ModuleType
 from typing import (
     Any,
     AsyncContextManager,
-    Optional,
     Awaitable,
     Sequence,
 )
@@ -45,41 +45,62 @@ import trio
 from trio.abc import ReceiveChannel
 from trio_typing import TaskStatus
 import tractor
-from tractor.trionics import (
-    maybe_open_context,
-    gather_contexts,
-)
+from tractor import trionics
 
-from ..brokers import get_brokermod
-from ..calc import humanize
+from piker.accounting import (
+    MktPair,
+    unpack_fqme,
+)
+from piker.types import Struct
+from piker.brokers import get_brokermod
+from piker.service import (
+    maybe_spawn_brokerd,
+)
+from piker.calc import humanize
 from ._util import (
     log,
     get_console_log,
 )
-from ..service import (
-    maybe_spawn_brokerd,
-)
 from .flows import Flume
 from .validate import (
     FeedInit,
     validate_backend,
 )
-from .history import (
+from ..tsp import (
     manage_history,
 )
 from .ingest import get_ingestormod
-from .types import Struct
-from ..accounting import (
-    MktPair,
-    unpack_fqme,
-)
-from ..ui import _search
 from ._sampling import (
     sample_and_broadcast,
     uniform_rate_send,
 )
 
 
+class Sub(Struct, frozen=True):
+    '''
+    A live feed subscription entry.
+
+    Contains meta-data on the remote-actor type (in functionality
+    terms) as well as refs to IPC streams and sampler runtime
+    params.
+
+    '''
+    ipc: tractor.MsgStream
+    send_chan: trio.abc.SendChannel | None = None
+
+    # tick throttle rate in Hz; determines how live
+    # quotes/ticks should be downsampled before relay
+    # to the receiving remote consumer (process).
+    throttle_rate:  float | None  = None
+    _throttle_cs: trio.CancelScope | None = None
+
+    # TODO: actually stash comms info for the far end to allow
+    # `.tsp`, `.fsp` and `.data._sampling` sub-systems to re-render
+    # the data view as needed via msging with the `._remote_ctl`
+    # ipc ctx.
+    rc_ui: bool = False
+
+
 class _FeedsBus(Struct):
     '''
     Data feeds broadcaster and persistence management.
@@ -104,13 +125,7 @@ class _FeedsBus(Struct):
 
     _subscribers: defaultdict[
         str,
-        set[
-            tuple[
-                tractor.MsgStream | trio.MemorySendChannel,
-                # tractor.Context,
-                float | None,  # tick throttle in Hz
-            ]
-        ]
+        set[Sub]
     ] = defaultdict(set)
 
     async def start_task(
@@ -125,6 +140,8 @@ class _FeedsBus(Struct):
                 trio.CancelScope] = trio.TASK_STATUS_IGNORED,
         ) -> None:
             with trio.CancelScope() as cs:
+                # TODO: shouldn't this be a direct await to avoid
+                # cancellation contagion to the bus nursery!?!?!
                 await self.nursery.start(
                     target,
                     *args,
@@ -142,31 +159,28 @@ class _FeedsBus(Struct):
     def get_subs(
         self,
         key: str,
-    ) -> set[
-        tuple[
-            tractor.MsgStream | trio.MemorySendChannel,
-            float | None,  # tick throttle in Hz
-        ]
-    ]:
+
+    ) -> set[Sub]:
         '''
         Get the ``set`` of consumer subscription entries for the given key.
 
         '''
         return self._subscribers[key]
 
+    def subs_items(self) -> abc.ItemsView[str, set[Sub]]:
+        return self._subscribers.items()
+
     def add_subs(
         self,
         key: str,
-        subs: set[tuple[
-            tractor.MsgStream | trio.MemorySendChannel,
-            float | None,  # tick throttle in Hz
-        ]],
-    ) -> set[tuple]:
+        subs: set[Sub],
+
+    ) -> set[Sub]:
         '''
         Add a ``set`` of consumer subscription entries for the given key.
 
         '''
-        _subs: set[tuple] = self._subscribers[key]
+        _subs: set[Sub] = self._subscribers.setdefault(key, set())
         _subs.update(subs)
         return _subs
 
@@ -331,7 +345,6 @@ async def allocate_persistent_feed(
     ) = await bus.nursery.start(
         manage_history,
         mod,
-        bus,
         mkt,
         some_data_ready,
         feed_is_live,
@@ -408,7 +421,13 @@ async def allocate_persistent_feed(
         rt_shm.array['time'][1] = ts + 1
 
     elif hist_shm.array.size == 0:
-        raise RuntimeError(f'History (1m) Shm for {fqme} is empty!?')
+        for i in range(100):
+            await trio.sleep(0.1)
+            if hist_shm.array.size > 0:
+                break
+        else:
+            await tractor.pause()
+            raise RuntimeError(f'History (1m) Shm for {fqme} is empty!?')
 
     # wait the spawning parent task to register its subscriber
     # send-stream entry before we start the sample loop.
@@ -438,8 +457,9 @@ async def open_feed_bus(
     symbols: list[str],  # normally expected to the broker-specific fqme
 
     loglevel: str = 'error',
-    tick_throttle: Optional[float] = None,
+    tick_throttle: float | None = None,
     start_stream: bool = True,
+    allow_remote_ctl_ui: bool = False,
 
 ) -> dict[
     str,  # fqme
@@ -454,8 +474,12 @@ async def open_feed_bus(
     if loglevel is None:
         loglevel = tractor.current_actor().loglevel
 
-    # XXX: required to propagate ``tractor`` loglevel to piker logging
-    get_console_log(loglevel or tractor.current_actor().loglevel)
+    # XXX: required to propagate ``tractor`` loglevel to piker
+    # logging
+    get_console_log(
+        loglevel
+        or tractor.current_actor().loglevel
+    )
 
     # local state sanity checks
     # TODO: check for any stale shm entries for this symbol
@@ -465,7 +489,7 @@ async def open_feed_bus(
     assert 'brokerd' in servicename
     assert brokername in servicename
 
-    bus = get_feed_bus(brokername)
+    bus: _FeedsBus = get_feed_bus(brokername)
     sub_registered = trio.Event()
 
     flumes: dict[str, Flume] = {}
@@ -512,10 +536,10 @@ async def open_feed_bus(
         # pack for ``.started()`` sync msg
         flumes[fqme] = flume
 
-        # we use the broker-specific fqme (bs_fqme) for the
-        # sampler subscription since the backend isn't (yet) expected to
-        # append it's own name to the fqme, so we filter on keys which
-        # *do not* include that name (e.g .ib) .
+        # we use the broker-specific fqme (bs_fqme) for the sampler
+        # subscription since the backend isn't (yet) expected to
+        # append it's own name to the fqme, so we filter on keys
+        # which *do not* include that name (e.g .ib) .
         bus._subscribers.setdefault(bs_fqme, set())
 
     # sync feed subscribers with flume handles
@@ -554,49 +578,60 @@ async def open_feed_bus(
             # that the ``sample_and_broadcast()`` task (spawned inside
             # ``allocate_persistent_feed()``) will push real-time quote
             # (ticks) to this new consumer.
-
+            cs: trio.CancelScope | None = None
+            send: trio.MemorySendChannel | None = None
             if tick_throttle:
                 flume.throttle_rate = tick_throttle
 
-                # open a bg task which receives quotes over a mem chan
-                # and only pushes them to the target actor-consumer at
-                # a max ``tick_throttle`` instantaneous rate.
+                # open a bg task which receives quotes over a mem
+                # chan and only pushes them to the target
+                # actor-consumer at a max ``tick_throttle``
+                # (instantaneous) rate.
                 send, recv = trio.open_memory_channel(2**10)
 
-                cs = await bus.start_task(
+                # NOTE: the ``.send`` channel here is a swapped-in
+                # trio mem chan which gets `.send()`-ed by the normal
+                # sampler task but instead of being sent directly
+                # over the IPC msg stream it's the throttle task
+                # does the work of incrementally forwarding to the
+                # IPC stream at the throttle rate.
+                cs: trio.CancelScope = await bus.start_task(
                     uniform_rate_send,
                     tick_throttle,
                     recv,
                     stream,
                 )
-                # NOTE: so the ``send`` channel here is actually a swapped
-                # in trio mem chan which gets pushed by the normal sampler
-                # task but instead of being sent directly over the IPC msg
-                # stream it's the throttle task does the work of
-                # incrementally forwarding to the IPC stream at the throttle
-                # rate.
-                send._ctx = ctx  # mock internal ``tractor.MsgStream`` ref
-                sub = (send, tick_throttle)
 
-            else:
-                sub = (stream, tick_throttle)
+            sub = Sub(
+                ipc=stream,
+                send_chan=send,
+                throttle_rate=tick_throttle,
+                _throttle_cs=cs,
+                rc_ui=allow_remote_ctl_ui,
+            )
 
             # TODO: add an api for this on the bus?
             # maybe use the current task-id to key the sub list that's
             # added / removed? Or maybe we can add a general
             # pause-resume by sub-key api?
             bs_fqme = fqme.removesuffix(f'.{brokername}')
-            local_subs.setdefault(bs_fqme, set()).add(sub)
-            bus.add_subs(bs_fqme, {sub})
+            local_subs.setdefault(
+                bs_fqme,
+                set()
+            ).add(sub)
+            bus.add_subs(
+                bs_fqme,
+                {sub}
+            )
 
         # sync caller with all subs registered state
         sub_registered.set()
 
-        uid = ctx.chan.uid
+        uid: tuple[str, str] = ctx.chan.uid
         try:
-            # ctrl protocol for start/stop of quote streams based on UI
-            # state (eg. don't need a stream when a symbol isn't being
-            # displayed).
+            # ctrl protocol for start/stop of live quote streams
+            # based on UI state (eg. don't need a stream when
+            # a symbol isn't being displayed).
             async for msg in stream:
 
                 if msg == 'pause':
@@ -734,6 +769,7 @@ async def install_brokerd_search(
                 except trio.EndOfChannel:
                     return {}
 
+            from piker.ui import _search
             async with _search.register_symbol_search(
 
                 provider_name=brokermod.name,
@@ -752,7 +788,7 @@ async def install_brokerd_search(
 async def maybe_open_feed(
 
     fqmes: list[str],
-    loglevel: Optional[str] = None,
+    loglevel: str | None = None,
 
     **kwargs,
 
@@ -768,7 +804,7 @@ async def maybe_open_feed(
     '''
     fqme = fqmes[0]
 
-    async with maybe_open_context(
+    async with trionics.maybe_open_context(
         acm_func=open_feed,
         kwargs={
             'fqmes': fqmes,
@@ -788,7 +824,7 @@ async def maybe_open_feed(
             # add a new broadcast subscription for the quote stream
             # if this feed is likely already in use
 
-            async with gather_contexts(
+            async with trionics.gather_contexts(
                 mngrs=[stream.subscribe() for stream in feed.streams.values()]
             ) as bstreams:
                 for bstream, flume in zip(bstreams, feed.flumes.values()):
@@ -812,6 +848,8 @@ async def open_feed(
     start_stream: bool = True,
     tick_throttle: float | None = None,  # Hz
 
+    allow_remote_ctl_ui: bool = False,
+
 ) -> Feed:
     '''
     Open a "data feed" which provides streamed real-time quotes.
@@ -848,7 +886,7 @@ async def open_feed(
         )
 
     portals: tuple[tractor.Portal]
-    async with gather_contexts(
+    async with trionics.gather_contexts(
         brokerd_ctxs,
     ) as portals:
 
@@ -894,13 +932,19 @@ async def open_feed(
                     # of these stream open sequences sequentially per
                     # backend? .. need some thot!
                     allow_overruns=True,
+
+                    # NOTE: UI actors (like charts) can allow
+                    # remote control of certain graphics rendering
+                    # capabilities via the
+                    # `.ui._remote_ctl.remote_annotate()` msg loop.
+                    allow_remote_ctl_ui=allow_remote_ctl_ui,
                 )
             )
 
         assert len(feed.mods) == len(feed.portals)
 
         async with (
-            gather_contexts(bus_ctxs) as ctxs,
+            trionics.gather_contexts(bus_ctxs) as ctxs,
         ):
             stream_ctxs: list[tractor.MsgStream] = []
             for (
@@ -942,7 +986,7 @@ async def open_feed(
             brokermod: ModuleType
             fqmes: list[str]
             async with (
-                gather_contexts(stream_ctxs) as streams,
+                trionics.gather_contexts(stream_ctxs) as streams,
             ):
                 for (
                     stream,
@@ -958,6 +1002,12 @@ async def open_feed(
                         if brokermod.name == flume.mkt.broker:
                             flume.stream = stream
 
-                assert len(feed.mods) == len(feed.portals) == len(feed.streams)
+                assert (
+                    len(feed.mods)
+                    ==
+                    len(feed.portals)
+                    ==
+                    len(feed.streams)
+                )
 
                 yield feed

piker/data/flows.py

@@ -30,53 +30,27 @@ import tractor
 import pendulum
 import numpy as np
 
-from ..accounting import MktPair
-from ._util import log
-from .types import Struct
+from piker.types import Struct
 from ._sharedmem import (
     attach_shm_array,
     ShmArray,
     _Token,
 )
-# from .._profile import (
-#     Profiler,
-#     pg_profile_enabled,
-# )
 
 if TYPE_CHECKING:
-    # from pyqtgraph import PlotItem
+    from ..accounting import MktPair
     from .feed import Feed
 
 
-# TODO: ideas for further abstractions as per
-# https://github.com/pikers/piker/issues/216 and
-# https://github.com/pikers/piker/issues/270:
-# - a ``Cascade`` would be the minimal "connection" of 2 ``Flumes``
-#   as per circuit parlance:
-#   https://en.wikipedia.org/wiki/Two-port_network#Cascade_connection
-#     - could cover the combination of our `FspAdmin` and the
-#       backend `.fsp._engine` related machinery to "connect" one flume
-#       to another?
-# - a (financial signal) ``Flow`` would be the a "collection" of such
-#    minmial cascades. Some engineering based jargon concepts:
-#     - https://en.wikipedia.org/wiki/Signal_chain
-#     - https://en.wikipedia.org/wiki/Daisy_chain_(electrical_engineering)
-#     - https://en.wikipedia.org/wiki/Audio_signal_flow
-#     - https://en.wikipedia.org/wiki/Digital_signal_processing#Implementation
-#     - https://en.wikipedia.org/wiki/Dataflow_programming
-#     - https://en.wikipedia.org/wiki/Signal_programming
-#     - https://en.wikipedia.org/wiki/Incremental_computing
-
-
 class Flume(Struct):
     '''
-    Composite reference type which points to all the addressing handles
-    and other meta-data necessary for the read, measure and management
-    of a set of real-time updated data flows.
+    Composite reference type which points to all the addressing
+    handles and other meta-data necessary for the read, measure and
+    management of a set of real-time updated data flows.
 
     Can be thought of as a "flow descriptor" or "flow frame" which
-    describes the high level properties of a set of data flows that can
-    be used seamlessly across process-memory boundaries.
+    describes the high level properties of a set of data flows that
+    can be used seamlessly across process-memory boundaries.
 
     Each instance's sub-components normally includes:
      - a msg oriented quote stream provided via an IPC transport
@@ -99,6 +73,7 @@ class Flume(Struct):
     # private shm refs loaded dynamically from tokens
     _hist_shm: ShmArray | None = None
     _rt_shm: ShmArray | None = None
+    _readonly: bool = True
 
     stream: tractor.MsgStream | None = None
     izero_hist: int = 0
@@ -115,7 +90,7 @@ class Flume(Struct):
         if self._rt_shm is None:
             self._rt_shm = attach_shm_array(
                 token=self._rt_shm_token,
-                readonly=True,
+                readonly=self._readonly,
             )
 
         return self._rt_shm
@@ -128,12 +103,10 @@ class Flume(Struct):
                 'No shm token has been set for the history buffer?'
             )
 
-        if (
-            self._hist_shm is None
-        ):
+        if self._hist_shm is None:
             self._hist_shm = attach_shm_array(
                 token=self._hist_shm_token,
-                readonly=True,
+                readonly=self._readonly,
             )
 
         return self._hist_shm
@@ -152,10 +125,10 @@ class Flume(Struct):
         period and ratio between them.
 
         '''
-        times = self.hist_shm.array['time']
-        end = pendulum.from_timestamp(times[-1])
-        start = pendulum.from_timestamp(times[times != times[-1]][-1])
-        hist_step_size_s = (end - start).seconds
+        times: np.ndarray = self.hist_shm.array['time']
+        end: float | int = pendulum.from_timestamp(times[-1])
+        start: float | int = pendulum.from_timestamp(times[times != times[-1]][-1])
+        hist_step_size_s: float = (end - start).seconds
 
         times = self.rt_shm.array['time']
         end = pendulum.from_timestamp(times[-1])
@@ -175,17 +148,25 @@ class Flume(Struct):
         msg = self.to_dict()
         msg['mkt'] = self.mkt.to_dict()
 
-        # can't serialize the stream or feed objects, it's expected
-        # you'll have a ref to it since this msg should be rxed on
-        # a stream on whatever far end IPC..
+        # NOTE: pop all un-msg-serializable fields:
+        # - `tractor.MsgStream`
+        # - `Feed`
+        # - `Shmarray`
+        # it's expected the `.from_msg()` on the other side
+        # will get instead some kind of msg-compat version
+        # that it can load.
         msg.pop('stream')
         msg.pop('feed')
+        msg.pop('_rt_shm')
+        msg.pop('_hist_shm')
+
         return msg
 
     @classmethod
     def from_msg(
         cls,
         msg: dict,
+        readonly: bool = True,
 
     ) -> dict:
         '''
@@ -194,8 +175,13 @@ class Flume(Struct):
 
         '''
         mkt_msg = msg.pop('mkt')
+        from ..accounting import MktPair  # cycle otherwise..
         mkt = MktPair.from_msg(mkt_msg)
-        return cls(mkt=mkt, **msg)
+        msg |= {'_readonly': readonly}
+        return cls(
+            mkt=mkt,
+            **msg,
+        )
 
     def get_index(
         self,
@@ -233,5 +219,3 @@ class Flume(Struct):
             np.all(np.isin(vlm, -1))
             or np.all(np.isnan(vlm))
         )
-
-

