Handle non-fqsn for derivs and don't put brokername in

Limit ohlc queries to 800k datums to avoid `purepc` size error
Get sync-to-marketstore-tsdb history retrieval workinnn
2022-04-13 08:11:57 -04:00 · 2022-04-13 08:11:57 -04:00 · 2022-04-13 08:11:57 -04:00 · 2022-04-13 08:11:57 -04:00 · 2022-04-13 08:11:57 -04:00 · 2022-04-13 08:11:57 -04:00
52 changed files with 4790 additions and 8016 deletions
--- a/README.rst
+++ b/README.rst
@ -74,57 +74,23 @@ for a development install::
 install for tinas
 *****************
-for windows peeps you can start by installing all the prerequisite software:
+for windows peeps you can start by getting `conda installed`_
-
+and the `C++ build toolz`_ on your system.
 - install git with all default settings - https://git-scm.com/download/win
 - install anaconda all default settings - https://www.anaconda.com/products/individual
 - install microsoft build tools (check the box for Desktop development for C++, you might be able to uncheck some optional downloads)  - https://visualstudio.microsoft.com/visual-cpp-build-tools/
 - install visual studio code default settings - https://code.visualstudio.com/download
 then, `crack a conda shell`_ and run the following commands::
-    mkdir code # create code directory
+    conda create piker --python=3.9
-    cd code # change directory to code
+    conda activate piker
-    git clone https://github.com/pikers/piker.git # downloads piker installation package from github
+    conda install pip
-    cd piker # change directory to piker
+    pip install --upgrade setuptools
-    
+    cd dIreCToRieZ\oF\cODez\piker\
-    conda create -n pikonda # creates conda environment named pikonda
+    pip install -r requirements -e .
    conda activate pikonda # activates pikonda
    conda install -c conda-forge python-levenshtein # in case it is not already installed
    conda install pip # may already be installed
    pip # will show if pip is installed
    pip install -e . -r requirements.txt # install piker in editable mode
 test Piker to see if it is working::
-    piker -b binance chart btcusdt.binance # formatting for loading a chart
+in order to look coolio in front of all ur tina friends (and maybe
-    piker -b kraken -b binance chart xbtusdt.kraken
+want to help us with testin, hackzing or configgin), install
-    piker -b kraken -b binance -b ib chart qqq.nasdaq.ib
+`vscode`_ and `setup a coolio tiled wm console`_ so you can start
-    piker -b ib chart tsla.nasdaq.ib
+living the life of the tech literate..
 potential error::
    FileNotFoundError: [Errno 2] No such file or directory: 'C:\\Users\\user\\AppData\\Roaming\\piker\\brokers.toml'
 solution:
 - navigate to file directory above (may be different on your machine, location should be listed in the error code)
 - copy and paste file from 'C:\\Users\\user\\code\\data/brokers.toml' or create a blank file using notepad at the location above
 Visual Studio Code setup:
 - now that piker is installed we can set up vscode as the default terminal for running piker and editing the code
 - open Visual Studio Code
 - file --> Add Folder to Workspace --> C:\Users\user\code\piker (adds piker directory where all piker files are located)
 - file --> Save Workspace As --> save it wherever you want and call it whatever you want, this is going to be your default workspace for running and editing piker code
 - ctrl + shift + p --> start typing Python: Select Interpetter --> when the option comes up select it --> Select at the workspace level --> select the one that shows ('pikonda')
 - change the default terminal to cmd.exe instead of powershell (default)
 - now when you create a new terminal VScode should automatically activate you conda env so that piker can be run as the first command after a new terminal is created
 also, try out fancyzones as part of powertoyz for a decent tiling windows manager to manage all the cool new software you are going to be running.
 .. _conda installed: https://
 .. _C++ build toolz: https://
@ -138,7 +104,7 @@ provider support
 ****************
 for live data feeds the in-progress set of supported brokers is:
- IB_ via ``ib_insync``, also see our `container docs`_
+- IB_ via ``ib_insync``
 - binance_ and kraken_ for crypto over their public websocket API
 - questrade_ (ish) which comes with effectively free L1
@ -150,7 +116,6 @@ coming soon...
 if you want your broker supported and they have an API let us know.
 .. _IB: https://interactivebrokers.github.io/tws-api/index.html
 .. _container docs: https://github.com/pikers/piker/tree/master/dockering/ib
 .. _questrade: https://www.questrade.com/api/documentation
 .. _kraken: https://www.kraken.com/features/api#public-market-data
 .. _binance: https://github.com/pikers/piker/pull/182
--- a/config/brokers.toml
+++ b/config/brokers.toml
@ -12,41 +12,16 @@ api_key = ""
 secret = ""
 [ib]
-hosts = [
+host = "127.0.0.1"
    "127.0.0.1",
 ]
 # XXX: the order in which ports will be scanned
 # (by the `brokerd` daemon-actor)
 # is determined # by the line order here.
 # TODO: when we eventually spawn gateways in our
 # container, we can just dynamically allocate these
 # using IBC.
 ports = [
    4002,  # gw
    7497,  # tws
 ]
-# XXX: for a paper account the flex web query service
+ports.gw = 4002
-# is not supported so you have to manually download
+ports.tws = 7497
-# and XML report and put it in a location that can be
+ports.order = ["gw", "tws",]
 # accessed by the ``brokerd.ib`` backend code for parsing.
 flex_token = '666666666666666666666666'
 flex_trades_query_id = '666666'  # live account
-# when clients are being scanned this determines
+accounts.margin = "X0000000"
-# which clients are preferred to be used for data
+accounts.ira = "X0000000"
-# feeds based on the order of account names, if
+accounts.paper = "XX0000000"
 # detected as active on an API client.
 prefer_data_account = [
    'paper',
    'margin',
    'ira',
 ]
-[ib.accounts]
+# the order in which accounts will be selected (if found through
-# the order in which accounts will be selectable
+# `brokerd`) when a new symbol is loaded
-# in the order mode UI (if found via clients during
+accounts_order = ['paper', 'margin', 'ira']
 # API-app scanning)when a new symbol is loaded.
 paper = "XX0000000"
 margin = "X0000000"
 ira = "X0000000"
--- a/dockering/ib/README.rst
+++ b/dockering/ib/README.rst
@ -1,30 +0,0 @@
 running ``ib`` gateway in ``docker``
 ------------------------------------
 We have a config based on the (now defunct)
 image from "waytrade":
 https://github.com/waytrade/ib-gateway-docker
 To startup this image with our custom settings
 simply run the command::
    docker compose up
 And you should have the following socket-available services:
 - ``x11vnc1@127.0.0.1:3003``
 - ``ib-gw@127.0.0.1:4002``
 You can attach to the container via a VNC client
 without password auth.
 SECURITY STUFF!?!?!
 -------------------
 Though "``ib``" claims they host filter connections outside
 localhost (aka ``127.0.0.1``) it's probably better if you filter
 the socket at the OS level using a stateless firewall rule::
    ip rule add not unicast iif lo to 0.0.0.0/0 dport 4002
 We will soon have this baked into our own custom image but for
 now you'll have to do it urself dawgy.
--- a/dockering/ib/docker-compose.yml
+++ b/dockering/ib/docker-compose.yml
@ -1,64 +0,0 @@
 # rework from the original @
 # https://github.com/waytrade/ib-gateway-docker/blob/master/docker-compose.yml
 version: "3.5"
 services:
  ib-gateway:
    # other image tags available:
    # https://github.com/waytrade/ib-gateway-docker#supported-tags
    image: waytrade/ib-gateway:981.3j
    restart: always
    network_mode: 'host'
    volumes:
      - type: bind
        source: ./jts.ini
        target: /root/Jts/jts.ini
        # don't let IBC clobber this file for
        # the main reason of not having a stupid
        # timezone set..
        read_only: true
      # force our own IBC config
      - type: bind
        source: ./ibc.ini
        target: /root/ibc/config.ini
      # force our noop script - socat isn't needed in host mode.
      - type: bind
        source: ./fork_ports_delayed.sh
        target: /root/scripts/fork_ports_delayed.sh
      # force our noop script - socat isn't needed in host mode.
      - type: bind
        source: ./run_x11_vnc.sh
        target: /root/scripts/run_x11_vnc.sh
        read_only: true
    # NOTE:to fill these out, define an `.env` file in the same dir as
    # this compose file which looks something like:
    # TWS_USERID='myuser'
    # TWS_PASSWORD='guest'
    # TRADING_MODE=paper (or live)
    # VNC_SERVER_PASSWORD='diggity'
    environment:
      TWS_USERID: ${TWS_USERID}
      TWS_PASSWORD: ${TWS_PASSWORD}
      TRADING_MODE: ${TRADING_MODE:-paper}
      VNC_SERVER_PASSWORD: ${VNC_SERVER_PASSWORD:-}
    # ports:
    #   - target: 4002
    #     host_ip: 127.0.0.1
    #     published: 4002
    #     protocol: tcp
      # original mappings for use in non-host-mode
      # which we won't really need going forward since
      # ideally we just pick the port to have ib-gw listen
      # on **when** we spawn the container - i.e. everything
      # will be driven by a ``brokers.toml`` def.
      # - "127.0.0.1:4001:4001"
      # - "127.0.0.1:4002:4002"
      # - "127.0.0.1:5900:5900"
--- a/dockering/ib/fork_ports_delayed.sh
+++ b/dockering/ib/fork_ports_delayed.sh
@ -1,6 +0,0 @@
 #!/bin/sh
 # we now just set this is to a noop script
 # since we can just run the container in
 # `network_mode: 'host'` and get literally
 # the exact same behaviour XD
--- a/dockering/ib/ibc.ini
+++ b/dockering/ib/ibc.ini
@ -1,711 +0,0 @@
 # Note that in the comments in this file, TWS refers to both the Trader
 # Workstation and the IB Gateway, unless explicitly stated otherwise.
 #
 # When referred to below, the default value for a setting is the value
 # assumed if either the setting is included but no value is specified, or
 # the setting is not included at all.
 #
 # IBC may also be used to start the FIX CTCI Gateway. All settings
 # relating to this have names prefixed with FIX.
 #
 # The IB API Gateway and the FIX CTCI Gateway share the same code. Which
 # gateway actually runs is governed by an option on the initial gateway
 # login screen. The FIX setting described under IBC Startup
 # Settings below controls this.
 # =============================================================================
 # 1.   IBC Startup Settings
 # =============================================================================
 # IBC may be used to start the IB Gateway for the FIX CTCI. This
 # setting must be set to 'yes' if you want to run the FIX CTCI gateway. The
 # default is 'no'.
 FIX=no
 # =============================================================================
 # 2.   Authentication Settings
 # =============================================================================
 # TWS and the IB API gateway require a single username and password.
 # You may specify the username and password using the following settings:
 #
 #	IbLoginId
 #	IbPassword
 #
 # Alternatively, you can specify the username and password in the command
 # files used to start TWS or the Gateway, but this is not recommended for
 # security reasons.
 #
 # If you don't specify them, you will be prompted for them in the usual
 # login dialog when TWS starts (but whatever you have specified will be
 # included in the dialog automatically: for example you may specify the
 # username but not the password, and then you will be prompted for the
 # password via the login dialog). Note that if you specify either
 # the username or the password (or both) in the command file, then
 # IbLoginId and IbPassword settings defined in this file are ignored.
 #
 #
 # The FIX CTCI gateway requires one username and password for FIX order
 # routing, and optionally a separate username and password for market
 # data connections. You may specify the usernames and passwords using
 # the following settings:
 #
 #	FIXLoginId
 #	FIXPassword
 #	IbLoginId	(optional - for market data connections)
 #	IbPassword	(optional - for market data connections)
 #
 # Alternatively you can specify the FIX username and password in the
 # command file used to start the FIX CTCI Gateway, but this is not
 # recommended for security reasons.
 #
 # If you don't specify them, you will be prompted for them in the usual
 # login dialog when FIX CTCI gateway starts (but whatever you have
 # specified will be included in the dialog automatically: for example
 # you may specify the usernames but not the passwords, and then you will
 # be prompted for the passwords via the login dialog). Note that if you
 # specify either the FIX username or the FIX password (or both) on the
 # command line, then FIXLoginId and FIXPassword settings defined in this
 # file are ignored; he same applies to the market data username and
 # password.
 # IB API Authentication Settings
 # ------------------------------
 # Your TWS username:
 IbLoginId=
 # Your TWS password:
 IbPassword=
 # FIX CTCI Authentication Settings
 # --------------------------------
 # Your FIX CTCI username:
 FIXLoginId=
 # Your FIX CTCI password:
 FIXPassword=
 # Second Factor Authentication Settings
 # -------------------------------------
 # If you have enabled more than one second factor authentication
 # device, TWS presents a list from which you must select the device
 # you want to use for this login. You can use this setting to
 # instruct IBC to select a particular item in the list on your
 # behalf. Note that you must spell this value exactly as it appears
 # in the list. If no value is set, you must manually select the
 # relevant list entry.
 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
 # Second Factor Authentication dialog is needed to complete the
 # login.
 #
 # Permitted values are 'yes' and 'no'. The default is 'no'.
 #
 # Note that the scripts provided with the IBC zips for Windows and
 # Linux provide options to automatically restart in these
 # circumstances, but only if this setting is also set to 'yes'.
 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.