piker/data/history.py

@@ -1,967 +0,0 @@
-# piker: trading gear for hackers
-# Copyright (C) Tyler Goodlet (in stewardship for pikers)
-
-# This program is free software: you can redistribute it and/or modify
-# it under the terms of the GNU Affero General Public License as published by
-# the Free Software Foundation, either version 3 of the License, or
-# (at your option) any later version.
-
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU Affero General Public License for more details.
-
-# You should have received a copy of the GNU Affero General Public License
-# along with this program.  If not, see <https://www.gnu.org/licenses/>.
-
-'''
-Historical data business logic for load, backfill and tsdb storage.
-
-'''
-from __future__ import annotations
-# from collections import (
-#     Counter,
-# )
-from datetime import datetime
-from functools import partial
-# import time
-from types import ModuleType
-from typing import (
-    Callable,
-    TYPE_CHECKING,
-)
-
-import trio
-from trio_typing import TaskStatus
-import tractor
-from pendulum import (
-    Duration,
-    from_timestamp,
-)
-import numpy as np
-
-from ..accounting import (
-    MktPair,
-)
-from ._util import (
-    log,
-)
-from ._sharedmem import (
-    maybe_open_shm_array,
-    ShmArray,
-)
-from ._source import def_iohlcv_fields
-from ._sampling import (
-    open_sample_stream,
-)
-from ..brokers._util import (
-    DataUnavailable,
-)
-
-if TYPE_CHECKING:
-    from bidict import bidict
-    from ..service.marketstore import StorageClient
-    from .feed import _FeedsBus
-
-
-# `ShmArray` buffer sizing configuration:
-_mins_in_day = int(60 * 24)
-# how much is probably dependent on lifestyle
-# but we reco a buncha times (but only on a
-# run-every-other-day kinda week).
-_secs_in_day = int(60 * _mins_in_day)
-_days_in_week: int = 7
-
-_days_worth: int = 3
-_default_hist_size: int = 6 * 365 * _mins_in_day
-_hist_buffer_start = int(
-    _default_hist_size - round(7 * _mins_in_day)
-)
-
-_default_rt_size: int = _days_worth * _secs_in_day
-# NOTE: start the append index in rt buffer such that 1 day's worth
-# can be appenened before overrun.
-_rt_buffer_start = int((_days_worth - 1) * _secs_in_day)
-
-
-def diff_history(
-    array: np.ndarray,
-    append_until_dt: datetime | None = None,
-    prepend_until_dt: datetime | None = None,
-
-) -> np.ndarray:
-
-    # no diffing with tsdb dt index possible..
-    if (
-        prepend_until_dt is None
-        and append_until_dt is None
-    ):
-        return array
-
-    times = array['time']
-
-    if append_until_dt:
-        return array[times < append_until_dt.timestamp()]
-    else:
-        return array[times >= prepend_until_dt.timestamp()]
-
-
-async def shm_push_in_between(
-    shm: ShmArray,
-    to_push: np.ndarray,
-    prepend_index: int,
-
-    update_start_on_prepend: bool = False,
-
-) -> int:
-    shm.push(
-        to_push,
-        prepend=True,
-
-        # XXX: only update the ._first index if no tsdb
-        # segment was previously prepended by the
-        # parent task.
-        update_first=update_start_on_prepend,
-
-        # XXX: only prepend from a manually calculated shm
-        # index if there was already a tsdb history
-        # segment prepended (since then the
-        # ._first.value is going to be wayyy in the
-        # past!)
-        start=(
-            prepend_index
-            if not update_start_on_prepend
-            else None
-        ),
-    )
-    # XXX: extremely important, there can be no checkpoints
-    # in the block above to avoid entering new ``frames``
-    # values while we're pipelining the current ones to
-    # memory...
-    array = shm.array
-    zeros = array[array['low'] == 0]
-    if (
-        0 < zeros.size < 1000
-    ):
-        tractor.breakpoint()
-
-
-
-async def start_backfill(
-    get_hist,
-    mod: ModuleType,
-    mkt: MktPair,
-    shm: ShmArray,
-    timeframe: float,
-
-    backfill_from_shm_index: int,
-    backfill_from_dt: datetime,
-
-    sampler_stream: tractor.MsgStream,
-
-    backfill_until_dt: datetime | None = None,
-    storage: StorageClient | None = None,
-
-    write_tsdb: bool = True,
-
-    task_status: TaskStatus[tuple] = trio.TASK_STATUS_IGNORED,
-
-) -> int:
-
-        # let caller unblock and deliver latest history frame
-        # and use to signal that backfilling the shm gap until
-        # the tsdb end is complete!
-        bf_done = trio.Event()
-        task_status.started(bf_done)
-
-        # based on the sample step size, maybe load a certain amount history
-        update_start_on_prepend: bool = False
-        if backfill_until_dt is None:
-
-            # TODO: drop this right and just expose the backfill
-            # limits inside a [storage] section in conf.toml?
-            # when no tsdb "last datum" is provided, we just load
-            # some near-term history.
-            # periods = {
-            #     1: {'days': 1},
-            #     60: {'days': 14},
-            # }
-
-            # do a decently sized backfill and load it into storage.
-            periods = {
-                1: {'days': 6},
-                60: {'years': 6},
-            }
-            period_duration: int = periods[timeframe]
-
-            update_start_on_prepend = True
-
-            # NOTE: manually set the "latest" datetime which we intend to
-            # backfill history "until" so as to adhere to the history
-            # settings above when the tsdb is detected as being empty.
-            backfill_until_dt = backfill_from_dt.subtract(**period_duration)
-
-
-        # TODO: can we drop this? without conc i don't think this
-        # is necessary any more?
-        # configure async query throttling
-        # rate = config.get('rate', 1)
-        # XXX: legacy from ``trimeter`` code but unsupported now.
-        # erlangs = config.get('erlangs', 1)
-        # avoid duplicate history frames with a set of datetime frame
-        # starts and associated counts of how many duplicates we see
-        # per time stamp.
-        # starts: Counter[datetime] = Counter()
-
-        # conduct "backward history gap filling" where we push to
-        # the shm buffer until we have history back until the
-        # latest entry loaded from the tsdb's table B)
-        last_start_dt: datetime = backfill_from_dt
-        next_prepend_index: int = backfill_from_shm_index
-
-        while last_start_dt > backfill_until_dt:
-
-            log.debug(
-                f'Requesting {timeframe}s frame ending in {last_start_dt}'
-            )
-
-            try:
-                (
-                    array,
-                    next_start_dt,
-                    next_end_dt,
-                ) = await get_hist(
-                    timeframe,
-                    end_dt=last_start_dt,
-                )
-
-            # broker says there never was or is no more history to pull
-            except DataUnavailable:
-                log.warning(
-                    f'NO-MORE-DATA: backend {mod.name} halted history!?'
-                )
-
-                # ugh, what's a better way?
-                # TODO: fwiw, we probably want a way to signal a throttle
-                # condition (eg. with ib) so that we can halt the
-                # request loop until the condition is resolved?
-                return
-
-            # TODO: drop this? see todo above..
-            # if (
-            #     next_start_dt in starts
-            #     and starts[next_start_dt] <= 6
-            # ):
-            #     start_dt = min(starts)
-            #     log.warning(
-            #         f"{mkt.fqme}: skipping duplicate frame @ {next_start_dt}"
-            #     )
-            #     starts[start_dt] += 1
-            #     await tractor.breakpoint()
-            #     continue
-
-            # elif starts[next_start_dt] > 6:
-            #     log.warning(
-            #         f'NO-MORE-DATA: backend {mod.name} before {next_start_dt}?'
-            #     )
-            #     return
-
-            # # only update new start point if not-yet-seen
-            # starts[next_start_dt] += 1
-
-            assert array['time'][0] == next_start_dt.timestamp()
-
-            diff = last_start_dt - next_start_dt
-            frame_time_diff_s = diff.seconds
-
-            # frame's worth of sample-period-steps, in seconds
-            frame_size_s = len(array) * timeframe
-            expected_frame_size_s = frame_size_s + timeframe
-            if frame_time_diff_s > expected_frame_size_s:
-
-                # XXX: query result includes a start point prior to our
-                # expected "frame size" and thus is likely some kind of
-                # history gap (eg. market closed period, outage, etc.)
-                # so just report it to console for now.
-                log.warning(
-                    f'History frame ending @ {last_start_dt} appears to have a gap:\n'
-                    f'{diff} ~= {frame_time_diff_s} seconds'
-                )
-
-            to_push = diff_history(
-                array,
-                prepend_until_dt=backfill_until_dt,
-            )
-            ln = len(to_push)
-            if ln:
-                log.info(f'{ln} bars for {next_start_dt} -> {last_start_dt}')
-
-            else:
-                log.warning(
-                    '0 BARS TO PUSH after diff!?\n'
-                    f'{next_start_dt} -> {last_start_dt}'
-                )
-
-            # bail gracefully on shm allocation overrun/full
-            # condition
-            try:
-                await shm_push_in_between(
-                    shm,
-                    to_push,
-                    prepend_index=next_prepend_index,
-                    update_start_on_prepend=update_start_on_prepend,
-                )
-                await sampler_stream.send({
-                    'broadcast_all': {
-                        'backfilling': (mkt.fqme, timeframe),
-                    },
-                })
-
-                # decrement next prepend point
-                next_prepend_index = next_prepend_index - ln
-                last_start_dt = next_start_dt
-
-            except ValueError as ve:
-                _ve = ve
-                log.error(
-                    f'Shm prepend OVERRUN on: {next_start_dt} -> {last_start_dt}?'
-                )
-
-                if next_prepend_index < ln:
-                    log.warning(
-                        f'Shm buffer can only hold {next_prepend_index} more rows..\n'
-                        f'Appending those from recent {ln}-sized frame, no more!'
-                    )
-
-                to_push = to_push[-next_prepend_index + 1:]
-                await shm_push_in_between(
-                    shm,
-                    to_push,
-                    prepend_index=next_prepend_index,
-                    update_start_on_prepend=update_start_on_prepend,
-                )
-                await sampler_stream.send({
-                    'broadcast_all': {
-                        'backfilling': (mkt.fqme, timeframe),
-                    },
-                })
-
-                # can't push the entire frame? so
-                # push only the amount that can fit..
-                break
-
-            log.info(
-                f'Shm pushed {ln} frame:\n'
-                f'{next_start_dt} -> {last_start_dt}'
-            )
-
-            # FINALLY, maybe write immediately to the tsdb backend for
-            # long-term storage.
-            if (
-                storage is not None
-                and write_tsdb
-            ):
-                log.info(
-                    f'Writing {ln} frame to storage:\n'
-                    f'{next_start_dt} -> {last_start_dt}'
-                )
-
-                if mkt.dst.atype not in {'crypto', 'crypto_currency'}:
-                    # for now, our table key schema is not including
-                    # the dst[/src] source asset token.
-                    col_sym_key: str = mkt.get_fqme(
-                        delim_char='',
-                        without_src=True,
-                    )
-                else:
-                    col_sym_key: str = mkt.get_fqme(delim_char='')
-
-                # TODO: implement parquet append!?
-                await storage.write_ohlcv(
-                    col_sym_key,
-                    shm.array,
-                    timeframe,
-                )
-        else:
-            # finally filled gap
-            log.info(
-                f'Finished filling gap to tsdb start @ {backfill_until_dt}!'
-            )
-            # conduct tsdb timestamp gap detection and backfill any
-            # seemingly missing sequence segments..
-            # TODO: ideally these never exist but somehow it seems
-            # sometimes we're writing zero-ed segments on certain
-            # (teardown) cases?
-            from ._timeseries import detect_null_time_gap
-
-            gap_indices: tuple | None = detect_null_time_gap(shm)
-            while gap_indices:
-                (
-                    istart,
-                    start,
-                    end,
-                    iend,
-                ) = gap_indices
-
-                start_dt = from_timestamp(start)
-                end_dt = from_timestamp(end)
-                (
-                    array,
-                    next_start_dt,
-                    next_end_dt,
-                ) = await get_hist(
-                    timeframe,
-                    start_dt=start_dt,
-                    end_dt=end_dt,
-                )
-
-                # XXX TODO: pretty sure if i plot tsla, btcusdt.binance
-                # and mnq.cme.ib this causes a Qt crash XXDDD
-
-                # make sure we don't overrun the buffer start
-                len_to_push: int = min(iend, array.size)
-                to_push: np.ndarray = array[-len_to_push:]
-                await shm_push_in_between(
-                    shm,
-                    to_push,
-                    prepend_index=iend,
-                    update_start_on_prepend=False,
-                )
-
-                # TODO: UI side needs IPC event to update..
-                # - make sure the UI actually always handles
-                #  this update!
-                # - remember that in the display side, only refersh this
-                #   if the respective history is actually "in view".
-                #   loop
-                await sampler_stream.send({
-                    'broadcast_all': {
-                        'backfilling': (mkt.fqme, timeframe),
-                    },
-                })
-                gap_indices: tuple | None = detect_null_time_gap(shm)
-
-        # XXX: extremely important, there can be no checkpoints
-        # in the block above to avoid entering new ``frames``
-        # values while we're pipelining the current ones to
-        # memory...
-        # await sampler_stream.send('broadcast_all')
-
-        # short-circuit (for now)
-        bf_done.set()
-
-
-async def back_load_from_tsdb(
-    storemod: ModuleType,
-    storage: StorageClient,
-
-    fqme: str,
-
-    tsdb_history: np.ndarray,
-
-    last_tsdb_dt: datetime,
-    latest_start_dt: datetime,
-    latest_end_dt: datetime,
-
-    bf_done: trio.Event,
-
-    timeframe: int,
-    shm: ShmArray,
-):
-    assert len(tsdb_history)
-
-    # sync to backend history task's query/load completion
-    # if bf_done:
-    #     await bf_done.wait()
-
-    # TODO: eventually it'd be nice to not require a shm array/buffer
-    # to accomplish this.. maybe we can do some kind of tsdb direct to
-    # graphics format eventually in a child-actor?
-    if storemod.name == 'nativedb':
-        return
-
-        await tractor.breakpoint()
-        assert shm._first.value == 0
-
-    array = shm.array
-
-    # if timeframe == 1:
-    #     times = shm.array['time']
-    #     assert (times[1] - times[0]) == 1
-
-    if len(array):
-        shm_last_dt = from_timestamp(
-            shm.array[0]['time']
-        )
-    else:
-        shm_last_dt = None
-
-    if last_tsdb_dt:
-        assert shm_last_dt >= last_tsdb_dt
-
-    # do diff against start index of last frame of history and only
-    # fill in an amount of datums from tsdb allows for most recent
-    # to be loaded into mem *before* tsdb data.
-    if (
-        last_tsdb_dt
-        and latest_start_dt
-    ):
-        backfilled_size_s = (
-            latest_start_dt - last_tsdb_dt
-        ).seconds
-        # if the shm buffer len is not large enough to contain
-        # all missing data between the most recent backend-queried frame
-        # and the most recent dt-index in the db we warn that we only
-        # want to load a portion of the next tsdb query to fill that
-        # space.
-        log.info(
-            f'{backfilled_size_s} seconds worth of {timeframe}s loaded'
-        )
-
-    # Load TSDB history into shm buffer (for display) if there is
-    # remaining buffer space.
-
-    time_key: str = 'time'
-    if getattr(storemod, 'ohlc_key_map', False):
-        keymap: bidict = storemod.ohlc_key_map
-        time_key: str = keymap.inverse['time']
-
-    # if (
-    #     not len(tsdb_history)
-    # ):
-    #     return
-
-    tsdb_last_frame_start: datetime = last_tsdb_dt
-    # load as much from storage into shm possible (depends on
-    # user's shm size settings).
-    while shm._first.value > 0:
-
-        tsdb_history = await storage.read_ohlcv(
-            fqme,
-            timeframe=timeframe,
-            end=tsdb_last_frame_start,
-        )
-
-        # # empty query
-        # if not len(tsdb_history):
-        #     break
-
-        next_start = tsdb_history[time_key][0]
-        if next_start >= tsdb_last_frame_start:
-            # no earlier data detected
-            break
-
-        else:
-            tsdb_last_frame_start = next_start
-
-        # TODO: see if there's faster multi-field reads:
-        # https://numpy.org/doc/stable/user/basics.rec.html#accessing-multiple-fields
-        # re-index  with a `time` and index field
-        prepend_start = shm._first.value
-
-        to_push = tsdb_history[-prepend_start:]
-        shm.push(
-            to_push,
-
-            # insert the history pre a "days worth" of samples
-            # to leave some real-time buffer space at the end.
-            prepend=True,
-            # update_first=False,
-            # start=prepend_start,
-            field_map=storemod.ohlc_key_map,
-        )
-
-        log.info(f'Loaded {to_push.shape} datums from storage')
-        tsdb_last_frame_start = tsdb_history[time_key][0]
-
-        # manually trigger step update to update charts/fsps
-        # which need an incremental update.
-        # NOTE: the way this works is super duper
-        # un-intuitive right now:
-        # - the broadcaster fires a msg to the fsp subsystem.
-        # - fsp subsys then checks for a sample step diff and
-        #   possibly recomputes prepended history.
-        # - the fsp then sends back to the parent actor
-        #   (usually a chart showing graphics for said fsp)
-        #   which tells the chart to conduct a manual full
-        #   graphics loop cycle.
-        # await sampler_stream.send('broadcast_all')
-
-
-async def tsdb_backfill(
-    mod: ModuleType,
-    storemod: ModuleType,
-    tn: trio.Nursery,
-
-    storage: StorageClient,
-    mkt: MktPair,
-    shm: ShmArray,
-    timeframe: float,
-
-    sampler_stream: tractor.MsgStream,
-
-    task_status: TaskStatus[
-        tuple[ShmArray, ShmArray]
-    ] = trio.TASK_STATUS_IGNORED,
-
-) -> None:
-
-    get_hist: Callable[
-        [int, datetime, datetime],
-        tuple[np.ndarray, str]
-    ]
-    config: dict[str, int]
-    async with mod.open_history_client(
-        mkt,
-    ) as (get_hist, config):
-        log.info(f'{mod} history client returned backfill config: {config}')
-
-        # get latest query's worth of history all the way
-        # back to what is recorded in the tsdb
-        try:
-            array, mr_start_dt, mr_end_dt = await get_hist(
-                timeframe,
-                end_dt=None,
-            )
-
-        # XXX: timeframe not supported for backend (since
-        # above exception type), terminate immediately since
-        # there's no backfilling possible.
-        except DataUnavailable:
-            task_status.started()
-            return
-
-        times: np.ndarray = array['time']
-
-        # sample period step size in seconds
-        step_size_s = (
-            from_timestamp(times[-1])
-            - from_timestamp(times[-2])
-        ).seconds
-
-        if step_size_s not in (1, 60):
-            log.error(f'Last 2 sample period is off!? -> {step_size_s}')
-            step_size_s = (
-                from_timestamp(times[-2])
-                - from_timestamp(times[-3])
-            ).seconds
-
-        # NOTE: on the first history, most recent history
-        # frame we PREPEND from the current shm ._last index
-        # and thus a gap between the earliest datum loaded here
-        # and the latest loaded from the tsdb may exist!
-        log.info(f'Pushing {array.size} to shm!')
-        shm.push(
-            array,
-            prepend=True,  # append on first frame
-        )
-        backfill_gap_from_shm_index: int = shm._first.value + 1
-
-        # tell parent task to continue
-        task_status.started()
-
-        # loads a (large) frame of data from the tsdb depending
-        # on the db's query size limit; our "nativedb" (using
-        # parquet) generally can load the entire history into mem
-        # but if not then below the remaining history can be lazy
-        # loaded?
-        fqme: str = mkt.fqme
-        tsdb_entry: tuple | None =  await storage.load(
-            fqme,
-            timeframe=timeframe,
-        )
-
-        last_tsdb_dt: datetime | None = None
-        if tsdb_entry:
-            (
-                tsdb_history,
-                first_tsdb_dt,
-                last_tsdb_dt,
-            ) = tsdb_entry
-
-            # calc the index from which the tsdb data should be
-            # prepended, presuming there is a gap between the
-            # latest frame (loaded/read above) and the latest
-            # sample loaded from the tsdb.
-            backfill_diff: Duration =  mr_start_dt - last_tsdb_dt
-            offset_s: float = backfill_diff.in_seconds()
-            offset_samples: int = round(offset_s / timeframe)
-
-            # TODO: see if there's faster multi-field reads:
-            # https://numpy.org/doc/stable/user/basics.rec.html#accessing-multiple-fields
-            # re-index  with a `time` and index field
-            prepend_start = shm._first.value - offset_samples + 1
-
-            # tsdb history is so far in the past we can't fit it in
-            # shm buffer space so simply don't load it!
-            if prepend_start > 0:
-                to_push = tsdb_history[-prepend_start:]
-                shm.push(
-                    to_push,
-
-                    # insert the history pre a "days worth" of samples
-                    # to leave some real-time buffer space at the end.
-                    prepend=True,
-                    # update_first=False,
-                    start=prepend_start,
-                    field_map=storemod.ohlc_key_map,
-                )
-
-                log.info(f'Loaded {to_push.shape} datums from storage')
-
-        # TODO: maybe start history anal and load missing "history
-        # gaps" via backend..
-
-        if timeframe not in (1, 60):
-            raise ValueError(
-                '`piker` only needs to support 1m and 1s sampling '
-                'but ur api is trying to deliver a longer '
-                f'timeframe of {timeframe} seconds..\n'
-                'So yuh.. dun do dat brudder.'
-            )
-        # if there is a gap to backfill from the first
-        # history frame until the last datum loaded from the tsdb
-        # continue that now in the background
-        bf_done = await tn.start(
-            partial(
-                start_backfill,
-                get_hist,
-                mod,
-                mkt,
-                shm,
-                timeframe,
-
-                backfill_from_shm_index=backfill_gap_from_shm_index,
-                backfill_from_dt=mr_start_dt,
-
-                sampler_stream=sampler_stream,
-
-                backfill_until_dt=last_tsdb_dt,
-                storage=storage,
-            )
-        )
-
-
-        # if len(hist_shm.array) < 2:
-        # TODO: there's an edge case here to solve where if the last
-        # frame before market close (at least on ib) was pushed and
-        # there was only "1 new" row pushed from the first backfill
-        # query-iteration, then the sample step sizing calcs will
-        # break upstream from here since you can't diff on at least
-        # 2 steps... probably should also add logic to compute from
-        # the tsdb series and stash that somewhere as meta data on
-        # the shm buffer?.. no se.
-
-        # backload any further data from tsdb (concurrently per
-        # timeframe) if not all data was able to be loaded (in memory)
-        # from the ``StorageClient.load()`` call above.
-        try:
-            await trio.sleep_forever()
-        finally:
-            return
-
-        # IF we need to continue backloading incrementall from the
-        # tsdb client..
-        tn.start_soon(
-            back_load_from_tsdb,
-
-            storemod,
-            storage,
-            fqme,
-
-            tsdb_history,
-            last_tsdb_dt,
-            mr_start_dt,
-            mr_end_dt,
-            bf_done,
-
-            timeframe,
-            shm,
-        )
-
-
-async def manage_history(
-    mod: ModuleType,
-    bus: _FeedsBus,
-    mkt: MktPair,
-    some_data_ready: trio.Event,
-    feed_is_live: trio.Event,
-    timeframe: float = 60,  # in seconds
-
-    task_status: TaskStatus[
-        tuple[ShmArray, ShmArray]
-    ] = trio.TASK_STATUS_IGNORED,
-
-) -> None:
-    '''
-    Load and manage historical data including the loading of any
-    available series from any connected tsdb as well as conduct
-    real-time update of both that existing db and the allocated
-    shared memory buffer.
-
-    Init sequence:
-    - allocate shm (numpy array) buffers for 60s & 1s sample rates
-    - configure "zero index" for each buffer: the index where
-      history will prepended *to* and new live data will be
-      appened *from*.
-    - open a ``.storage.StorageClient`` and load any existing tsdb
-      history as well as (async) start a backfill task which loads
-      missing (newer) history from the data provider backend:
-      - tsdb history is loaded first and pushed to shm ASAP.
-      - the backfill task loads the most recent history before
-        unblocking its parent task, so that the `ShmArray._last` is
-        up to date to allow the OHLC sampler to begin writing new
-        samples as the correct buffer index once the provider feed
-        engages.
-
-    '''
-    # TODO: is there a way to make each shm file key
-    # actor-tree-discovery-addr unique so we avoid collisions
-    # when doing tests which also allocate shms for certain instruments
-    # that may be in use on the system by some other running daemons?
-    # from tractor._state import _runtime_vars
-    # port = _runtime_vars['_root_mailbox'][1]
-
-    uid = tractor.current_actor().uid
-    name, uuid = uid
-    service = name.rstrip(f'.{mod.name}')
-
-    fqme: str = mkt.get_fqme(delim_char='')
-
-    # (maybe) allocate shm array for this broker/symbol which will
-    # be used for fast near-term history capture and processing.
-    hist_shm, opened = maybe_open_shm_array(
-        size=_default_hist_size,
-        append_start_index=_hist_buffer_start,
-
-        key=f'piker.{service}[{uuid[:16]}].{fqme}.hist',
-
-        # use any broker defined ohlc dtype:
-        dtype=getattr(mod, '_ohlc_dtype', def_iohlcv_fields),
-
-        # we expect the sub-actor to write
-        readonly=False,
-    )
-    hist_zero_index = hist_shm.index - 1
-
-    # TODO: history validation
-    if not opened:
-        raise RuntimeError(
-            "Persistent shm for sym was already open?!"
-        )
-
-    rt_shm, opened = maybe_open_shm_array(
-        size=_default_rt_size,
-        append_start_index=_rt_buffer_start,
-        key=f'piker.{service}[{uuid[:16]}].{fqme}.rt',
-
-        # use any broker defined ohlc dtype:
-        dtype=getattr(mod, '_ohlc_dtype', def_iohlcv_fields),
-
-        # we expect the sub-actor to write
-        readonly=False,
-    )
-
-    # (for now) set the rt (hft) shm array with space to prepend
-    # only a few days worth of 1s history.
-    days = 2
-    start_index = days*_secs_in_day
-    rt_shm._first.value = start_index
-    rt_shm._last.value = start_index
-    rt_zero_index = rt_shm.index - 1
-
-    if not opened:
-        raise RuntimeError(
-            "Persistent shm for sym was already open?!"
-        )
-
-    open_history_client = getattr(
-        mod,
-        'open_history_client',
-        None,
-    )
-    assert open_history_client
-
-    # TODO: maybe it should be a subpkg of `.data`?
-    from piker import storage
-
-    async with (
-        storage.open_storage_client() as (storemod, client),
-        trio.open_nursery() as tn,
-    ):
-        log.info(
-            f'Connecting to storage backend `{storemod.name}`:\n'
-            f'location: {client.address}\n'
-            f'db cardinality: {client.cardinality}\n'
-            # TODO: show backend config, eg:
-            # - network settings
-            # - storage size with compression
-            # - number of loaded time series?
-        )
-
-        # NOTE: this call ONLY UNBLOCKS once the latest-most frame
-        # (i.e. history just before the live feed latest datum) of
-        # history has been loaded and written to the shm buffer:
-        # - the backfiller task can write in reverse chronological
-        #   to the shm and tsdb
-        # - the tsdb data can be loaded immediately and the
-        #   backfiller can do a single append from it's end datum and
-        #   then prepends backward to that from the current time
-        #   step.
-        tf2mem: dict = {
-            1: rt_shm,
-            60: hist_shm,
-        }
-        async with open_sample_stream(
-            period_s=1.,
-            shms_by_period={
-                1.: rt_shm.token,
-                60.: hist_shm.token,
-            },
-
-            # NOTE: we want to only open a stream for doing
-            # broadcasts on backfill operations, not receive the
-            # sample index-stream (since there's no code in this
-            # data feed layer that needs to consume it).
-            open_index_stream=True,
-            sub_for_broadcasts=False,
-
-        ) as sample_stream:
-            # register 1s and 1m buffers with the global incrementer task
-            log.info(f'Connected to sampler stream: {sample_stream}')
-
-            for timeframe in [60, 1]:
-                await tn.start(
-                    tsdb_backfill,
-                    mod,
-                    storemod,
-                    tn,
-                    # bus,
-                    client,
-                    mkt,
-                    tf2mem[timeframe],
-                    timeframe,
-
-                    sample_stream,
-                )
-
-            # indicate to caller that feed can be delivered to
-            # remote requesting client since we've loaded history
-            # data that can be used.
-            some_data_ready.set()
-
-            # wait for a live feed before starting the sampler.
-            await feed_is_live.wait()
-
-            # yield back after client connect with filled shm
-            task_status.started((
-                hist_zero_index,
-                hist_shm,
-                rt_zero_index,
-                rt_shm,
-            ))
-
-            # history retreival loop depending on user interaction
-            # and thus a small RPC-prot for remotely controllinlg
-            # what data is loaded for viewing.
-            await trio.sleep_forever()

piker/data/types.py

@@ -1,104 +0,0 @@
-# piker: trading gear for hackers
-# Copyright (C) (in stewardship for pikers)
-#  - Tyler Goodlet
-#  - Guillermo Rodriguez
-
-# This program is free software: you can redistribute it and/or modify
-# it under the terms of the GNU Affero General Public License as published by
-# the Free Software Foundation, either version 3 of the License, or
-# (at your option) any later version.
-
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU Affero General Public License for more details.
-
-# You should have received a copy of the GNU Affero General Public License
-# along with this program.  If not, see <https://www.gnu.org/licenses/>.
-
-'''
-Extensions to built-in or (heavily used but 3rd party) friend-lib
-types.
-
-'''
-from pprint import pformat
-
-from msgspec import (
-    msgpack,
-    Struct,
-    structs,
-)
-
-
-class Struct(
-    Struct,
-
-    # https://jcristharif.com/msgspec/structs.html#tagged-unions
-    # tag='pikerstruct',
-    # tag=True,
-):
-    '''
-    A "human friendlier" (aka repl buddy) struct subtype.
-
-    '''
-    def to_dict(self) -> dict:
-        '''
-        Like it sounds.. direct delegation to:
-        https://jcristharif.com/msgspec/api.html#msgspec.structs.asdict
-
-        TODO: probably just drop this method since it's now a built-int method?
-
-        '''
-        return structs.asdict(self)
-
-    def pformat(self) -> str:
-        return f'Struct({pformat(self.to_dict())})'
-
-    def copy(
-        self,
-        update: dict | None = None,
-
-    ) -> Struct:
-        '''
-        Validate-typecast all self defined fields, return a copy of
-        us with all such fields.
-
-        NOTE: This is kinda like the default behaviour in
-        `pydantic.BaseModel` except a copy of the object is
-        returned making it compat with `frozen=True`.
-
-        '''
-        if update:
-            for k, v in update.items():
-                setattr(self, k, v)
-
-        # NOTE: roundtrip serialize to validate
-        # - enode to msgpack binary format,
-        # - decode that back to a struct.
-        return msgpack.Decoder(type=type(self)).decode(
-            msgpack.Encoder().encode(self)
-        )
-
-    def typecast(
-        self,
-
-        # TODO: allow only casting a named subset?
-        # fields: set[str] | None = None,
-
-    ) -> None:
-        '''
-        Cast all fields using their declared type annotations
-        (kinda like what `pydantic` does by default).
-
-        NOTE: this of course won't work on frozen types, use
-        ``.copy()`` above in such cases.
-
-        '''
-        # https://jcristharif.com/msgspec/api.html#msgspec.structs.fields
-        fi: structs.FieldInfo
-        for fi in structs.fields(self):
-            setattr(
-                self,
-                fi.name,
-                fi.type(getattr(self, fi.name)),
-            )

piker/data/validate.py

@@ -18,6 +18,7 @@ Data feed synchronization protocols, init msgs, and general
 data-provider-backend-agnostic schema definitions.
 
 '''
+from __future__ import annotations
 from decimal import Decimal
 from pprint import pformat
 from types import ModuleType
@@ -28,8 +29,8 @@ from typing import (
 
 from msgspec import field
 
-from .types import Struct
-from ..accounting import (
+from piker.types import Struct
+from piker.accounting import (
     Asset,
     MktPair,
 )
@@ -81,8 +82,8 @@ _eps: dict[str, list[str]] = {
     # live order control and trading
     'brokerd': [
         'trades_dialogue',
-        # TODO: ledger normalizer helper?
-        # norm_trades(records: dict[str, Any]) -> TransactionLedger)
+        'open_trade_dialog',  # live order ctl
+        'norm_trade', # ledger normalizer for txns
     ],
 }
 

piker/fsp/__init__.py

@@ -22,17 +22,40 @@ from typing import AsyncIterator
 
 import numpy as np
 
-from ._engine import cascade
+from ._api import (
+    maybe_mk_fsp_shm,
+    Fsp,
+)
+from ._engine import (
+    cascade,
+    Cascade,
+)
+from ._volume import (
+    dolla_vlm,
+    flow_rates,
+    tina_vwap,
+)
 
-__all__ = ['cascade']
+__all__: list[str] = [
+    'cascade',
+    'Cascade',
+    'maybe_mk_fsp_shm',
+    'Fsp',
+    'dolla_vlm',
+    'flow_rates',
+    'tina_vwap',
+]
 
 
 async def latency(
     source: 'TickStream[Dict[str, float]]',  # noqa
     ohlcv: np.ndarray
+
 ) -> AsyncIterator[np.ndarray]:
-    """Latency measurements, broker to piker.
-    """
+    '''
+    Latency measurements, broker to piker.
+
+    '''
     # TODO: do we want to offer yielding this async
     # before the rt data connection comes up?
 

piker/fsp/_engine.py

@@ -18,13 +18,12 @@
 core task logic for processing chains
 
 '''
-from dataclasses import dataclass
+from __future__ import annotations
+from contextlib import asynccontextmanager as acm
 from functools import partial
 from typing import (
     AsyncIterator,
     Callable,
-    Optional,
-    Union,
 )
 
 import numpy as np
@@ -33,9 +32,9 @@ from trio_typing import TaskStatus
 import tractor
 from tractor.msg import NamespacePath
 
+from piker.types import Struct
 from ..log import get_logger, get_console_log
 from .. import data
-from ..data import attach_shm_array
 from ..data.feed import (
     Flume,
     Feed,
@@ -51,17 +50,11 @@ from ._api import (
     _load_builtins,
     _Token,
 )
-from .._profile import Profiler
+from ..toolz import Profiler
 
 log = get_logger(__name__)
 
 
-@dataclass
-class TaskTracker:
-    complete: trio.Event
-    cs: trio.CancelScope
-
-
 async def filter_quotes_by_sym(
 
     sym: str,
@@ -82,30 +75,168 @@ async def filter_quotes_by_sym(
         if quote:
             yield quote
 
+# TODO: unifying the abstractions in this FSP subsys/layer:
+# -[ ] move the `.data.flows.Flume` type into this
+#   module/subsys/pkg?
+# -[ ] ideas for further abstractions as per
+#   - https://github.com/pikers/piker/issues/216,
+#   - https://github.com/pikers/piker/issues/270:
+#   - a (financial signal) ``Flow`` would be the a "collection" of such
+#     minmial cascades. Some engineering based jargon concepts:
+#     - https://en.wikipedia.org/wiki/Signal_chain
+#     - https://en.wikipedia.org/wiki/Daisy_chain_(electrical_engineering)
+#     - https://en.wikipedia.org/wiki/Audio_signal_flow
+#     - https://en.wikipedia.org/wiki/Digital_signal_processing#Implementation
+#     - https://en.wikipedia.org/wiki/Dataflow_programming
+#     - https://en.wikipedia.org/wiki/Signal_programming
+#     - https://en.wikipedia.org/wiki/Incremental_computing
+#     - https://en.wikipedia.org/wiki/Signal-flow_graph
+#     - https://en.wikipedia.org/wiki/Signal-flow_graph#Basic_components
 
-async def fsp_compute(
+# -[ ] we probably want to eval THE BELOW design and unify with the
+#   proto `TaskManager` in the `tractor` dev branch as well as with
+#   our below idea for `Cascade`:
+#   - https://github.com/goodboy/tractor/pull/363
+class Cascade(Struct):
+    '''
+    As per sig-proc engineering parlance, this is a chaining of
+    `Flume`s, which are themselves collections of "Streams"
+    implemented currently via `ShmArray`s.
 
+    A `Cascade` is be the minimal "connection" of 2 `Flumes`
+    as per circuit parlance:
+    https://en.wikipedia.org/wiki/Two-port_network#Cascade_connection
+
+    TODO:
+      -[ ] could cover the combination of our `FspAdmin` and the
+        backend `.fsp._engine` related machinery to "connect" one flume
+        to another?
+
+    '''
+    # TODO: make these `Flume`s
+    src: Flume
+    dst: Flume
+    tn: trio.Nursery
+    fsp: Fsp  # UI-side middleware ctl API
+
+    # filled during cascade/.bind_func() (fsp_compute) init phases
+    bind_func: Callable | None = None
+    complete: trio.Event | None = None
+    cs: trio.CancelScope | None = None
+    client_stream: tractor.MsgStream | None = None
+
+    async def resync(self) -> int:
+        # TODO: adopt an incremental update engine/approach
+        # where possible here eventually!
+        log.info(f're-syncing fsp {self.fsp.name} to source')
+        self.cs.cancel()
+        await self.complete.wait()
+        index: int = await self.tn.start(self.bind_func)
+
+        # always trigger UI refresh after history update,
+        # see ``piker.ui._fsp.FspAdmin.open_chain()`` and
+        # ``piker.ui._display.trigger_update()``.
+        dst_shm: ShmArray = self.dst.rt_shm
+        await self.client_stream.send({
+            'fsp_update': {
+                'key': dst_shm.token,
+                'first': dst_shm._first.value,
+                'last': dst_shm._last.value,
+            }
+        })
+        return index
+
+    def is_synced(self) -> tuple[bool, int, int]:
+        '''
+        Predicate to dertmine if a destination FSP
+        output array is aligned to its source array.
+
+        '''
+        src_shm: ShmArray = self.src.rt_shm
+        dst_shm: ShmArray = self.dst.rt_shm
+        step_diff = src_shm.index - dst_shm.index
+        len_diff = abs(len(src_shm.array) - len(dst_shm.array))
+        synced: bool = not (
+            # the source is likely backfilling and we must
+            # sync history calculations
+            len_diff > 2
+
+            # we aren't step synced to the source and may be
+            # leading/lagging by a step
+            or step_diff > 1
+            or step_diff < 0
+        )
+        if not synced:
+            fsp: Fsp = self.fsp
+            log.warning(
+                '***DESYNCED FSP***\n'
+                f'{fsp.ns_path}@{src_shm.token}\n'
+                f'step_diff: {step_diff}\n'
+                f'len_diff: {len_diff}\n'
+            )
+        return (
+            synced,
+            step_diff,
+            len_diff,
+        )
+
+    async def poll_and_sync_to_step(self) -> int:
+        synced, step_diff, _ = self.is_synced()
+        while not synced:
+            await self.resync()
+            synced, step_diff, _ = self.is_synced()
+
+        return step_diff
+
+    @acm
+    async def open_edge(
+        self,
+        bind_func: Callable,
+    ) -> int:
+        self.bind_func = bind_func
+        index = await self.tn.start(bind_func)
+        yield index
+        # TODO: what do we want on teardown/error?
+        # -[ ] dynamic reconnection after update?
+
+
+async def connect_streams(
+    casc: Cascade,
     mkt: MktPair,
-    flume: Flume,
     quote_stream: trio.abc.ReceiveChannel,
+    src: Flume,
+    dst: Flume,
 
-    src: ShmArray,
-    dst: ShmArray,
-
-    func: Callable,
+    edge_func: Callable,
 
     # attach_stream: bool = False,
     task_status: TaskStatus[None] = trio.TASK_STATUS_IGNORED,
 
 ) -> None:
+    '''
+    Stream and per-sample compute and write the cascade of
+    2 `Flumes`/streams given some operating `func`.
 
+    https://en.wikipedia.org/wiki/Signal-flow_graph#Basic_components
+
+    Not literally, but something like:
+
+        edge_func(Flume_in) -> Flume_out
+
+    '''
     profiler = Profiler(
         delayed=False,
         disabled=True
     )
 
-    fqme = mkt.fqme
-    out_stream = func(
+    # TODO: just pull it from src.mkt.fqme no?
+    # fqme: str = mkt.fqme
+    fqme: str = src.mkt.fqme
+
+    # TODO: dynamic introspection of what the underlying (vertex)
+    # function actually requires from input node (flumes) then
+    # deliver those inputs as part of a graph "compilation" step?
+    out_stream = edge_func(
 
         # TODO: do we even need this if we do the feed api right?
         # shouldn't a local stream do this before we get a handle
@@ -113,20 +244,21 @@ async def fsp_compute(
         # async itertools style?
         filter_quotes_by_sym(fqme, quote_stream),
 
-        # XXX: currently the ``ohlcv`` arg
-        flume.rt_shm,
+        # XXX: currently the ``ohlcv`` arg, but we should allow
+        # (dynamic) requests for src flume (node) streams?
+        src.rt_shm,
     )
 
     # HISTORY COMPUTE PHASE
     # conduct a single iteration of fsp with historical bars input
     # and get historical output.
-    history_output: Union[
-        dict[str, np.ndarray],  # multi-output case
-        np.ndarray,  # single output case
-    ]
+    history_output: (
+        dict[str, np.ndarray]  # multi-output case
+        | np.ndarray,  # single output case
+    )
     history_output = await anext(out_stream)
 
-    func_name = func.__name__
+    func_name = edge_func.__name__
     profiler(f'{func_name} generated history')
 
     # build struct array with an 'index' field to push as history
@@ -134,10 +266,12 @@ async def fsp_compute(
     # TODO: push using a[['f0', 'f1', .., 'fn']] = .. syntax no?
     # if the output array is multi-field then push
     # each respective field.
-    fields = getattr(dst.array.dtype, 'fields', None).copy()
+    dst_shm: ShmArray = dst.rt_shm
+    fields = getattr(dst_shm.array.dtype, 'fields', None).copy()
     fields.pop('index')
-    history_by_field: Optional[np.ndarray] = None
-    src_time = src.array['time']
+    history_by_field: np.ndarray | None = None
+    src_shm: ShmArray = src.rt_shm
+    src_time = src_shm.array['time']
 
     if (
         fields and
@@ -156,7 +290,7 @@ async def fsp_compute(
                 if history_by_field is None:
 
                     if output is None:
-                        length = len(src.array)
+                        length = len(src_shm.array)
                     else:
                         length = len(output)
 
@@ -165,7 +299,7 @@ async def fsp_compute(
                     # will be pushed to shm.
                     history_by_field = np.zeros(
                         length,
-                        dtype=dst.array.dtype
+                        dtype=dst_shm.array.dtype
                     )
 
                 if output is None:
@@ -182,13 +316,13 @@ async def fsp_compute(
             )
         history_by_field = np.zeros(
             len(history_output),
-            dtype=dst.array.dtype
+            dtype=dst_shm.array.dtype
         )
         history_by_field[func_name] = history_output
 
     history_by_field['time'] = src_time[-len(history_by_field):]
 
-    history_output['time'] = src.array['time']
+    history_output['time'] = src_shm.array['time']
 
     # TODO: XXX:
     # THERE'S A BIG BUG HERE WITH THE `index` field since we're
@@ -201,11 +335,11 @@ async def fsp_compute(
     #   is `index` aware such that historical data can be indexed
     #   relative to the true first datum? Not sure if this is sane
     #   for incremental compuations.
-    first = dst._first.value = src._first.value
+    first = dst_shm._first.value = src_shm._first.value
 
     # TODO: can we use this `start` flag instead of the manual
     # setting above?
-    index = dst.push(
+    index = dst_shm.push(
         history_by_field,
         start=first,
     )
@@ -216,12 +350,9 @@ async def fsp_compute(
     # setup a respawn handle
     with trio.CancelScope() as cs:
 
-        # TODO: might be better to just make a "restart" method where
-        # the target task is spawned implicitly and then the event is
-        # set via some higher level api? At that poing we might as well
-        # be writing a one-cancels-one nursery though right?
-        tracker = TaskTracker(trio.Event(), cs)
-        task_status.started((tracker, index))
+        casc.cs = cs
+        casc.complete = trio.Event()
+        task_status.started(index)
 
         profiler(f'{func_name} yield last index')
 
@@ -235,12 +366,12 @@ async def fsp_compute(
                 log.debug(f"{func_name}: {processed}")
                 key, output = processed
                 # dst.array[-1][key] = output
-                dst.array[[key, 'time']][-1] = (
+                dst_shm.array[[key, 'time']][-1] = (
                     output,
                     # TODO: what about pushing ``time.time_ns()``
                     # in which case we'll need to round at the graphics
                     # processing / sampling layer?
-                    src.array[-1]['time']
+                    src_shm.array[-1]['time']
                 )
 
                 # NOTE: for now we aren't streaming this to the consumer
@@ -252,7 +383,7 @@ async def fsp_compute(
                 # N-consumers who subscribe for the real-time output,
                 # which we'll likely want to implement using local-mem
                 # chans for the fan out?
-                # index = src.index
+                # index = src_shm.index
                 # if attach_stream:
                 #     await client_stream.send(index)
 
@@ -262,7 +393,7 @@ async def fsp_compute(
                 #     log.info(f'FSP quote too fast: {hz}')
                 # last = time.time()
         finally:
-            tracker.complete.set()
+            casc.complete.set()
 
 
 @tractor.context
@@ -273,15 +404,15 @@ async def cascade(
     # data feed key
     fqme: str,
 
-    src_shm_token: dict,
-    dst_shm_token: tuple[str, np.dtype],
-
+    # flume pair cascaded using an "edge function"
+    src_flume_addr: dict,
+    dst_flume_addr: dict,
     ns_path: NamespacePath,
 
     shm_registry: dict[str, _Token],
 
     zero_on_step: bool = False,
-    loglevel: Optional[str] = None,
+    loglevel: str | None = None,
 
 ) -> None:
     '''
@@ -297,8 +428,14 @@ async def cascade(
     if loglevel:
         get_console_log(loglevel)
 
-    src = attach_shm_array(token=src_shm_token)
-    dst = attach_shm_array(readonly=False, token=dst_shm_token)
+    src: Flume = Flume.from_msg(src_flume_addr)
+    dst: Flume = Flume.from_msg(
+        dst_flume_addr,
+        readonly=False,
+    )
+
+    # src: ShmArray = attach_shm_array(token=src_shm_token)
+    # dst: ShmArray = attach_shm_array(readonly=False, token=dst_shm_token)
 
     reg = _load_builtins()
     lines = '\n'.join([f'{key.rpartition(":")[2]} => {key}' for key in reg])
@@ -306,11 +443,11 @@ async def cascade(
         f'Registered FSP set:\n{lines}'
     )
 
-    # update actorlocal flows table which registers
-    # readonly "instances" of this fsp for symbol/source
-    # so that consumer fsps can look it up by source + fsp.
-    # TODO: ugh i hate this wind/unwind to list over the wire
-    # but not sure how else to do it.
+    # NOTE XXX: update actorlocal flows table which registers
+    # readonly "instances" of this fsp for symbol/source so that
+    # consumer fsps can look it up by source + fsp.
+    # TODO: ugh i hate this wind/unwind to list over the wire but
+    # not sure how else to do it.
     for (token, fsp_name, dst_token) in shm_registry:
         Fsp._flow_registry[(
             _Token.from_msg(token),
@@ -320,12 +457,15 @@ async def cascade(
     fsp: Fsp = reg.get(
         NamespacePath(ns_path)
     )
-    func = fsp.func
+    func: Callable = fsp.func
 
     if not func:
         # TODO: assume it's a func target path
         raise ValueError(f'Unknown fsp target: {ns_path}')
 
+    _fqme: str = src.mkt.fqme
+    assert _fqme == fqme
+
     # open a data feed stream with requested broker
     feed: Feed
     async with data.feed.maybe_open_feed(
@@ -339,177 +479,142 @@ async def cascade(
 
     ) as feed:
 
-        flume = feed.flumes[fqme]
-        mkt = flume.mkt
-        assert src.token == flume.rt_shm.token
+        flume: Flume = feed.flumes[fqme]
+        # XXX: can't do this since flume.feed will be set XD
+        # assert flume == src
+        assert flume.mkt == src.mkt
+        mkt: MktPair = flume.mkt
+
+        # NOTE: FOR NOW, sanity checks around the feed as being
+        # always the src flume (until we get to fancier/lengthier
+        # chains/graphs.
+        assert src.rt_shm.token == flume.rt_shm.token
+
+        # XXX: won't work bc the _hist_shm_token value will be
+        # list[list] after IPC..
+        # assert flume.to_msg() == src_flume_addr
+
         profiler(f'{func}: feed up')
 
-        func_name = func.__name__
+        func_name: str = func.__name__
         async with (
-            trio.open_nursery() as n,
+            trio.open_nursery() as tn,
         ):
+            # TODO: might be better to just make a "restart" method where
+            # the target task is spawned implicitly and then the event is
+            # set via some higher level api? At that poing we might as well
+            # be writing a one-cancels-one nursery though right?
+            casc = Cascade(
+                src,
+                dst,
+                tn,
+                fsp,
+            )
 
+            # TODO: this seems like it should be wrapped somewhere?
             fsp_target = partial(
-
-                fsp_compute,
+                connect_streams,
+                casc=casc,
                 mkt=mkt,
-                flume=flume,
                 quote_stream=flume.stream,
 
-                # shm
+                # flumes and shm passthrough
                 src=src,
                 dst=dst,
 
-                # target
-                func=func
+                # chain function which takes src flume input(s)
+                # and renders dst flume output(s)
+                edge_func=func
             )
+            async with casc.open_edge(
+                bind_func=fsp_target,
+            ) as index:
+                # casc.bind_func = fsp_target
+                # index = await tn.start(fsp_target)
+                dst_shm: ShmArray = dst.rt_shm
+                src_shm: ShmArray = src.rt_shm
 
-            tracker, index = await n.start(fsp_target)
+                if zero_on_step:
+                    last = dst.rt_shm.array[-1:]
+                    zeroed = np.zeros(last.shape, dtype=last.dtype)
 
-            if zero_on_step:
-                last = dst.array[-1:]
-                zeroed = np.zeros(last.shape, dtype=last.dtype)
+                profiler(f'{func_name}: fsp up')
 
-            profiler(f'{func_name}: fsp up')
+                # sync to client-side actor
+                await ctx.started(index)
 
-            # sync client
-            await ctx.started(index)
+                # XXX:  rt stream with client which we MUST
+                # open here (and keep it open) in order to make
+                # incremental "updates" as history prepends take
+                # place.
+                async with ctx.open_stream() as client_stream:
+                    casc.client_stream: tractor.MsgStream = client_stream
 
-            # XXX:  rt stream with client which we MUST
-            # open here (and keep it open) in order to make
-            # incremental "updates" as history prepends take
-            # place.
-            async with ctx.open_stream() as client_stream:
+                    s, step, ld = casc.is_synced()
 
-                # TODO: these likely should all become
-                # methods of this ``TaskLifetime`` or wtv
-                # abstraction..
-                async def resync(
-                    tracker: TaskTracker,
+                    # detect sample period step for subscription to increment
+                    # signal
+                    times = src.rt_shm.array['time']
+                    if len(times) > 1:
+                        last_ts = times[-1]
+                        delay_s: float = float(last_ts - times[times != last_ts][-1])
+                    else:
+                        # our default "HFT" sample rate.
+                        delay_s: float = _default_delay_s
 
-                ) -> tuple[TaskTracker, int]:
-                    # TODO: adopt an incremental update engine/approach
-                    # where possible here eventually!
-                    log.info(f're-syncing fsp {func_name} to source')
-                    tracker.cs.cancel()
-                    await tracker.complete.wait()
-                    tracker, index = await n.start(fsp_target)
+                    # sub and increment the underlying shared memory buffer
+                    # on every step msg received from the global `samplerd`
+                    # service.
+                    async with open_sample_stream(
+                        float(delay_s)
+                    ) as istream:
 
-                    # always trigger UI refresh after history update,
-                    # see ``piker.ui._fsp.FspAdmin.open_chain()`` and
-                    # ``piker.ui._display.trigger_update()``.
-                    await client_stream.send({
-                        'fsp_update': {
-                            'key': dst_shm_token,
-                            'first': dst._first.value,
-                            'last': dst._last.value,
-                        }
-                    })
-                    return tracker, index
+                        profiler(f'{func_name}: sample stream up')
+                        profiler.finish()
 
-                def is_synced(
-                    src: ShmArray,
-                    dst: ShmArray
-                ) -> tuple[bool, int, int]:
-                    '''
-                    Predicate to dertmine if a destination FSP
-                    output array is aligned to its source array.
+                        async for i in istream:
+                            # print(f'FSP incrementing {i}')
 
-                    '''
-                    step_diff = src.index - dst.index
-                    len_diff = abs(len(src.array) - len(dst.array))
-                    return not (
-                        # the source is likely backfilling and we must
-                        # sync history calculations
-                        len_diff > 2
+                            # respawn the compute task if the source
+                            # array has been updated such that we compute
+                            # new history from the (prepended) source.
+                            synced, step_diff, _ = casc.is_synced()
+                            if not synced:
+                                step_diff: int = await casc.poll_and_sync_to_step()
 
-                        # we aren't step synced to the source and may be
-                        # leading/lagging by a step
-                        or step_diff > 1
-                        or step_diff < 0
-                    ), step_diff, len_diff
+                                # skip adding a last bar since we should already
+                                # be step alinged
+                                if step_diff == 0:
+                                    continue
 
-                async def poll_and_sync_to_step(
-                    tracker: TaskTracker,
-                    src: ShmArray,
-                    dst: ShmArray,
+                            # read out last shm row, copy and write new row
+                            array = dst_shm.array
 
-                ) -> tuple[TaskTracker, int]:
+                            # some metrics like vlm should be reset
+                            # to zero every step.
+                            if zero_on_step:
+                                last = zeroed
+                            else:
+                                last = array[-1:].copy()
 
-                    synced, step_diff, _ = is_synced(src, dst)
-                    while not synced:
-                        tracker, index = await resync(tracker)
-                        synced, step_diff, _ = is_synced(src, dst)
+                            dst.rt_shm.push(last)
 
-                    return tracker, step_diff
+                            # sync with source buffer's time step
+                            src_l2 = src_shm.array[-2:]
+                            src_li, src_lt = src_l2[-1][['index', 'time']]
+                            src_2li, src_2lt = src_l2[-2][['index', 'time']]
+                            dst_shm._array['time'][src_li] = src_lt
+                            dst_shm._array['time'][src_2li] = src_2lt
 
-                s, step, ld = is_synced(src, dst)
-
-                # detect sample period step for subscription to increment
-                # signal
-                times = src.array['time']
-                if len(times) > 1:
-                    last_ts = times[-1]
-                    delay_s = float(last_ts - times[times != last_ts][-1])
-                else:
-                    # our default "HFT" sample rate.
-                    delay_s = _default_delay_s
-
-                # sub and increment the underlying shared memory buffer
-                # on every step msg received from the global `samplerd`
-                # service.
-                async with open_sample_stream(float(delay_s)) as istream:
-
-                    profiler(f'{func_name}: sample stream up')
-                    profiler.finish()
-
-                    async for i in istream:
-                        # print(f'FSP incrementing {i}')
-
-                        # respawn the compute task if the source
-                        # array has been updated such that we compute
-                        # new history from the (prepended) source.
-                        synced, step_diff, _ = is_synced(src, dst)
-                        if not synced:
-                            tracker, step_diff = await poll_and_sync_to_step(
-                                tracker,
-                                src,
-                                dst,
-                            )
-
-                            # skip adding a last bar since we should already
-                            # be step alinged
-                            if step_diff == 0:
-                                continue
-
-                        # read out last shm row, copy and write new row
-                        array = dst.array
-
-                        # some metrics like vlm should be reset
-                        # to zero every step.
-                        if zero_on_step:
-                            last = zeroed
-                        else:
-                            last = array[-1:].copy()
-
-                        dst.push(last)
-
-                        # sync with source buffer's time step
-                        src_l2 = src.array[-2:]
-                        src_li, src_lt = src_l2[-1][['index', 'time']]
-                        src_2li, src_2lt = src_l2[-2][['index', 'time']]
-                        dst._array['time'][src_li] = src_lt
-                        dst._array['time'][src_2li] = src_2lt
-
-                        # last2 = dst.array[-2:]
-                        # if (
-                        #     last2[-1]['index'] != src_li
-                        #     or last2[-2]['index'] != src_2li
-                        # ):
-                        #     dstl2 = list(last2)
-                        #     srcl2 = list(src_l2)
-                        #     print(
-                        #         # f'{dst.token}\n'
-                        #         f'src: {srcl2}\n'
-                        #         f'dst: {dstl2}\n'
-                        #     )
+                            # last2 = dst.array[-2:]
+                            # if (
+                            #     last2[-1]['index'] != src_li
+                            #     or last2[-2]['index'] != src_2li
+                            # ):
+                            #     dstl2 = list(last2)
+                            #     srcl2 = list(src_l2)
+                            #     print(
+                            #         # f'{dst.token}\n'
+                            #         f'src: {srcl2}\n'
+                            #         f'dst: {dstl2}\n'
+                            #     )

piker/service/__init__.py

@@ -14,49 +14,45 @@
 # You should have received a copy of the GNU Affero General Public License
 # along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
-"""
-Actor-runtime service orchestration machinery.
+'''
+Actor runtime primtives and (distributed) service APIs for,
 
-"""
-from __future__ import annotations
+- daemon-service mgmt: `_daemon` (i.e. low-level spawn and supervise machinery
+  for sub-actors like `brokerd`, `emsd`, datad`, etc.)
 
-from ._mngr import Services
-from ._registry import (  # noqa
-    _tractor_kwargs,
-    _default_reg_addr,
-    _default_registry_host,
-    _default_registry_port,
-    open_registry,
-    find_service,
-    check_for_service,
+- service-actor supervision (via `trio` tasks) API: `._mngr`
+
+- discovery interface (via light wrapping around `tractor`'s built-in
+  prot): `._registry`
+
+- `docker` cntr SC supervision for use with `trio`: `_ahab`
+  - wrappers for marketstore and elasticsearch dbs
+  => TODO: maybe to (re)move elsewhere?
+
+'''
+from ._mngr import Services as Services
+from ._registry import (
+    _tractor_kwargs as _tractor_kwargs,
+    _default_reg_addr as _default_reg_addr,
+    _default_registry_host as _default_registry_host,
+    _default_registry_port as _default_registry_port,
+
+    open_registry as open_registry,
+    find_service as find_service,
+    check_for_service as check_for_service,
 )
-from ._daemon import (  # noqa
-    maybe_spawn_daemon,
-    spawn_emsd,
-    maybe_open_emsd,
+from ._daemon import (
+    maybe_spawn_daemon as maybe_spawn_daemon,
+    spawn_emsd as spawn_emsd,
+    maybe_open_emsd as maybe_open_emsd,
 )
 from ._actor_runtime import (
-    open_piker_runtime,
-    maybe_open_pikerd,
-    open_pikerd,
-    get_tractor_runtime_kwargs,
+    open_piker_runtime as open_piker_runtime,
+    maybe_open_pikerd as maybe_open_pikerd,
+    open_pikerd as open_pikerd,
+    get_runtime_vars as get_runtime_vars,
 )
 from ..brokers._daemon import (
-    spawn_brokerd,
-    maybe_spawn_brokerd,
+    spawn_brokerd as spawn_brokerd,
+    maybe_spawn_brokerd as maybe_spawn_brokerd,
 )
-
-
-__all__ = [
-    'check_for_service',
-    'Services',
-    'maybe_spawn_daemon',
-    'spawn_brokerd',
-    'maybe_spawn_brokerd',
-    'spawn_emsd',
-    'maybe_open_emsd',
-    'open_piker_runtime',
-    'maybe_open_pikerd',
-    'open_pikerd',
-    'get_tractor_runtime_kwargs',
-]

piker/service/_actor_runtime.py

@@ -45,7 +45,7 @@ from ._registry import (  # noqa
 )
 
 
-def get_tractor_runtime_kwargs() -> dict[str, Any]:
+def get_runtime_vars() -> dict[str, Any]:
     '''
     Deliver ``tractor`` related runtime variables in a `dict`.
 
@@ -56,6 +56,8 @@ def get_tractor_runtime_kwargs() -> dict[str, Any]:
 @acm
 async def open_piker_runtime(
     name: str,
+    registry_addrs: list[tuple[str, int]] = [],
+
     enable_modules: list[str] = [],
     loglevel: Optional[str] = None,
 
@@ -63,8 +65,6 @@ async def open_piker_runtime(
     # for data daemons when running in production.
     debug_mode: bool = False,
 
-    registry_addr: None | tuple[str, int] = None,
-
     # TODO: once we have `rsyscall` support we will read a config
     # and spawn the service tree distributed per that.
     start_method: str = 'trio',
@@ -74,7 +74,7 @@ async def open_piker_runtime(
 
 ) -> tuple[
     tractor.Actor,
-    tuple[str, int],
+    list[tuple[str, int]],
 ]:
     '''
     Start a piker actor who's runtime will automatically sync with
@@ -84,21 +84,31 @@ async def open_piker_runtime(
     a root actor.
 
     '''
+    # check for existing runtime, boot it
+    # if not already running.
     try:
-        # check for existing runtime
-        actor = tractor.current_actor().uid
-
+        actor = tractor.current_actor()
     except tractor._exceptions.NoRuntime:
         tractor._state._runtime_vars[
-            'piker_vars'] = tractor_runtime_overrides
+            'piker_vars'
+        ] = tractor_runtime_overrides
 
-        registry_addr = registry_addr or _default_reg_addr
+        # NOTE: if no registrar list passed used the default of just
+        # setting it as the root actor on localhost.
+        registry_addrs = (
+            registry_addrs
+            or [_default_reg_addr]
+        )
+
+        if ems := tractor_kwargs.pop('enable_modules', None):
+            # import pdbp; pdbp.set_trace()
+            enable_modules.extend(ems)
 
         async with (
             tractor.open_root_actor(
 
                 # passed through to ``open_root_actor``
-                arbiter_addr=registry_addr,
+                registry_addrs=registry_addrs,
                 name=name,
                 loglevel=loglevel,
                 debug_mode=debug_mode,
@@ -110,24 +120,30 @@ async def open_piker_runtime(
                 enable_modules=enable_modules,
 
                 **tractor_kwargs,
-            ) as _,
+            ) as actor,
 
-            open_registry(registry_addr, ensure_exists=False) as addr,
+            open_registry(
+                registry_addrs,
+                ensure_exists=False,
+            ) as addrs,
         ):
-            yield (
-                tractor.current_actor(),
-                addr,
-            )
-    else:
-        async with open_registry(registry_addr) as addr:
+            assert actor is tractor.current_actor()
             yield (
                 actor,
-                addr,
+                addrs,
+            )
+    else:
+        async with open_registry(
+            registry_addrs
+        ) as addrs:
+            yield (
+                actor,
+                addrs,
             )
 
 
-_root_dname = 'pikerd'
-_root_modules = [
+_root_dname: str = 'pikerd'
+_root_modules: list[str] = [
     __name__,
     'piker.service._daemon',
     'piker.brokers._daemon',
@@ -141,13 +157,13 @@ _root_modules = [
 
 @acm
 async def open_pikerd(
+    registry_addrs: list[tuple[str, int]],
 
     loglevel: str | None = None,
 
     # XXX: you should pretty much never want debug mode
     # for data daemons when running in production.
     debug_mode: bool = False,
-    registry_addr: None | tuple[str, int] = None,
 
     **kwargs,
 
@@ -159,29 +175,39 @@ async def open_pikerd(
     alive underling services (see below).
 
     '''
+    # NOTE: for the root daemon we always enable the root
+    # mod set and we `list.extend()` it into wtv the
+    # caller requested.
+    # TODO: make this mod set more strict?
+    # -[ ] eventually we should be able to avoid
+    #    having the root have more then permissions to spawn other
+    #    specialized daemons I think?
+    ems: list[str] = kwargs.setdefault('enable_modules', [])
+    ems.extend(_root_modules)
+
     async with (
         open_piker_runtime(
 
             name=_root_dname,
-            # TODO: eventually we should be able to avoid
-            # having the root have more then permissions to
-            # spawn other specialized daemons I think?
-            enable_modules=_root_modules,
             loglevel=loglevel,
             debug_mode=debug_mode,
-            registry_addr=registry_addr,
+            registry_addrs=registry_addrs,
 
             **kwargs,
 
-        ) as (root_actor, reg_addr),
+        ) as (
+            root_actor,
+            reg_addrs,
+        ),
         tractor.open_nursery() as actor_nursery,
         trio.open_nursery() as service_nursery,
     ):
-        if root_actor.accept_addr != reg_addr:
-            raise RuntimeError(
-                f'`pikerd` failed to bind on {reg_addr}!\n'
-                'Maybe you have another daemon already running?'
-            )
+        for addr in reg_addrs:
+            if addr not in root_actor.accept_addrs:
+                raise RuntimeError(
+                    f'`pikerd` failed to bind on {addr}!\n'
+                    'Maybe you have another daemon already running?'
+                )
 
         # assign globally for future daemon/task creation
         Services.actor_n = actor_nursery
@@ -225,9 +251,9 @@ async def open_pikerd(
 
 @acm
 async def maybe_open_pikerd(
-    loglevel: Optional[str] = None,
-    registry_addr: None | tuple = None,
+    registry_addrs: list[tuple[str, int]] | None = None,
 
+    loglevel: str | None = None,
     **kwargs,
 
 ) -> tractor._portal.Portal | ClassVar[Services]:
@@ -253,32 +279,51 @@ async def maybe_open_pikerd(
     #     async with open_portal(chan) as arb_portal:
     #         yield arb_portal
 
+    registry_addrs: list[tuple[str, int]] = (
+        registry_addrs
+        or [_default_reg_addr]
+    )
+
+    pikerd_portal: tractor.Portal | None
     async with (
         open_piker_runtime(
             name=query_name,
-            registry_addr=registry_addr,
+            registry_addrs=registry_addrs,
             loglevel=loglevel,
             **kwargs,
-        ) as _,
-
-        tractor.find_actor(
-            _root_dname,
-            arbiter_sockaddr=registry_addr,
-        ) as portal
+        ) as (actor, addrs),
     ):
-        # connect to any existing daemon presuming
-        # its registry socket was selected.
-        if (
-            portal is not None
-        ):
-            yield portal
+        if _root_dname in actor.uid:
+            yield None
             return
 
+        # NOTE: IFF running in disti mode, try to attach to any
+        # existing (host-local) `pikerd`.
+        else:
+            async with tractor.find_actor(
+                _root_dname,
+                registry_addrs=registry_addrs,
+                only_first=True,
+                # raise_on_none=True,
+            ) as pikerd_portal:
+
+                # connect to any existing remote daemon presuming its
+                # registry socket was selected.
+                if pikerd_portal is not None:
+
+                    # sanity check that we are actually connecting to
+                    # a remote process and not ourselves.
+                    assert actor.uid != pikerd_portal.channel.uid
+                    assert registry_addrs
+
+                    yield pikerd_portal
+                    return
+
     # presume pikerd role since no daemon could be found at
     # configured address
     async with open_pikerd(
         loglevel=loglevel,
-        registry_addr=registry_addr,
+        registry_addrs=registry_addrs,
 
         # passthrough to ``tractor`` init
         **kwargs,

piker/service/_ahab.py

@@ -15,8 +15,8 @@
 # along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
 '''
-Supervisor for ``docker`` with included async and SC wrapping
-to ensure a cancellable container lifetime system.
+Supervisor for ``docker`` with included async and SC wrapping to
+ensure a cancellable container lifetime system.
 
 '''
 from __future__ import annotations

piker/service/_daemon.py

@@ -70,7 +70,10 @@ async def maybe_spawn_daemon(
     lock = Services.locks[service_name]
     await lock.acquire()
 
-    async with find_service(service_name) as portal:
+    async with find_service(
+        service_name,
+        registry_addrs=[('127.0.0.1', 6116)],
+    ) as portal:
         if portal is not None:
             lock.release()
             yield portal

piker/service/_mngr.py

@@ -27,6 +27,12 @@ from typing import (
 import trio
 from trio_typing import TaskStatus
 import tractor
+from tractor import (
+    current_actor,
+    ContextCancelled,
+    Context,
+    Portal,
+)
 
 from ._util import (
     log,  # sub-sys logger
@@ -38,6 +44,8 @@ from ._util import (
 #   library.
 # - wrap a "remote api" wherein you can get a method proxy
 #   to the pikerd actor for starting services remotely!
+# - prolly rename this to ActorServicesNursery since it spawns
+#   new actors and supervises them to completion?
 class Services:
 
     actor_n: tractor._supervise.ActorNursery
@@ -47,7 +55,7 @@ class Services:
         str,
         tuple[
             trio.CancelScope,
-            tractor.Portal,
+            Portal,
             trio.Event,
         ]
     ] = {}
@@ -57,12 +65,12 @@ class Services:
     async def start_service_task(
         self,
         name: str,
-        portal: tractor.Portal,
+        portal: Portal,
         target: Callable,
         allow_overruns: bool = False,
         **ctx_kwargs,
 
-    ) -> (trio.CancelScope, tractor.Context):
+    ) -> (trio.CancelScope, Context):
         '''
         Open a context in a service sub-actor, add to a stack
         that gets unwound at ``pikerd`` teardown.
@@ -101,13 +109,30 @@ class Services:
                         # wait on any context's return value
                         # and any final portal result from the
                         # sub-actor.
-                        ctx_res = await ctx.result()
+                        ctx_res: Any = await ctx.result()
 
                         # NOTE: blocks indefinitely until cancelled
                         # either by error from the target context
                         # function or by being cancelled here by the
                         # surrounding cancel scope.
                         return (await portal.result(), ctx_res)
+                    except ContextCancelled as ctxe:
+                        canceller: tuple[str, str] = ctxe.canceller
+                        our_uid: tuple[str, str] = current_actor().uid
+                        if (
+                            canceller != portal.channel.uid
+                            and
+                            canceller != our_uid
+                        ):
+                            log.cancel(
+                                f'Actor-service {name} was remotely cancelled?\n'
+                                f'remote canceller: {canceller}\n'
+                                f'Keeping {our_uid} alive, ignoring sub-actor cancel..\n'
+                            )
+                        else:
+                            raise
+
+
 
                     finally:
                         await portal.cancel_actor()

piker/service/_registry.py

@@ -27,6 +27,7 @@ from typing import (
 )
 
 import tractor
+from tractor import Portal
 
 from ._util import (
     log,  # sub-sys logger
@@ -46,7 +47,9 @@ _registry: Registry | None = None
 
 
 class Registry:
-    addr: None | tuple[str, int] = None
+    # TODO: should this be a set or should we complain
+    # on duplicates?
+    addrs: list[tuple[str, int]] = []
 
     # TODO: table of uids to sockaddrs
     peers: dict[
@@ -60,69 +63,115 @@ _tractor_kwargs: dict[str, Any] = {}
 
 @acm
 async def open_registry(
-    addr: None | tuple[str, int] = None,
+    addrs: list[tuple[str, int]],
     ensure_exists: bool = True,
 
-) -> tuple[str, int]:
+) -> list[tuple[str, int]]:
+    '''
+    Open the service-actor-discovery registry by returning a set of
+    tranport socket-addrs to registrar actors which may be
+    contacted and queried for similar addresses for other
+    non-registrar actors.
 