 TradingMode=
 # Paper-trading Account Warning
 # -----------------------------
 #
 # Logging in to a paper-trading account results in TWS displaying
 # a dialog asking the user to confirm that they are aware that this
 # is not a brokerage account. Until this dialog has been accepted,
 # TWS will not allow API connections to succeed. Setting this
 # to 'yes' (the default) will cause IBC to automatically
 # confirm acceptance. Setting it to 'no' will leave the dialog
 # on display, and the user will have to deal with it manually.
 AcceptNonBrokerageAccountWarning=yes
 # Login Dialog Display Timeout
 #-----------------------------
 #
 # In some circumstances, starting TWS may result in failure to display
 # the login dialog. Restarting TWS may help to resolve this situation,
 # and IBC does this automatically.
 #
 # This setting controls how long (in seconds) IBC waits for the login
 # dialog to appear before restarting TWS.
 #
 # Note that in normal circumstances with a reasonably specified 
 # computer the time to displaying the login dialog is typically less
 # than 20 seconds, and frequently much less. However many factors can
 # influence this, and it is unwise to set this value too low.
 #
 # The default value is 60.
 LoginDialogDisplayTimeout = 60
 # =============================================================================
 # 3.   TWS Startup Settings
 # =============================================================================
 # Path to settings store
 # ----------------------
 #
 # Path to the directory where TWS should store its settings. This is
 # normally the folder in which TWS is installed. However you may set
 # it to some other location if you wish (for example if you want to
 # run multiple instances of TWS with different settings).
 #
 # It is recommended for clarity that you use an absolute path. The
 # effect of using a relative path is undefined.
 #
 # Linux and macOS users should use the appropriate path syntax.
 #
 # Note that, for Windows users, you MUST use double separator
 # characters to separate the elements of the folder path: for
 # example, IbDir=C:\\IBLiveSettings is valid, but
 # IbDir=C:\IBLiveSettings is NOT valid and will give unexpected
 # results. Linux and macOS users need not use double separators,
 # but they are acceptable.
 #
 # The default is the current working directory when IBC is
 # started.
 IbDir=/root/Jts
 # Store settings on server
 # ------------------------
 #
 # If you wish to store a copy of your TWS settings on IB's
 # servers as well as locally on your computer, set this to
 # 'yes': this enables you to run TWS on different computers
 # with the same configuration, market data lines, etc. If set
 # to 'no', running TWS on different computers will not share the
 # same settings. If no value is specified, TWS will obtain its
 # settings from the same place as the last time this user logged
 # in (whether manually or using IBC).
 StoreSettingsOnServer=
 # Minimize TWS on startup
 # -----------------------
 #
 # Set to 'yes' to minimize TWS when it starts:
 MinimizeMainWindow=no
 # Existing Session Detected Action
 # --------------------------------
 #
 # When a user logs on to an IBKR account for trading purposes by any means, the
 # IBKR account server checks to see whether the account is already logged in
 # elsewhere. If so, a dialog is displayed to both the users that enables them
 # to determine what happens next. The 'ExistingSessionDetectedAction' setting
 # instructs TWS how to proceed when it displays this dialog:
 #
 #   * If the new TWS session is set to 'secondary', the existing session continues
 #     and the new session terminates. Thus a secondary TWS session can never
 #     override any other session.
 #
 #   * If the existing TWS session is set to 'primary', the existing session
 #     continues and the new session terminates (even if the new session is also
 #     set to primary). Thus a primary TWS session can never be overridden by
 #     any new session).
 #
 #   * If both the existing and the new TWS sessions are set to 'primaryoverride',
 #     the existing session terminates and the new session proceeds.
 #
 #   * If the existing TWS session is set to 'manual', the user must handle the
 #     dialog.
 #
 # The difference between 'primary' and 'primaryoverride' is that a
 # 'primaryoverride' session can be overriden over by a new 'primary' session,
 # but a 'primary' session cannot be overriden by any other session.
 #
 # When set to 'primary', if another TWS session is started and manually told to
 # end the 'primary' session, the 'primary' session is automatically reconnected.
 #
 # The default is 'manual'.
 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
 # 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
 # Read-only Login
 # ---------------
 #
 # If ReadOnlyLogin is set to 'yes', and the user is enrolled in IB's
 # 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. 
 # If the user is not enrolled in IB's account security programme,
 # this setting is ignored. The default is 'no'.
 ReadOnlyLogin=no
 # Read-only API
 # -------------
 #
 # If ReadOnlyApi is set to 'yes', API programs cannot submit, modify
 # or cancel orders. If set to 'no', API programs can do these things.
 # If not set, the existing TWS/Gateway configuration is unchanged.
 # NB: this setting is really only supplied for the benefit of new TWS
 # or Gateway instances that are being automatically installed and
 # started without user intervention (eg Docker containers). Where
 # a user is involved, they should use the Global Configuration to
 # set the relevant checkbox (this only needs to be done once) and
 # not provide a value for this setting.
 ReadOnlyApi=no
 # Market data size for US stocks - lots or shares
 # -----------------------------------------------
 #
 # Since IB introduced the option of market data for US stocks showing
 # bid, ask and last sizes in shares rather than lots, TWS and Gateway
 # display a dialog immediately after login notifying the user about
 # this and requiring user input before allowing market data to be
 # accessed. The user can request that the dialog not be shown again.
 #
 # It is recommended that the user should handle this dialog manually
 # rather than using these settings, which are provided for situations 
 # where the user interface is not easily accessible, or where user
 # settings are not preserved between sessions (eg some Docker images).
 #
 # - If this setting is set to 'accept', the dialog will be handled
 #   automatically and the option to not show it again will be
 #   selected.
 #
 #   Note that in this case, the only way to allow the dialog to be
 #   displayed again is to manually enable the 'Bid, Ask and Last
 #   Size Display Update' message in the 'Messages' section of the TWS
 #   configuration dialog. So you should only use 'Accept' if you are
 #   sure you really don't want the dialog to be displayed again, or
 #   you have easy access to the user interface.
 #
 # - If set to 'defer', the dialog will be handled automatically (so
 #   that market data will start), but the option to not show it again
 #   will not be selected, and it will be shown again after the next
 #   login.
 #
 # - If set to 'ignore', the user has to deal with the dialog manually.
 #
 # The default value is 'ignore'.
 #
 # Note if set to 'accept' or 'defer', TWS also automatically sets
 # the API settings checkbox labelled 'Send market data in lots for
 # US stocks for dual-mode API clients'. IBC cannot prevent this.
 # However you can change this immmediately by setting
 # SendMarketDataInLotsForUSstocks (see below) to 'no' .
 AcceptBidAskLastSizeDisplayUpdateNotification=accept
 # This setting determines whether the API settings checkbox labelled
 # 'Send market data in lots for US stocks for dual-mode API clients'
 # is set or cleared. If set to 'yes', the checkbox is set. If set to
 # 'no' the checkbox is cleared. If defaulted, the checkbox is
 # unchanged.
 SendMarketDataInLotsForUSstocks=
 # =============================================================================
 # 4.   TWS Auto-Closedown
 # =============================================================================
 #
 # IMPORTANT NOTE: Starting with TWS 974, this setting no longer
 # works properly, because IB have changed the way TWS handles its
 # autologoff mechanism. 
 #
 # You should now configure the TWS autologoff time to something
 # convenient for you, and restart IBC each day.
 #
 # 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).
 # 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
 # =============================================================================
 # 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.
 #
 # 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.
 #
 # To tell IBC to tidily close TWS at a specified time every
 # day, set this value to <hh:mm>, for example:
 # ClosedownAt=22:00
 #
 # To tell IBC to tidily close TWS at a specified day and time
 # each week, set this value to <dayOfWeek hh:mm>, for example:
 # ClosedownAt=Friday 22:00
 #
 # Note that the day of the week must be specified using your
 # default locale. Also note that Java will only accept
 # characters encoded to ISO 8859-1 (Latin-1). This means that
 # if the day name in your default locale uses any non-Latin-1
 # characters you need to encode them using Unicode escapes
 # (see http://java.sun.com/docs/books/jls/third_edition/html/lexical.html#3.3
 # for details). For example, to tidily close TWS at 12:00 on
 # Saturday where the default locale is Simplified Chinese,
 # use the following:
 # #ClosedownAt=\u661F\u671F\u516D 12:00
 ClosedownAt=
 # =============================================================================
 # 6.   Other TWS Settings
 # =============================================================================
 # Accept Incoming Connection
 # --------------------------
 #
 # If set to 'accept', IBC automatically accepts incoming
 # API connection dialogs. If set to 'reject', IBC
 # automatically rejects incoming API connection dialogs. If
 # set to 'manual', the user must decide whether to accept or reject
 # incoming API connection dialogs. The default is 'manual'.
 # NB: it is recommended to set this to 'reject', and to explicitly
 # configure which IP addresses can connect to the API in TWS's API
 # configuration page, as this is much more secure (in this case, no
 # incoming API connection dialogs will occur for those IP addresses).
 AcceptIncomingConnectionAction=reject
 # Allow Blind Trading
 # -------------------
 #
 # If you attempt to place an order for a contract for which
 # you have no market data subscription, TWS displays a dialog
 # to warn you against such blind trading.
 #
 #   yes   means the dialog is dismissed as though the user had
 # 	  clicked the 'Ok' button: this means that you accept
 # 	  the risk and want the order to be submitted.
 #
 #   no    means the dialog remains on display and must be
 #         handled by the user.
 AllowBlindTrading=yes
 # Save Settings on a Schedule
 # ---------------------------
 #
 # You can tell TWS to automatically save its settings on a schedule
 # of your choosing. You can specify one or more specific times,
 # like this:
 #
 # SaveTwsSettingsAt=HH:MM [ HH:MM]...
 #
 # for example:
 # SaveTwsSettingsAt=08:00   12:30 17:30
 #
 # Or you can specify an interval at which settings are to be saved,
 # optionally starting at a specific time and continuing until another
 # time, like this:
 #
 #SaveTwsSettingsAt=Every n [{mins | hours}] [hh:mm] [hh:mm]