+    '''
     global _tractor_kwargs
     actor = tractor.current_actor()
     uid = actor.uid
+    preset_reg_addrs: list[tuple[str, int]] = Registry.addrs
     if (
-        Registry.addr is not None
-        and addr
+        preset_reg_addrs
+        and addrs
     ):
-        raise RuntimeError(
-            f'`{uid}` registry addr already bound @ {_registry.sockaddr}'
-        )
+        if preset_reg_addrs != addrs:
+            # if any(addr in preset_reg_addrs for addr in addrs):
+            diff: set[tuple[str, int]] = set(preset_reg_addrs) - set(addrs)
+            if diff:
+                log.warning(
+                    f'`{uid}` requested only subset of registrars: {addrs}\n'
+                    f'However there are more @{diff}'
+                )
+            else:
+                raise RuntimeError(
+                    f'`{uid}` has non-matching registrar addresses?\n'
+                    f'request: {addrs}\n'
+                    f'already set: {preset_reg_addrs}'
+                )
 
     was_set: bool = False
 
     if (
         not tractor.is_root_process()
-        and Registry.addr is None
+        and not Registry.addrs
     ):
-        Registry.addr = actor._arb_addr
+        Registry.addrs.extend(actor.reg_addrs)
 
     if (
         ensure_exists
-        and Registry.addr is None
+        and not Registry.addrs
     ):
         raise RuntimeError(
-            f"`{uid}` registry should already exist bug doesn't?"
+            f"`{uid}` registry should already exist but doesn't?"
         )
 
     if (
-        Registry.addr is None
+        not Registry.addrs
     ):
         was_set = True
-        Registry.addr = addr or _default_reg_addr
+        Registry.addrs = addrs or [_default_reg_addr]
 
-    _tractor_kwargs['arbiter_addr'] = Registry.addr
+    # NOTE: only spot this seems currently used is inside
+    # `.ui._exec` which is the (eventual qtloops) bootstrapping
+    # with guest mode.
+    _tractor_kwargs['registry_addrs'] = Registry.addrs
 
     try:
-        yield Registry.addr
+        yield Registry.addrs
     finally:
         # XXX: always clear the global addr if we set it so that the
         # next (set of) calls will apply whatever new one is passed
         # in.
         if was_set:
-            Registry.addr = None
+            Registry.addrs = None
 
 
 @acm
 async def find_service(
     service_name: str,
-) -> tractor.Portal | None:
+    registry_addrs: list[tuple[str, int]] | None = None,
 
-    async with open_registry() as reg_addr:
+    first_only: bool = True,
+
+) -> (
+    Portal
+    | list[Portal]
+    | None
+):
+
+    reg_addrs: list[tuple[str, int]]
+    async with open_registry(
+        addrs=(
+            registry_addrs
+            # NOTE: if no addr set is passed assume the registry has
+            # already been opened and use the previously applied
+            # startup set.
+            or Registry.addrs
+        ),
+    ) as reg_addrs:
         log.info(f'Scanning for service `{service_name}`')
+
+        maybe_portals: list[Portal] | Portal | None
+
         # attach to existing daemon by name if possible
         async with tractor.find_actor(
             service_name,
-            arbiter_sockaddr=reg_addr,
-        ) as maybe_portal:
-            yield maybe_portal
+            registry_addrs=reg_addrs,
+            only_first=first_only,  # if set only returns single ref
+        ) as maybe_portals:
+            if not maybe_portals:
+                yield None
+                return
+
+            yield maybe_portals
 
 
 async def check_for_service(
@@ -133,9 +182,11 @@ async def check_for_service(
     Service daemon "liveness" predicate.
 
     '''
-    async with open_registry(ensure_exists=False) as reg_addr:
-        async with tractor.query_actor(
+    async with (
+        open_registry(ensure_exists=False) as reg_addr,
+        tractor.query_actor(
             service_name,
             arbiter_sockaddr=reg_addr,
-        ) as sockaddr:
-            return sockaddr
+        ) as sockaddr,
+    ):
+        return sockaddr

piker/storage/__init__.py

@@ -139,6 +139,13 @@ class StorageClient(
         ...
 
 
+class TimeseriesNotFound(Exception):
+    '''
+    No timeseries entry can be found for this backend.
+
+    '''
+
+
 class StorageConnectionError(ConnectionError):
     '''
     Can't connect to the desired tsdb subsys/service.
@@ -169,10 +176,13 @@ async def open_storage_client(
     tsdb_host: str = 'localhost'
 
     # load root config and any tsdb user defined settings
-    conf, path = config.load('conf', touch_if_dne=True)
+    conf, path = config.load(
+        conf_name='conf',
+        touch_if_dne=True,
+    )
 
     # TODO: maybe not under a "network" section.. since
-    # no more chitty mkts..
+    # no more chitty `marketstore`..
     tsdbconf: dict = {}
     service_section = conf.get('service')
     if (
@@ -183,8 +193,11 @@ async def open_storage_client(
 
         # lookup backend tsdb module by name and load any user service
         # settings for connecting to the tsdb service.
-        backend: str = tsdbconf.pop('backend')
-        tsdb_host: str = tsdbconf['host']
+        backend: str = tsdbconf.pop(
+            'name',
+            def_backend,
+        )
+        tsdb_host: str = tsdbconf.get('maddrs', [])
 
     if backend is None:
         backend: str = def_backend
@@ -246,7 +259,7 @@ async def open_tsdb_client(
     #   * the original data feed arch blurb:
     #     - https://github.com/pikers/piker/issues/98
     #
-    from .._profile import Profiler
+    from ..toolz import Profiler
     profiler = Profiler(
         disabled=True,  # not pg_profile_enabled(),
         delayed=False,

piker/storage/cli.py

@@ -1,5 +1,5 @@
 # piker: trading gear for hackers
-# Copyright (C) 2018-present  Tyler Goodlet (in stewardship of piker0)
+# Copyright (C) 2018-present  Tyler Goodlet (in stewardship of pikers)
 
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU Affero General Public License as published by
@@ -19,10 +19,18 @@ Storage middle-ware CLIs.
 
 """
 from __future__ import annotations
+# from datetime import datetime
+# from contextlib import (
+#     AsyncExitStack,
+# )
 from pathlib import Path
+from math import copysign
 import time
-from typing import Generator
-# from typing import TYPE_CHECKING
+from types import ModuleType
+from typing import (
+    Any,
+    TYPE_CHECKING,
+)
 
 import polars as pl
 import numpy as np
@@ -35,24 +43,21 @@ import typer
 
 from piker.service import open_piker_runtime
 from piker.cli import cli
-from piker.config import get_conf_dir
 from piker.data import (
-    maybe_open_shm_array,
-    def_iohlcv_fields,
     ShmArray,
 )
-from piker.data.history import (
-    _default_hist_size,
-    _default_rt_size,
-)
-from . import (
-    log,
-)
+from piker import tsp
+from piker.data._formatters import BGM
+from . import log
 from . import (
     __tsdbs__,
     open_storage_client,
+    StorageClient,
 )
 
+if TYPE_CHECKING:
+    from piker.ui._remote_ctl import AnnotCtl
+
 
 store = typer.Typer()
 
@@ -77,7 +82,6 @@ def ls(
         async with (
             open_piker_runtime(
                 'tsdb_storage',
-                enable_modules=['piker.service._ahab'],
             ),
         ):
             for i, backend in enumerate(backends):
@@ -99,6 +103,18 @@ def ls(
     trio.run(query_all)
 
 
+# TODO: like ls but takes in a pattern and matches
+# @store.command()
+# def search(
+#     patt: str,
+#     backends: list[str] = typer.Argument(
+#         default=None,
+#         help='Storage backends to query, default is all.'
+#     ),
+# ):
+#     ...
+
+
 @store.command()
 def delete(
     symbols: list[str],
@@ -121,7 +137,6 @@ def delete(
         async with (
             open_piker_runtime(
                 'tsdb_storage',
-                enable_modules=['piker.service._ahab']
             ),
             open_storage_client(backend) as (_, client),
             trio.open_nursery() as n,
@@ -142,20 +157,33 @@ def delete(
 def anal(
     fqme: str,
     period: int = 60,
+    pdb: bool = False,
 
 ) -> np.ndarray:
+    '''
+    Anal-ysis is when you take the data do stuff to it.
 
+    NOTE: This ONLY loads the offline timeseries data (by default
+    from a parquet file) NOT the in-shm version you might be seeing
+    in a chart.
+
+    '''
     async def main():
         async with (
             open_piker_runtime(
+                # are you a bear or boi?
                 'tsdb_polars_anal',
-                # enable_modules=['piker.service._ahab']
+                debug_mode=pdb,
+            ),
+            open_storage_client() as (
+                mod,
+                client,
             ),
-            open_storage_client() as (mod, client),
         ):
             syms: list[str] = await client.list_keys()
-            print(f'{len(syms)} FOUND for {mod.name}')
+            log.info(f'{len(syms)} FOUND for {mod.name}')
 
+            history: ShmArray  # np buffer format
             (
                 history,
                 first_dt,
@@ -166,166 +194,357 @@ def anal(
             )
             assert first_dt < last_dt
 
-            src_df = await client.as_df(fqme, period)
-            from piker.data import _timeseries as tsmod
-            df = tsmod.with_dts(src_df)
-            gaps: pl.DataFrame = tsmod.detect_time_gaps(df)
+            null_segs: tuple = tsp.get_null_segs(
+                frame=history,
+                period=period,
+            )
+            # TODO: do tsp queries to backcend to fill i missing
+            # history and then prolly write it to tsdb!
 
-            if gaps:
-                print(f'Gaps found:\n{gaps}')
+            shm_df: pl.DataFrame = await client.as_df(
+                fqme,
+                period,
+            )
 
-            # TODO: something better with tab completion..
-            # is there something more minimal but nearly as
-            # functional as ipython?
-            await tractor.breakpoint()
+            df: pl.DataFrame  # with dts
+            deduped: pl.DataFrame  # deduplicated dts
+            (
+                df,
+                deduped,
+                diff,
+            ) = tsp.dedupe(
+                shm_df,
+                period=period,
+            )
+
+            write_edits: bool = True
+            if (
+                write_edits
+                and (
+                    diff
+                    or null_segs
+                )
+            ):
+                await tractor.pause()
+                await client.write_ohlcv(
+                    fqme,
+                    ohlcv=deduped,
+                    timeframe=period,
+                )
+
+            else:
+                # TODO: something better with tab completion..
+                # is there something more minimal but nearly as
+                # functional as ipython?
+                await tractor.pause()
+                assert not null_segs
 
     trio.run(main)
 
 
-def iter_dfs_from_shms(fqme: str) -> Generator[
-    tuple[Path, ShmArray, pl.DataFrame],
-    None,
-    None,
-]:
-    # shm buffer size table based on known sample rates
-    sizes: dict[str, int] = {
-        'hist': _default_hist_size,
-        'rt': _default_rt_size,
-    }
+async def markup_gaps(
+    fqme: str,
+    timeframe: float,
+    actl: AnnotCtl,
+    wdts: pl.DataFrame,
+    gaps: pl.DataFrame,
 
-    # load all detected shm buffer files which have the
-    # passed FQME pattern in the file name.
-    shmfiles: list[Path] = []
-    shmdir = Path('/dev/shm/')
+) -> dict[int, dict]:
+    '''
+    Remote annotate time-gaps in a dt-fielded ts (normally OHLC)
+    with rectangles.
 
-    for shmfile in shmdir.glob(f'*{fqme}*'):
-        filename: str = shmfile.name
+    '''
+    aids: dict[int] = {}
+    for i in range(gaps.height):
 
-        # skip index files
-        if (
-            '_first' in filename
-            or '_last' in filename
-        ):
-            continue
+        row: pl.DataFrame = gaps[i]
 
-        assert shmfile.is_file()
-        log.debug(f'Found matching shm buffer file: {filename}')
-        shmfiles.append(shmfile)
+        # the gap's RIGHT-most bar's OPEN value
+        # at that time (sample) step.
+        iend: int = row['index'][0]
+        # dt: datetime = row['dt'][0]
+        # dt_prev: datetime = row['dt_prev'][0]
+        # dt_end_t: float = dt.timestamp()
 
-    for shmfile in shmfiles:
 
-        # lookup array buffer size based on file suffix
-        # being either .rt or .hist
-        size: int = sizes[shmfile.name.rsplit('.')[-1]]
+        # TODO: can we eventually remove this
+        # once we figure out why the epoch cols
+        # don't match?
+        # TODO: FIX HOW/WHY these aren't matching
+        # and are instead off by 4hours (EST
+        # vs. UTC?!?!)
+        # end_t: float = row['time']
+        # assert (
+        #     dt.timestamp()
+        #     ==
+        #     end_t
+        # )
 
-        # attach to any shm buffer, load array into polars df,
-        # write to local parquet file.
-        shm, opened = maybe_open_shm_array(
-            key=shmfile.name,
-            size=size,
-            dtype=def_iohlcv_fields,
-            readonly=True,
+        # the gap's LEFT-most bar's CLOSE value
+        # at that time (sample) step.
+        prev_r: pl.DataFrame = wdts.filter(
+            pl.col('index') == iend - 1
         )
-        assert not opened
-        ohlcv = shm.array
+        # XXX: probably a gap in the (newly sorted or de-duplicated)
+        # dt-df, so we might need to re-index first..
+        if prev_r.is_empty():
+            await tractor.pause()
 
-        start = time.time()
+        istart: int = prev_r['index'][0]
+        # dt_start_t: float = dt_prev.timestamp()
 
-        # XXX: thanks to this SO answer for this conversion tip:
-        # https://stackoverflow.com/a/72054819
-        df = pl.DataFrame({
-            field_name: ohlcv[field_name]
-            for field_name in ohlcv.dtype.fields
-        })
-        delay: float = round(
-            time.time() - start,
-            ndigits=6,
+        # start_t: float = prev_r['time']
+        # assert (
+        #     dt_start_t
+        #     ==
+        #     start_t
+        # )
+
+        # TODO: implement px-col width measure
+        # and ensure at least as many px-cols
+        # shown per rect as configured by user.
+        # gap_w: float = abs((iend - istart))
+        # if gap_w < 6:
+        #     margin: float = 6
+        #     iend += margin
+        #     istart -= margin
+
+        rect_gap: float = BGM*3/8
+        opn: float = row['open'][0]
+        ro: tuple[float, float] = (
+            # dt_end_t,
+            iend + rect_gap + 1,
+            opn,
         )
-        log.info(
-            f'numpy -> polars conversion took {delay} secs\n'
-            f'polars df: {df}'
+        cls: float = prev_r['close'][0]
+        lc: tuple[float, float] = (
+            # dt_start_t,
+            istart - rect_gap, # + 1 ,
+            cls,
         )
 
-        yield (
-            shmfile,
-            shm,
-            df,
+        color: str = 'dad_blue'
+        diff: float = cls - opn
+        sgn: float = copysign(1, diff)
+        color: str = {
+            -1: 'buy_green',
+            1: 'sell_red',
+        }[sgn]
+
+        rect_kwargs: dict[str, Any] = dict(
+            fqme=fqme,
+            timeframe=timeframe,
+            start_pos=lc,
+            end_pos=ro,
+            color=color,
         )
 
+        aid: int = await actl.add_rect(**rect_kwargs)
+        assert aid
+        aids[aid] = rect_kwargs
+
+    # tell chart to redraw all its
+    # graphics view layers Bo
+    await actl.redraw(
+        fqme=fqme,
+        timeframe=timeframe,
+    )
+    return aids
+
 
 @store.command()
 def ldshm(
     fqme: str,
-
-    write_parquet: bool = False,
+    write_parquet: bool = True,
+    reload_parquet_to_shm: bool = True,
 
 ) -> None:
     '''
     Linux ONLY: load any fqme file name matching shm buffer from
     /dev/shm/ into an OHLCV numpy array and polars DataFrame,
-    optionally write to .parquet file.
+    optionally write to offline storage via `.parquet` file.
 
     '''
     async def main():
+        from piker.ui._remote_ctl import (
+            open_annot_ctl,
+        )
+        actl: AnnotCtl
+        mod: ModuleType
+        client: StorageClient
         async with (
             open_piker_runtime(
                 'polars_boi',
                 enable_modules=['piker.data._sharedmem'],
+                debug_mode=True,
             ),
+            open_storage_client() as (
+                mod,
+                client,
+            ),
+            open_annot_ctl() as actl,
         ):
+            shm_df: pl.DataFrame | None = None
+            tf2aids: dict[float, dict] = {}
 
-            df: pl.DataFrame | None = None
-            for shmfile, shm, df in iter_dfs_from_shms(fqme):
+            for (
+                shmfile,
+                shm,
+                # parquet_path,
+                shm_df,
+            ) in tsp.iter_dfs_from_shms(fqme):
 
-                # compute ohlc properties for naming
                 times: np.ndarray = shm.array['time']
-                secs: float = times[-1] - times[-2]
-                if secs < 1.:
-                    breakpoint()
+                d1: float = float(times[-1] - times[-2])
+                d2: float = float(times[-2] - times[-3])
+                med: float = np.median(np.diff(times))
+                if (
+                    d1 < 1.
+                    and d2 < 1.
+                    and med < 1.
+                ):
                     raise ValueError(
                         f'Something is wrong with time period for {shm}:\n{times}'
                     )
 
-                # TODO: maybe only optionally enter this depending
-                # on some CLI flags and/or gap detection?
-                await tractor.breakpoint()
+                period_s: float = float(max(d1, d2, med))
 
-                # write to parquet file?
-                if write_parquet:
-                    timeframe: str = f'{secs}s'
+                null_segs: tuple = tsp.get_null_segs(
+                    frame=shm.array,
+                    period=period_s,
+                )
 
-                    datadir: Path = get_conf_dir() / 'nativedb'
-                    if not datadir.is_dir():
-                        datadir.mkdir()
+                # TODO: call null-seg fixer somehow?
+                if null_segs:
+                    await tractor.pause()
+                #     async with (
+                #         trio.open_nursery() as tn,
+                #         mod.open_history_client(
+                #             mkt,
+                #         ) as (get_hist, config),
+                #     ):
+                #         nulls_detected: trio.Event = await tn.start(partial(
+                #             tsp.maybe_fill_null_segments,
 
-                    path: Path = datadir / f'{fqme}.{timeframe}.parquet'
+                #             shm=shm,
+                #             timeframe=timeframe,
+                #             get_hist=get_hist,
+                #             sampler_stream=sampler_stream,
+                #             mkt=mkt,
+                #         ))
 
-                    # write to fs
-                    start = time.time()
-                    df.write_parquet(path)
-                    delay: float = round(
-                        time.time() - start,
-                        ndigits=6,
-                    )
-                    log.info(
-                        f'parquet write took {delay} secs\n'
-                        f'file path: {path}'
+                # over-write back to shm?
+                wdts: pl.DataFrame  # with dts
+                deduped: pl.DataFrame  # deduplicated dts
+                (
+                    wdts,
+                    deduped,
+                    diff,
+                ) = tsp.dedupe(
+                    shm_df,
+                    period=period_s,
+                )
+
+                # detect gaps from in expected (uniform OHLC) sample period
+                step_gaps: pl.DataFrame = tsp.detect_time_gaps(
+                    deduped,
+                    expect_period=period_s,
+                )
+
+                # TODO: by default we always want to mark these up
+                # with rects showing up/down gaps Bo
+                venue_gaps: pl.DataFrame = tsp.detect_time_gaps(
+                    deduped,
+                    expect_period=period_s,
+
+                    # TODO: actually pull the exact duration
+                    # expected for each venue operational period?
+                    gap_dt_unit='days',
+                    gap_thresh=1,
+                )
+
+                # TODO: find the disjoint set of step gaps from
+                # venue (closure) set!
+                # -[ ] do a set diff by checking for the unique
+                #    gap set only in the step_gaps?
+                if (
+                    not venue_gaps.is_empty()
+                    or (
+                        period_s < 60
+                        and not step_gaps.is_empty()
                     )
+                ):
+                    # write repaired ts to parquet-file?
+                    if write_parquet:
+                        start: float = time.time()
+                        path: Path = await client.write_ohlcv(
+                            fqme,
+                            ohlcv=deduped,
+                            timeframe=period_s,
+                        )
+                        write_delay: float = round(
+                            time.time() - start,
+                            ndigits=6,
+                        )
 
-                    # read back from fs
-                    start = time.time()
-                    read_df: pl.DataFrame = pl.read_parquet(path)
-                    delay: float = round(
-                        time.time() - start,
-                        ndigits=6,
-                    )
-                    print(
-                        f'parquet read took {delay} secs\n'
-                        f'polars df: {read_df}'
-                    )
+                        # read back from fs
+                        start: float = time.time()
+                        read_df: pl.DataFrame = pl.read_parquet(path)
+                        read_delay: float = round(
+                            time.time() - start,
+                            ndigits=6,
+                        )
+                        log.info(
+                            f'parquet write took {write_delay} secs\n'
+                            f'file path: {path}'
+                            f'parquet read took {read_delay} secs\n'
+                            f'polars df: {read_df}'
+                        )
 
-            if df is None:
-                log.error(f'No matching shm buffers for {fqme} ?')
+                        if reload_parquet_to_shm:
+                            new = tsp.pl2np(
+                                deduped,
+                                dtype=shm.array.dtype,
+                            )
+                            # since normally readonly
+                            shm._array.setflags(
+                                write=int(1),
+                            )
+                            shm.push(
+                                new,
+                                prepend=True,
+                                start=new['index'][-1],
+                                update_first=False,  # don't update ._first
+                            )
+
+                    do_markup_gaps: bool = True
+                    if do_markup_gaps:
+                        new_df: pl.DataFrame = tsp.np2pl(new)
+                        aids: dict = await markup_gaps(
+                            fqme,
+                            period_s,
+                            actl,
+                            new_df,
+                            step_gaps,
+                        )
+                        # last chance manual overwrites in REPL
+                        # await tractor.pause()
+                        assert aids
+                        tf2aids[period_s] = aids
+
+                else:
+                    # allow interaction even when no ts problems.
+                    assert not diff
+
+            await tractor.pause()
+            log.info('Exiting TSP shm anal-izer!')
+
+            if shm_df is None:
+                log.error(
+                    f'No matching shm buffers for {fqme} ?'
+
+                )
 
     trio.run(main)
 

piker/storage/marketstore/__init__.py

@@ -59,7 +59,6 @@ from anyio_marketstore import (  # noqa
     Params,
 )
 from piker.log import get_logger
-# from .._profile import Profiler
 
 
 log = get_logger(__name__)
@@ -205,7 +204,7 @@ class MktsStorageClient:
         #         break
         #     except purerpc.grpclib.exceptions.UnknownError as err:
         #         if 'snappy' in err.args:
-        #             await tractor.breakpoint()
+        #             await tractor.pause()
 
         #         # indicate there is no history for this timeframe
         #         log.exception(
@@ -233,7 +232,7 @@ class MktsStorageClient:
                     'YOUR DATABASE LIKELY CONTAINS BAD DATA FROM AN OLD BUG '
                     f'WIPING HISTORY FOR {ts}s'
                 )
-                await tractor.breakpoint()
+                await tractor.pause()
                 # await self.delete_ts(fqme, timeframe)
 
                 # try reading again..

piker/storage/nativedb.py

@@ -19,7 +19,8 @@
             call a poor man's tsdb).
 
 AKA a `piker`-native file-system native "time series database"
-without needing an extra process and no standard TSDB features, YET!
+without needing an extra process and no standard TSDB features,
+YET!
 
 '''
 # TODO: like there's soo much..
@@ -55,8 +56,6 @@ from datetime import datetime
 from pathlib import Path
 import time
 
-# from bidict import bidict
-# import tractor
 import numpy as np
 import polars as pl
 from pendulum import (
@@ -64,46 +63,18 @@ from pendulum import (
 )
 
 from piker import config
-from piker.data import def_iohlcv_fields
-from piker.data import ShmArray
+from piker import tsp
+from piker.data import (
+    def_iohlcv_fields,
+    ShmArray,
+)
 from piker.log import get_logger
-# from .._profile import Profiler
+from . import TimeseriesNotFound
 
 
 log = get_logger('storage.nativedb')
 
 
-# NOTE: thanks to this SO answer for the below conversion routines
-# to go from numpy struct-arrays to polars dataframes and back:
-# https://stackoverflow.com/a/72054819
-def np2pl(array: np.ndarray) -> pl.DataFrame:
-    return pl.DataFrame({
-        field_name: array[field_name]
-        for field_name in array.dtype.fields
-    })
-
-
-def pl2np(
-    df: pl.DataFrame,
-    dtype: np.dtype,
-
-) -> np.ndarray:
-
-    # Create numpy struct array of the correct size and dtype
-    # and loop through df columns to fill in array fields.
-    array = np.empty(
-        df.height,
-        dtype,
-    )
-    for field, col in zip(
-        dtype.fields,
-        df.columns,
-    ):
-        array[field] = df.get_column(col).to_numpy()
-
-    return array
-
-
 def detect_period(shm: ShmArray) -> float:
     '''
     Attempt to detect the series time step sampling period
@@ -124,16 +95,19 @@ def detect_period(shm: ShmArray) -> float:
 
 def mk_ohlcv_shm_keyed_filepath(
     fqme: str,
-    period: float,  # ow known as the "timeframe"
+    period: float | int,  # ow known as the "timeframe"
     datadir: Path,
 
-) -> str:
+) -> Path:
 
     if period < 1.:
         raise ValueError('Sample period should be >= 1.!?')
 
-    period_s: str = f'{period}s'
-    path: Path = datadir / f'{fqme}.ohlcv{period_s}.parquet'
+    path: Path = (
+        datadir
+        /
+        f'{fqme}.ohlcv{int(period)}s.parquet'
+    )
     return path
 
 
@@ -187,7 +161,13 @@ class NativeStorageClient:
 
     def index_files(self):
         for path in self._datadir.iterdir():
-            if 'borked' in path.name:
+            if (
+                path.is_dir()
+                or
+                '.parquet' not in str(path)
+                # or
+                # path.name in {'borked', 'expired',}
+            ):
                 continue
 
             key: str = path.name.rstrip('.parquet')
@@ -229,8 +209,21 @@ class NativeStorageClient:
                 fqme,
                 timeframe,
             )
-        except FileNotFoundError:
-            return None
+        except FileNotFoundError as fnfe:
+
+            bs_fqme, _, *_ = fqme.rpartition('.')
+
+            possible_matches: list[str] = []
+            for tskey in self._index:
+                if bs_fqme in tskey:
+                    possible_matches.append(tskey)
+
+            match_str: str = '\n'.join(sorted(possible_matches))
+            raise TimeseriesNotFound(
+                f'No entry for `{fqme}`?\n'
+                f'Maybe you need a more specific fqme-key like:\n\n'
+                f'{match_str}'
+            ) from fnfe
 
         times = array['time']
         return (
@@ -243,6 +236,7 @@ class NativeStorageClient:
         self,
         fqme: str,
         period: float,
+
     ) -> Path:
         return mk_ohlcv_shm_keyed_filepath(
             fqme=fqme,
@@ -250,6 +244,23 @@ class NativeStorageClient:
             datadir=self._datadir,
         )
 
+    def _cache_df(
+        self,
+        fqme: str,
+        df: pl.DataFrame,
+        timeframe: float,
+
+    ) -> None:
+        # cache df for later usage since we (currently) need to
+        # convert to np.ndarrays to push to our `ShmArray` rt
+        # buffers subsys but later we may operate entirely on
+        # pyarrow arrays/buffers so keeping the dfs around for
+        # a variety of purposes is handy.
+        self._dfs.setdefault(
+            timeframe,
+            {},
+        )[fqme] = df
+
     async def read_ohlcv(
         self,
         fqme: str,
@@ -258,13 +269,20 @@ class NativeStorageClient:
         # limit: int = int(200e3),
 
     ) -> np.ndarray:
-        path: Path = self.mk_path(fqme, period=int(timeframe))
+        path: Path = self.mk_path(
+            fqme,
+            period=int(timeframe),
+        )
         df: pl.DataFrame = pl.read_parquet(path)
-        self._dfs.setdefault(timeframe, {})[fqme] = df
 
+        self._cache_df(
+            fqme=fqme,
+            df=df,
+            timeframe=timeframe,
+        )
         # TODO: filter by end and limit inputs
         # times: pl.Series = df['time']
-        array: np.ndarray = pl2np(
+        array: np.ndarray = tsp.pl2np(
             df,
             dtype=np.dtype(def_iohlcv_fields),
         )
@@ -274,11 +292,15 @@ class NativeStorageClient:
         self,
         fqme: str,
         period: int = 60,
+        load_from_offline: bool = True,
 
     ) -> pl.DataFrame:
         try:
             return self._dfs[period][fqme]
         except KeyError:
+            if not load_from_offline:
+                raise
+
             await self.read_ohlcv(fqme, period)
             return self._dfs[period][fqme]
 
@@ -300,32 +322,39 @@ class NativeStorageClient:
             datadir=self._datadir,
         )
         if isinstance(ohlcv, np.ndarray):
-            df: pl.DataFrame = np2pl(ohlcv)
+            df: pl.DataFrame = tsp.np2pl(ohlcv)
         else:
             df = ohlcv
 
+        self._cache_df(
+            fqme=fqme,
+            df=df,
+            timeframe=timeframe,
+        )
+
         # TODO: in terms of managing the ultra long term data
-        # - use a proper profiler to measure all this IO and
+        # -[ ] use a proper profiler to measure all this IO and
         #   roundtripping!
-        # - try out ``fastparquet``'s append writing:
-        # https://fastparquet.readthedocs.io/en/latest/api.html#fastparquet.write
+        # -[ ] implement parquet append!? see issue:
+        #   https://github.com/pikers/piker/issues/536
+        #   -[ ] try out ``fastparquet``'s append writing:
+        #     https://fastparquet.readthedocs.io/en/latest/api.html#fastparquet.write
         start = time.time()
         df.write_parquet(path)
         delay: float = round(
             time.time() - start,
             ndigits=6,
         )
-        print(
+        log.info(
             f'parquet write took {delay} secs\n'
             f'file path: {path}'
         )
         return path
 
-
     async def write_ohlcv(
         self,
         fqme: str,
-        ohlcv: np.ndarray,
+        ohlcv: np.ndarray | pl.DataFrame,
         timeframe: int,
 
     ) -> Path:
@@ -377,6 +406,8 @@ class NativeStorageClient:
     #     ...
 