 #
 # where the first hh:mm is the start time and the second is the end
 # time. If you don't specify the end time, settings are saved regularly
 # from the start time till midnight. If you don't specify the start time.
 # settings are saved regularly all day, beginning at 00:00. Note that
 # settings will always be saved at the end time, even if that is not
 # exactly one interval later than the previous time. If neither 'mins'
 # nor 'hours' is specified, 'mins' is assumed. Examples:
 #
 # To save every 30 minutes all day starting at 00:00
 #SaveTwsSettingsAt=Every 30
 #SaveTwsSettingsAt=Every 30 mins
 #
 # To save every hour starting at 08:00 and ending at midnight
 #SaveTwsSettingsAt=Every 1 hours 08:00
 #SaveTwsSettingsAt=Every 1 hours 08:00 00:00
 #
 # To save every 90 minutes starting at 08:00 up to and including 17:43
 #SaveTwsSettingsAt=Every 90 08:00 17:43
 SaveTwsSettingsAt=
 # =============================================================================
 # 7.   Settings Specific to Indian Versions of TWS
 # =============================================================================
 # Indian versions of TWS may display a password expiry
 # notification dialog and a NSE Compliance dialog. These can be
 # dismissed by setting the following to yes. By default the
 # password expiry notice is not dismissed, but the NSE Compliance
 # notice is dismissed.
 # Warning: setting DismissPasswordExpiryWarning=yes will mean
 # you will not be notified when your password is about to expire.
 # You must then take other measures to ensure that your password
 # is changed within the expiry period, otherwise IBC will
 # not be able to login successfully.
 DismissPasswordExpiryWarning=no
 DismissNSEComplianceNotice=yes
 # =============================================================================
 # 8.   IBC Command Server Settings
 # =============================================================================
 # Do NOT CHANGE THE FOLLOWING SETTINGS unless you
 # intend to issue commands to IBC (for example
 # using telnet). Note that these settings have nothing to
 # do with running programs that use the TWS API.
 # Command Server Port Number
 # --------------------------
 #
 # 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
 #CommandServerPort=7462
 # Permitted Command Sources
 # -------------------------
 #
 # A comma separated list of IP addresses, or host names,
 # which are allowed addresses for sending commands to
 # IBC.  Commands can always be sent from the
 # same host as IBC is running on.
 ControlFrom=127.0.0.1
 # 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
 # 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
 # Command Prompt
 # --------------
 #
 # The specified string is output by the server when
 # the connection is first opened and after the completion
 # of each command. This can be useful if sending commands
 # using an interactive program such as telnet. The default
 # is that no prompt is output.
 # For example:
 #
 # CommandPrompt=>
 CommandPrompt=
 # Suppress Command Server Info Messages
 # -------------------------------------
 #
 # Some commands can return intermediate information about
 # their progress. This setting controls whether such
 # information is sent. The default is that such information
 # is not sent.
 SuppressInfoMessages=no
 # =============================================================================
 # 9.   Diagnostic Settings
 # =============================================================================
 #
 # IBC can log information about the structure of windows
 # displayed by TWS. This information is useful when adding
 # new features to IBC or when behaviour is not as expected. 
 #
 # The logged information shows the hierarchical organisation
 # of all the components of the window, and includes the
 # current values of text boxes and labels.
 #
 # Note that this structure logging has a small performance
 # impact, and depending on the settings can cause the logfile
 # size to be significantly increased. It is therefore
 # recommended that the LogStructureWhen setting be set to
 # 'never' (the default) unless there is a specific reason
 # that this information is needed.
 # Scope of Structure Logging
 # --------------------------
 #
 # 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;
 #
 #    - if set to 'unknown', only windows that IBC does not
 #      recognize are eligible. Most windows displayed by
 #      TWS fall into this category;
 #
 #    - if set to 'untitled', only windows that IBC does not
 #      recognize and that have no title are eligible. These
 #      are usually message boxes or similar small windows,
 #
 #    - if set to 'all', then every window displayed by TWS
 #      is eligible.
 #
 # The default value is 'known'.
 LogStructureScope=all
 # When to Log Window Structure
 # ----------------------------
 #
 # The LogStructureWhen setting specifies the circumstances
 # when eligible TWS windows have their structure logged:
 #
 #     - if set to 'open' or 'yes' or 'true', IBC logs the
 #       structure of an eligible window the first time it
 #       is encountered;
 #
 #    - 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.
 #
 # The default value is 'never'.
 LogStructureWhen=never
 # DEPRECATED SETTING
 # ------------------
 #
 # LogComponents - THIS SETTING WILL BE REMOVED IN A FUTURE
 # RELEASE
 #
 # If LogComponents is set to any value, this is equivalent
 # to setting LogStructureWhen to that same value and
 # LogStructureScope to 'all': the actual values of those
 # settings are ignored. The default is that the values
 # of LogStructureScope and LogStructureWhen are honoured.
 #LogComponents=
--- a/dockering/ib/jts.ini
+++ b/dockering/ib/jts.ini
@ -1,33 +0,0 @@
 [IBGateway]
 ApiOnly=true
 LocalServerPort=4002
 # NOTE: must be set if using IBC's "reject" mode
 TrustedIPs=127.0.0.1
 ; RemoteHostOrderRouting=ndc1.ibllc.com
 ; WriteDebug=true
 ; RemotePortOrderRouting=4001
 ; useRemoteSettings=false
 ; tradingMode=p
 ; Steps=8
 ; colorPalletName=dark
 # window geo, this may be useful for sending `xdotool` commands?
 ; MainWindow.Width=1986
 ; screenHeight=3960
 [Logon]
 Locale=en
 # most markets are oriented around this zone
 # so might as well hard code it.
 TimeZone=America/New_York
 UseSSL=true
 displayedproxymsg=1
 os_titlebar=true
 s3store=true
 useRemoteSettings=false
 [Communication]
 ctciAutoEncrypt=true
 Region=usr
 ; Peer=cdc1.ibllc.com:4001
--- a/dockering/ib/run_x11_vnc.sh
+++ b/dockering/ib/run_x11_vnc.sh
@ -1,16 +0,0 @@
 #!/bin/sh
 # start VNC server
 x11vnc \
    -ncache_cr \
    -listen localhost \
    -display :1 \
    -forever \
    -shared \
    -logappend /var/log/x11vnc.log \
    -bg \
    -noipv6 \
    -autoport 3003 \
    # can't use this because of ``asyncvnc`` issue:
    # https://github.com/barneygale/asyncvnc/issues/1
    # -passwd 'ibcansmbz'
--- a/piker/_daemon.py
+++ b/piker/_daemon.py
@ -35,7 +35,7 @@ log = get_logger(__name__)
 _root_dname = 'pikerd'
-_registry_addr = ('127.0.0.1', 6116)
+_registry_addr = ('127.0.0.1', 1616)
 _tractor_kwargs: dict[str, Any] = {
    # use a different registry addr then tractor's default
    'arbiter_addr': _registry_addr
@ -426,19 +426,9 @@ async def spawn_brokerd(
    # ask `pikerd` to spawn a new sub-actor and manage it under its
    # actor nursery
    modpath = brokermod.__name__
    broker_enable = [modpath]
    for submodname in getattr(
        brokermod,
        '__enable_modules__',
        [],
    ):
        subpath = f'{modpath}.{submodname}'
        broker_enable.append(subpath)
    portal = await _services.actor_n.start_actor(
        dname,
-        enable_modules=_data_mods + broker_enable,
+        enable_modules=_data_mods + [brokermod.__name__],
        loglevel=loglevel,
        debug_mode=_services.debug_mode,
        **tractor_kwargs
--- a/piker/brokers/_util.py
+++ b/piker/brokers/_util.py
@ -33,41 +33,7 @@ class SymbolNotFound(BrokerError):
 class NoData(BrokerError):
-    '''
+    "Symbol data not permitted"
    Symbol data not permitted or no data
    for time range found.
    '''
    def __init__(
        self,
        *args,
        frame_size: int = 1000,
    ) -> None:
        super().__init__(*args)
        # when raised, machinery can check if the backend
        # set a "frame size" for doing datetime calcs.
        self.frame_size: int = 1000
 class DataUnavailable(BrokerError):
    '''
    Signal storage requests to terminate.
    '''
    # TODO: add in a reason that can be displayed in the
    # UI (for eg. `kraken` is bs and you should complain
    # to them that you can't pull more OHLC data..)
 class DataThrottle(BrokerError):
    '''
    Broker throttled request rate for data.
    '''
    # TODO: add in throttle metrics/feedback
 def resproc(
@ -84,12 +50,12 @@ def resproc(
    if not resp.status_code == 200:
        raise BrokerError(resp.body)
    try:
-        msg = resp.json()
+        json = resp.json()
    except json.decoder.JSONDecodeError:
        log.exception(f"Failed to process {resp}:\n{resp.text}")
        raise BrokerError(resp.text)
    if log_resp:
-        log.debug(f"Received json contents:\n{colorize_json(msg)}")
+        log.debug(f"Received json contents:\n{colorize_json(json)}")
-    return msg if return_json else resp
+    return json if return_json else resp
--- a/piker/brokers/binance.py
+++ b/piker/brokers/binance.py
@ -19,7 +19,6 @@ Binance backend
 """
 from contextlib import asynccontextmanager as acm
 from datetime import datetime
 from typing import (
    Any, Union, Optional,
    AsyncGenerator, Callable,
@ -28,7 +27,7 @@ import time
 import trio
 from trio_typing import TaskStatus
-import pendulum
+import arrow
 import asks
 from fuzzywuzzy import process as fuzzy
 import numpy as np
@ -133,7 +132,7 @@ class OHLC:
    bar_wap: float = 0.0
-# convert datetime obj timestamp to unixtime in milliseconds
+# convert arrow timestamp to unixtime in miliseconds
 def binance_timestamp(when):
    return int((when.timestamp() * 1000) + (when.microsecond / 1000))
@ -222,22 +221,20 @@ class Client:
    async def bars(
        self,
        symbol: str,
-        start_dt: Optional[datetime] = None,
+        start_time: int = None,
-        end_dt: Optional[datetime] = None,
+        end_time: int = None,
        limit: int = 1000,  # <- max allowed per query
        as_np: bool = True,
    ) -> dict:
-        if end_dt is None:
+        if start_time is None:
-            end_dt = pendulum.now('UTC')
+            start_time = binance_timestamp(
                arrow.utcnow().floor('minute').shift(minutes=-limit)
            )
-        if start_dt is None:
+        if end_time is None:
-            start_dt = end_dt.start_of(
+            end_time = binance_timestamp(arrow.utcnow())
                'minute').subtract(minutes=limit)
        start_time = binance_timestamp(start_dt)
        end_time = binance_timestamp(end_dt)
        # https://binance-docs.github.io/apidocs/spot/en/#kline-candlestick-data
        bars = await self._api(
@ -382,27 +379,7 @@ async def open_history_client(
    # TODO implement history getter for the new storage layer.
    async with open_cached_client('binance') as client:
-
+        yield client
        async def get_ohlc(
            end_dt: Optional[datetime] = None,
            start_dt: Optional[datetime] = None,
        ) -> tuple[
            np.ndarray,
            datetime,  # start
            datetime,  # end
        ]:
            array = await client.bars(
                symbol,
                start_dt=start_dt,
                end_dt=end_dt,
            )
            start_dt = pendulum.from_timestamp(array[0]['time'])
            end_dt = pendulum.from_timestamp(array[-1]['time'])
            return array, start_dt, end_dt
        yield get_ohlc, {'erlangs': 3, 'rate': 3}
 async def backfill_bars(
@ -452,8 +429,8 @@ async def stream_quotes(
            # XXX: after manually inspecting the response format we
            # just directly pick out the info we need
-            si['price_tick_size'] = float(syminfo.filters[0]['tickSize'])
+            si['price_tick_size'] = syminfo.filters[0]['tickSize']
-            si['lot_tick_size'] = float(syminfo.filters[2]['stepSize'])
+            si['lot_tick_size'] = syminfo.filters[2]['stepSize']
            si['asset_type'] = 'crypto'
        symbol = symbols[0]
--- a/piker/brokers/cli.py
+++ b/piker/brokers/cli.py
@ -23,6 +23,7 @@ from operator import attrgetter
 from operator import itemgetter
 import click
 import pandas as pd
 import trio
 import tractor
@ -46,10 +47,8 @@ _watchlists_data_path = os.path.join(_config_dir, 'watchlists.json')
@click.argument('kwargs', nargs=-1)
@click.pass_obj
 def api(config, meth, kwargs, keys):
-    '''
+    """Make a broker-client API method call
-    Make a broker-client API method call
+    """
    '''
    # global opts
    broker = config['brokers'][0]
@ -80,13 +79,13 @@ def api(config, meth, kwargs, keys):
@cli.command()
@click.option('--df-output', '-df', flag_value=True,
              help='Output in `pandas.DataFrame` format')
@click.argument('tickers', nargs=-1, required=True)
@click.pass_obj
-def quote(config, tickers):
+def quote(config, tickers, df_output):
-    '''
+    """Print symbol quotes to the console
-    Print symbol quotes to the console
+    """
    '''
    # global opts
    brokermod = config['brokermods'][0]
@ -101,19 +100,28 @@ def quote(config, tickers):
            if ticker not in syms:
                brokermod.log.warn(f"Could not find symbol {ticker}?")
-    click.echo(colorize_json(quotes))
+    if df_output:
        cols = next(filter(bool, quotes)).copy()
        cols.pop('symbol')
        df = pd.DataFrame(
            (quote or {} for quote in quotes),
            columns=cols,
        )
        click.echo(df)
    else:
        click.echo(colorize_json(quotes))
@cli.command()
@click.option('--df-output', '-df', flag_value=True,
              help='Output in `pandas.DataFrame` format')
@click.option('--count', '-c', default=1000,
              help='Number of bars to retrieve')
@click.argument('symbol', required=True)
@click.pass_obj
-def bars(config, symbol, count):
+def bars(config, symbol, count, df_output):
-    '''
+    """Retreive 1m bars for symbol and print on the console
-    Retreive 1m bars for symbol and print on the console
+    """
    '''
    # global opts
    brokermod = config['brokermods'][0]
@ -125,7 +133,7 @@ def bars(config, symbol, count):
            brokermod,
            symbol,
            count=count,
-            as_np=False,
+            as_np=df_output
        )
    )
@ -133,7 +141,10 @@ def bars(config, symbol, count):
        log.error(f"No quotes could be found for {symbol}?")
        return
-    click.echo(colorize_json(bars))
+    if df_output:
        click.echo(pd.DataFrame(bars))
    else:
        click.echo(colorize_json(bars))
@cli.command()
@ -145,10 +156,8 @@ def bars(config, symbol, count):
@click.argument('name', nargs=1, required=True)
@click.pass_obj
 def record(config, rate, name, dhost, filename):
-    '''
+    """Record client side quotes to a file on disk
-    Record client side quotes to a file on disk
+    """
    '''
    # global opts
    brokermod = config['brokermods'][0]
    loglevel = config['loglevel']
@ -186,10 +195,8 @@ def record(config, rate, name, dhost, filename):
@click.argument('symbol', required=True)
@click.pass_context
 def contracts(ctx, loglevel, broker, symbol, ids):
-    '''
+    """Get list of all option contracts for symbol
-    Get list of all option contracts for symbol
+    """
    '''
    brokermod = get_brokermod(broker)
    get_console_log(loglevel)
@ -206,14 +213,14 @@ def contracts(ctx, loglevel, broker, symbol, ids):
@cli.command()
@click.option('--df-output', '-df', flag_value=True,
              help='Output in `pandas.DataFrame` format')
@click.option('--date', '-d', help='Contracts expiry date')
@click.argument('symbol', required=True)
@click.pass_obj
-def optsquote(config, symbol, date):
+def optsquote(config, symbol, df_output, date):
-    '''
+    """Retreive symbol option quotes on the console
-    Retreive symbol option quotes on the console
+    """
    '''
    # global opts
    brokermod = config['brokermods'][0]
@ -226,17 +233,22 @@ def optsquote(config, symbol, date):
        log.error(f"No option quotes could be found for {symbol}?")
        return
-    click.echo(colorize_json(quotes))
+    if df_output:
        df = pd.DataFrame(
            (quote.values() for quote in quotes),
            columns=quotes[0].keys(),
        )
        click.echo(df)
    else:
        click.echo(colorize_json(quotes))
@cli.command()
@click.argument('tickers', nargs=-1, required=True)
@click.pass_obj
 def symbol_info(config, tickers):
-    '''
+    """Print symbol quotes to the console
-    Print symbol quotes to the console
+    """
    '''
    # global opts
    brokermod = config['brokermods'][0]
@ -258,10 +270,8 @@ def symbol_info(config, tickers):
@click.argument('pattern', required=True)
@click.pass_obj
 def search(config, pattern):
-    '''
+    """Search for symbols from broker backend(s).
-    Search for symbols from broker backend(s).
+    """
    '''
    # global opts
    brokermods = config['brokermods']
--- a/piker/brokers/ib.py
+++ b/piker/brokers/ib.py
--- a/piker/brokers/ib/init.py
+++ b/piker/brokers/ib/init.py
@ -1,67 +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/>.
 """
 Interactive Brokers API backend.
 Sub-modules within break into the core functionalities:
 - ``broker.py`` part for orders / trading endpoints
 - ``data.py`` for real-time data feed endpoints
 - ``client.py`` for the core API machinery which is ``trio``-ized
  wrapping around ``ib_insync``.
 - ``report.py`` for the hackery to build manual pp calcs
  to avoid ib's absolute bullshit FIFO style position
  tracking..
 """
 from .api import (
    get_client,
 )
 from .feed import (
    open_history_client,
    open_symbol_search,
    stream_quotes,
 )
 from .broker import trades_dialogue
 __all__ = [
    'get_client',
    'trades_dialogue',
    'open_history_client',
    'open_symbol_search',
    'stream_quotes',
 ]
 # tractor RPC enable arg
 __enable_modules__: list[str] = [
    'api',
    'feed',
    'broker',
 ]
 # passed to ``tractor.ActorNursery.start_actor()``
 _spawn_kwargs = {
    'infect_asyncio': True,
 }
 # annotation to let backend agnostic code
 # know if ``brokerd`` should be spawned with
 # ``tractor``'s aio mode.
 _infect_asyncio: bool = True
--- a/piker/brokers/ib/api.py
+++ b/piker/brokers/ib/api.py
--- a/piker/brokers/ib/broker.py
+++ b/piker/brokers/ib/broker.py
@ -1,590 +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/>.
 """
 Order and trades endpoints for use with ``piker``'s EMS.
 """
 from __future__ import annotations
 from dataclasses import asdict
 from functools import partial
 from pprint import pformat
 import time
 from typing import (
    Any,
    Optional,
    AsyncIterator,
 )
 import trio
 from trio_typing import TaskStatus
 import tractor
 from ib_insync.contract import (
    Contract,
    Option,
 )
 from ib_insync.order import (
    Trade,
    OrderStatus,
 )
 from ib_insync.objects import (
    Fill,
    Execution,
 )
 from ib_insync.objects import Position
 from piker import config
 from piker.log import get_console_log
 from piker.clearing._messages import (
    BrokerdOrder,
    BrokerdOrderAck,
    BrokerdStatus,
    BrokerdPosition,
    BrokerdCancel,
    BrokerdFill,
    BrokerdError,
 )
 from .api import (
    _accounts2clients,
    _adhoc_futes_set,
    log,
    get_config,
    open_client_proxies,
    Client,
 )
 def pack_position(
    pos: Position
 ) -> dict[str, Any]:
    con = pos.contract
    if isinstance(con, Option):
        # TODO: option symbol parsing and sane display:
        symbol = con.localSymbol.replace(' ', '')
    else:
        # TODO: lookup fqsn even for derivs.
        symbol = con.symbol.lower()
    exch = (con.primaryExchange or con.exchange).lower()
    symkey = '.'.join((symbol, exch))
    if not exch:
        # attempt to lookup the symbol from our
        # hacked set..
        for sym in _adhoc_futes_set:
            if symbol in sym:
                symkey = sym
                break
    expiry = con.lastTradeDateOrContractMonth
    if expiry:
        symkey += f'.{expiry}'
    # TODO: options contracts into a sane format..
    return BrokerdPosition(
        broker='ib',
        account=pos.account,
        symbol=symkey,
        currency=con.currency,
        size=float(pos.position),
        avg_price=float(pos.avgCost) / float(con.multiplier or 1.0),
    )
 async def handle_order_requests(
    ems_order_stream: tractor.MsgStream,
    accounts_def: dict[str, str],
 ) -> None:
    request_msg: dict
    async for request_msg in ems_order_stream:
        log.info(f'Received order request {request_msg}')
        action = request_msg['action']
        account = request_msg['account']
        acct_number = accounts_def.get(account)
        if not acct_number:
            log.error(
                f'An IB account number for name {account} is not found?\n'
                'Make sure you have all TWS and GW instances running.'
            )
            await ems_order_stream.send(BrokerdError(
                oid=request_msg['oid'],
                symbol=request_msg['symbol'],
                reason=f'No account found: `{account}` ?',
            ).dict())
            continue
        client = _accounts2clients.get(account)
        if not client:
            log.error(
                f'An IB client for account name {account} is not found.\n'
                'Make sure you have all TWS and GW instances running.'
            )
            await ems_order_stream.send(BrokerdError(
                oid=request_msg['oid'],
                symbol=request_msg['symbol'],
                reason=f'No api client loaded for account: `{account}` ?',
            ).dict())
            continue
        if action in {'buy', 'sell'}:
            # validate
            order = BrokerdOrder(**request_msg)
            # call our client api to submit the order
            reqid = client.submit_limit(
                oid=order.oid,
                symbol=order.symbol,
                price=order.price,
                action=order.action,
                size=order.size,
                account=acct_number,
                # XXX: by default 0 tells ``ib_insync`` methods that
                # there is no existing order so ask the client to create
                # a new one (which it seems to do by allocating an int
                # counter - collision prone..)
                reqid=order.reqid,
            )
            if reqid is None:
                await ems_order_stream.send(BrokerdError(
                    oid=request_msg['oid'],
                    symbol=request_msg['symbol'],
                    reason='Order already active?',
                ).dict())
            # deliver ack that order has been submitted to broker routing
            await ems_order_stream.send(
                BrokerdOrderAck(
                    # ems order request id
                    oid=order.oid,
                    # broker specific request id
                    reqid=reqid,
                    time_ns=time.time_ns(),
                    account=account,
                ).dict()
            )
        elif action == 'cancel':
            msg = BrokerdCancel(**request_msg)
            client.submit_cancel(reqid=msg.reqid)
        else:
            log.error(f'Unknown order command: {request_msg}')
 async def recv_trade_updates(
    client: Client,
    to_trio: trio.abc.SendChannel,
 ) -> None:
    """Stream a ticker using the std L1 api.
    """
    client.inline_errors(to_trio)
    # sync with trio task
    to_trio.send_nowait(None)
    def push_tradesies(eventkit_obj, obj, fill=None):
        """Push events to trio task.
        """
        if fill is not None:
            # execution details event
            item = ('fill', (obj, fill))
        elif eventkit_obj.name() == 'positionEvent':
            item = ('position', obj)
        else:
            item = ('status', obj)
        log.info(f'eventkit event ->\n{pformat(item)}')
        try:
            to_trio.send_nowait(item)
        except trio.BrokenResourceError:
            log.exception(f'Disconnected from {eventkit_obj} updates')
            eventkit_obj.disconnect(push_tradesies)
    # hook up to the weird eventkit object - event stream api
    for ev_name in [
        'orderStatusEvent',  # all order updates
        'execDetailsEvent',  # all "fill" updates
        'positionEvent',  # avg price updates per symbol per account
        # 'commissionReportEvent',
        # XXX: ugh, it is a separate event from IB and it's
        # emitted as follows:
        # self.ib.commissionReportEvent.emit(trade, fill, report)
        # XXX: not sure yet if we need these
        # 'updatePortfolioEvent',
        # XXX: these all seem to be weird ib_insync intrernal
        # events that we probably don't care that much about
        # given the internal design is wonky af..
        # 'newOrderEvent',
        # 'orderModifyEvent',
        # 'cancelOrderEvent',
        # 'openOrderEvent',
    ]:
        eventkit_obj = getattr(client.ib, ev_name)
        handler = partial(push_tradesies, eventkit_obj)
        eventkit_obj.connect(handler)
    # let the engine run and stream
    await client.ib.disconnectedEvent
@tractor.context
 async def trades_dialogue(
    ctx: tractor.Context,
    loglevel: str = None,
 ) -> AsyncIterator[dict[str, Any]]:
    # XXX: required to propagate ``tractor`` loglevel to piker logging
    get_console_log(loglevel or tractor.current_actor().loglevel)
    accounts_def = config.load_accounts(['ib'])
    global _client_cache
    # deliver positions to subscriber before anything else
    all_positions = []
    accounts = set()
    clients: list[tuple[Client, trio.MemoryReceiveChannel]] = []
    async with (
        trio.open_nursery() as nurse,
        open_client_proxies() as (proxies, aioclients),
    ):
        for account, proxy in proxies.items():
            client = aioclients[account]
            async def open_stream(
                task_status: TaskStatus[
                    trio.abc.ReceiveChannel
                ] = trio.TASK_STATUS_IGNORED,
            ):
                # each api client has a unique event stream
                async with tractor.to_asyncio.open_channel_from(
                    recv_trade_updates,
                    client=client,
                ) as (first, trade_event_stream):
                    task_status.started(trade_event_stream)
                    await trio.sleep_forever()
            trade_event_stream = await nurse.start(open_stream)
            clients.append((client, trade_event_stream))
            assert account in accounts_def
            accounts.add(account)
        for client in aioclients.values():
            for pos in client.positions():
                msg = pack_position(pos)
                msg.account = accounts_def.inverse[msg.account]
                assert msg.account in accounts, (
                    f'Position for unknown account: {msg.account}')
                all_positions.append(msg.dict())
        trades: list[dict] = []
        for proxy in proxies.values():
            trades.append(await proxy.trades())
        log.info(f'Loaded {len(trades)} from this session')
        # TODO: write trades to local ``trades.toml``
        # - use above per-session trades data and write to local file
        # - get the "flex reports" working and pull historical data and
        # also save locally.
        await ctx.started((
            all_positions,
            tuple(name for name in accounts_def if name in accounts),
        ))
        async with (
            ctx.open_stream() as ems_stream,
            trio.open_nursery() as n,
        ):
            # start order request handler **before** local trades event loop
            n.start_soon(handle_order_requests, ems_stream, accounts_def)
            # allocate event relay tasks for each client connection
            for client, stream in clients:
                n.start_soon(
                    deliver_trade_events,
                    stream,
                    ems_stream,
                    accounts_def
                )
            # block until cancelled
            await trio.sleep_forever()
 async def deliver_trade_events(
    trade_event_stream: trio.MemoryReceiveChannel,
    ems_stream: tractor.MsgStream,
    accounts_def: dict[str, str],
 ) -> None:
    '''Format and relay all trade events for a given client to the EMS.
    '''
    action_map = {'BOT': 'buy', 'SLD': 'sell'}
    # TODO: for some reason we can receive a ``None`` here when the
    # ib-gw goes down? Not sure exactly how that's happening looking
    # at the eventkit code above but we should probably handle it...
    async for event_name, item in trade_event_stream:
        log.info(f'ib sending {event_name}:\n{pformat(item)}')
        # TODO: templating the ib statuses in comparison with other
        # brokers is likely the way to go:
        # https://interactivebrokers.github.io/tws-api/interfaceIBApi_1_1EWrapper.html#a17f2a02d6449710b6394d0266a353313
        # short list:
        # - PendingSubmit
        # - PendingCancel
        # - PreSubmitted (simulated orders)
        # - ApiCancelled (cancelled by client before submission
        #                 to routing)
        # - Cancelled
        # - Filled
        # - Inactive (reject or cancelled but not by trader)
        # XXX: here's some other sucky cases from the api
        # - short-sale but securities haven't been located, in this
        #   case we should probably keep the order in some kind of
        #   weird state or cancel it outright?
        # status='PendingSubmit', message=''),
        # status='Cancelled', message='Error 404,
        #   reqId 1550: Order held while securities are located.'),
        # status='PreSubmitted', message='')],
        if event_name == 'status':
            # XXX: begin normalization of nonsense ib_insync internal
            # object-state tracking representations...
            # unwrap needed data from ib_insync internal types
            trade: Trade = item
            status: OrderStatus = trade.orderStatus
            # skip duplicate filled updates - we get the deats
            # from the execution details event
            msg = BrokerdStatus(
                reqid=trade.order.orderId,
                time_ns=time.time_ns(),  # cuz why not
                account=accounts_def.inverse[trade.order.account],
                # everyone doin camel case..
                status=status.status.lower(),  # force lower case
                filled=status.filled,
                reason=status.whyHeld,
                # this seems to not be necessarily up to date in the
                # execDetails event.. so we have to send it here I guess?
                remaining=status.remaining,
                broker_details={'name': 'ib'},
            )
        elif event_name == 'fill':
            # for wtv reason this is a separate event type
            # from IB, not sure why it's needed other then for extra
            # complexity and over-engineering :eyeroll:.
            # we may just end up dropping these events (or
            # translating them to ``Status`` msgs) if we can
            # show the equivalent status events are no more latent.
            # unpack ib_insync types
            # pep-0526 style:
            # https://www.python.org/dev/peps/pep-0526/#global-and-local-variable-annotations
            trade: Trade
            fill: Fill
            trade, fill = item
            execu: Execution = fill.execution
            # TODO: normalize out commissions details?
            details = {
                'contract': asdict(fill.contract),
                'execution': asdict(fill.execution),
                'commissions': asdict(fill.commissionReport),
                'broker_time': execu.time,   # supposedly server fill time
                'name': 'ib',
            }
            msg = BrokerdFill(
                # should match the value returned from `.submit_limit()`
                reqid=execu.orderId,
                time_ns=time.time_ns(),  # cuz why not
                action=action_map[execu.side],
                size=execu.shares,
                price=execu.price,
                broker_details=details,
                # XXX: required by order mode currently
                broker_time=details['broker_time'],
            )
        elif event_name == 'error':
            err: dict = item
            # f$#$% gawd dammit insync..
            con = err['contract']
            if isinstance(con, Contract):
                err['contract'] = asdict(con)
            if err['reqid'] == -1:
                log.error(f'TWS external order error:\n{pformat(err)}')
            # TODO: what schema for this msg if we're going to make it
            # portable across all backends?
            # msg = BrokerdError(**err)
            continue
        elif event_name == 'position':
            msg = pack_position(item)
            msg.account = accounts_def.inverse[msg.account]
        elif event_name == 'event':
            # it's either a general system status event or an external
            # trade event?
            log.info(f"TWS system status: \n{pformat(item)}")
            # TODO: support this again but needs parsing at the callback
            # level...
            # reqid = item.get('reqid', 0)
            # if getattr(msg, 'reqid', 0) < -1:
            # log.info(f"TWS triggered trade\n{pformat(msg.dict())}")
            continue
            # msg.reqid = 'tws-' + str(-1 * reqid)
            # mark msg as from "external system"
            # TODO: probably something better then this.. and start
            # considering multiplayer/group trades tracking
            # msg.broker_details['external_src'] = 'tws'
        # XXX: we always serialize to a dict for msgpack
        # translations, ideally we can move to an msgspec (or other)
        # encoder # that can be enabled in ``tractor`` ahead of
        # time so we can pass through the message types directly.
        await ems_stream.send(msg.dict())
 def load_flex_trades(
    path: Optional[str] = None,
 ) -> dict[str, str]:
    from pprint import pprint
    from ib_insync import flexreport, util
    conf = get_config()
    if not path:
        # load ``brokers.toml`` and try to get the flex
        # token and query id that must be previously defined
        # by the user.
        token = conf.get('flex_token')
        if not token:
            raise ValueError(
               'You must specify a ``flex_token`` field in your'
               '`brokers.toml` in order load your trade log, see our'
               'intructions for how to set this up here:\n'
               'PUT LINK HERE!'