 
+# TODO: does this need to be async on average?
+# I guess for any IPC connected backend yes?
 @acm
 async def get_client(
 
@@ -394,7 +425,7 @@ async def get_client(
     '''
     datadir: Path = config.get_conf_dir() / 'nativedb'
     if not datadir.is_dir():
-        log.info(f'Creating `nativedb` director: {datadir}')
+        log.info(f'Creating `nativedb` dir: {datadir}')
         datadir.mkdir()
 
     client = NativeStorageClient(datadir)

piker/toolz/__init__.py

@@ -0,0 +1,29 @@
+# piker: trading gear for hackers
+# Copyright (C) Tyler Goodlet (in stewardship for pikers)
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+'''
+Toolz for debug, profile and trace of the distributed runtime :surfer:
+
+'''
+from tractor.devx import (
+    open_crash_handler as open_crash_handler,
+)
+from .profile import (
+    Profiler as Profiler,
+    pg_profile_enabled as pg_profile_enabled,
+    ms_slower_then as ms_slower_then,
+    timeit as timeit,
+)

piker/trionics.py

@@ -1,80 +0,0 @@
-# piker: trading gear for hackers
-# Copyright (C) Tyler Goodlet (in stewardship of piker0)
-
-# This program is free software: you can redistribute it and/or modify
-# it under the terms of the GNU Affero General Public License as published by
-# the Free Software Foundation, either version 3 of the License, or
-# (at your option) any later version.
-
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU Affero General Public License for more details.
-
-# You should have received a copy of the GNU Affero General Public License
-# along with this program.  If not, see <https://www.gnu.org/licenses/>.
-
-'''
-sugarz for trio/tractor conc peeps.
-
-'''
-from typing import AsyncContextManager
-from typing import TypeVar
-from contextlib import asynccontextmanager as acm
-
-import trio
-
-
-# A regular invariant generic type
-T = TypeVar("T")
-
-
-async def _enter_and_sleep(
-
-    mngr: AsyncContextManager[T],
-    to_yield: dict[int, T],
-    all_entered: trio.Event,
-    # task_status: TaskStatus[T] = trio.TASK_STATUS_IGNORED,
-
-) -> T:
-    '''Open the async context manager deliver it's value
-    to this task's spawner and sleep until cancelled.
-
-    '''
-    async with mngr as value:
-        to_yield[id(mngr)] = value
-
-        if all(to_yield.values()):
-            all_entered.set()
-
-        # sleep until cancelled
-        await trio.sleep_forever()
-
-
-@acm
-async def async_enter_all(
-
-    *mngrs: list[AsyncContextManager[T]],
-
-) -> tuple[T]:
-
-    to_yield = {}.fromkeys(id(mngr) for mngr in mngrs)
-
-    all_entered = trio.Event()
-
-    async with trio.open_nursery() as n:
-        for mngr in mngrs:
-            n.start_soon(
-                _enter_and_sleep,
-                mngr,
-                to_yield,
-                all_entered,
-            )
-
-        # deliver control once all managers have started up
-        await all_entered.wait()
-        yield tuple(to_yield.values())
-
-        # tear down all sleeper tasks thus triggering individual
-        # mngr ``__aexit__()``s.
-        n.cancel_scope.cancel()

piker/tsp/_anal.py

@@ -0,0 +1,746 @@
+# piker: trading gear for hackers
+# Copyright (C) 2018-present  Tyler Goodlet (in stewardship of pikers)
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+'''
+Financial time series processing utilities usually
+pertaining to OHLCV style sampled data.
+
+Routines are generally implemented in either ``numpy`` or
+``polars`` B)
+
+'''
+from __future__ import annotations
+from functools import partial
+from math import (
+    ceil,
+    floor,
+)
+import time
+from typing import (
+    Literal,
+    # AsyncGenerator,
+    Generator,
+)
+
+import numpy as np
+import polars as pl
+from pendulum import (
+    DateTime,
+    from_timestamp,
+)
+
+from ..toolz.profile import (
+    Profiler,
+    pg_profile_enabled,
+    ms_slower_then,
+)
+from ..log import (
+    get_logger,
+    get_console_log,
+)
+# for "time series processing"
+subsys: str = 'piker.tsp'
+
+log = get_logger(subsys)
+get_console_log = partial(
+    get_console_log,
+    name=subsys,
+)
+
+# NOTE: union type-defs to handle generic `numpy` and `polars` types
+# side-by-side Bo
+# |_ TODO: schema spec typing?
+#   -[ ] nptyping!
+#   -[ ] wtv we can with polars?
+Frame = pl.DataFrame | np.ndarray
+Seq = pl.Series | np.ndarray
+
+
+def slice_from_time(
+    arr: np.ndarray,
+    start_t: float,
+    stop_t: float,
+    step: float,  # sampler period step-diff
+
+) -> slice:
+    '''
+    Calculate array indices mapped from a time range and return them in
+    a slice.
+
+    Given an input array with an epoch `'time'` series entry, calculate
+    the indices which span the time range and return in a slice. Presume
+    each `'time'` step increment is uniform and when the time stamp
+    series contains gaps (the uniform presumption is untrue) use
+    ``np.searchsorted()`` binary search to look up the appropriate
+    index.
+
+    '''
+    profiler = Profiler(
+        msg='slice_from_time()',
+        disabled=not pg_profile_enabled(),
+        ms_threshold=ms_slower_then,
+    )
+
+    times = arr['time']
+    t_first = floor(times[0])
+    t_last = ceil(times[-1])
+
+    # the greatest index we can return which slices to the
+    # end of the input array.
+    read_i_max = arr.shape[0]
+
+    # compute (presumed) uniform-time-step index offsets
+    i_start_t = floor(start_t)
+    read_i_start = floor(((i_start_t - t_first) // step)) - 1
+
+    i_stop_t = ceil(stop_t)
+
+    # XXX: edge case -> always set stop index to last in array whenever
+    # the input stop time is detected to be greater then the equiv time
+    # stamp at that last entry.
+    if i_stop_t >= t_last:
+        read_i_stop = read_i_max
+    else:
+        read_i_stop = ceil((i_stop_t - t_first) // step) + 1
+
+    # always clip outputs to array support
+    # for read start:
+    # - never allow a start < the 0 index
+    # - never allow an end index > the read array len
+    read_i_start = min(
+        max(0, read_i_start),
+        read_i_max - 1,
+    )
+    read_i_stop = max(
+        0,
+        min(read_i_stop, read_i_max),
+    )
+
+    # check for larger-then-latest calculated index for given start
+    # time, in which case we do a binary search for the correct index.
+    # NOTE: this is usually the result of a time series with time gaps
+    # where it is expected that each index step maps to a uniform step
+    # in the time stamp series.
+    t_iv_start = times[read_i_start]
+    if (
+        t_iv_start > i_start_t
+    ):
+        # do a binary search for the best index mapping to ``start_t``
+        # given we measured an overshoot using the uniform-time-step
+        # calculation from above.
+
+        # TODO: once we start caching these per source-array,
+        # we can just overwrite ``read_i_start`` directly.
+        new_read_i_start = np.searchsorted(
+            times,
+            i_start_t,
+            side='left',
+        )
+
+        # TODO: minimize binary search work as much as possible:
+        # - cache these remap values which compensate for gaps in the
+        #   uniform time step basis where we calc a later start
+        #   index for the given input ``start_t``.
+        # - can we shorten the input search sequence by heuristic?
+        #   up_to_arith_start = index[:read_i_start]
+
+        if (
+            new_read_i_start <= read_i_start
+        ):
+            # t_diff = t_iv_start - start_t
+            # print(
+            #     f"WE'RE CUTTING OUT TIME - STEP:{step}\n"
+            #     f'start_t:{start_t} -> 0index start_t:{t_iv_start}\n'
+            #     f'diff: {t_diff}\n'
+            #     f'REMAPPED START i: {read_i_start} -> {new_read_i_start}\n'
+            # )
+            read_i_start = new_read_i_start
+
+    t_iv_stop = times[read_i_stop - 1]
+    if (
+        t_iv_stop > i_stop_t
+    ):
+        # t_diff = stop_t - t_iv_stop
+        # print(
+        #     f"WE'RE CUTTING OUT TIME - STEP:{step}\n"
+        #     f'calced iv stop:{t_iv_stop} -> stop_t:{stop_t}\n'
+        #     f'diff: {t_diff}\n'
+        #     # f'SHOULD REMAP STOP: {read_i_start} -> {new_read_i_start}\n'
+        # )
+        new_read_i_stop = np.searchsorted(
+            times[read_i_start:],
+            # times,
+            i_stop_t,
+            side='right',
+        )
+
+        if (
+            new_read_i_stop <= read_i_stop
+        ):
+            read_i_stop = read_i_start + new_read_i_stop + 1
+
+    # sanity checks for range size
+    # samples = (i_stop_t - i_start_t) // step
+    # index_diff = read_i_stop - read_i_start + 1
+    # if index_diff > (samples + 3):
+    #     breakpoint()
+
+    # read-relative indexes: gives a slice where `shm.array[read_slc]`
+    # will be the data spanning the input time range `start_t` ->
+    # `stop_t`
+    read_slc = slice(
+        int(read_i_start),
+        int(read_i_stop),
+    )
+
+    profiler(
+        'slicing complete'
+        # f'{start_t} -> {abs_slc.start} | {read_slc.start}\n'
+        # f'{stop_t} -> {abs_slc.stop} | {read_slc.stop}\n'
+    )
+
+    # NOTE: if caller needs absolute buffer indices they can
+    # slice the buffer abs index like so:
+    # index = arr['index']
+    # abs_indx = index[read_slc]
+    # abs_slc = slice(
+    #     int(abs_indx[0]),
+    #     int(abs_indx[-1]),
+    # )
+
+    return read_slc
+
+
+def get_null_segs(
+    frame: Frame,
+    period: float,  # sampling step in seconds
+    imargin: int = 1,
+    col: str = 'time',
+
+) -> tuple[
+    # Seq,  # TODO: can we make it an array-type instead?
+    list[
+        list[int, int],
+    ],
+    Seq,
+    Frame
+] | None:
+    '''
+    Detect if there are any zero(-epoch stamped) valued
+    rows in for the provided `col: str` column; by default
+    presume the 'time' field/column.
+
+    Filter to all such zero (time) segments and return
+    the corresponding frame zeroed segment's,
+
+      - gap absolute (in buffer terms) indices-endpoints as
+        `absi_zsegs`
+      - abs indices of all rows with zeroed `col` values as `absi_zeros`
+      - the corresponding frame's row-entries (view) which are
+        zeroed for the `col` as `zero_t`
+
+    '''
+    times: Seq = frame['time']
+    zero_pred: Seq = (times == 0)
+
+    if isinstance(frame, np.ndarray):
+        tis_zeros: int = zero_pred.any()
+    else:
+        tis_zeros: int = zero_pred.any()
+
+    if not tis_zeros:
+        return None
+
+    # TODO: use ndarray for this?!
+    absi_zsegs: list[list[int, int]] = []
+
+    if isinstance(frame, np.ndarray):
+        # view of ONLY the zero segments as one continuous chunk
+        zero_t: np.ndarray = frame[zero_pred]
+        # abs indices of said zeroed rows
+        absi_zeros = zero_t['index']
+        # diff of abs index steps between each zeroed row
+        absi_zdiff: np.ndarray = np.diff(absi_zeros)
+
+        # scan for all frame-indices where the
+        # zeroed-row-abs-index-step-diff is greater then the
+        # expected increment of 1.
+        # data  1st zero seg  data  zeros
+        # ----  ------------  ----  -----  ------  ----
+        # ||||..000000000000..||||..00000..||||||..0000
+        # ----  ------------  ----  -----  ------  ----
+        #       ^zero_t[0]                            ^zero_t[-1]
+        #                           ^fi_zgaps[0]   ^fi_zgaps[1]
+        #       ^absi_zsegs[0][0]   ^---^ => absi_zsegs[1]: tuple
+        #  absi_zsegs[0][1]^
+        #
+        # NOTE: the first entry in `fi_zgaps` is where
+        # the first (absolute) index step diff is > 1.
+        # and it is a frame-relative index into `zero_t`.
+        fi_zgaps = np.argwhere(
+            absi_zdiff > 1
+            # NOTE: +1 here is ensure we index to the "start" of each
+            # segment (if we didn't the below loop needs to be
+            # re-written to expect `fi_end_rows`!
+        ) + 1
+        # the rows from the contiguous zeroed segments which have
+        # abs-index steps >1 compared to the previous zero row
+        # (indicating an end of zeroed segment).
+        fi_zseg_start_rows = zero_t[fi_zgaps]
+
+    # TODO: equiv for pl.DataFrame case!
+    else:
+        izeros: pl.Series = zero_pred.arg_true()
+        zero_t: pl.DataFrame = frame[izeros]
+
+        absi_zeros = zero_t['index']
+        absi_zdiff: pl.Series = absi_zeros.diff()
+        fi_zgaps = (absi_zdiff > 1).arg_true()
+
+    # XXX: our goal (in this func) is to select out slice index
+    # pairs (zseg0_start, zseg_end) in abs index units for each
+    # null-segment portion detected throughout entire input frame.
+
+    # only up to one null-segment in entire frame?
+    num_gaps: int = fi_zgaps.size + 1
+    if num_gaps < 1:
+        if absi_zeros.size > 1:
+            absi_zsegs = [[
+                # TODO: maybe mk these max()/min() limits func
+                # consts instead of called more then once?
+                max(
+                    absi_zeros[0] - 1,
+                    0,
+                ),
+                # NOTE: need the + 1 to guarantee we index "up to"
+                # the next non-null row-datum.
+                min(
+                    absi_zeros[-1] + 1,
+                    frame['index'][-1],
+                ),
+            ]]
+        else:
+            # XXX EDGE CASE: only one null-datum found so
+            # mark the start abs index as None to trigger
+            # a full frame-len query to the respective backend?
+            absi_zsegs = [[
+                # see `get_hist()` in backend, should ALWAYS be
+                # able to handle a `start_dt=None`!
+                # None,
+                None,
+                absi_zeros[0] + 1,
+            ]]
+
+    # XXX NOTE XXX: if >= 2 zeroed segments are found, there should
+    # ALWAYS be more then one zero-segment-abs-index-step-diff row
+    # in `absi_zdiff`, so loop through all such
+    # abs-index-step-diffs >1 (i.e. the entries of `absi_zdiff`)
+    # and add them as the "end index" entries for each segment.
+    # Then, iif NOT iterating the first such segment end, look back
+    # for the prior segments zero-segment start indext by relative
+    # indexing the `zero_t` frame by -1 and grabbing the abs index
+    # of what should be the prior zero-segment abs start index.
+    else:
+        # NOTE: since `absi_zdiff` will never have a row
+        # corresponding to the first zero-segment's row, we add it
+        # manually here.
+        absi_zsegs.append([
+            max(
+                absi_zeros[0] - 1,
+                0,
+            ),
+            None,
+        ])
+
+        # TODO: can we do it with vec ops?
+        for i, (
+            fi,  # frame index of zero-seg start
+            zseg_start_row,  # full row for ^
+        ) in enumerate(zip(
+            fi_zgaps,
+            fi_zseg_start_rows,
+        )):
+            assert (zseg_start_row == zero_t[fi]).all()
+            iabs: int = zseg_start_row['index'][0]
+            absi_zsegs.append([
+                iabs - 1,
+                None,  # backfilled on next iter
+            ])
+
+            # final iter case, backfill FINAL end iabs!
+            if (i + 1) == fi_zgaps.size:
+                absi_zsegs[-1][1] = absi_zeros[-1] + 1
+
+            # NOTE: only after the first segment (due to `.diff()`
+            # usage above) can we do a lookback to the prior
+            # segment's end row and determine it's abs index to
+            # retroactively insert to the prior
+            # `absi_zsegs[i-1][1]` entry Bo
+            last_end: int = absi_zsegs[i][1]
+            if last_end is None:
+                prev_zseg_row = zero_t[fi - 1]
+                absi_post_zseg = prev_zseg_row['index'][0] + 1
+                # XXX: MUST BACKFILL previous end iabs!
+                absi_zsegs[i][1] = absi_post_zseg
+
+        else:
+            if 0 < num_gaps < 2:
+                absi_zsegs[-1][1] = min(
+                    absi_zeros[-1] + 1,
+                    frame['index'][-1],
+                )
+
+            iabs_first: int = frame['index'][0]
+            for start, end in absi_zsegs:
+
+                ts_start: float = times[start - iabs_first]
+                ts_end: float = times[end - iabs_first]
+                if (
+                    (ts_start == 0 and not start == 0)
+                    or
+                    ts_end == 0
+                ):
+                    import pdbp
+                    pdbp.set_trace()
+
+                assert end
+                assert start < end
+
+    log.warning(
+        f'Frame has {len(absi_zsegs)} NULL GAPS!?\n'
+        f'period: {period}\n'
+        f'total null samples: {len(zero_t)}\n'
+    )
+
+    return (
+        absi_zsegs,  # [start, end] abs slice indices of seg
+        absi_zeros,  # all abs indices within all null-segs
+        zero_t,  # sliced-view of all null-segment rows-datums
+    )
+
+
+def iter_null_segs(
+    timeframe: float,
+    frame: Frame | None = None,
+    null_segs: tuple | None = None,
+
+) -> Generator[
+    tuple[
+        int, int,
+        int, int,
+        float, float,
+        float, float,
+
+        # Seq,  # TODO: can we make it an array-type instead?
+        # list[
+        #     list[int, int],
+        # ],
+        # Seq,
+        # Frame
+    ],
+    None,
+]:
+    if not (
+        null_segs := get_null_segs(
+            frame,
+            period=timeframe,
+        )
+    ):
+        return
+
+    absi_pairs_zsegs: list[list[float, float]]
+    izeros: Seq
+    zero_t: Frame
+    (
+        absi_pairs_zsegs,
+        izeros,
+        zero_t,
+    ) = null_segs
+
+    absi_first: int = frame[0]['index']
+    for (
+        absi_start,
+        absi_end,
+    ) in absi_pairs_zsegs:
+
+        fi_end: int = absi_end - absi_first
+        end_row: Seq = frame[fi_end]
+        end_t: float = end_row['time']
+        end_dt: DateTime = from_timestamp(end_t)
+
+        fi_start = None
+        start_row = None
+        start_t = None
+        start_dt = None
+        if (
+            absi_start is not None
+            and start_t != 0
+        ):
+            fi_start: int = absi_start - absi_first
+            start_row: Seq = frame[fi_start]
+            start_t: float = start_row['time']
+            start_dt: DateTime = from_timestamp(start_t)
+
+        if absi_start < 0:
+            import pdbp
+            pdbp.set_trace()
+
+        yield (
+            absi_start, absi_end,  # abs indices
+            fi_start, fi_end,  # relative "frame" indices
+            start_t, end_t,
+            start_dt, end_dt,
+        )
+
+
+def with_dts(
+    df: pl.DataFrame,
+    time_col: str = 'time',
+
+) -> pl.DataFrame:
+    '''
+    Insert datetime (casted) columns to a (presumably) OHLC sampled
+    time series with an epoch-time column keyed by `time_col: str`.
+
+    '''
+    return df.with_columns([
+        pl.col(time_col).shift(1).suffix('_prev'),
+        pl.col(time_col).diff().alias('s_diff'),
+        pl.from_epoch(pl.col(time_col)).alias('dt'),
+    ]).with_columns([
+        pl.from_epoch(
+            column=pl.col(f'{time_col}_prev'),
+        ).alias('dt_prev'),
+        pl.col('dt').diff().alias('dt_diff'),
+    ])
+
+
+t_unit: Literal = Literal[
+    'days',
+    'hours',
+    'minutes',
+    'seconds',
+    'miliseconds',
+    'microseconds',
+    'nanoseconds',
+]
+
+
+def detect_time_gaps(
+    w_dts: pl.DataFrame,
+
+    time_col: str = 'time',
+    # epoch sampling step diff
+    expect_period: float = 60,
+
+    # NOTE: legacy stock mkts have venue operating hours
+    # and thus gaps normally no more then 1-2 days at
+    # a time.
+    gap_thresh: float = 1.,
+
+    # TODO: allow passing in a frame of operating hours?
+    # -[ ] durations/ranges for faster legit gap checks?
+    # XXX -> must be valid ``polars.Expr.dt.<name>``
+    # like 'days' which a sane default for venue closures
+    # though will detect weekend gaps which are normal :o
+    gap_dt_unit: t_unit | None = None,
+
+) -> pl.DataFrame:
+    '''
+    Filter to OHLC datums which contain sample step gaps.
+
+    For eg. legacy markets which have venue close gaps and/or
+    actual missing data segments.
+
+    '''
+    # first select by any sample-period (in seconds unit) step size
+    # greater then expected.
+    step_gaps: pl.DataFrame = w_dts.filter(
+        pl.col('s_diff').abs() > expect_period
+    )
+
+    if gap_dt_unit is None:
+        return step_gaps
+
+    # NOTE: this flag is to indicate that on this (sampling) time
+    # scale we expect to only be filtering against larger venue
+    # closures-scale time gaps.
+    return step_gaps.filter(
+        # Second by an arbitrary dt-unit step size
+        getattr(
+            pl.col('dt_diff').dt,
+            gap_dt_unit,
+        )().abs() > gap_thresh
+    )
+
+
+def detect_price_gaps(
+    df: pl.DataFrame,
+    gt_multiplier: float = 2.,
+    price_fields: list[str] = ['high', 'low'],
+
+) -> pl.DataFrame:
+    '''
+    Detect gaps in clearing price over an OHLC series.
+
+    2 types of gaps generally exist; up gaps and down gaps:
+
+    - UP gap: when any next sample's lo price is strictly greater
+      then the current sample's hi price.
+
+    - DOWN gap: when any next sample's hi price is strictly
+      less then the current samples lo price.
+
+    '''
+    # return df.filter(
+    #     pl.col('high') - ) > expect_period,
+    # ).select([
+    #     pl.dt.datetime(pl.col(time_col).shift(1)).suffix('_previous'),
+    #     pl.all(),
+    # ]).select([
+    #     pl.all(),
+    #     (pl.col(time_col) - pl.col(f'{time_col}_previous')).alias('diff'),
+    # ])
+    ...
+
+# TODO: probably just use the null_segs impl above?
+def detect_vlm_gaps(
+    df: pl.DataFrame,
+    col: str = 'volume',
+
+) -> pl.DataFrame:
+
+    vnull: pl.DataFrame = w_dts.filter(
+        pl.col(col) == 0
+    )
+    return vnull
+
+
+def dedupe(
+    src_df: pl.DataFrame,
+
+    time_gaps: pl.DataFrame | None = None,
+    sort: bool = True,
+    period: float = 60,
+
+) -> tuple[
+    pl.DataFrame,  # with dts
+    pl.DataFrame,  # with deduplicated dts (aka gap/repeat removal)
+    int,  # len diff between input and deduped
+]:
+    '''
+    Check for time series gaps and if found
+    de-duplicate any datetime entries, check for
+    a frame height diff and return the newly
+    dt-deduplicated frame.
+
+    '''
+    wdts: pl.DataFrame = with_dts(src_df)
+
+    deduped = wdts
+
+    # remove duplicated datetime samples/sections
+    deduped: pl.DataFrame = wdts.unique(
+        # subset=['dt'],
+        subset=['time'],
+        maintain_order=True,
+    )
+
+    # maybe sort on any time field
+    if sort:
+        deduped = deduped.sort(by='time')
+        # TODO: detect out-of-order segments which were corrected!
+        # -[ ] report in log msg
+        # -[ ] possibly return segment sections which were moved?
+
+    diff: int = (
+        wdts.height
+        -
+        deduped.height
+    )
+    return (
+        wdts,
+        deduped,
+        diff,
+    )
+
+
+def sort_diff(
+    src_df: pl.DataFrame,
+    col: str = 'time',
+
+) -> tuple[
+    pl.DataFrame,  # with dts
+    pl.DataFrame,  # sorted
+    list[int],  # indices of segments that are out-of-order
+]:
+    ser: pl.Series = src_df[col]
+    sortd: pl.DataFrame = ser.sort()
+    diff: pl.Series = ser.diff()
+
+    sortd_diff: pl.Series = sortd.diff()
+    i_step_diff = (diff != sortd_diff).arg_true()
+    frame_reorders: int = i_step_diff.len()
+    if frame_reorders:
+        log.warn(
+            f'Resorted frame on col: {col}\n'
+            f'{frame_reorders}'
+
+        )
+        # import pdbp; pdbp.set_trace()
+
+# NOTE: thanks to this SO answer for the below conversion routines
+# to go from numpy struct-arrays to polars dataframes and back:
+# https://stackoverflow.com/a/72054819
+def np2pl(array: np.ndarray) -> pl.DataFrame:
+    start: float = time.time()
+
+    # XXX: thanks to this SO answer for this conversion tip:
+    # https://stackoverflow.com/a/72054819
+    df = pl.DataFrame({
+        field_name: array[field_name]
+        for field_name in array.dtype.fields
+    })
+    delay: float = round(
+        time.time() - start,
+        ndigits=6,
+    )
+    log.info(
+        f'numpy -> polars conversion took {delay} secs\n'
+        f'polars df: {df}'
+    )
+    return df
+
+
+def pl2np(
+    df: pl.DataFrame,
+    dtype: np.dtype,
+
+) -> np.ndarray:
+
+    # Create numpy struct array of the correct size and dtype
+    # and loop through df columns to fill in array fields.
+    array = np.empty(
+        df.height,
+        dtype,
+    )
+    for field, col in zip(
+        dtype.fields,
+        df.columns,
+    ):
+        array[field] = df.get_column(col).to_numpy()
+
+    return array

piker/types.py

@@ -0,0 +1,250 @@
+# piker: trading gear for hackers
+# Copyright (C) (in stewardship for pikers)
+#  - Tyler Goodlet
+#  - Guillermo Rodriguez
+
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+'''
+Extensions to built-in or (heavily used but 3rd party) friend-lib
+types.
+
+'''
+from __future__ import annotations
+from collections import UserList
+from pprint import (
+    saferepr,
+)
+from typing import Any
+
+from msgspec import (
+    msgpack,
+    Struct as _Struct,
+    structs,
+)
+
+
+class DiffDump(UserList):
+    '''
+    Very simple list delegator that repr() dumps (presumed) tuple
+    elements of the form `tuple[str, Any, Any]` in a nice
+    multi-line readable form for analyzing `Struct` diffs.
+
+    '''
+    def __repr__(self) -> str:
+        if not len(self):
+            return super().__repr__()
+
+        # format by displaying item pair's ``repr()`` on multiple,
+        # indented lines such that they are more easily visually
+        # comparable when printed to console when printed to
+        # console.
+        repstr: str = '[\n'
+        for k, left, right in self:
+            repstr += (
+                f'({k},\n'
+                f'\t{repr(left)},\n'
+                f'\t{repr(right)},\n'
+                ')\n'
+            )
+        repstr += ']\n'
+        return repstr
+
+
+class Struct(
+    _Struct,
+
+    # https://jcristharif.com/msgspec/structs.html#tagged-unions
+    # tag='pikerstruct',
+    # tag=True,
+):
+    '''
+    A "human friendlier" (aka repl buddy) struct subtype.
+
+    '''
+    def _sin_props(self) -> Iterator[
+        tuple[
+            structs.FieldIinfo,
+            str,
+            Any,
+        ]
+    ]:
+        '''
+        Iterate over all non-@property fields of this struct.
+
+        '''
+        fi: structs.FieldInfo
+        for fi in structs.fields(self):
+            key: str = fi.name
+            val: Any = getattr(self, key)
+            yield fi, key, val
+
+    def to_dict(
+        self,
+        include_non_members: bool = True,
+
+    ) -> dict:
+        '''
+        Like it sounds.. direct delegation to:
+        https://jcristharif.com/msgspec/api.html#msgspec.structs.asdict
+
+        BUT, by default we pop all non-member (aka not defined as
+        struct fields) fields by default.
+
+        '''
+        asdict: dict = structs.asdict(self)
+        if include_non_members:
+            return asdict
+
+        # only return a dict of the struct members
+        # which were provided as input, NOT anything
+        # added as type-defined `@property` methods!
+        sin_props: dict = {}
+        fi: structs.FieldInfo
+        for fi, k, v in self._sin_props():
+            sin_props[k] = asdict[k]
+
+        return sin_props
+
+    def pformat(
+        self,
+        field_indent: int = 2,
+        indent: int = 0,
+
+    ) -> str:
+        '''
+        Recursion-safe `pprint.pformat()` style formatting of
+        a `msgspec.Struct` for sane reading by a human using a REPL.
+
+        '''
+        # global whitespace indent
+        ws: str = ' '*indent
+
+        # field whitespace indent
+        field_ws: str = ' '*(field_indent + indent)
+
+        # qtn: str = ws + self.__class__.__qualname__
+        qtn: str = self.__class__.__qualname__
+
+        obj_str: str = ''  # accumulator
+        fi: structs.FieldInfo
+        k: str
+        v: Any
+        for fi, k, v in self._sin_props():
+
+            # TODO: how can we prefer `Literal['option1',  'option2,
+            # ..]` over .__name__ == `Literal` but still get only the
+            # latter for simple types like `str | int | None` etc..?
+            ft: type = fi.type
+            typ_name: str = getattr(ft, '__name__', str(ft))
+
+            # recurse to get sub-struct's `.pformat()` output Bo
+            if isinstance(v, Struct):
+                val_str: str =  v.pformat(
+                    indent=field_indent + indent,
+                    field_indent=indent + field_indent,
+                )
+
+            else:  # the `pprint` recursion-safe format:
+                # https://docs.python.org/3.11/library/pprint.html#pprint.saferepr
+                val_str: str = saferepr(v)
+
+            obj_str += (field_ws + f'{k}: {typ_name} = {val_str},\n')
+
+        return (
+            f'{qtn}(\n'
+            f'{obj_str}'
+            f'{ws})'
+        )
+
+    # TODO: use a pprint.PrettyPrinter instance around ONLY rendering
+    # inside a known tty?
+    # def __repr__(self) -> str:
+    #     ...
+
+    # __str__ = __repr__ = pformat
+    __repr__ = pformat
+
+    def copy(
+        self,
+        update: dict | None = None,
+
+    ) -> Struct:
+        '''
+        Validate-typecast all self defined fields, return a copy of
+        us with all such fields.
+
+        NOTE: This is kinda like the default behaviour in
+        `pydantic.BaseModel` except a copy of the object is
+        returned making it compat with `frozen=True`.
+
+        '''
+        if update:
+            for k, v in update.items():
+                setattr(self, k, v)
+
+        # NOTE: roundtrip serialize to validate
+        # - enode to msgpack binary format,
+        # - decode that back to a struct.
+        return msgpack.Decoder(type=type(self)).decode(
+            msgpack.Encoder().encode(self)
+        )
+
+    def typecast(
+        self,
+
+        # TODO: allow only casting a named subset?
+        # fields: set[str] | None = None,
+
+    ) -> None:
+        '''
+        Cast all fields using their declared type annotations
+        (kinda like what `pydantic` does by default).
+
+        NOTE: this of course won't work on frozen types, use
+        ``.copy()`` above in such cases.
+
+        '''
+        # https://jcristharif.com/msgspec/api.html#msgspec.structs.fields
+        fi: structs.FieldInfo
+        for fi in structs.fields(self):
+            setattr(
+                self,
+                fi.name,
+                fi.type(getattr(self, fi.name)),
+            )
+
+    def __sub__(
+        self,
+        other: Struct,
+
+    ) -> DiffDump[tuple[str, Any, Any]]:
+        '''
+        Compare fields/items key-wise and return a ``DiffDump``
+        for easy visual REPL comparison B)
+
+        '''
+        diffs: DiffDump[tuple[str, Any, Any]] = DiffDump()
+        for fi in structs.fields(self):
+            attr_name: str = fi.name
+            ours: Any = getattr(self, attr_name)
+            theirs: Any = getattr(other, attr_name)
+            if ours != theirs:
+                diffs.append((
+                    attr_name,
+                    ours,
+                    theirs,
+                ))
+
+        return diffs

piker/ui/__init__.py

@@ -14,9 +14,8 @@
 # You should have received a copy of the GNU Affero General Public License
 # along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
-"""
-Stuff for your eyes, aka super hawt Qt UI components.
+'''
+UI components built using `Qt` with major versions swapped in via
+the import indirection in the `.qt` sub-mod.
 