            )
        qid = conf['flex_trades_query_id']
        # TODO: hack this into our logging
        # system like we do with the API client..
        util.logToConsole()
        # TODO: rewrite the query part of this with async..httpx?
        report = flexreport.FlexReport(
            token=token,
            queryId=qid,
        )
    else:
        # XXX: another project we could potentially look at,
        # https://pypi.org/project/ibflex/
        report = flexreport.FlexReport(path=path)
    trade_entries = report.extract('Trade')
    trades = {
        # XXX: LOL apparently ``toml`` has a bug
        # where a section key error will show up in the write
        # if you leave this as an ``int``?
        str(t.__dict__['tradeID']): t.__dict__
        for t in trade_entries
    }
    ln = len(trades)
    log.info(f'Loaded {ln} trades from flex query')
    trades_by_account = {}
    for tid, trade in trades.items():
        trades_by_account.setdefault(
            # oddly for some so-called "BookTrade" entries
            # this field seems to be blank, no cuckin clue.
            # trade['ibExecID']
            str(trade['accountId']), {}
        )[tid] = trade
    section = {'ib': trades_by_account}
    pprint(section)
    # TODO: load the config first and append in
    # the new trades loaded here..
    try:
        config.write(section, 'trades')
    except KeyError:
        import pdbpp; pdbpp.set_trace()  # noqa
 if __name__ == '__main__':
    load_flex_trades()
--- a/piker/brokers/ib/feed.py
+++ b/piker/brokers/ib/feed.py
@ -1,938 +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/>.
 """
 Data feed endpoints pre-wrapped and ready for use with ``tractor``/``trio``.
 """
 from __future__ import annotations
 import asyncio
 from contextlib import asynccontextmanager as acm
 from dataclasses import asdict
 from datetime import datetime
 from math import isnan
 import time
 from typing import (
    Callable,
    Optional,
    Awaitable,
 )
 from async_generator import aclosing
 from fuzzywuzzy import process as fuzzy
 import numpy as np
 import pendulum
 import tractor
 import trio
 from trio_typing import TaskStatus
 from piker.data._sharedmem import ShmArray
 from .._util import SymbolNotFound, NoData
 from .api import (
    _adhoc_futes_set,
    log,
    load_aio_clients,
    ibis,
    MethodProxy,
    open_client_proxies,
    get_preferred_data_client,
    Ticker,
    RequestError,
    Contract,
 )
 # https://interactivebrokers.github.io/tws-api/tick_types.html
 tick_types = {
    77: 'trade',
    # a "utrade" aka an off exchange "unreportable" (dark) vlm:
    # https://interactivebrokers.github.io/tws-api/tick_types.html#rt_volume
    48: 'dark_trade',
    # standard L1 ticks
    0: 'bsize',
    1: 'bid',
    2: 'ask',
    3: 'asize',
    4: 'last',
    5: 'size',
    8: 'volume',
    # ``ib_insync`` already packs these into
    # quotes under the following fields.
    # 55: 'trades_per_min',  # `'tradeRate'`
    # 56: 'vlm_per_min',  # `'volumeRate'`
    # 89: 'shortable',  # `'shortableShares'`
 }
@acm
 async def open_data_client() -> MethodProxy:
    '''
    Open the first found preferred "data client" as defined in the
    user's ``brokers.toml`` in the ``ib.prefer_data_account`` variable
    and deliver that client wrapped in a ``MethodProxy``.
    '''
    async with (
        open_client_proxies() as (proxies, clients),
    ):
        account_name, client = get_preferred_data_client(clients)
        proxy = proxies.get(f'ib.{account_name}')
        if not proxy:
            raise ValueError(
                f'No preferred data client could be found for {account_name}!'
            )
        yield proxy
@acm
 async def open_history_client(
    symbol: str,
 ) -> tuple[Callable, int]:
    '''
    History retreival endpoint - delivers a historical frame callble
    that takes in ``pendulum.datetime`` and returns ``numpy`` arrays.
    '''
    async with open_data_client() as proxy:
        async def get_hist(
            end_dt: Optional[datetime] = None,
            start_dt: Optional[datetime] = None,
        ) -> tuple[np.ndarray, str]:
            out, fails = await get_bars(proxy, symbol, end_dt=end_dt)
            # TODO: add logic here to handle tradable hours and only grab
            # valid bars in the range
            if out is None:
                # could be trying to retreive bars over weekend
                log.error(f"Can't grab bars starting at {end_dt}!?!?")
                raise NoData(
                    f'{end_dt}',
                    frame_size=2000,
                )
            bars, bars_array, first_dt, last_dt = out
            # volume cleaning since there's -ve entries,
            # wood luv to know what crookery that is..
            vlm = bars_array['volume']
            vlm[vlm < 0] = 0
            return bars_array, first_dt, last_dt
        # TODO: it seems like we can do async queries for ohlc
        # but getting the order right still isn't working and I'm not
        # 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': 6}
 _pacing: str = (
    'Historical Market Data Service error '
    'message:Historical data request pacing violation'
 )
 async def get_bars(
    proxy: MethodProxy,
    fqsn: str,
    # blank to start which tells ib to look up the latest datum
    end_dt: str = '',
 ) -> (dict, np.ndarray):
    '''
    Retrieve historical data from a ``trio``-side task using
    a ``MethoProxy``.
    '''
    fails = 0
    bars: Optional[list] = None
    first_dt: datetime = None
    last_dt: datetime = None
    if end_dt:
        last_dt = pendulum.from_timestamp(end_dt.timestamp())
    for _ in range(10):
        try:
            out = await proxy.bars(
                fqsn=fqsn,
                end_dt=end_dt,
            )
            if out:
                bars, bars_array = out
            else:
                await tractor.breakpoint()
            if bars_array is None:
                raise SymbolNotFound(fqsn)
            first_dt = pendulum.from_timestamp(
                bars[0].date.timestamp())
            last_dt = pendulum.from_timestamp(
                bars[-1].date.timestamp())
            time = bars_array['time']
            assert time[-1] == last_dt.timestamp()
            assert time[0] == first_dt.timestamp()
            log.info(
                f'{len(bars)} bars retreived for {first_dt} -> {last_dt}'
            )
            return (bars, bars_array, first_dt, last_dt), fails
        except RequestError as err:
            msg = err.message
            # why do we always need to rebind this?
            # _err = err
            if 'No market data permissions for' in msg:
                # TODO: signalling for no permissions searches
                raise NoData(
                    f'Symbol: {fqsn}',
                )
            elif (
                err.code == 162
                and 'HMDS query returned no data' in err.message
            ):
                # XXX: this is now done in the storage mgmt layer
                # and we shouldn't implicitly decrement the frame dt
                # index since the upper layer may be doing so
                # concurrently and we don't want to be delivering frames
                # that weren't asked for.
                log.warning(
                    f'NO DATA found ending @ {end_dt}\n'
                )
                # try to decrement start point and look further back
                # end_dt = last_dt = last_dt.subtract(seconds=2000)
                raise NoData(
                    f'Symbol: {fqsn}',
                    frame_size=2000,
                )
            elif _pacing in msg:
                log.warning(
                    'History throttle rate reached!\n'
                    'Resetting farms with `ctrl-alt-f` hack\n'
                )
                # TODO: we might have to put a task lock around this
                # method..
                hist_ev = proxy.status_event(
                    'HMDS data farm connection is OK:ushmds'
                )
                # XXX: other event messages we might want to try and
                # wait for but i wasn't able to get any of this
                # reliable..
                # reconnect_start = proxy.status_event(
                #     'Market data farm is connecting:usfuture'
                # )
                # live_ev = proxy.status_event(
                #     'Market data farm connection is OK:usfuture'
                # )
                # try to wait on the reset event(s) to arrive, a timeout
                # will trigger a retry up to 6 times (for now).
                tries: int = 2
                timeout: float = 10
                # try 3 time with a data reset then fail over to
                # a connection reset.
                for i in range(1, tries):
                    log.warning('Sending DATA RESET request')
                    await data_reset_hack(reset_type='data')
                    with trio.move_on_after(timeout) as cs:
                        for name, ev in [
                            # TODO: not sure if waiting on other events
                            # is all that useful here or not. in theory
                            # you could wait on one of the ones above
                            # first to verify the reset request was
                            # sent?
                            ('history', hist_ev),
                        ]:
                            await ev.wait()
                            log.info(f"{name} DATA RESET")
                            break
                    if cs.cancelled_caught:
                        fails += 1
                        log.warning(
                            f'Data reset {name} timeout, retrying {i}.'
                        )
                        continue
                else:
                    log.warning('Sending CONNECTION RESET')
                    await data_reset_hack(reset_type='connection')
                    with trio.move_on_after(timeout) as cs:
                        for name, ev in [
                            # TODO: not sure if waiting on other events
                            # is all that useful here or not. in theory
                            # you could wait on one of the ones above
                            # first to verify the reset request was
                            # sent?
                            ('history', hist_ev),
                        ]:
                            await ev.wait()
                            log.info(f"{name} DATA RESET")
                    if cs.cancelled_caught:
                        fails += 1
                        log.warning('Data CONNECTION RESET timeout!?')
            else:
                raise
    return None, None
    # else:  # throttle wasn't fixed so error out immediately
    #     raise _err
 async def backfill_bars(
    fqsn: str,
    shm: ShmArray,  # type: ignore # noqa
    # TODO: we want to avoid overrunning the underlying shm array buffer
    # and we should probably calc the number of calls to make depending
    # on that until we have the `marketstore` daemon in place in which
    # case the shm size will be driven by user config and available sys
    # memory.
    count: int = 16,
    task_status: TaskStatus[trio.CancelScope] = trio.TASK_STATUS_IGNORED,
 ) -> None:
    '''
    Fill historical bars into shared mem / storage afap.
    TODO: avoid pacing constraints:
    https://github.com/pikers/piker/issues/128
    '''
    # last_dt1 = None
    last_dt = None
    with trio.CancelScope() as cs:
        async with open_data_client() as proxy:
            out, fails = await get_bars(proxy, fqsn)
            if out is None:
                raise RuntimeError("Could not pull currrent history?!")
            (first_bars, bars_array, first_dt, last_dt) = out
            vlm = bars_array['volume']
            vlm[vlm < 0] = 0
            last_dt = first_dt
            # write historical data to buffer
            shm.push(bars_array)
            task_status.started(cs)
            i = 0
            while i < count:
                out, fails = await get_bars(proxy, fqsn, end_dt=first_dt)
                if out is None:
                    # could be trying to retreive bars over weekend
                    # TODO: add logic here to handle tradable hours and
                    # only grab valid bars in the range
                    log.error(f"Can't grab bars starting at {first_dt}!?!?")
                    # XXX: get_bars() should internally decrement dt by
                    # 2k seconds and try again.
                    continue
                (first_bars, bars_array, first_dt, last_dt) = out
                # last_dt1 = last_dt
                # last_dt = first_dt
                # volume cleaning since there's -ve entries,
                # wood luv to know what crookery that is..
                vlm = bars_array['volume']
                vlm[vlm < 0] = 0
                # TODO we should probably dig into forums to see what peeps
                # think this data "means" and then use it as an indicator of
                # sorts? dinkus has mentioned that $vlms for the day dont'
                # match other platforms nor the summary stat tws shows in
                # the monitor - it's probably worth investigating.
                shm.push(bars_array, prepend=True)
                i += 1
 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',
    # 'NEWS': 'news',
 }
 _quote_streams: dict[str, trio.abc.ReceiveStream] = {}
 async def _setup_quote_stream(
    from_trio: asyncio.Queue,
    to_trio: trio.abc.SendChannel,
    symbol: str,
    opts: tuple[int] = (
        '375',  # RT trade volume (excludes utrades)
        '233',  # RT trade volume (includes utrades)
        '236',  # Shortable shares
        # these all appear to only be updated every 25s thus
        # 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
    ),
    contract: Optional[Contract] = None,
 ) -> trio.abc.ReceiveChannel:
    '''
    Stream a ticker using the std L1 api.
    This task is ``asyncio``-side and must be called from
    ``tractor.to_asyncio.open_channel_from()``.
    '''
    global _quote_streams
    to_trio.send_nowait(None)
    async with load_aio_clients() as accts2clients:
        caccount_name, client = get_preferred_data_client(accts2clients)
        contract = contract or (await client.find_contract(symbol))
        ticker: Ticker = client.ib.reqMktData(contract, ','.join(opts))
        # NOTE: it's batch-wise and slow af but I guess could
        # be good for backchecking? Seems to be every 5s maybe?
        # ticker: Ticker = client.ib.reqTickByTickData(
        #     contract, 'Last',
        # )
        # # define a simple queue push routine that streams quote packets
        # # to trio over the ``to_trio`` memory channel.
        # to_trio, from_aio = trio.open_memory_channel(2**8)  # type: ignore
        def teardown():
            ticker.updateEvent.disconnect(push)
            log.error(f"Disconnected stream for `{symbol}`")
            client.ib.cancelMktData(contract)
            # decouple broadcast mem chan
            _quote_streams.pop(symbol, None)
        def push(t: Ticker) -> None:
            """
            Push quotes to trio task.
            """
            # log.debug(t)
            try:
                to_trio.send_nowait(t)
            except (
                trio.BrokenResourceError,
                # XXX: HACK, not sure why this gets left stale (probably
                # due to our terrible ``tractor.to_asyncio``
                # implementation for streams.. but if the mem chan
                # gets left here and starts blocking just kill the feed?
                # trio.WouldBlock,
            ):
                # XXX: eventkit's ``Event.emit()`` for whatever redic
                # reason will catch and ignore regular exceptions
                # resulting in tracebacks spammed to console..
                # Manually do the dereg ourselves.
                teardown()
            except trio.WouldBlock:
                log.warning(
                    f'channel is blocking symbol feed for {symbol}?'
                    f'\n{to_trio.statistics}'
                )
            # except trio.WouldBlock:
            #     # for slow debugging purposes to avoid clobbering prompt
            #     # with log msgs
            #     pass
        ticker.updateEvent.connect(push)
        try:
            await asyncio.sleep(float('inf'))
        finally:
            teardown()
        # return from_aio
@acm
 async def open_aio_quote_stream(
    symbol: str,
    contract: Optional[Contract] = None,
 ) -> trio.abc.ReceiveStream:
    from tractor.trionics import broadcast_receiver
    global _quote_streams
    from_aio = _quote_streams.get(symbol)
    if from_aio:
        # if we already have a cached feed deliver a rx side clone to consumer
        async with broadcast_receiver(
            from_aio,
            2**6,
        ) as from_aio:
            yield from_aio
            return
    async with tractor.to_asyncio.open_channel_from(
        _setup_quote_stream,
        symbol=symbol,
        contract=contract,
    ) as (first, from_aio):
        # cache feed for later consumers
        _quote_streams[symbol] = from_aio
        yield from_aio
 # TODO: cython/mypyc/numba this!
 def normalize(
    ticker: Ticker,
    calc_price: bool = False
 ) -> dict:
    # should be real volume for this contract by default
    calc_price = False
    # check for special contract types
    con = ticker.contract
    if type(con) in (
        ibis.Commodity,
        ibis.Forex,
    ):
        # 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
    else:
        suffix = con.primaryExchange
        if not suffix:
            suffix = con.exchange
        # append a `.<suffix>` to the returned symbol
        # key for derivatives that normally is the expiry
        # date key.
        expiry = con.lastTradeDateOrContractMonth
        if expiry:
            suffix += f'.{expiry}'
    # convert named tuples to dicts so we send usable keys
    new_ticks = []
    for tick in ticker.ticks:
        if tick and not isinstance(tick, dict):
            td = tick._asdict()
            td['type'] = tick_types.get(
                td['tickType'],
                'n/a',
            )
            new_ticks.append(td)
            tbt = ticker.tickByTicks
            if tbt:
                print(f'tickbyticks:\n {ticker.tickByTicks}')
    ticker.ticks = new_ticks
    # some contracts don't have volume so we may want to calculate
    # a midpoint price based on data we can acquire (such as bid / ask)
    if calc_price:
        ticker.ticks.append(
            {'type': 'trade', 'price': ticker.marketPrice()}
        )
    # serialize for transport
    data = asdict(ticker)
    # generate fqsn with possible specialized suffix
    # for derivatives, note the lowercase.
    data['symbol'] = data['fqsn'] = '.'.join(
        (con.symbol, suffix)
    ).lower()
    # convert named tuples to dicts for transport
    tbts = data.get('tickByTicks')
    if tbts:
        data['tickByTicks'] = [tbt._asdict() for tbt in tbts]
    # add time stamps for downstream latency measurements
    data['brokerd_ts'] = time.time()
    # stupid stupid shit...don't even care any more..
    # leave it until we do a proper latency study
    # if ticker.rtTime is not None:
    #     data['broker_ts'] = data['rtTime_s'] = float(
    #         ticker.rtTime.timestamp) / 1000.
    data.pop('rtTime')
    return data
 async def stream_quotes(
    send_chan: trio.abc.SendChannel,
    symbols: list[str],
    feed_is_live: trio.Event,
    loglevel: str = None,
    # startup sync
    task_status: TaskStatus[tuple[dict, dict]] = trio.TASK_STATUS_IGNORED,
 ) -> None:
    '''
    Stream symbol quotes.
    This is a ``trio`` callable routine meant to be invoked
    once the brokerd is up.
    '''
    # TODO: support multiple subscriptions
    sym = symbols[0]
    log.info(f'request for real-time quotes: {sym}')
    async with open_data_client() as proxy:
        con, first_ticker, details = await proxy.get_sym_details(symbol=sym)
        first_quote = normalize(first_ticker)
        # print(f'first quote: {first_quote}')
        def mk_init_msgs() -> dict[str, dict]:
            '''
            Collect a bunch of meta-data useful for feed startup and
            pack in a `dict`-msg.
            '''
            # pass back some symbol info like min_tick, trading_hours, etc.
            syminfo = asdict(details)
            syminfo.update(syminfo['contract'])
            # nested dataclass we probably don't need and that won't IPC
            # serialize
            syminfo.pop('secIdList')
            # TODO: more consistent field translation
            atype = syminfo['asset_type'] = asset_type_map[syminfo['secType']]
            # for stocks it seems TWS reports too small a tick size
            # such that you can't submit orders with that granularity?
            min_tick = 0.01 if atype == 'stock' else 0
            syminfo['price_tick_size'] = max(syminfo['minTick'], min_tick)
            # for "traditional" assets, volume is normally discreet, not
            # a float
            syminfo['lot_tick_size'] = 0.0
            ibclient = proxy._aio_ns.ib.client
            host, port = ibclient.host, ibclient.port
            # TODO: for loop through all symbols passed in
            init_msgs = {
                # pass back token, and bool, signalling if we're the writer
                # and that history has been written
                sym: {
                    'symbol_info': syminfo,
                    'fqsn': first_quote['fqsn'],
                },
                'status': {
                    'data_ep': f'{host}:{port}',
                },
            }
            return init_msgs
        init_msgs = mk_init_msgs()
        # 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.
        with trio.move_on_after(1):
            contract, first_ticker, details = await proxy.get_quote(symbol=sym)
        # it might be outside regular trading hours so see if we can at
        # least grab history.
        if isnan(first_ticker.last):
            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?
            feed_is_live.set()
            # block and let data history backfill code run.
            await trio.sleep_forever()
            return  # we never expect feed to come up?
        async with open_aio_quote_stream(
            symbol=sym,
            contract=con,
        ) as stream:
            # ugh, clear ticks since we've consumed them
            # (ahem, ib_insync is stateful trash)
            first_ticker.ticks = []
            task_status.started((init_msgs, first_quote))
            async with aclosing(stream):
                if type(first_ticker.contract) not in (
                    ibis.Commodity,
                    ibis.Forex
                ):
                    # wait for real volume on feed (trading might be closed)
                    while True:
                        ticker = await stream.receive()
                        # for a real volume contract we rait for the first
                        # "real" trade to take place
                        if (
                            # not calc_price
                            # and not ticker.rtTime
                            not ticker.rtTime
                        ):
                            # spin consuming tickers until we get a real
                            # market datum
                            log.debug(f"New unsent ticker: {ticker}")
                            continue
                        else:
                            log.debug("Received first real volume tick")
                            # ugh, clear ticks since we've consumed them
                            # (ahem, ib_insync is truly stateful trash)
                            ticker.ticks = []
                            # XXX: this works because we don't use
                            # ``aclosing()`` above?
                            break
                    quote = normalize(ticker)
                    log.debug(f"First ticker received {quote}")
                # tell caller quotes are now coming in live
                feed_is_live.set()
                # last = time.time()
                async for ticker in stream:
                    quote = normalize(ticker)
                    await send_chan.send({quote['fqsn']: quote})
                    # ugh, clear ticks since we've consumed them
                    ticker.ticks = []
                    # last = time.time()
 async def data_reset_hack(
    reset_type: str = 'data',
 ) -> None:
    '''
    Run key combos for resetting data feeds and yield back to caller
    when complete.
    This is a linux-only hack around:
    https://interactivebrokers.github.io/tws-api/historical_limitations.html#pacing_violations
    TODOs:
        - a return type that hopefully determines if the hack was
          successful.
        - other OS support?
        - integration with ``ib-gw`` run in docker + Xorg?
    '''
    async def vnc_click_hack(
        reset_type: str = 'data'
    ) -> None:
        '''
        Reset the data or netowork connection for the VNC attached
        ib gateway using magic combos.
        '''
        key = {'data': 'f', 'connection': 'r'}[reset_type]
        import asyncvnc
        async with asyncvnc.connect(
            'localhost',
            port=3003,
            # password='ibcansmbz',
        ) as client:
            # move to middle of screen
            # 640x1800
            client.mouse.move(
                x=500,
                y=500,
            )
            client.mouse.click()
            client.keyboard.press('Ctrl', 'Alt', key)  # keys are stacked
    await tractor.to_asyncio.run_task(vnc_click_hack)
    # we don't really need the ``xdotool`` approach any more B)
    return True
@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_data_client() as proxy:
        async with ctx.open_stream() as stream:
            last = time.time()
            async for pattern in stream:
                log.debug(f'received {pattern}')
                now = time.time()
                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():
                    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.debug(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]):
                    stock_results.extend(await target)
                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)
                    # 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 | {
                    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)
--- a/piker/brokers/kraken.py
+++ b/piker/brokers/kraken.py
@ -20,14 +20,12 @@ Kraken backend.
 '''
 from contextlib import asynccontextmanager as acm
 from dataclasses import asdict, field
-from datetime import datetime
+from typing import Any, Optional, AsyncIterator, Callable
 from pprint import pformat
 from typing import Any, Optional, AsyncIterator, Callable, Union
 import time
 from trio_typing import TaskStatus
 import trio
-import pendulum
+import arrow
 import asks
 from fuzzywuzzy import process as fuzzy
 import numpy as np
@ -35,6 +33,7 @@ import tractor
 from pydantic.dataclasses import dataclass
 from pydantic import BaseModel
 import wsproto
 import itertools
 import urllib.parse
 import hashlib
 import hmac
@ -42,13 +41,7 @@ import base64
 from .. import config
 from .._cacheables import open_cached_client
-from ._util import (
+from ._util import resproc, SymbolNotFound, BrokerError
    resproc,
    SymbolNotFound,
    BrokerError,
    DataThrottle,
    DataUnavailable,
 )
 from ..log import get_logger, get_console_log
 from ..data import ShmArray
 from ..data._web_bs import open_autorecon_ws, NoBsWs
@ -168,6 +161,13 @@ def get_config() -> dict[str, Any]:
    conf, path = config.load()
    section = conf.get('kraken')
    if section:
        log.warning(
            'Kraken order mode is currently disabled due to bug!\n'
            'See https://github.com/pikers/piker/issues/299'
        )
        return {}
    if section is None:
        log.warning(f'No config section found for kraken in {path}')
        return {}
@ -308,38 +308,34 @@ class Client:
    async def submit_limit(
        self,
        oid: str,
        symbol: str,
        price: float,
        action: str,
        size: float,
-        reqid: str = None,
+        reqid: int = None,
-        validate: bool = False  # set True test call without a real submission
+    ) -> int:
    ) -> dict:
        '''
        Place an order and return integer request id provided by client.
        '''
-        # Build common data dict for common keys from both endpoints
+        # Build order data for kraken api
        data = {
            "userref": reqid,
            "ordertype": "limit",
            "type": action,
            "volume": str(size),
            "pair": symbol,
            "price": str(price),
-            "validate": validate
+            # set to True test AddOrder call without a real submission
            "validate": False
        }
-        if reqid is None:
+        return await self.endpoint('AddOrder', data)
            # Build order data for kraken api
            data |= {
                "ordertype": "limit", "type": action, "volume": str(size)
            }
            return await self.endpoint('AddOrder', data)
        else:
            # Edit order data for kraken api
            data["txid"] = reqid
            return await self.endpoint('EditOrder', data)
    async def submit_cancel(
        self,
        reqid: str,
-    ) -> dict:
+    ) -> None:
        '''
        Send cancel request for order id ``reqid``.