-Currently we only support PyQt5 due to this issue in Pyside2:
-https://bugreports.qt.io/projects/PYSIDE/issues/PYSIDE-1313
-"""
+'''

piker/ui/_anchors.py

@@ -21,8 +21,10 @@ Anchor funtions for UI placement of annotions.
 from __future__ import annotations
 from typing import Callable, TYPE_CHECKING
 
-from PyQt5.QtCore import QPointF
-from PyQt5.QtWidgets import QGraphicsPathItem
+from piker.ui.qt import (
+    QPointF,
+    QGraphicsPathItem,
+)
 
 if TYPE_CHECKING:
     from ._chart import ChartPlotWidget

piker/ui/_annotate.py

@@ -20,12 +20,22 @@ Annotations for ur faces.
 """
 from typing import Callable
 
-from PyQt5 import QtCore, QtGui, QtWidgets
-from PyQt5.QtCore import QPointF, QRectF
-from PyQt5.QtWidgets import QGraphicsPathItem
-from pyqtgraph import Point, functions as fn, Color
+from pyqtgraph import (
+    Point,
+    functions as fn,
+    Color,
+)
 import numpy as np
 
+from piker.ui.qt import (
+    QtCore,
+    QtGui,
+    QtWidgets,
+    QPointF,
+    QRectF,
+    QGraphicsPathItem,
+)
+
 
 def mk_marker_path(
 

piker/ui/_app.py

@@ -21,13 +21,16 @@ Main app startup and run.
 from functools import partial
 from types import ModuleType
 
-from PyQt5.QtCore import QEvent
 import trio
 
+from piker.ui.qt import (
+    QEvent,
+)
 from ..service import maybe_spawn_brokerd
 from . import _event
 from ._exec import run_qtractor
 from ..data.feed import install_brokerd_search
+from ..data._symcache import open_symcache
 from ..accounting import unpack_fqme
 from . import _search
 from ._chart import GodWidget
@@ -56,7 +59,13 @@ async def load_provider_search(
             portal,
             brokermod,
         ),
+        open_symcache(brokermod) as symcache,
     ):
+        if not symcache.mktmaps:
+            log.warning(
+                f'BACKEND DOES NOT (yet) support symcaching: `{brokermod.name}`'
+            )
+
         # keep search engine stream up until cancelled
         await trio.sleep_forever()
 
@@ -99,6 +108,8 @@ async def _async_main(
     sbar = godwidget.window.status_bar
     starting_done = sbar.open_status('starting ze sexy chartz')
 
+    # NOTE: by default we load all "builtin" backends for search
+    # and that includes loading their symcaches if possible B)
     needed_brokermods: dict[str, ModuleType] = {}
     for fqme in syms:
         brokername, *_ = unpack_fqme(fqme)

piker/ui/_axes.py

@@ -23,16 +23,24 @@ from functools import lru_cache
 from typing import Callable
 from math import floor
 
-import numpy as np
+import polars as pl
 import pyqtgraph as pg
-from PyQt5 import QtCore, QtGui, QtWidgets
-from PyQt5.QtCore import QPointF
 
+from piker.ui.qt import (
+    QtCore,
+    QtGui,
+    QtWidgets,
+    QPointF,
+    txt_flag,
+    align_flag,
+    px_cache_mode,
+)
 from . import _pg_overrides as pgo
 from ..accounting._mktinfo import float_digits
 from ._label import Label
 from ._style import DpiAwareFont, hcolor, _font
 from ._interaction import ChartView
+from ._dataviz import Viz
 
 _axis_pen = pg.mkPen(hcolor('bracket'))
 
@@ -287,9 +295,7 @@ class DynamicDateAxis(Axis):
     # time formats mapped by seconds between bars
     tick_tpl = {
         60 * 60 * 24: '%Y-%b-%d',
-        60: '%H:%M',
-        30: '%H:%M:%S',
-        5: '%H:%M:%S',
+        60: '%Y-%b-%d(%H:%M)',
         1: '%H:%M:%S',
     }
 
@@ -305,10 +311,10 @@ class DynamicDateAxis(Axis):
         # XX: ARGGGGG AG:LKSKDJF:LKJSDFD
         chart = self.pi.chart_widget
 
-        viz = chart._vizs[chart.name]
+        viz: Viz = chart._vizs[chart.name]
         shm = viz.shm
         array = shm.array
-        ifield = viz.index_field
+        ifield: str = viz.index_field
         index = array[ifield]
         i_0, i_l = index[0], index[-1]
 
@@ -329,7 +335,7 @@ class DynamicDateAxis(Axis):
             arr_len = index.shape[0]
             first = shm._first.value
             times = array['time']
-            epochs = times[
+            epochs: list[int] = times[
                 list(
                     map(
                         int,
@@ -341,23 +347,30 @@ class DynamicDateAxis(Axis):
                 )
             ]
         else:
-            epochs = list(map(int, indexes))
+            epochs: list[int] = list(map(int, indexes))
 
         # TODO: **don't** have this hard coded shift to EST
-        # delay = times[-1] - times[-2]
-        dts = np.array(
+        delay: float = viz.time_step()
+        if delay > 1:
+            # NOTE: use less granular dt-str when using 1M+ OHLC
+            fmtstr: str = self.tick_tpl[delay]
+        else:
+            fmtstr: str = '%Y-%m-%d(%H:%M:%S)'
+
+        # https://pola-rs.github.io/polars/py-polars/html/reference/expressions/api/polars.from_epoch.html#polars-from-epoch
+        pl_dts: pl.Series = pl.from_epoch(
             epochs,
-            dtype='datetime64[s]',
+            time_unit='s',
+        # NOTE: kinda weird we can pass it to `.from_epoch()` no?
+        ).dt.replace_time_zone(
+            time_zone='UTC'
+        ).dt.convert_time_zone(
+            # TODO: pull this from either:
+            # -[ ] the mkt venue tz by default
+            # -[ ] the user's config under `sys.mkt_timezone: str`
+            'EST'
         )
-
-        # see units listing:
-        # https://numpy.org/devdocs/reference/arrays.datetime.html#datetime-units
-        return list(np.datetime_as_string(dts))
-
-        # TODO: per timeframe formatting?
-        # - we probably need this based on zoom now right?
-        # prec = self.np_dt_precision[delay]
-        # return dts.strftime(self.tick_tpl[delay])
+        return pl_dts.dt.to_string(fmtstr).to_list()
 
     def tickStrings(
         self,
@@ -408,11 +421,15 @@ class AxisLabel(pg.GraphicsObject):
         super().__init__()
         self.setParentItem(parent)
 
-        self.setFlag(self.ItemIgnoresTransformations)
+        self.setFlag(
+            self.GraphicsItemFlag.ItemIgnoresTransformations
+        )
         self.setZValue(100)
 
         # XXX: pretty sure this is faster
-        self.setCacheMode(QtWidgets.QGraphicsItem.DeviceCoordinateCache)
+        self.setCacheMode(
+            px_cache_mode.DeviceCoordinateCache
+        )
 
         self._parent = parent
 
@@ -549,21 +566,14 @@ class AxisLabel(pg.GraphicsObject):
 
         return (self.rect.width(), self.rect.height())
 
-# _common_text_flags = (
-#     QtCore.Qt.TextDontClip |
-#     QtCore.Qt.AlignCenter |
-#     QtCore.Qt.AlignTop |
-#     QtCore.Qt.AlignHCenter |
-#     QtCore.Qt.AlignVCenter
-# )
 
 
 class XAxisLabel(AxisLabel):
     _x_margin = 8
 
     text_flags = (
-        QtCore.Qt.TextDontClip
-        | QtCore.Qt.AlignCenter
+        align_flag.AlignCenter
+        | txt_flag.TextDontClip
     )
 
     def size_hint(self) -> tuple[float, float]:
@@ -620,10 +630,10 @@ class YAxisLabel(AxisLabel):
     _y_margin: int = 4
 
     text_flags = (
-        QtCore.Qt.AlignLeft
-        # QtCore.Qt.AlignHCenter
-        | QtCore.Qt.AlignVCenter
-        | QtCore.Qt.TextDontClip
+        align_flag.AlignLeft
+        | align_flag.AlignVCenter
+        # | align_flag.AlignHCenter
+        | txt_flag.TextDontClip
     )
 
     def __init__(

piker/ui/_chart.py

@@ -28,22 +28,20 @@ from typing import (
     TYPE_CHECKING,
 )
 
-from PyQt5 import QtCore, QtWidgets
-from PyQt5.QtCore import (
+import pyqtgraph as pg
+import trio
+
+from piker.ui.qt import (
+    QtCore,
+    QtWidgets,
     Qt,
     QLineF,
-    # QPointF,
-)
-from PyQt5.QtWidgets import (
     QFrame,
     QWidget,
     QHBoxLayout,
     QVBoxLayout,
     QSplitter,
 )
-import pyqtgraph as pg
-import trio
-
 from ._axes import (
     DynamicDateAxis,
     PriceAxis,
@@ -282,7 +280,7 @@ class GodWidget(QWidget):
             # TODO: probably stick this in some kinda `LooknFeel` API?
             for tracker in self.rt_linked.mode.trackers.values():
                 pp_nav = tracker.nav
-                if tracker.live_pp.size:
+                if tracker.live_pp.cumsize:
                     pp_nav.show()
                     pp_nav.hide_info()
                 else:
@@ -406,6 +404,7 @@ class ChartnPane(QFrame):
         )
         self._sidepane = sidepane
 
+    @property
     def sidepane(self) -> FieldsForm | SearchWidget:
         return self._sidepane
 
@@ -495,7 +494,7 @@ class LinkedSplits(QWidget):
         Set the proportion of space allocated for linked subcharts.
 
         '''
-        ln = len(self.subplots) or 1
+        ln: int = len(self.subplots) or 1
 
         # proportion allocated to consumer subcharts
         if not prop:
@@ -569,8 +568,8 @@ class LinkedSplits(QWidget):
 
         # style?
         self.chart.setFrameStyle(
-            QFrame.StyledPanel |
-            QFrame.Plain
+            QFrame.Shape.StyledPanel |
+            QFrame.Shadow.Plain
         )
 
         return self.chart
@@ -688,8 +687,8 @@ class LinkedSplits(QWidget):
 
         cpw.plotItem.vb.linked = self
         cpw.setFrameStyle(
-            QtWidgets.QFrame.StyledPanel
-            # | QtWidgets.QFrame.Plain
+            QFrame.Shape.StyledPanel
+            # | QFrame.Shadow.Plain
         )
 
         # don't show the little "autoscale" A label.
@@ -925,6 +924,7 @@ class ChartPlotWidget(pg.PlotWidget):
         self.useOpenGL(use_open_gl)
         self.name = name
         self.data_key = data_key or name
+        self.qframe: ChartnPane | None = None
 
         # scene-local placeholder for book graphics
         # sizing to avoid overlap with data contents

piker/ui/_cursor.py

@@ -28,9 +28,14 @@ from typing import (
 import inspect
 import numpy as np
 import pyqtgraph as pg
-from PyQt5 import QtCore, QtWidgets
-from PyQt5.QtCore import QPointF, QRectF
 
+from piker.ui.qt import (
+    QPointF,
+    QRectF,
+    QtCore,
+    QtWidgets,
+    px_cache_mode,
+)
 from ._style import (
     _xaxis_at,
     hcolor,
@@ -104,7 +109,9 @@ class LineDot(pg.CurvePoint):
         dot.setParentItem(self)
 
         # keep a static size
-        self.setFlag(self.ItemIgnoresTransformations)
+        self.setFlag(
+            self.GraphicsItemFlag.ItemIgnoresTransformations
+        )
 
     def event(
         self,
@@ -207,9 +214,10 @@ class ContentsLabel(pg.LabelItem):
         # this being "html" is the dumbest shit :eyeroll:
 
         self.setText(
-            "<b>i</b>:{index}<br/>"
+            "<b>i_arr</b>:{index}<br/>"
             # NB: these fields must be indexed in the correct order via
             # the slice syntax below.
+            "<b>i_shm</b>:{}<br/>"
             "<b>epoch</b>:{}<br/>"
             "<b>O</b>:{}<br/>"
             "<b>H</b>:{}<br/>"
@@ -219,6 +227,7 @@ class ContentsLabel(pg.LabelItem):
             # "<b>wap</b>:{}".format(
                 *array[ix][
                     [
+                        'index',
                         'time',
                         'open',
                         'high',
@@ -270,10 +279,15 @@ class ContentsLabels:
         x_in: int,
 
     ) -> None:
-        for chart, name, label, update in self._labels:
+        for (
+            chart,
+            name,
+            label,
+            update,
+        )in self._labels:
 
             viz = chart.get_viz(name)
-            array = viz.shm.array
+            array: np.ndarray = viz.shm._array
             index = array[viz.index_field]
             start = index[0]
             stop = index[-1]
@@ -284,7 +298,7 @@ class ContentsLabels:
             ):
                 # out of range
                 print('WTF out of range?')
-                continue
+                # continue
 
             # call provided update func with data point
             try:
@@ -292,6 +306,7 @@ class ContentsLabels:
                 ix = np.searchsorted(index, x_in)
                 if ix > len(array):
                     breakpoint()
+
                 update(ix, array)
 
             except IndexError:
@@ -416,10 +431,10 @@ class Cursor(pg.GraphicsObject):
         # vertical and horizonal lines and a y-axis label
 
         vl = plot.addLine(x=0, pen=self.lines_pen, movable=False)
-        vl.setCacheMode(QtWidgets.QGraphicsItem.DeviceCoordinateCache)
+        vl.setCacheMode(px_cache_mode.DeviceCoordinateCache)
 
         hl = plot.addLine(y=0, pen=self.lines_pen, movable=False)
-        hl.setCacheMode(QtWidgets.QGraphicsItem.DeviceCoordinateCache)
+        hl.setCacheMode(px_cache_mode.DeviceCoordinateCache)
         hl.hide()
 
         yl = YAxisLabel(
@@ -503,7 +518,10 @@ class Cursor(pg.GraphicsObject):
             plot=chart
         )
         chart.addItem(cursor)
-        self.graphics[chart].setdefault('cursors', []).append(cursor)
+        self.graphics[chart].setdefault(
+            'cursors',
+            [],
+        ).append(cursor)
         return cursor
 
     def mouseAction(

piker/ui/_curve.py

@@ -19,42 +19,47 @@ Fast, smooth, sexy curves.
 
 """
 from contextlib import contextmanager as cm
+from enum import EnumType
 from typing import Callable
 
 import numpy as np
 import pyqtgraph as pg
-from PyQt5 import QtWidgets
-from PyQt5.QtWidgets import QGraphicsItem
-from PyQt5.QtCore import (
+
+from piker.ui.qt import (
+    QtWidgets,
+    QGraphicsItem,
     Qt,
     QLineF,
     QRectF,
-)
-from PyQt5.QtGui import (
     QPainter,
     QPainterPath,
+    px_cache_mode,
 )
-from .._profile import pg_profile_enabled, ms_slower_then
 from ._style import hcolor
 from ..log import get_logger
-from .._profile import Profiler
-
+from ..toolz.profile import (
+    Profiler,
+    pg_profile_enabled,
+    ms_slower_then,
+)
 
 log = get_logger(__name__)
 
 
+pen_style: EnumType = Qt.PenStyle
+
 _line_styles: dict[str, int] = {
-    'solid': Qt.PenStyle.SolidLine,
-    'dash': Qt.PenStyle.DashLine,
-    'dot': Qt.PenStyle.DotLine,
-    'dashdot': Qt.PenStyle.DashDotLine,
+    'solid': pen_style.SolidLine,
+    'dash': pen_style.DashLine,
+    'dot': pen_style.DotLine,
+    'dashdot': pen_style.DashDotLine,
 }
 
 
 class FlowGraphic(pg.GraphicsObject):
     '''
-    Base class with minimal interface for `QPainterPath` implemented,
-    real-time updated "data flow" graphics.
+    Base class with minimal interface for `QPainterPath`
+    implemented, real-time updated "data flow" graphics.
 
     See subtypes below.
 
@@ -66,12 +71,12 @@ class FlowGraphic(pg.GraphicsObject):
     # XXX-NOTE-XXX: graphics caching B)
     # see explanation for different caching modes:
     # https://stackoverflow.com/a/39410081
-    cache_mode: int = QGraphicsItem.DeviceCoordinateCache
+    cache_mode: int = px_cache_mode.DeviceCoordinateCache
     # XXX: WARNING item caching seems to only be useful
     # if we don't re-generate the entire QPainterPath every time
     # don't ever use this - it's a colossal nightmare of artefacts
     # and is disastrous for performance.
-    # QGraphicsItem.ItemCoordinateCache
+    # cache_mode.ItemCoordinateCache
     # TODO: still questions todo with coord-cacheing that we should
     # probably talk to a core dev about:
     # - if this makes trasform interactions slower (such as zooming)
@@ -164,15 +169,16 @@ class FlowGraphic(pg.GraphicsObject):
         return None
 
     # XXX: due to a variety of weird jitter bugs and "smearing"
-    # artifacts when click-drag panning and viewing history time series,
-    # we offer this ctx-mngr interface to allow temporarily disabling
-    # Qt's graphics caching mode; this is now currently used from
-    # ``ChartView.start/signal_ic()`` methods which also disable the
-    # rt-display loop when the user is moving around a view.
+    # artifacts when click-drag panning and viewing history time
+    # series, we offer this ctx-mngr interface to allow temporarily
+    # disabling Qt's graphics caching mode; this is now currently
+    # used from ``ChartView.start/signal_ic()`` methods which also
+    # disable the rt-display loop when the user is moving around
+    # a view.
     @cm
     def reset_cache(self) -> None:
         try:
-            none = QGraphicsItem.NoCache
+            none = px_cache_mode.NoCache
             log.debug(
                 f'{self._name} -> CACHE DISABLE: {none}'
             )

piker/ui/_dataviz.py

@@ -36,9 +36,12 @@ from msgspec import (
     field,
 )
 import numpy as np
+from numpy import (
+    ndarray,
+)
 import pyqtgraph as pg
-from PyQt5.QtCore import QLineF
 
+from piker.ui.qt import QLineF
 from ..data._sharedmem import (
     ShmArray,
 )
@@ -49,7 +52,7 @@ from ..data._formatters import (
     OHLCBarsAsCurveFmtr,  # OHLC converted to line
     StepCurveFmtr,  # "step" curve (like for vlm)
 )
-from ..data._timeseries import (
+from ..tsp import (
     slice_from_time,
 )
 from ._ohlc import (
@@ -62,7 +65,7 @@ from ._curve import (
 )
 from ._render import Renderer
 from ..log import get_logger
-from .._profile import (
+from ..toolz.profile import (
     Profiler,
     pg_profile_enabled,
     ms_slower_then,
@@ -82,10 +85,11 @@ def render_baritems(
     viz: Viz,
     graphics: BarItems,
     read: tuple[
-        int, int, np.ndarray,
-        int, int, np.ndarray,
+        int, int, ndarray,
+        int, int, ndarray,
     ],
     profiler: Profiler,
+    force_redraw: bool = False,
     **kwargs,
 
 ) -> None:
@@ -216,9 +220,11 @@ def render_baritems(
     viz._in_ds = should_line
 
     should_redraw = (
-        changed_to_line
+        force_redraw
+        or changed_to_line
         or not should_line
     )
+    # print(f'should_redraw: {should_redraw}')
     return (
         graphics,
         r,
@@ -250,7 +256,7 @@ class ViewState(Struct):
     ] | None = None
 
     # last in view ``ShmArray.array[read_slc]`` data
-    in_view: np.ndarray | None = None
+    in_view: ndarray | None = None
 
 
 class Viz(Struct):
@@ -313,6 +319,7 @@ class Viz(Struct):
     _last_uppx: float = 0
     _in_ds: bool = False
     _index_step: float | None = None
+    _time_step: float | None = None
 
     # map from uppx -> (downsampled data, incremental graphics)
     _src_r: Renderer | None = None
@@ -359,7 +366,8 @@ class Viz(Struct):
 
     def index_step(
         self,
-        reset: bool = False,
+        index_field: str | None = None,
+
     ) -> float:
         '''
         Return the size between sample steps in the units of the
@@ -367,12 +375,17 @@ class Viz(Struct):
         epoch time in seconds.
 
         '''
-        # attempt to dectect the best step size by scanning a sample of
-        # the source data.
-        if self._index_step is None:
-
-            index: np.ndarray = self.shm.array[self.index_field]
-            isample: np.ndarray = index[-16:]
+        # attempt to detect the best step size by scanning a sample
+        # of the source data.
+        if (
+            self._index_step is None
+            or index_field is not None
+        ):
+            index: ndarray = self.shm.array[
+                index_field
+                or self.index_field
+            ]
+            isample: ndarray = index[-16:]
 
             mxdiff: None | float = None
             for step in np.diff(isample):
@@ -386,7 +399,15 @@ class Viz(Struct):
                         )
                     mxdiff = step
 
-            self._index_step = max(mxdiff, 1)
+            step: float = max(mxdiff, 1)
+
+            # only SET the internal index step if an explicit
+            # field name is NOT passed, since in such cases this
+            # is likely just being called from `.time_step()`.
+            if index_field is not None:
+                return step
+
+            self._index_step = step
             if (
                 mxdiff < 1
                 or 1 < mxdiff < 60
@@ -397,6 +418,17 @@ class Viz(Struct):
 
         return self._index_step
 
+    def time_step(self) -> float:
+        '''
+        Attempt to determine the per-sample time-step period by 
+        forcing an epoch-index and calling `.index_step()`.
+
+        '''
+        if self._time_step is None:
+            self._time_step: float = self.index_step(index_field='time')
+
+        return self._time_step
+
     def maxmin(
         self,
 
@@ -404,6 +436,9 @@ class Viz(Struct):
         i_read_range: tuple[int, int] | None = None,
         use_caching: bool = True,
 
+        # XXX: internal debug
+        _do_print: bool = False
+
     ) -> tuple[float, float] | None:
         '''
         Compute the cached max and min y-range values for a given
@@ -423,15 +458,14 @@ class Viz(Struct):
         if shm is None:
             return None
 
-        do_print: bool = False
-        arr = shm.array
+        arr: ndarray = shm.array
 
         if i_read_range is not None:
             read_slc = slice(*i_read_range)
-            index = arr[read_slc][self.index_field]
+            index: float | int = arr[read_slc][self.index_field]
             if not index.size:
                 return None
-            ixrng = (index[0], index[-1])
+            ixrng: tuple[int, int] = (index[0], index[-1])
 
         else:
             if x_range is None:
@@ -449,15 +483,24 @@ class Viz(Struct):
 
             # TODO: hash the slice instead maybe?
             # https://stackoverflow.com/a/29980872
-            ixrng = lbar, rbar = round(x_range[0]), round(x_range[1])
+            ixrng = lbar, rbar = (
+                round(x_range[0]),
+                round(x_range[1]),
+            )
 
         if (
             use_caching
             and self._mxmn_cache_enabled
         ):
+            # TODO: is there a way to ONLY clear ranges containing
+            # a certain sub-range?
+            # -[ ] currently we have a problem where a previously
+            #    cached mxmn will persist even if the viz is "hard
+            #    re-rendered" (usually bc underlying data was
+            #    corrected)
             cached_result = self._mxmns.get(ixrng)
             if cached_result:
-                if do_print:
+                if _do_print:
                     print(
                         f'{self.name} CACHED maxmin\n'
                         f'{ixrng} -> {cached_result}'
@@ -487,7 +530,7 @@ class Viz(Struct):
                     (rbar - ifirst) + 1
                 )
 
-        slice_view = arr[read_slc]
+        slice_view: ndarray = arr[read_slc]
 
         if not slice_view.size:
             log.warning(
@@ -498,7 +541,7 @@ class Viz(Struct):
 
         elif self.ds_yrange:
             mxmn = self.ds_yrange
-            if do_print:
+            if _do_print:
                 print(
                     f'{self.name} M4 maxmin:\n'
                     f'{ixrng} -> {mxmn}'
@@ -515,7 +558,7 @@ class Viz(Struct):
 
             mxmn = ylow, yhigh
             if (
-                do_print
+                _do_print
             ):
                 s = 3
                 print(
@@ -529,14 +572,23 @@ class Viz(Struct):
 
         # cache result for input range
         ylow, yhi = mxmn
+        diff: float = yhi - ylow
+
+        # order-of-magnitude check
+        # TODO: really we should be checking the hi or low
+        # against the previous sample to catch stuff like,
+        # - rando stock (reverse-)split
+        # - null-segments written by some prior
+        #   crash-during-backfil
+        if diff > 0:
+            omg: float = abs(logf(diff, 10))
+        else:
+            omg: float = 0
 
         try:
             prolly_anomaly: bool = (
-                (
-                    abs(logf(ylow, 10)) > 16
-                    if ylow
-                    else False
-                )
+                # diff == 0
+                (ylow and omg > 10)
                 or (
                     isnan(ylow) or isnan(yhi)
                 )
@@ -563,7 +615,8 @@ class Viz(Struct):
 
     def view_range(self) -> tuple[int, int]:
         '''
-        Return the start and stop x-indexes for the managed ``ViewBox``.
+        Return the start and stop x-indexes for the managed
+        ``ViewBox``.
 