@ -399,26 +395,17 @@ class Client:
    async def bars(
        self,
        symbol: str = 'XBTUSD',
        # UTC 2017-07-02 12:53:20
-        since: Optional[Union[int, datetime]] = None,
+        since: int = None,
        count: int = 720,  # <- max allowed per query
        as_np: bool = True,
    ) -> dict:
        if since is None:
-            since = pendulum.now('UTC').start_of('minute').subtract(
+            since = arrow.utcnow().floor('minute').shift(
-                minutes=count).timestamp()
+                minutes=-count).timestamp()
        elif isinstance(since, int):
            since = pendulum.from_timestamp(since).timestamp()
        else:  # presumably a pendulum datetime
            since = since.timestamp()
        # UTC 2017-07-02 12:53:20 is oldest seconds value
-        since = str(max(1499000000, int(since)))
+        since = str(max(1499000000, since))
        json = await self._public(
            'OHLC',
            data={
@ -462,16 +449,7 @@ class Client:
            array = np.array(new_bars, dtype=_ohlc_dtype) if as_np else bars
            return array
        except KeyError:
-            errmsg = json['error'][0]
+            raise SymbolNotFound(json['error'][0] + f': {symbol}')
            if 'not found' in errmsg:
                raise SymbolNotFound(errmsg + f': {symbol}')
            elif 'Too many requests' in errmsg:
                raise DataThrottle(f'{symbol}')
            else:
                raise BrokerError(errmsg)
@acm
@ -568,12 +546,10 @@ async def handle_order_requests(
    request_msg: dict
    order: BrokerdOrder
    userref_counter = itertools.count()
    async for request_msg in ems_order_stream:
-        log.info(
+        log.info(f'Received order request {request_msg}')
            'Received order request:\n'
            f'{pformat(request_msg)}'
        )
        action = request_msg['action']
@ -599,14 +575,17 @@ async def handle_order_requests(
                continue
            # validate
            temp_id = next(userref_counter)
            order = BrokerdOrder(**request_msg)
            # call our client api to submit the order
            resp = await client.submit_limit(
                oid=order.oid,
                symbol=order.symbol,
                price=order.price,
                action=order.action,
                size=order.size,
-                reqid=order.reqid,
+                reqid=temp_id,
            )
            err = resp['error']
@ -617,22 +596,16 @@ async def handle_order_requests(
                await ems_order_stream.send(
                    BrokerdError(
                        oid=order.oid,
-                        reqid=order.reqid,
+                        reqid=temp_id,
                        symbol=order.symbol,
                        reason="Failed order submission",
                        broker_details=resp
                    ).dict()
                )
            else:
-                # TODO: handle multiple orders (cancels?)
+                # TODO: handle multiple cancels
                #       txid is an array of strings
-                if order.reqid is None:
+                reqid = resp['result']['txid'][0]
                    reqid = resp['result']['txid'][0]
                else:
                    # update the internal pairing of oid to krakens
                    # txid with the new txid that is returned on edit
                    reqid = resp['result']['txid']
                # deliver ack that order has been submitted to broker routing
                await ems_order_stream.send(
                    BrokerdOrderAck(
@ -669,9 +642,9 @@ async def handle_order_requests(
                await ems_order_stream.send(
                    BrokerdError(
-                        oid=msg.oid,
+                        oid=order.oid,
-                        reqid=msg.reqid,
+                        reqid=temp_id,
-                        symbol=msg.symbol,
+                        symbol=order.symbol,
                        reason="Failed order cancel",
                        broker_details=resp
                    ).dict()
@ -695,11 +668,9 @@ async def handle_order_requests(
                        await ems_order_stream.send(
                            BrokerdError(
-                                oid=msg.oid,
+                                oid=order.oid,
-                                reqid=msg.reqid,
+                                reqid=temp_id,
-                                symbol=msg.symbol,
+                                symbol=order.symbol,
                                # TODO: maybe figure out if pending
                                # cancels will eventually get cancelled
                                reason="Order cancel is still pending?",
                                broker_details=resp
                            ).dict()
@ -793,10 +764,7 @@ async def trades_dialogue(
        # Get websocket token for authenticated data stream
        # Assert that a token was actually received.
        resp = await client.endpoint('GetWebSocketsToken', {})
        # lol wtf is this..
        assert resp['error'] == []
        token = resp['result']['token']
        async with (
@ -1036,45 +1004,7 @@ async def open_history_client(
    # TODO implement history getter for the new storage layer.
    async with open_cached_client('kraken') as client:
-
+        yield client
        # lol, kraken won't send any more then the "last"
        # 720 1m bars.. so we have to just ignore further
        # requests of this type..
        queries: int = 0
        async def get_ohlc(
            end_dt: Optional[datetime] = None,
            start_dt: Optional[datetime] = None,
        ) -> tuple[
            np.ndarray,
            datetime,  # start
            datetime,  # end
        ]:
            nonlocal queries
            if queries > 0:
                raise DataUnavailable
            count = 0
            while count <= 3:
                try:
                    array = await client.bars(
                        symbol,
                        since=end_dt,
                    )
                    count += 1
                    queries += 1
                    break
                except DataThrottle:
                    log.warning(f'kraken OHLC throttle for {symbol}')
                    await trio.sleep(1)
            start_dt = pendulum.from_timestamp(array[0]['time'])
            end_dt = pendulum.from_timestamp(array[-1]['time'])
            return array, start_dt, end_dt
        yield get_ohlc, {'erlangs': 1, 'rate': 1}
 async def backfill_bars(
--- a/piker/brokers/questrade.py
+++ b/piker/brokers/questrade.py
@ -19,6 +19,7 @@ Questrade API backend.
 """
 from __future__ import annotations
 import inspect
 import contextlib
 import time
 from datetime import datetime
 from functools import partial
@ -31,10 +32,11 @@ from typing import (
    Callable,
 )
-import pendulum
+import arrow
 import trio
 import tractor
 from async_generator import asynccontextmanager
 import pandas as pd
 import numpy as np
 import wrapt
 import asks
@ -44,6 +46,7 @@ from .._cacheables import open_cached_client, async_lifo_cache
 from .. import config
 from ._util import resproc, BrokerError, SymbolNotFound
 from ..log import get_logger, colorize_json, get_console_log
 from . import get_brokermod
 log = get_logger(__name__)
@ -598,16 +601,12 @@ class Client:
        sid = sids[symbol]
        # get last market open end time
-        est_end = now = pendulum.now('UTC').in_timezoe(
+        est_end = now = arrow.utcnow().to('US/Eastern').floor('minute')
            'America/New_York').start_of('minute')
        # on non-paid feeds we can't retreive the first 15 mins
        wd = now.isoweekday()
        if wd > 5:
            quotes = await self.quote([symbol])
-            est_end = pendulum.parse(
+            est_end = arrow.get(quotes[0]['lastTradeTime'])
                quotes[0]['lastTradeTime']
            )
            if est_end.hour == 0:
                # XXX don't bother figuring out extended hours for now
                est_end = est_end.replace(hour=17)
@ -668,7 +667,7 @@ def get_OHLCV(
    """
    del bar['end']
    del bar['VWAP']
-    bar['start'] = pendulum.from_timestamp(bar['start']) / 10**9
+    bar['start'] = pd.Timestamp(bar['start']).value/10**9
    return tuple(bar.values())
--- a/piker/clearing/_ems.py
+++ b/piker/clearing/_ems.py
@ -80,9 +80,7 @@ def mk_check(
        return check_lt
-    raise ValueError(
+    raise ValueError('trigger: {trigger_price}, last: {known_last}')
        f'trigger: {trigger_price}, last: {known_last}'
    )
@dataclass
@ -563,10 +561,7 @@ async def translate_and_relay_brokerd_events(
        name = brokerd_msg['name']
-        log.info(
+        log.info(f'Received broker trade event:\n{pformat(brokerd_msg)}')
            f'Received broker trade event:\n'
            f'{pformat(brokerd_msg)}'
        )
        if name == 'position':
@ -618,28 +613,19 @@ async def translate_and_relay_brokerd_events(
            # packed at submission since we already know it ahead of
            # time
            paper = brokerd_msg['broker_details'].get('paper_info')
            ext = brokerd_msg['broker_details'].get('external')
            if paper:
                # paperboi keeps the ems id up front
                oid = paper['oid']
-            elif ext:
+            else:
                # may be an order msg specified as "external" to the
                # piker ems flow (i.e. generated by some other
                # external broker backend client (like tws for ib)
-                log.error(f"External trade event {ext}")
+                ext = brokerd_msg['broker_details'].get('external')
                if ext:
                    log.error(f"External trade event {ext}")
                continue
            else:
                # something is out of order, we don't have an oid for
                # this broker-side message.
                log.error(
                    'Unknown oid:{oid} for msg:\n'
                    f'{pformat(brokerd_msg)}'
                    'Unable to relay message to client side!?'
                )
        else:
            # check for existing live flow entry
            entry = book._ems_entries.get(oid)
@ -837,9 +823,7 @@ async def process_client_order_cmds(
                if reqid:
                    # send cancel to brokerd immediately!
-                    log.info(
+                    log.info("Submitting cancel for live order {reqid}")
                        f'Submitting cancel for live order {reqid}'
                    )
                    await brokerd_order_stream.send(msg.dict())
--- a/piker/clearing/_messages.py
+++ b/piker/clearing/_messages.py
@ -242,7 +242,7 @@ class BrokerdError(BaseModel):
    # if no brokerd order request was actually submitted (eg. we errored
    # at the ``pikerd`` layer) then there will be ``reqid`` allocated.
-    reqid: Optional[Union[int, str]] = None
+    reqid: Union[int, str] = ''
    symbol: str
    reason: str
--- a/piker/cli/init.py
+++ b/piker/cli/init.py
@ -1,25 +1,7 @@
-# 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/>.
 '''
 CLI commons.
-
+"""
 '''
 import os
 from pprint import pformat
 import click
 import trio
@ -72,22 +54,17 @@ def pikerd(loglevel, host, tl, pdb, tsdb):
            trio.open_nursery() as n,
        ):
            if tsdb:
-                from piker.data._ahab import start_ahab
+                # TODO:
-                from piker.data.marketstore import start_marketstore
+                # async with maybe_open_marketstored():
                from piker.data._ahab import start_ahab
                log.info('Spawning `marketstore` supervisor')
-                ctn_ready, config, (cid, pid) = await n.start(
+                ctn_ready = await n.start(
                    start_ahab,
                    'marketstored',
                    start_marketstore,
                )
                log.info(
                    f'`marketstore` up!\n'
                    f'`marketstored` pid: {pid}\n'
                    f'docker container id: {cid}\n'
                    f'config: {pformat(config)}'
                )
                await ctn_ready.wait()
                log.info('`marketstore` container:{uid} up')
            await trio.sleep_forever()
--- a/piker/config.py
+++ b/piker/config.py
@ -1,5 +1,5 @@
 # piker: trading gear for hackers
-# Copyright (C) 2018-present  Tyler Goodlet (in stewardship for pikers)
+# Copyright (C) 2018-present  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
@ -16,7 +16,6 @@
 """
 Broker configuration mgmt.
 """
 import platform
 import sys
@ -51,7 +50,7 @@ def get_app_dir(app_name, roaming=True, force_posix=False):
    Unix (POSIX):
      ``~/.foo-bar``
    Win XP (roaming):
-      ``C:\Documents and Settings\<user>\Local Settings\Application Data\Foo``
+      ``C:\Documents and Settings\<user>\Local Settings\Application Data\Foo Bar``
    Win XP (not roaming):
      ``C:\Documents and Settings\<user>\Application Data\Foo Bar``
    Win 7 (roaming):
@ -82,8 +81,7 @@ def get_app_dir(app_name, roaming=True, force_posix=False):
            folder = os.path.expanduser("~")
        return os.path.join(folder, app_name)
    if force_posix:
-        return os.path.join(
+        return os.path.join(os.path.expanduser("~/.{}".format(_posixify(app_name))))
            os.path.expanduser("~/.{}".format(_posixify(app_name))))
    if sys.platform == "darwin":
        return os.path.join(
            os.path.expanduser("~/Library/Application Support"), app_name
@ -109,12 +107,7 @@ if _parent_user:
        ]
    )
-_conf_names: set[str] = {
+_file_name = 'brokers.toml'
    'brokers',
    'trades',
    'watchlists',
 }
 _watchlists_data_path = os.path.join(_config_dir, 'watchlists.json')
 _context_defaults = dict(
    default_map={
@ -136,43 +129,23 @@ def _override_config_dir(
    _config_dir = path
-def _conf_fn_w_ext(
+def get_broker_conf_path():
    name: str,
 ) -> str:
    # change this if we ever change the config file format.
    return f'{name}.toml'
 def get_conf_path(
    conf_name: str = 'brokers',
 ) -> str:
    """Return the default config path normally under
    ``~/.config/piker`` on linux.
    Contains files such as:
    - brokers.toml
    - watchlists.toml
    - trades.toml
    # maybe coming soon ;)
    - signals.toml
    - strats.toml
    """
-    assert conf_name in _conf_names
+    return os.path.join(_config_dir, _file_name)
    fn = _conf_fn_w_ext(conf_name)
    return os.path.join(
        _config_dir,
        fn,
    )
 def repodir():
-    '''
+    """Return the abspath to the repo directory.
-    Return the abspath to the repo directory.
+    """
    '''
    dirpath = os.path.abspath(
        # we're 3 levels down in **this** module file
        dirname(dirname(os.path.realpath(__file__)))
@ -181,27 +154,16 @@ def repodir():
 def load(
    conf_name: str = 'brokers',
    path: str = None
 ) -> (dict, str):
-    '''
+    """Load broker config.
-    Load config file by name.