         '''
         vr = self.plot.viewRect()
@@ -576,7 +629,7 @@ class Viz(Struct):
         self,
         view_range: None | tuple[float, float] = None,
         index_field: str | None = None,
-        array: np.ndarray | None = None,
+        array: ndarray | None = None,
 
     ) -> tuple[
         int, int, int, int, int, int
@@ -647,8 +700,8 @@ class Viz(Struct):
         profiler: None | Profiler = None,
 
     ) -> tuple[
-        int, int, np.ndarray,
-        int, int, np.ndarray,
+        int, int, ndarray,
+        int, int, ndarray,
     ]:
         '''
         Read the underlying shm array buffer and
@@ -818,6 +871,10 @@ class Viz(Struct):
                 graphics,
                 read,
                 profiler,
+
+                # NOTE: only set when caller says to
+                force_redraw=should_redraw,
+
                 **kwargs,
             )
 
@@ -980,6 +1037,39 @@ class Viz(Struct):
             graphics,
         )
 
+    def reset_graphics(
+        self,
+
+        # TODO: allow only resetting within some x-domain range?
+        # ixrng: tuple[int, int] | None = None,
+
+    ) -> None:
+        '''
+        Hard reset all graphics (rendering) layers for this
+        data viz including clearing the mxmn auto-y-range
+        cache.
+
+        Normally called when the underlying data set is modified
+        (probably by some `.tsp` correcting/editing routine) and
+        the (now cached) graphics need to be fully re-rendered from
+        source.
+
+        '''
+        log.warning(
+            f'Forcing hard Viz graphihcs RESET:\n'
+            f'.name: {self.name}\n'
+            f'.index_field: {self.index_field}\n'
+            f'.index_step(): {self.index_step()}\n'
+            f'.time_step(): {self.time_step()}\n'
+        )
+        # XXX: always clear the mxn y-range cache
+        # to avoid old data (anomalies) from being
+        # retained in auto-yrange output.
+        self._mxmn_cache_enabled = False
+        self._mxmns.clear()
+        self.update_graphics(force_redraw=True)
+        self._mxmn_cache_enabled = True
+
     def draw_last(
         self,
         array_key: str | None = None,
@@ -1072,7 +1162,7 @@ class Viz(Struct):
 
         '''
         shm: ShmArray = self.shm
-        array: np.ndarray = shm.array
+        array: ndarray = shm.array
         view: ChartView = self.plot.vb
         (
             vl,

piker/ui/_display.py

@@ -36,25 +36,28 @@ import pyqtgraph as pg
 from msgspec import field
 
 # from .. import brokers
-from ..accounting import (
+from piker.accounting import (
     MktPair,
 )
-from ..data import (
+from piker.types import Struct
+from piker.data import (
     open_feed,
     Feed,
     Flume,
+    open_sample_stream,
+    ShmArray,
 )
-from ..data.ticktools import (
+from piker.data.ticktools import (
     _tick_groups,
     _auction_ticks,
 )
-from ..data.types import Struct
-from ..data._sharedmem import (
-    ShmArray,
-)
-from ..data._sampling import (
-    open_sample_stream,
+from piker.toolz import (
+    pg_profile_enabled,
+    ms_slower_then,
+    Profiler,
 )
+from piker.log import get_logger
+from piker import config
 # from ..data._source import tf_in_1s
 from ._axes import YAxisLabel
 from ._chart import (
@@ -79,12 +82,6 @@ from .order_mode import (
     open_order_mode,
     OrderMode,
 )
-from .._profile import (
-    pg_profile_enabled,
-    ms_slower_then,
-)
-from ..log import get_logger
-from .._profile import Profiler
 
 if TYPE_CHECKING:
     from ._interaction import ChartView
@@ -214,9 +211,9 @@ async def increment_history_view(
 ):
     hist_chart: ChartPlotWidget = ds.hist_chart
     hist_viz: Viz = ds.hist_viz
-    viz: Viz = ds.viz
+    # viz: Viz = ds.viz
     assert 'hist' in hist_viz.shm.token['shm_name']
-    name: str = hist_viz.name
+    # name: str = hist_viz.name
 
     # TODO: seems this is more reliable at keeping the slow
     # chart incremented in view more correctly?
@@ -229,7 +226,8 @@ async def increment_history_view(
     # draw everything from scratch on first entry!
     for curve_name, hist_viz in hist_chart._vizs.items():
         log.info(f'Forcing hard redraw -> {curve_name}')
-        hist_viz.update_graphics(force_redraw=True)
+        hist_viz.reset_graphics()
+        # hist_viz.update_graphics(force_redraw=True)
 
     async with open_sample_stream(1.) as min_istream:
         async for msg in min_istream:
@@ -252,17 +250,27 @@ async def increment_history_view(
             # - samplerd could emit the actual update range via
             #   tuple and then we only enter the below block if that
             #   range is detected as in-view?
-            if (
-                (bf_wut := msg.get('backfilling', False))
-            ):
-                viz_name, timeframe = bf_wut
-                if viz_name == name:
-                    log.info(f'Forcing hard redraw -> {name}@{timeframe}')
-                    match timeframe:
-                        case 60:
-                            hist_viz.update_graphics(force_redraw=True)
-                        case 1:
-                            viz.update_graphics(force_redraw=True)
+            # match msg:
+            #     case {
+            #         'backfilling': (viz_name, timeframe),
+            #     } if (
+            #         viz_name == name
+            #     ):
+            #         log.warning(
+            #             f'Forcing HARD REDRAW:\n'
+            #             f'name: {name}\n'
+            #             f'timeframe: {timeframe}\n'
+            #         )
+            #         # TODO: only allow this when the data is IN VIEW!
+            #         # also, we probably can do this more efficiently
+            #         # / smarter by only redrawing the portion of the
+            #         # path necessary?
+            #         {
+            #             60: hist_viz,
+            #             1: viz,
+            #         }[timeframe].update_graphics(
+            #             force_redraw=True
+            #         )
 
             # check if slow chart needs an x-domain shift and/or
             # y-range resize.
@@ -303,6 +311,7 @@ async def increment_history_view(
 
 async def graphics_update_loop(
 
+    dss: dict[str, DisplayState],
     nurse: trio.Nursery,
     godwidget: GodWidget,
     feed: Feed,
@@ -344,8 +353,6 @@ async def graphics_update_loop(
         'i_last_slow_t':  0,  # multiview-global slow (1m) step index
     }
 
-    dss: dict[str, DisplayState] = {}
-
     for fqme, flume in feed.flumes.items():
         ohlcv = flume.rt_shm
         hist_ohlcv = flume.hist_shm
@@ -462,12 +469,20 @@ async def graphics_update_loop(
         await trio.sleep(0)
 
         if ds.hist_vars['i_last'] < ds.hist_vars['i_last_append']:
-            await tractor.breakpoint()
+            await tractor.pause()
+
+    # try:
+
+    # XXX TODO: we need to do _dss UPDATE here so that when
+    # a feed-view is switched you can still remote annotate the
+    # prior view..
+    from . import _remote_ctl
+    _remote_ctl._dss.update(dss)
 
     # main real-time quotes update loop
     stream: tractor.MsgStream
     async with feed.open_multi_stream() as stream:
-        assert stream
+        # assert stream
         async for quotes in stream:
             quote_period = time.time() - last_quote_s
             quote_rate = round(
@@ -483,7 +498,7 @@ async def graphics_update_loop(
                 pass
                 # log.warning(f'High quote rate {mkt.fqme}: {quote_rate}')
 
-            last_quote_s = time.time()
+            last_quote_s: float = time.time()
 
             for fqme, quote in quotes.items():
                 ds = dss[fqme]
@@ -513,6 +528,12 @@ async def graphics_update_loop(
                     quote,
                 )
 
+    # finally:
+    #     # XXX: cancel any remote annotation control ctxs
+    #     _remote_ctl._dss = None
+    #     for cid, (ctx, aids) in _remote_ctl._ctxs.items():
+    #         await ctx.cancel()
+
 
 def graphics_update_cycle(
     ds: DisplayState,
@@ -1211,6 +1232,8 @@ async def link_views_with_region(
     # region.sigRegionChangeFinished.connect(update_pi_from_region)
 
 
+# NOTE: default is set to 60 FPS until the runtime delivers the
+# discoverd hw value below.
 _quote_throttle_rate: int = 60 - 6
 
 
@@ -1229,7 +1252,7 @@ async def display_symbol_data(
     fast from a cached watch-list.
 
     '''
-    sbar = godwidget.window.status_bar
+    # sbar = godwidget.window.status_bar
     # historical data fetch
     # brokermod = brokers.get_brokermod(provider)
 
@@ -1239,11 +1262,11 @@ async def display_symbol_data(
     #     group_key=loading_sym_key,
     # )
 
-    for fqme in fqmes:
-        loading_sym_key = sbar.open_status(
-            f'loading {fqme} ->',
-            group_key=True
-        )
+    # for fqme in fqmes:
+    #     loading_sym_key = sbar.open_status(
+    #         f'loading {fqme} ->',
+    #         group_key=True
+    #     )
 
     # (TODO: make this not so shit XD)
     # close group status once a symbol feed fully loads to view.
@@ -1252,26 +1275,54 @@ async def display_symbol_data(
     # TODO: ctl over update loop's maximum frequency.
     # - load this from a config.toml!
     # - allow dyanmic configuration from chart UI?
+    (
+        conf,
+        path,
+    ) = config.load()
+    ui_conf: dict = conf['ui']
+
     global _quote_throttle_rate
     from ._window import main_window
-    display_rate = main_window().current_screen().refreshRate()
-    _quote_throttle_rate = floor(display_rate) - 6
+
+    display_rate: int = floor(
+        main_window().current_screen().refreshRate()
+    ) - 6
+
+    mx_redraw_rate: int = ui_conf.get(
+        'max_redraw_rate',
+        _quote_throttle_rate,
+    )
+
+    if mx_redraw_rate < display_rate:
+        log.info(
+            'Down-throttling redraw rate to config setting\n'
+            f'display FPS: {display_rate}\n'
+            'max_redraw_rate: {max_redraw_rate}\n'
+        )
+    else:
+        _quote_throttle_rate = display_rate
 
     # TODO: we should be able to increase this if we use some
     # `mypyc` speedups elsewhere? 22ish seems to be the sweet
     # spot for single-feed chart.
     num_of_feeds = len(fqmes)
-    mx: int = 22
-    if num_of_feeds > 1:
-        # there will be more ctx switches with more than 1 feed so we
-        # max throttle down a bit more.
-        mx = 16
+    # if num_of_feeds > 1:
+
+    # there will be more ctx switches with more than 1 feed so we
+    # max throttle down a bit more.
+    mx_per_feed: int = (
+        ui_conf.get(
+        'per_feed_redraw_rate',
+        mx_redraw_rate,
+        )
+        or 16
+    )
 
     # limit to at least display's FPS
     # avoiding needless Qt-in-guest-mode context switches
     cycles_per_feed = min(
         round(_quote_throttle_rate/num_of_feeds),
-        mx,
+        mx_per_feed,
     )
 
     feed: Feed
@@ -1416,7 +1467,7 @@ async def display_symbol_data(
                 start_fsp_displays,
                 rt_linked,
                 flume,
-                loading_sym_key,
+                # loading_sym_key,
                 loglevel,
             )
 
@@ -1535,8 +1586,10 @@ async def display_symbol_data(
             )
 
             # start update loop task
+            dss: dict[str, DisplayState] = {}
             ln.start_soon(
                 graphics_update_loop,
+                dss,
                 ln,
                 godwidget,
                 feed,
@@ -1550,15 +1603,31 @@ async def display_symbol_data(
             order_ctl_fqme: str = fqmes[0]
             mode: OrderMode
             async with (
+
                 open_order_mode(
                     feed,
                     godwidget,
                     order_ctl_fqme,
                     order_mode_started,
                     loglevel=loglevel
-                ) as mode
-            ):
+                ) as mode,
 
+                # TODO: maybe have these startup sooner before
+                # order mode fully boots? but we gotta,
+                # -[ ] decouple the order mode bindings until
+                #    the mode has fully booted..
+                #    -[ ] maybe do an Event to sync?
+
+                # start input handling for ``ChartView`` input
+                # (i.e. kb + mouse handling loops)
+                rt_chart.view.open_async_input_handler(
+                    dss=dss,
+                ),
+                hist_chart.view.open_async_input_handler(
+                    dss=dss,
+                ),
+
+            ):
                 rt_linked.mode = mode
 
                 rt_viz = rt_chart.get_viz(order_ctl_fqme)

piker/ui/_editors.py

@@ -21,7 +21,8 @@ Higher level annotation editors.
 from __future__ import annotations
 from collections import defaultdict
 from typing import (
-    TYPE_CHECKING
+    Sequence,
+    TYPE_CHECKING,
 )
 
 import pyqtgraph as pg
@@ -31,24 +32,34 @@ from pyqtgraph import (
     QtCore,
     QtWidgets,
 )
-from PyQt5.QtGui import (
-    QColor,
-)
-from PyQt5.QtWidgets import (
-    QLabel,
-)
 
 from pyqtgraph import functions as fn
-from PyQt5.QtCore import QPointF
 import numpy as np
 
-from ._style import hcolor, _font
+from piker.types import Struct
+from piker.ui.qt import (
+    Qt,
+    QPointF,
+    QRectF,
+    QGraphicsProxyWidget,
+    QGraphicsScene,
+    QLabel,
+    QColor,
+    QTransform,
+)
+from ._style import (
+    hcolor,
+    _font,
+)
 from ._lines import LevelLine
 from ..log import get_logger
-from ..data.types import Struct
 
 if TYPE_CHECKING:
-    from ._chart import GodWidget
+    from ._chart import (
+        GodWidget,
+        ChartPlotWidget,
+    )
+    from ._interaction import ChartView
 
 
 log = get_logger(__name__)
@@ -65,7 +76,7 @@ class ArrowEditor(Struct):
         uid: str,
         x: float,
         y: float,
-        color='default',
+        color: str = 'default',
         pointing: str | None = None,
 
     ) -> pg.ArrowItem:
@@ -251,43 +262,75 @@ class LineEditor(Struct):
         return lines
 
 
-class SelectRect(QtWidgets.QGraphicsRectItem):
+def as_point(
+    pair: Sequence[float, float] | QPointF,
+) -> list[QPointF, QPointF]:
+    '''
+    Case any input tuple of floats to a a list of `QPoint` objects
+    for use in Qt geometry routines.
 
+    '''
+    if isinstance(pair, QPointF):
+        return pair
+
+    return QPointF(pair[0], pair[1])
+
+
+# TODO: maybe implement better, something something RectItemProxy??
+# -[ ] dig into details of how proxy's work?
+#    https://doc.qt.io/qt-5/qgraphicsscene.html#addWidget
+# -[ ] consider using `.addRect()` maybe?
+
+class SelectRect(QtWidgets.QGraphicsRectItem):
+    '''
+    A data-view "selection rectangle": the most fundamental
+    geometry for annotating data views.
+
+    - https://doc.qt.io/qt-5/qgraphicsrectitem.html
+    - https://doc.qt.io/qt-6/qgraphicsrectitem.html
+
+    '''
     def __init__(
         self,
         viewbox: ViewBox,
-        color: str = 'dad_blue',
+        color: str | None = None,
     ) -> None:
         super().__init__(0, 0, 1, 1)
 
         # self.rbScaleBox = QGraphicsRectItem(0, 0, 1, 1)
-        self.vb = viewbox
-        self._chart: 'ChartPlotWidget' = None  # noqa
+        self.vb: ViewBox = viewbox
 
-        # override selection box color
+        self._chart: ChartPlotWidget | None = None  # noqa
+
+        # TODO: maybe allow this to be dynamic via a method?
+        #l override selection box color
+        color: str = color or 'dad_blue'
         color = QColor(hcolor(color))
+
         self.setPen(fn.mkPen(color, width=1))
         color.setAlpha(66)
         self.setBrush(fn.mkBrush(color))
         self.setZValue(1e9)
-        self.hide()
-        self._label = None
 
         label = self._label = QLabel()
-        label.setTextFormat(0)  # markdown
+        label.setTextFormat(
+            Qt.TextFormat.MarkdownText
+        )
         label.setFont(_font.font)
         label.setMargin(0)
         label.setAlignment(
             QtCore.Qt.AlignLeft
             # | QtCore.Qt.AlignVCenter
         )
+        label.hide()  # always right after init
 
         # proxy is created after containing scene is initialized
-        self._label_proxy = None
-        self._abs_top_right = None
+        self._label_proxy: QGraphicsProxyWidget | None = None
+        self._abs_top_right: Point | None = None
 
-        # TODO: "swing %" might be handy here (data's max/min # % change)
-        self._contents = [
+        # TODO: "swing %" might be handy here (data's max/min
+        # # % change)?
+        self._contents: list[str] = [
             'change: {pchng:.2f} %',
             'range: {rng:.2f}',
             'bars: {nbars}',
@@ -297,12 +340,31 @@ class SelectRect(QtWidgets.QGraphicsRectItem):
             'sigma: {std:.2f}',
         ]
 
+        self.add_to_view(viewbox)
+        self.hide()
+
+    def add_to_view(
+        self,
+        view: ChartView,
+    ) -> None:
+        '''
+        Self-defined view hookup impl which will
+        also re-assign the internal ref.
+
+        '''
+        view.addItem(
+            self,
+            ignoreBounds=True,
+        )
+        if self.vb is not view:
+            self.vb = view
+
     @property
-    def chart(self) -> 'ChartPlotWidget':  # noqa
+    def chart(self) -> ChartPlotWidget:  # noqa
         return self._chart
 
     @chart.setter
-    def chart(self, chart: 'ChartPlotWidget') -> None:  # noqa
+    def chart(self, chart: ChartPlotWidget) -> None:  # noqa
         self._chart = chart
         chart.sigRangeChanged.connect(self.update_on_resize)
         palette = self._label.palette()
@@ -315,57 +377,155 @@ class SelectRect(QtWidgets.QGraphicsRectItem):
         )
 
     def update_on_resize(self, vr, r):
-        """Re-position measure label on view range change.
+        '''
+        Re-position measure label on view range change.
 
-        """
+        '''
         if self._abs_top_right:
             self._label_proxy.setPos(
                 self.vb.mapFromView(self._abs_top_right)
             )
 
-    def mouse_drag_released(
+    def set_scen_pos(
         self,
-        p1: QPointF,
-        p2: QPointF
+        scen_p1: QPointF,
+        scen_p2: QPointF,
+
+        update_label: bool = True,
+
     ) -> None:
-        """Called on final button release for mouse drag with start and
-        end positions.
+        '''
+        Set position from scene coords of selection rect (normally
+        from mouse position) and accompanying label, move label to
+        match.
 
-        """
-        self.set_pos(p1, p2)
+        '''
+        # NOTE XXX: apparently just setting it doesn't work!?
+        # i have no idea why but it's pretty weird we have to do
+        # this transform thing which was basically pulled verbatim
+        # from the `pg.ViewBox.updateScaleBox()` method.
+        view_rect: QRectF = self.vb.childGroup.mapRectFromScene(
+            QRectF(
+                scen_p1, 
+                scen_p2,
+            )
+        )
+        self.setPos(view_rect.topLeft())
+        # XXX: does not work..!?!?
+        # https://doc.qt.io/qt-5/qgraphicsrectitem.html#setRect
+        # self.setRect(view_rect)
 
-    def set_pos(
-        self,
-        p1: QPointF,
-        p2: QPointF
-    ) -> None:
-        """Set position of selection rect and accompanying label, move
-        label to match.
+        tr = QTransform.fromScale(
+            view_rect.width(),
+            view_rect.height(),
+        )
+        self.setTransform(tr)
 
-        """
-        if self._label_proxy is None:
-            # https://doc.qt.io/qt-5/qgraphicsproxywidget.html
-            self._label_proxy = self.vb.scene().addWidget(self._label)
-
-        start_pos = self.vb.mapToView(p1)
-        end_pos = self.vb.mapToView(p2)
-
-        # map to view coords and update area
-        r = QtCore.QRectF(start_pos, end_pos)
-
-        # old way; don't need right?
-        # lr = QtCore.QRectF(p1, p2)
-        # r = self.vb.childGroup.mapRectFromParent(lr)
-
-        self.setPos(r.topLeft())
-        self.resetTransform()
-        self.setRect(r)
+        # XXX: never got this working, was always offset
+        # / transformed completely wrong (and off to the far right
+        # from the cursor?)
+        # self.set_view_pos(
+        #     view_rect=view_rect,
+        #     # self.vwqpToView(p1),
+        #     # self.vb.mapToView(p2),
+        #     # start_pos=self.vb.mapToScene(p1),
+        #     # end_pos=self.vb.mapToScene(p2),
+        # )
         self.show()
 
-        y1, y2 = start_pos.y(), end_pos.y()
-        x1, x2 = start_pos.x(), end_pos.x()
+        if update_label:
+            self.init_label(view_rect)
 
-        # TODO: heh, could probably use a max-min streamin algo here too
+    def set_view_pos(
+        self,
+
+        start_pos: QPointF | Sequence[float, float] | None = None,
+        end_pos: QPointF | Sequence[float, float] | None = None,
+        view_rect: QRectF | None = None,
+
+        update_label: bool = True,
+
+    ) -> None:
+        '''
+        Set position from `ViewBox` coords (i.e. from the actual
+        data domain) of rect (and any accompanying label which is
+        moved to match).
+
+        '''
+        if self._chart is None:
+            raise RuntimeError(
+                'You MUST assign a `SelectRect.chart: ChartPlotWidget`!'
+            )
+
+        if view_rect is None:
+            # ensure point casting
+            start_pos: QPointF = as_point(start_pos)
+            end_pos: QPointF = as_point(end_pos)
+
+            # map to view coords and update area
+            view_rect = QtCore.QRectF(
+                start_pos,
+                end_pos,
+            )
+
+        self.setPos(view_rect.topLeft())
+
+        # NOTE: SERIOUSLY NO IDEA WHY THIS WORKS...
+        # but it does and all the other commented stuff above
+        # dint, dawg..
+
+        # self.resetTransform()
+        # self.setRect(view_rect)
+
+        tr = QTransform.fromScale(
+            view_rect.width(),
+            view_rect.height(),
+        )
+        self.setTransform(tr)
+
+        if update_label:
+            self.init_label(view_rect)
+
+        print(
+            'SelectRect modify:\n'
+            f'QRectF: {view_rect}\n'
+            f'start_pos: {start_pos}\n'
+            f'end_pos: {end_pos}\n'
+        )
+        self.show()
+
+    def init_label(
+        self,
+        view_rect: QRectF,
+   ) -> QLabel:
+
+        # should be init-ed in `.__init__()`
+        label: QLabel = self._label
+        cv: ChartView = self.vb
+
+        # https://doc.qt.io/qt-5/qgraphicsproxywidget.html
+        if self._label_proxy is None:
+            scen: QGraphicsScene = cv.scene()
+            # NOTE: specifically this is passing a widget
+            # pointer to the scene's `.addWidget()` as per,
+            # https://doc.qt.io/qt-5/qgraphicsproxywidget.html#embedding-a-widget-with-qgraphicsproxywidget
+            self._label_proxy: QGraphicsProxyWidget = scen.addWidget(label)
+
+        # get label startup coords
+        tl: QPointF = view_rect.topLeft()
+        br: QPointF = view_rect.bottomRight()
+
+        x1, y1 = tl.x(), tl.y()
+        x2, y2 = br.x(), br.y()
+
+        # TODO: to remove, previous label corner point unpacking
+        # x1, y1 = start_pos.x(), start_pos.y()
+        # x2, y2 = end_pos.x(), end_pos.y()
+        # y1, y2 = start_pos.y(), end_pos.y()
+        # x1, x2 = start_pos.x(), end_pos.x()
+
+        # TODO: heh, could probably use a max-min streamin algo
+        # here too?
         _, xmn = min(y1, y2), min(x1, x2)
         ymx, xmx = max(y1, y2), max(x1, x2)
 
@@ -375,26 +535,35 @@ class SelectRect(QtWidgets.QGraphicsRectItem):
         ixmn, ixmx = round(xmn), round(xmx)
         nbars = ixmx - ixmn + 1
 
-        chart = self._chart
-        data = chart.get_viz(chart.name).shm.array[ixmn:ixmx]
+        chart: ChartPlotWidget = self._chart
+        data: np.ndarray = chart.get_viz(
+            chart.name
+        ).shm.array[ixmn:ixmx]
 
         if len(data):
-            std = data['close'].std()
-            dmx = data['high'].max()
-            dmn = data['low'].min()
+            std: float = data['close'].std()
+            dmx: float = data['high'].max()
+            dmn: float = data['low'].min()
         else:
             dmn = dmx = std = np.nan
 
         # update label info
-        self._label.setText('\n'.join(self._contents).format(
-            pchng=pchng, rng=rng, nbars=nbars,
-            std=std, dmx=dmx, dmn=dmn,
+        label.setText('\n'.join(self._contents).format(
+            pchng=pchng,
+            rng=rng,
+            nbars=nbars,
+            std=std,
+            dmx=dmx,
+            dmn=dmn,
         ))
 
         # print(f'x2, y2: {(x2, y2)}')
         # print(f'xmn, ymn: {(xmn, ymx)}')
 
-        label_anchor = Point(xmx + 2, ymx)
+        label_anchor = Point(
+            xmx + 2,
+            ymx,
+        )
 
         # XXX: in the drag bottom-right -> top-left case we don't
         # want the label to overlay the box.
@@ -403,13 +572,40 @@ class SelectRect(QtWidgets.QGraphicsRectItem):
         #     # label_anchor = Point(x2, y2 + self._label.height())
         #     label_anchor = Point(xmn, ymn)
 
-        self._abs_top_right = label_anchor
-        self._label_proxy.setPos(self.vb.mapFromView(label_anchor))
-        # self._label.show()
+        self._abs_top_right: Point = label_anchor
+        self._label_proxy.setPos(
+            cv.mapFromView(label_anchor)
+        )
+        label.show()
 
-    def clear(self):
-        """Clear the selection box from view.
+    def hide(self):
+        '''
+        Clear the selection box from its graphics scene but
+        don't delete it permanently.
 
-        """
+        '''
+        super().hide()
         self._label.hide()
-        self.hide()
+
+    # TODO: ensure noone else using dis.
+    clear = hide
+
+    def delete(self) -> None:
+        '''
+        De-allocate this rect from its rendering graphics scene.
+
+        Like a permanent hide.
+
+        '''
+        scen: QGraphicsScene = self.scene()
+        if scen is None:
+            return
+
+        scen.removeItem(self)
+        if (
+            self._label
+            and
+            self._label_proxy
+
+        ):
+            scen.removeItem(self._label_proxy)

piker/ui/_event.py

@@ -23,28 +23,29 @@ from typing import Callable
 
 import trio
 from tractor.trionics import gather_contexts
-from PyQt5 import QtCore
-from PyQt5.QtCore import QEvent, pyqtBoundSignal
-from PyQt5.QtWidgets import QWidget
-from PyQt5.QtWidgets import (
-    QGraphicsSceneMouseEvent as gs_mouse,
-)
 
-from ..data.types import Struct
+from piker.ui.qt import (
+    QtCore,
+    QWidget,
+    QEvent,
+    keys,
+    gs_keys,
+    pyqtBoundSignal,
+)
+from piker.types import Struct
 
 
 MOUSE_EVENTS = {
-    gs_mouse.GraphicsSceneMousePress,
-    gs_mouse.GraphicsSceneMouseRelease,
-    QEvent.MouseButtonPress,
-    QEvent.MouseButtonRelease,
+    gs_keys.GraphicsSceneMousePress,
+    gs_keys.GraphicsSceneMouseRelease,
+    keys.MouseButtonPress,
+    keys.MouseButtonRelease,
     # QtGui.QMouseEvent,
 }
 
 
 # TODO: maybe consider some constrained ints down the road?
 # https://pydantic-docs.helpmanual.io/usage/types/#constrained-types
-
 class KeyboardMsg(Struct):
     '''Unpacked Qt keyboard event data.
 
@@ -114,7 +115,10 @@ class EventRelay(QtCore.QObject):
         # something to do with Qt internals and calling the
         # parent handler?
 
-        if etype in {QEvent.KeyPress, QEvent.KeyRelease}:
+        if etype in {
+            QEvent.Type.KeyPress,
+            QEvent.Type.KeyRelease,
+        }:
 
             msg = KeyboardMsg(
                 event=ev,
@@ -160,7 +164,9 @@ class EventRelay(QtCore.QObject):
 async def open_event_stream(
 
     source_widget: QWidget,
-    event_types: set[QEvent] = {QEvent.KeyPress},
+    event_types: set[QEvent] = {
+        QEvent.Type.KeyPress,
+    },
     filter_auto_repeats: bool = True,
 
 ) -> trio.abc.ReceiveChannel:
@@ -201,8 +207,8 @@ async def open_signal_handler(
         async for args in recv:
             await async_handler(*args)
 
-    async with trio.open_nursery() as n:
-        n.start_soon(proxy_to_handler)
+    async with trio.open_nursery() as tn:
+        tn.start_soon(proxy_to_handler)
         async with send:
             yield
 
@@ -212,18 +218,48 @@ async def open_handlers(
 
     source_widgets: list[QWidget],
     event_types: set[QEvent],
-    async_handler: Callable[[QWidget, trio.abc.ReceiveChannel], None],
-    **kwargs,
+
+    # NOTE: if you want to bind in additional kwargs to the handler
+    # pass in a `partial()` instead!
+    async_handler: Callable[
+        [QWidget, trio.abc.ReceiveChannel],  # required handler args
+        None
+    ],
+
+    # XXX: these are ONLY inputs available to the
+    # `open_event_stream()` event-relay to mem-chan factor above!
+    **open_ev_stream_kwargs,
 
 ) -> None:
+    '''
+    Connect and schedule an async handler function to receive an
+    arbitrary `QWidget`'s events with kb/mouse msgs repacked into
+    structs (see above) and shuttled over a mem-chan to the input
+    `async_handler` to allow interaction-IO processing from
+    a `trio` func-as-task.
+
+    '''
+    widget: QWidget
+    streams: list[trio.abc.ReceiveChannel]
     async with (
-        trio.open_nursery() as n,
+        trio.open_nursery() as tn,
         gather_contexts([
-            open_event_stream(widget, event_types, **kwargs)
+            open_event_stream(
+                widget,
+                event_types,
+                **open_ev_stream_kwargs,
+            )
             for widget in source_widgets
         ]) as streams,
     ):
-        for widget, event_recv_stream in zip(source_widgets, streams):
-            n.start_soon(async_handler, widget, event_recv_stream)
+        for widget, event_recv_stream in zip(
+            source_widgets,
+            streams,
+        ):
+            tn.start_soon(
+                async_handler,
+                widget,
+                event_recv_stream,
+            )
 
         yield

piker/ui/_exec.py

@@ -30,34 +30,35 @@ from typing import (
 import platform
 import traceback
 
-# Qt specific
-import PyQt5  # noqa
-from PyQt5.QtWidgets import (
-    QWidget,
-    QMainWindow,
-    QApplication,
-)
-from PyQt5 import QtCore
-from PyQt5.QtCore import (
-    pyqtRemoveInputHook,
-    Qt,
-    QCoreApplication,
-)
 import qdarkstyle
 from qdarkstyle import DarkPalette
 # import qdarkgraystyle  # TODO: play with it
 import trio
 from outcome import Error
 
+# Qt version-agnostic
+from .qt import (
+    QWidget,
+    QMainWindow,
+    QApplication,
+    QtCore,
+    pyqtRemoveInputHook,
+    Qt,
+    QCoreApplication,
+)
 from ..service import (
     maybe_open_pikerd,
-    get_tractor_runtime_kwargs,
+    get_runtime_vars,
 )
 from ..log import get_logger
 from ._pg_overrides import _do_overrides
 from . import _style
 
 
+if TYPE_CHECKING:
+    from ._chart import GodWidget
+
+
 log = get_logger(__name__)
 
 # pyqtgraph global config
@@ -146,7 +147,7 @@ def run_qtractor(
 
     # load dark theme
     stylesheet = qdarkstyle.load_stylesheet(
-        qt_api='pyqt5',
+        qt_api='pyqt6',
         palette=DarkPalette,
     )
     app.setStyleSheet(stylesheet)
@@ -173,7 +174,9 @@ def run_qtractor(
     instance.window = window
 
     # override tractor's defaults
-    tractor_kwargs.update(get_tractor_runtime_kwargs())
+    tractor_kwargs.update(
+        get_runtime_vars()
+    )
 
     # define tractor entrypoint
     async def main():

piker/ui/_forms.py

@@ -21,7 +21,6 @@ Text entry "forms" widgets (mostly for configuration and UI user input).
 from __future__ import annotations
 from contextlib import asynccontextmanager
 from functools import partial
-from math import floor
 from typing import (
     Any,
     Callable,
@@ -29,9 +28,15 @@ from typing import (
 )
 
 import trio
-from PyQt5 import QtGui
-from PyQt5.QtCore import QSize, QModelIndex, Qt, QEvent
-from PyQt5.QtWidgets import (
+
+from piker.ui.qt import (
+    keys,
+    size_policy,
+    QtGui,
+    QSize,
+    QModelIndex,
+    Qt,
+    QEvent,
     QWidget,
     QLabel,
     QComboBox,
@@ -40,7 +45,6 @@ from PyQt5.QtWidgets import (
     QVBoxLayout,
     QFormLayout,
     QProgressBar,
-    QSizePolicy,
     QStyledItemDelegate,
     QStyleOptionViewItem,
 )
@@ -72,14 +76,14 @@ class Edit(QLineEdit):
 
         if width_in_chars:
             self._chars = int(width_in_chars)
-            x_size_policy = QSizePolicy.Fixed
+            x_size_policy = size_policy.Fixed
 
         else:
             # chart count which will be used to calculate
             # width of input field.
             self._chars: int = 6
             # fit to surroundingn frame width
-            x_size_policy = QSizePolicy.Expanding
+            x_size_policy = size_policy.Expanding
 
         super().__init__(parent)
 
@@ -87,7 +91,7 @@ class Edit(QLineEdit):
         # https://doc.qt.io/qt-5/qsizepolicy.html#Policy-enum
         self.setSizePolicy(
             x_size_policy,
-            QSizePolicy.Fixed,
+            size_policy.Fixed,
         )
         self.setFont(font.font)
 
@@ -181,11 +185,13 @@ class Selection(QComboBox):
 
         self._items: dict[str, int] = {}
         super().__init__(parent=parent)
-        self.setSizeAdjustPolicy(QComboBox.AdjustToContents)
+        self.setSizeAdjustPolicy(
+            QComboBox.SizeAdjustPolicy.AdjustToContents,
+        )
         # make line edit expand to surrounding frame
         self.setSizePolicy(
-            QSizePolicy.Expanding,
-            QSizePolicy.Fixed,
+            size_policy.Expanding,
+            size_policy.Fixed,
         )
         view = self.view()
         view.setUniformItemSizes(True)
@@ -309,8 +315,8 @@ class FieldsForm(QWidget):
 
         # size it as we specify
         self.setSizePolicy(
-            QSizePolicy.Expanding,
-            QSizePolicy.Expanding,
+            size_policy.Expanding,
+            size_policy.Expanding,
         )
 
         # XXX: not sure why we have to create this here exactly
@@ -417,8 +423,8 @@ class FieldsForm(QWidget):
         select.set_items(values)
 
         self.setSizePolicy(
-            QSizePolicy.Fixed,
-            QSizePolicy.Fixed,
+            size_policy.Fixed,
+            size_policy.Fixed,
         )
         select.show()
         self.form.addRow(label, select)
@@ -438,7 +444,10 @@ async def handle_field_input(
 
     async for kbmsg in recv_chan:
 
-        if kbmsg.etype in {QEvent.KeyPress, QEvent.KeyRelease}:
+        if kbmsg.etype in {
+            keys.KeyPress,
+            keys.KeyRelease,
+        }:
             event, etype, key, mods, txt = kbmsg.to_tuple()
             print(f'key: {kbmsg.key}, mods: {kbmsg.mods}, txt: {kbmsg.txt}')
 
@@ -704,7 +713,8 @@ def mk_fill_status_bar(
     )
 
     bottom_label = form.add_field_label(
-        'x: {step_size}',
+        # 'x: {step_size}',
+       '{unit_prefix}: {step_size}',
         font_size=bar_label_font_size,
         font_color='gunmetal',
     )

piker/ui/_fsp.py

@@ -36,16 +36,27 @@ from tractor.trionics import maybe_open_context
 import trio
 from trio_typing import TaskStatus
 
-from piker.data.types import Struct
-from ._axes import PriceAxis
-from ..calc import humanize
-from ..data._sharedmem import (
+from piker.accounting import MktPair
+from piker.fsp import (
+    cascade,
+    maybe_mk_fsp_shm,
+    Fsp,
+    dolla_vlm,
+    flow_rates,
+)
+from piker.data import (
+    Flume,
     ShmArray,
+)
+from piker.data._sharedmem import (
     _Token,
     try_read,
 )
-from ..data.feed import Flume
-from ..accounting import MktPair
+from piker.log import get_logger
+from piker.toolz import Profiler
+from piker.types import Struct
+from ._axes import PriceAxis
+from ..calc import humanize
 from ._chart import (
     ChartPlotWidget,
     LinkedSplits,
@@ -55,18 +66,6 @@ from ._forms import (
     mk_form,
     open_form_input_handling,
 )
-from ..fsp._api import (
-    maybe_mk_fsp_shm,
-    Fsp,
-)
-from ..fsp import cascade
-from ..fsp._volume import (
-    # tina_vwap,
-    dolla_vlm,
-    flow_rates,
-)
-from ..log import get_logger
-from .._profile import Profiler
 
 log = get_logger(__name__)
 
@@ -182,7 +181,10 @@ async def open_fsp_sidepane(
 async def open_fsp_actor_cluster(
     names: list[str] = ['fsp_0', 'fsp_1'],
 
-) -> AsyncGenerator[int, dict[str, tractor.Portal]]:
+) -> AsyncGenerator[
+    int,
+    dict[str, tractor.Portal]
+]:
 
     from tractor._clustering import open_actor_cluster
 
@@ -283,6 +285,7 @@ async def run_fsp_ui(
             name,
             array_key=array_key,
         )
+        assert chart.qframe
 
         chart.linked.focus()
 
@@ -390,7 +393,7 @@ class FspAdmin:
         complete: trio.Event,
         started: trio.Event,
         fqme: str,
-        dst_fsp_flume: Flume,
+        dst_flume: Flume,
         conf: dict,
         target: Fsp,
         loglevel: str,
@@ -408,16 +411,14 @@ class FspAdmin:
                 # chaining entrypoint
                 cascade,
 
+                # TODO: can't we just drop this and expect
+                # far end to read the src flume's .mkt.fqme?
                 # data feed key
                 fqme=fqme,
 
-                # TODO: pass `Flume.to_msg()`s here?
-                # mems
-                src_shm_token=self.flume.rt_shm.token,
-                dst_shm_token=dst_fsp_flume.rt_shm.token,
-
-                # target
-                ns_path=ns_path,
+                src_flume_addr=self.flume.to_msg(),
+                dst_flume_addr=dst_flume.to_msg(),
+                ns_path=ns_path,  # edge-bind-func
 
                 loglevel=loglevel,
                 zero_on_step=conf.get('zero_on_step', False),
@@ -431,14 +432,14 @@ class FspAdmin:
             ctx.open_stream() as stream,
         ):
 
-            dst_fsp_flume.stream: tractor.MsgStream = stream
+            dst_flume.stream: tractor.MsgStream = stream
 
             # register output data
             self._registry[
                 (fqme, ns_path)
             ] = (
                 stream,
-                dst_fsp_flume.rt_shm,
+                dst_flume.rt_shm,
                 complete
             )
 
@@ -515,7 +516,7 @@ class FspAdmin:
             broker='piker',
             _atype='fsp',
         )
-        dst_fsp_flume = Flume(
+        dst_flume = Flume(
             mkt=mkt,
             _rt_shm_token=dst_shm.token,
             first_quote={},
@@ -543,13 +544,13 @@ class FspAdmin:
             complete,
             started,
             fqme,
-            dst_fsp_flume,
+            dst_flume,
             conf,
             target,
             loglevel,
         )
 
-        return dst_fsp_flume, started
+        return dst_flume, started
 
     async def open_fsp_chart(
         self,
@@ -559,7 +560,7 @@ class FspAdmin:
         conf: dict,  # yeah probably dumb..
         loglevel: str = 'error',
 
-    ) -> (trio.Event, ChartPlotWidget):
+    ) -> trio.Event:
 
         flume, started = await self.start_engine_task(
             target,
@@ -926,7 +927,7 @@ async def start_fsp_displays(
 
     linked: LinkedSplits,
     flume: Flume,
-    group_status_key: str,
+    # group_status_key: str,
     loglevel: str,
 
 ) -> None:
@@ -973,21 +974,23 @@ async def start_fsp_displays(
             flume,
         ) as admin,
     ):
-        statuses = []
+        statuses: list[trio.Event] = []
         for target, conf in fsp_conf.items():
-            started = await admin.open_fsp_chart(
+            started: trio.Event = await admin.open_fsp_chart(
                 target,
                 conf,
             )
-            done = linked.window().status_bar.open_status(
-                f'loading fsp, {target}..',
-                group_key=group_status_key,
-            )
-            statuses.append((started, done))
+            # done = linked.window().status_bar.open_status(
+            #     f'loading fsp, {target}..',
+            #     group_key=group_status_key,
+            # )
+            # statuses.append((started, done))
+            statuses.append(started)
 
-        for fsp_loaded, status_cb in statuses:
+        # for fsp_loaded, status_cb in statuses:
+        for fsp_loaded in statuses:
             await fsp_loaded.wait()
             profiler(f'attached to fsp portal: {target}')
-            status_cb()
+            # status_cb()
 
     # blocks on nursery until all fsp actors complete

piker/ui/_icons.py

@@ -15,15 +15,18 @@
 # along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
 '''
-``QIcon`` hackery.
+`QIcon` hackery.
+
+Mostly dynamically loading pixmaps for use with `QGraphicsScene`.
 
 '''
-from PyQt5.QtWidgets import QStyle
-from PyQt5.QtGui import (
-    QIcon, QPixmap, QColor
+from piker.ui.qt import (
+    QSize,
+    QStyle,
+    QIcon,
+    QPixmap,
+    QColor,
 )
-from PyQt5.QtCore import QSize
-
 from ._style import hcolor
 
 # https://www.pythonguis.com/faq/built-in-qicons-pyqt/
@@ -44,7 +47,8 @@ def mk_icons(
     size: QSize,
 
 ) -> dict[str, QIcon]:
-    '''This helper is indempotent.
+    '''
+    This helper is indempotent.
 
     '''
     global _icons, _icon_names
@@ -56,7 +60,11 @@ def mk_icons(
     # load account selection using current style
     for name, icon_name in _icon_names.items():
 
-        stdpixmap = getattr(QStyle, icon_name)
+        stdpixmap = getattr(
+            # https://www.pythonguis.com/faq/built-in-qicons-pyqt/
+            QStyle.StandardPixmap,  # pyqt/pyside6
+            icon_name,
+        )
         stdicon = style.standardIcon(stdpixmap)
         pixmap = stdicon.pixmap(size)
 

piker/ui/_interaction.py

@@ -23,6 +23,7 @@ from contextlib import (
     asynccontextmanager,
     ExitStack,
 )
+from functools import partial
 import time
 from typing import (
     Callable,
@@ -30,46 +31,70 @@ from typing import (
 )
 
 import pyqtgraph as pg
-# from pyqtgraph.GraphicsScene import mouseEvents
-from PyQt5.QtWidgets import QGraphicsSceneMouseEvent as gs_mouse
-from PyQt5.QtCore import Qt, QEvent
-from pyqtgraph import ViewBox, Point, QtCore
-from pyqtgraph import functions as fn
+# NOTE XXX: pg is super annoying and re-implements it's own mouse
+# event subsystem.. we should really look into re-working/writing
+# this down the road.. Bo
+from pyqtgraph.GraphicsScene import mouseEvents as mevs
+# from pyqtgraph.GraphicsScene.mouseEvents import MouseDragEvent
+from pyqtgraph import (
+    ViewBox,
+    Point,
+    QtCore,
+    functions as fn,
+)
 import numpy as np
 import trio
 
+from piker.ui.qt import (
+    QWheelEvent,
+    QGraphicsSceneMouseEvent as gs_mouse,
+    Qt,
+    QEvent,
+)
 from ..log import get_logger
-from .._profile import Profiler
-from .._profile import pg_profile_enabled, ms_slower_then
+from ..toolz import (
+    Profiler,
+    pg_profile_enabled,
+    ms_slower_then,
+)
 from .view_mode import overlay_viewlists
 # from ._style import _min_points_to_show
 from ._editors import SelectRect
 from . import _event
 
 if TYPE_CHECKING:
-    from ._chart import ChartPlotWidget
+    # from ._search import (
+    #     SearchWidget,
+    # )
+    from ._chart import (
+        ChartnPane,
+        ChartPlotWidget,
+        GodWidget,
+    )
     from ._dataviz import Viz
+    from .order_mode import OrderMode
+    from ._display import DisplayState
 
 
 log = get_logger(__name__)
 
 NUMBER_LINE = {
-    Qt.Key_1,
-    Qt.Key_2,
-    Qt.Key_3,
-    Qt.Key_4,
-    Qt.Key_5,
-    Qt.Key_6,
-    Qt.Key_7,
-    Qt.Key_8,
-    Qt.Key_9,
-    Qt.Key_0,
+    Qt.Key.Key_1,
+    Qt.Key.Key_2,
+    Qt.Key.Key_3,
+    Qt.Key.Key_4,
+    Qt.Key.Key_5,
+    Qt.Key.Key_6,
+    Qt.Key.Key_7,
+    Qt.Key.Key_8,
+    Qt.Key.Key_9,
+    Qt.Key.Key_0,
 }
 
 ORDER_MODE = {
-    Qt.Key_A,
-    Qt.Key_F,
-    Qt.Key_D,
+    Qt.Key.Key_A,
+    Qt.Key.Key_F,
+    Qt.Key.Key_D,
 }
 
 
@@ -77,10 +102,12 @@ async def handle_viewmode_kb_inputs(
 
     view: ChartView,
     recv_chan: trio.abc.ReceiveChannel,
+    dss: dict[str, DisplayState],
 
 ) -> None:
 
-    order_mode = view.order_mode
+    order_mode: OrderMode = view.order_mode
+    godw: GodWidget = order_mode.godw  # noqa
 
     # track edge triggered keys
     # (https://en.wikipedia.org/wiki/Interrupt#Triggering_methods)
@@ -144,29 +171,55 @@ async def handle_viewmode_kb_inputs(
             if mods == Qt.ControlModifier:
                 ctrl = True
 
-            # UI REPL-shell
+            # UI REPL-shell, with ctrl-p (for "pause")
             if (
-                ctrl and key in {
-                    Qt.Key_U,
+                ctrl
+                and key in {
+                    Qt.Key_P,
                 }
             ):
-                import tractor
-                god = order_mode.godw  # noqa
                 feed = order_mode.feed  # noqa
                 chart = order_mode.chart  # noqa
                 viz = chart.main_viz  # noqa
                 vlm_chart = chart.linked.subplots['volume']  # noqa
                 vlm_viz = vlm_chart.main_viz  # noqa
                 dvlm_pi = vlm_chart._vizs['dolla_vlm'].plot  # noqa
-                await tractor.breakpoint()
+                import tractor
+                await tractor.pause()
                 view.interact_graphics_cycle()
 
-            # SEARCH MODE #
+            # FORCE graphics reset-and-render of all currently
+            # shown data `Viz`s for the current chart app.
+            if (
+                ctrl
+                and key in {
+                    Qt.Key_R,
+                }
+            ):
+                fqme: str
+                ds: DisplayState
+                for fqme, ds in dss.items():
+
+                    viz: Viz
+                    for tf, viz in {
+                        60: ds.hist_viz,
+                        1: ds.viz,
+                    }.items():
+                        # TODO: only allow this when the data is IN VIEW!
+                        # also, we probably can do this more efficiently
+                        # / smarter by only redrawing the portion of the
+                        # path necessary?
+                        viz.reset_graphics()
+
+            # ------ - ------
+            # SEARCH MODE
+            # ------ - ------
             # ctlr-<space>/<l> for "lookup", "search" -> open search tree
             if (
-                ctrl and key in {
+                ctrl
+                and key in {
                     Qt.Key_L,
-                    Qt.Key_Space,
+                    # Qt.Key_Space,
                 }
             ):
                 godw = view._chart.linked.godwidget
@@ -174,20 +227,56 @@ async def handle_viewmode_kb_inputs(
                 godw.search.focus()
 
             # esc and ctrl-c
-            if key == Qt.Key_Escape or (ctrl and key == Qt.Key_C):
+            if (
+                key == Qt.Key_Escape
+                or (
+                    ctrl
+                    and key == Qt.Key_C
+                )
+            ):
                 # ctrl-c as cancel
                 # https://forum.qt.io/topic/532/how-to-catch-ctrl-c-on-a-widget/9
                 view.select_box.clear()
                 view.linked.focus()
 
             # cancel order or clear graphics
-            if key == Qt.Key_C or key == Qt.Key_Delete:
+            if (
+                key == Qt.Key_C
+                or key == Qt.Key_Delete
+            ):
 
                 order_mode.cancel_orders_under_cursor()
 
             # View modes
-            if key == Qt.Key_R:
+            if (
+                ctrl
+                and (
+                    key == Qt.Key_Equal
+                    or key == Qt.Key_I
+                )
+            ):
+                view.wheelEvent(
+                    ev=None,
+                    axis=None,
+                    delta=view.def_delta,
+                )
+            elif (
+                ctrl
+                and (
+                    key == Qt.Key_Minus
+                    or key == Qt.Key_O
+                )
+            ):
+                view.wheelEvent(
+                    ev=None,
+                    axis=None,
+                    delta=-view.def_delta,
+                )
 
+            elif (
+                not ctrl
+                and key == Qt.Key_R
+            ):
                 # NOTE: seems that if we don't yield a Qt render
                 # cycle then the m4 downsampled curves will show here
                 # without another reset..
@@ -232,15 +321,47 @@ async def handle_viewmode_kb_inputs(
 
         # Toggle position config pane
         if (
-            ctrl and key in {
-                Qt.Key_P,
+            ctrl
+            and key in {
+                Qt.Key_Space,
             }
         ):
-            pp_pane = order_mode.current_pp.pane
-            if pp_pane.isHidden():
-                pp_pane.show()
+            # searchw: SearchWidget = godw.search
+            # pp_pane = order_mode.current_pp.pane
+            qframes: list[ChartnPane] = []
+
+            for linked in (
+                godw.rt_linked,
+                godw.hist_linked,
+            ):
+                for chartw in (
+                    [linked.chart]
+                    +
+                    list(linked.subplots.values())
+                ):
+                    qframes.append(
+                        chartw.qframe
+                    )
+
+            # NOTE: place priority on FIRST hiding all
+            # panes before showing them.
+            # TODO: make this more "fancy"?
+            # - maybe look at majority of hidden states and then
+            #   flip based on that?
+            # - move these loops into the chart APIs?
+            # - store the UX-state for a given feed/symbol and
+            #   apply when opening a new one (eg. if panes were
+            #   hidden then also hide them on newly loaded mkt
+            #   feeds).
+            if not any(
+                qf.sidepane.isHidden() for qf in qframes
+            ):
+                for qf in qframes:
+                    qf.sidepane.hide()
+
             else:
-                pp_pane.hide()
+                for qf in qframes:
+                    qf.sidepane.show()
 
         # ORDER MODE
         # ----------
@@ -258,7 +379,7 @@ async def handle_viewmode_kb_inputs(
             # show the pp size label only if there is
             # a non-zero pos existing
             tracker = order_mode.current_pp
-            if tracker.live_pp.size:
+            if tracker.live_pp.cumsize:
                 tracker.nav.show()
 
             # TODO: show pp config mini-params in status bar widget
@@ -338,6 +459,7 @@ async def handle_viewmode_mouse(
 
     view: ChartView,
     recv_chan: trio.abc.ReceiveChannel,
+    dss: dict[str, DisplayState],
 
 ) -> None:
 
@@ -375,6 +497,9 @@ class ChartView(ViewBox):
 
     '''
     mode_name: str = 'view'
+    def_delta: float = 616 * 6
+    def_scale_factor: float = 1.016 ** (def_delta * -1 / 20)
+    # annots: dict[int, GraphicsObject] = {}
 
     def __init__(
         self,
@@ -395,6 +520,7 @@ class ChartView(ViewBox):
             # defaultPadding=0.,
             **kwargs
         )
+
         # for "known y-range style"
         self._static_yrange = static_yrange
 
@@ -409,7 +535,11 @@ class ChartView(ViewBox):
 
         # add our selection box annotator
         self.select_box = SelectRect(self)
-        self.addItem(self.select_box, ignoreBounds=True)
+        # self.select_box.add_to_view(self)
+        # self.addItem(
+        #     self.select_box,
+        #     ignoreBounds=True,
+        # )
 
         self.mode = None
         self.order_mode: bool = False
@@ -466,6 +596,7 @@ class ChartView(ViewBox):
     @asynccontextmanager
     async def open_async_input_handler(
         self,
+        **handler_kwargs,
 
     ) -> ChartView:
 
@@ -476,14 +607,20 @@ class ChartView(ViewBox):
                     QEvent.KeyPress,
                     QEvent.KeyRelease,
                 },
-                async_handler=handle_viewmode_kb_inputs,
+                async_handler=partial(
+                    handle_viewmode_kb_inputs,
+                    **handler_kwargs,
+                ),
             ),
             _event.open_handlers(
                 [self],
                 event_types={
                     gs_mouse.GraphicsSceneMousePress,
                 },
-                async_handler=handle_viewmode_mouse,
+                async_handler=partial(
+                    handle_viewmode_mouse,
+                    **handler_kwargs,
+                ),
             ),
         ):
             yield self
@@ -499,8 +636,9 @@ class ChartView(ViewBox):
 
     def wheelEvent(
         self,
-        ev,
-        axis=None,
+        ev: QWheelEvent | None = None,
+        axis: int | None = None,
+        delta: float | None = None,
     ):
         '''
         Override "center-point" location for scrolling.
@@ -511,6 +649,12 @@ class ChartView(ViewBox):
         TODO: PR a method into ``pyqtgraph`` to make this configurable
 
         '''
+        # NOTE: certain operations are only avail when this handler is
+        # actually called on events.
+        if ev is None:
+            assert delta
+            assert axis is None
+
         linked = self.linked
         if (
             not linked
@@ -521,7 +665,7 @@ class ChartView(ViewBox):
             mask = [False, False]
             mask[axis] = self.state['mouseEnabled'][axis]
         else:
-            mask = self.state['mouseEnabled'][:]
+            mask: list[bool] = self.state['mouseEnabled'][:]
 
         chart = self.linked.chart
 
@@ -542,8 +686,15 @@ class ChartView(ViewBox):
         #     return
 
         # actual scaling factor
-        s = 1.016 ** (ev.delta() * -1 / 20)  # self.state['wheelScaleFactor'])
-        s = [(None if m is False else s) for m in mask]
+        delta: float = ev.delta() if ev else delta
+        scale_factor: float = 1.016 ** (delta * -1 / 20)
+
+        # NOTE: if elem is False -> None meaning "do not scale that
+        # axis".
+        scales: list[float | bool] = [
+            (None if m is False else scale_factor)
+            for m in mask
+        ]
 
         if (
             # zoom happened on axis
@@ -566,7 +717,7 @@ class ChartView(ViewBox):
                 ).map(ev.pos())
             )
             # scale_y = 1.3 ** (center.y() * -1 / 20)
-            self.scaleBy(s, center)
+            self.scaleBy(scales, center)
 
         # zoom in view-box area
         else:
@@ -581,7 +732,7 @@ class ChartView(ViewBox):
 
             # NOTE: scroll "around" the right most datum-element in view
             # gives the feeling of staying "pinned" in place.
-            self.scaleBy(s, focal)
+            self.scaleBy(scales, focal)
 
             # XXX: the order of the next 2 lines i'm pretty sure
             # matters, we want the resize to trigger before the graphics
@@ -601,21 +752,23 @@ class ChartView(ViewBox):
             self.interact_graphics_cycle()
             self.interact_graphics_cycle()
 
-            ev.accept()
+            if ev:
+                ev.accept()
 
     def mouseDragEvent(
         self,
-        ev,
+        ev: mevs.MouseDragEvent,
         axis: int | None = None,
 
     ) -> None:
-        pos = ev.pos()
-        lastPos = ev.lastPos()
-        dif = pos - lastPos
-        dif = dif * -1
+        pos: Point = ev.pos()
+        lastPos: Point = ev.lastPos()
+        dif: Point = (pos - lastPos) * -1
+        # dif: Point = pos - lastPos
+        # dif: Point = dif * -1
 
         # NOTE: if axis is specified, event will only affect that axis.
-        button = ev.button()
+        btn = ev.button()
 
         # Ignore axes if mouse is disabled
         mouseEnabled = np.array(
@@ -627,7 +780,7 @@ class ChartView(ViewBox):
             mask[1-axis] = 0.0
 
         # Scale or translate based on mouse button
-        if button & (
+        if btn & (
             QtCore.Qt.LeftButton | QtCore.Qt.MidButton
         ):
             # zoom y-axis ONLY when click-n-drag on it
@@ -650,34 +803,55 @@ class ChartView(ViewBox):
                 # XXX: WHY
                 ev.accept()
 
-                down_pos = ev.buttonDownPos()
+                down_pos: Point = ev.buttonDownPos(
+                    btn=btn,
+                )
+                scen_pos: Point = ev.scenePos()
+                scen_down_pos: Point = ev.buttonDownScenePos(
+                    btn=btn,
+                )
 
                 # This is the final position in the drag
                 if ev.isFinish():
 
-                    self.select_box.mouse_drag_released(down_pos, pos)
+                    # import pdbp; pdbp.set_trace()
 
-                    ax = QtCore.QRectF(down_pos, pos)
-                    ax = self.childGroup.mapRectFromParent(ax)
+                    # NOTE: think of this as a `.mouse_drag_release()`
+                    # (bc HINT that's what i called the shit ass
+                    # method that wrapped this call [yes, as a single
+                    # fucking call] originally.. you bish, guille)
+                    # Bo.. oraleeee
+                    self.select_box.set_scen_pos(
+                        # down_pos,
+                        # pos,
+                        scen_down_pos,
+                        scen_pos,
+                    )
 
                     # this is the zoom transform cmd
-                    self.showAxRect(ax)
-
+                    ax = QtCore.QRectF(down_pos, pos)
+                    ax = self.childGroup.mapRectFromParent(ax)
+                    # self.showAxRect(ax)
                     # axis history tracking
                     self.axHistoryPointer += 1
                     self.axHistory = self.axHistory[
                         :self.axHistoryPointer] + [ax]
 
                 else:
-                    print('drag finish?')
-                    self.select_box.set_pos(down_pos, pos)
+                    self.select_box.set_scen_pos(
+                        # down_pos,
+                        # pos,
+                        scen_down_pos,
+                        scen_pos,
+                    )
 
                     # update shape of scale box
                     # self.updateScaleBox(ev.buttonDownPos(), ev.pos())
-                    self.updateScaleBox(
-                        down_pos,
-                        ev.pos(),
-                    )
+                    # breakpoint()
+                    # self.updateScaleBox(
+                    #     down_pos,
+                    #     ev.pos(),
+                    # )
 
             # PANNING MODE
             else:
@@ -716,7 +890,7 @@ class ChartView(ViewBox):
                 # ev.accept()
 
         # WEIRD "RIGHT-CLICK CENTER ZOOM" MODE
-        elif button & QtCore.Qt.RightButton:
+        elif btn & QtCore.Qt.RightButton:
 
             if self.state['aspectLocked'] is not False:
                 mask[0] = 0