+    """
-
+    path = path or get_broker_conf_path()
    '''
    path = path or get_conf_path(conf_name)
    if not os.path.isfile(path):
-        fn = _conf_fn_w_ext(conf_name)
+        shutil.copyfile(
-
+            os.path.join(repodir(), 'config', 'brokers.toml'),
-        template = os.path.join(
+            path,
            repodir(),
            'config',
            fn
        )
        # try to copy in a template config to the user's directory
        # if one exists.
        if os.path.isfile(template):
            shutil.copyfile(template, path)
    config = toml.load(path)
    log.debug(f"Read config file {path}")
@ -210,17 +172,13 @@ def load(
 def write(
    config: dict,  # toml config as dict
    name: str = 'brokers',
    path: str = None,
 ) -> None:
-    ''''
+    """Write broker config to disk.
    Write broker config to disk.
    Create a ``brokers.ini`` file if one does not exist.
-
+    """
-    '''
+    path = path or get_broker_conf_path()
    path = path or get_conf_path(name)
    dirname = os.path.dirname(path)
    if not os.path.isdir(dirname):
        log.debug(f"Creating config dir {_config_dir}")
@ -230,10 +188,7 @@ def write(
        raise ValueError(
            "Watch out you're trying to write a blank config!")
-    log.debug(
+    log.debug(f"Writing config file {path}")
        f"Writing config `{name}` file to:\n"
        f"{path}"
    )
    with open(path, 'w') as cf:
        return toml.dump(config, cf)
@ -263,5 +218,4 @@ def load_accounts(
    # our default paper engine entry
    accounts['paper'] = None
    return accounts
--- a/piker/data/_ahab.py
+++ b/piker/data/_ahab.py
@ -1,5 +1,5 @@
 # piker: trading gear for hackers
-# Copyright (C) 2018-present  Tyler Goodlet (in stewardship of pikers)
+# Copyright (C) 2018-present  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
@ -19,25 +19,19 @@ Supervisor for docker with included specific-image service helpers.
 '''
 import os
 import time
 from typing import (
    Optional,
-    Callable,
+    # Any,
    Any,
 )
 from contextlib import asynccontextmanager as acm
 import trio
 from trio_typing import TaskStatus
 import tractor
 from tractor.msg import NamespacePath
 import docker
 import json
-from docker.models.containers import Container as DockerContainer
+from docker.models.containers import Container
-from docker.errors import (
+from docker.errors import DockerException, APIError
    DockerException,
    APIError,
 )
 from requests.exceptions import ConnectionError, ReadTimeout
 from ..log import get_logger, get_console_log
@ -46,14 +40,49 @@ from .. import config
 log = get_logger(__name__)
 _config = '''
 # piker's ``marketstore`` config.
 # mount this config using:
 # sudo docker run --mount \
 # type=bind,source="$HOME/.config/piker/",target="/etc" -i -p \
 # 5993:5993 alpacamarkets/marketstore:latest
 root_directory: data
 listen_port: 5993
 grpc_listen_port: 5995
 log_level: debug
 queryable: true
 stop_grace_period: 0
 wal_rotate_interval: 5
 stale_threshold: 5
 enable_add: true
 enable_remove: false
 triggers:
  - module: ondiskagg.so
    on: "*/1Sec/OHLCV"
    config:
        # filter: "nasdaq"
        destinations:
            - 1Min
            - 5Min
            - 15Min
            - 1H
            - 1D
  - module: stream.so
    on: '*/*/*'
    # config:
    #     filter: "nasdaq"
 '''
 class DockerNotStarted(Exception):
    'Prolly you dint start da daemon bruh'
 class ContainerError(RuntimeError):
    'Error reported via app-container logging level'
@acm
 async def open_docker(
    url: Optional[str] = None,
@ -99,216 +128,145 @@ async def open_docker(
    finally:
        if client:
            client.close()
-
+            # client.api._custom_adapter.close()
-
+            for c in client.containers.list():
-class Container:
+                c.kill()
    '''
    Wrapper around a ``docker.models.containers.Container`` to include
    log capture and relay through our native logging system and helper
    method(s) for cancellation/teardown.
    '''
    def __init__(
        self,
        cntr: DockerContainer,
    ) -> None:
        self.cntr = cntr
        # log msg de-duplication
        self.seen_so_far = set()
    async def process_logs_until(
        self,
        patt: str,
        bp_on_msg: bool = False,
    ) -> bool:
        '''
        Attempt to capture container log messages and relay through our
        native logging system.
        '''
        seen_so_far = self.seen_so_far
        while True:
            logs = self.cntr.logs()
            entries = logs.decode().split('\n')
            for entry in entries:
                # ignore null lines
                if not entry:
                    continue
                try:
                    record = json.loads(entry.strip())
                except json.JSONDecodeError:
                    if 'Error' in entry:
                        raise RuntimeError(entry)
                    raise
                msg = record['msg']
                level = record['level']
                if msg and entry not in seen_so_far:
                    seen_so_far.add(entry)
                    if bp_on_msg:
                        await tractor.breakpoint()
                    getattr(log, level, log.error)(f'{msg}')
                    # print(f'level: {level}')
                    if level in ('error', 'fatal'):
                        raise ContainerError(msg)
                if patt in msg:
                    return True
                # do a checkpoint so we don't block if cancelled B)
                await trio.sleep(0.01)
        return False
    def try_signal(
        self,
        signal: str = 'SIGINT',
    ) -> bool:
        try:
            # XXX: market store doesn't seem to shutdown nicely all the
            # time with this (maybe because there are still open grpc
            # connections?) noticably after client connections have been
            # made or are in use/teardown. It works just fine if you
            # just start and stop the container tho?..
            log.cancel(f'SENDING {signal} to {self.cntr.id}')
            self.cntr.kill(signal)
            return True
        except docker.errors.APIError as err:
            if 'is not running' in err.explanation:
                return False
    async def cancel(
        self,
        stop_msg: str,
    ) -> None:
        cid = self.cntr.id
        # first try a graceful cancel
        log.cancel(
            f'SIGINT cancelling container: {cid}\n'
            f'waiting on stop msg: "{stop_msg}"'
        )
        self.try_signal('SIGINT')
        start = time.time()
        for _ in range(30):
            with trio.move_on_after(0.5) as cs:
                cs.shield = True
                await self.process_logs_until(stop_msg)
                # if we aren't cancelled on above checkpoint then we
                # assume we read the expected stop msg and terminated.
                break
            try:
                log.info(f'Polling for container shutdown:\n{cid}')
                if self.cntr.status not in {'exited', 'not-running'}:
                    self.cntr.wait(
                        timeout=0.1,
                        condition='not-running',
                    )
                break
            except (
                ReadTimeout,
            ):
                log.info(f'Still waiting on container:\n{cid}')
                continue
            except (
                docker.errors.APIError,
                ConnectionError,
            ):
                log.exception('Docker connection failure')
                break
        else:
            delay = time.time() - start
            log.error(
                f'Failed to kill container {cid} after {delay}s\n'
                'sending SIGKILL..'
            )
            # get out the big guns, bc apparently marketstore
            # doesn't actually know how to terminate gracefully
            # :eyeroll:...
            self.try_signal('SIGKILL')
            self.cntr.wait(
                timeout=3,
                condition='not-running',
            )
        log.cancel(f'Container stopped: {cid}')
@tractor.context
-async def open_ahabd(
+async def open_marketstored(
    ctx: tractor.Context,
    endpoint: str,  # ns-pointer str-msg-type
    **kwargs,
 ) -> None:
-    get_console_log('info', name=__name__)
+    '''
    Start and supervise a marketstore instance with its config bind-mounted
    in from the piker config directory on the system.
    The equivalent cli cmd to this code is:
        sudo docker run --mount \
        type=bind,source="$HOME/.config/piker/",target="/etc" -i -p \
        5993:5993 alpacamarkets/marketstore:latest
    '''
    log = get_console_log('info', name=__name__)
    async with open_docker() as client:
-        # TODO: eventually offer a config-oriented API to do the mounts,
+        # create a mount from user's local piker config dir into container
-        # params, etc. passing to ``Containter.run()``?
+        config_dir_mnt = docker.types.Mount(
-        # call into endpoint for container config/init
+            target='/etc',
-        ep_func = NamespacePath(endpoint).load_ref()
+            source=config._config_dir,
-        (
+            type='bind',
-            dcntr,
+        )
            cntr_config,
            start_msg,
            stop_msg,
        ) = ep_func(client)
        cntr = Container(dcntr)
-        with trio.move_on_after(1):
+        # create a user config subdir where the marketstore
-            found = await cntr.process_logs_until(start_msg)
+        # backing filesystem database can be persisted.
        persistent_data_dir = os.path.join(
            config._config_dir, 'data',
        )
        if not os.path.isdir(persistent_data_dir):
            os.mkdir(persistent_data_dir)
-            if not found and cntr not in client.containers.list():
+        data_dir_mnt = docker.types.Mount(
-                raise RuntimeError(
+            target='/data',
-                    'Failed to start `marketstore` check logs deats'
+            source=persistent_data_dir,
            type='bind',
        )
        cntr: Container = client.containers.run(
            'alpacamarkets/marketstore:latest',
            # do we need this for cmds?
            # '-i',
            # '-p 5993:5993',
            ports={
                '5993/tcp': 5993,  # jsonrpc
                '5995/tcp': 5995,  # grpc
            },
            mounts=[config_dir_mnt, data_dir_mnt],
            detach=True,
            # stop_signal='SIGINT',
            init=True,
            # remove=True,
        )
        try:
            seen_so_far = set()
            async def process_logs_until(
                match: str,
                bp_on_msg: bool = False,
            ):
                logs = cntr.logs(stream=True)
                for entry in logs:
                    entry = entry.decode()
                    try:
                        record = json.loads(entry.strip())
                    except json.JSONDecodeError:
                        if 'Error' in entry:
                            raise RuntimeError(entry)
                    msg = record['msg']
                    level = record['level']
                    if msg and entry not in seen_so_far:
                        seen_so_far.add(entry)
                        if bp_on_msg:
                            await tractor.breakpoint()
                        getattr(log, level, log.error)(f'{msg}')
                    # if "launching tcp listener for all services..." in msg:
                    if match in msg:
                        return True
                    # do a checkpoint so we don't block if cancelled B)
                    await trio.sleep(0)
                return False
            with trio.move_on_after(0.5):
                found = await process_logs_until(
                    "launching tcp listener for all services...",
                )
-        await ctx.started((
+                if not found and cntr not in client.containers.list():
-            cntr.cntr.id,
+                    raise RuntimeError(
-            os.getpid(),
+                        'Failed to start `marketstore` check logs deats'
-            cntr_config,
+                    )
        ))
-        try:
+            await ctx.started(cntr.id)
-            # TODO: we might eventually want a proxy-style msg-prot here
+            # block for the expected "teardown log msg"..
-            # to allow remote control of containers without needing
+            await process_logs_until('exiting...',)
-            # callers to have root perms?
+
-            await trio.sleep_forever()
+        except (
            BaseException,
            # trio.Cancelled,
            # KeyboardInterrupt,
        ):
            cntr.kill('SIGINT')
            with trio.move_on_after(0.5) as cs:
                cs.shield = True
                await process_logs_until('exiting...',)
            raise
        finally:
-            with trio.CancelScope(shield=True):
+            try:
-                await cntr.cancel(stop_msg)
+                cntr.wait(
                    timeout=0.5,
                    condition='not-running',
                )
            except (
                ReadTimeout,
                ConnectionError,
            ):
                cntr.kill()
 async def start_ahab(
    service_name: str,
-    endpoint: Callable[docker.DockerClient, DockerContainer],
+    task_status: TaskStatus[trio.Event] = trio.TASK_STATUS_IGNORED,
    task_status: TaskStatus[
        tuple[
            trio.Event,
            dict[str, Any],
        ],
    ] = trio.TASK_STATUS_IGNORED,
 ) -> None:
    '''
@ -347,19 +305,14 @@ async def start_ahab(
                    )[2]  # named user's uid
                )
            task_status.started(cn_ready)
            async with portal.open_context(
-                open_ahabd,
+                open_marketstored,
                endpoint=str(NamespacePath.from_ref(endpoint)),
            ) as (ctx, first):
-                cid, pid, cntr_config = first
+                assert str(first)
-
+                # run till cancelled
                task_status.started((
                    cn_ready,
                    cntr_config,
                    (cid, pid),
                ))
                await trio.sleep_forever()
    # since we demoted root perms in this parent
@ -371,6 +324,7 @@ async def start_ahab(
    # TODO: we could also consider adding
    # a ``tractor.ZombieDetected`` or something that we could raise
    # if we find the child didn't terminate.
    # await tractor.breakpoint()
    except PermissionError:
        log.warning('Failed to cancel root permsed container')
@ -383,3 +337,12 @@ async def start_ahab(
                return
        else:
            raise
 async def main():
    await start_ahab()
    await trio.sleep_forever()
 if __name__ == '__main__':
    trio.run(main)
--- a/piker/data/_sampling.py
+++ b/piker/data/_sampling.py
@ -22,7 +22,7 @@ financial data flows.
 from __future__ import annotations
 from collections import Counter
 import time
-from typing import TYPE_CHECKING, Optional, Union
+from typing import TYPE_CHECKING
 import tractor
 import trio
@ -32,7 +32,6 @@ from ..log import get_logger
 if TYPE_CHECKING:
    from ._sharedmem import ShmArray
    from .feed import _FeedsBus
 log = get_logger(__name__)
@ -91,7 +90,6 @@ async def increment_ohlc_buffer(
    total_s = 0  # total seconds counted
    lowest = min(sampler.ohlcv_shms.keys())
    lowest_shm = sampler.ohlcv_shms[lowest][0]
    ad = lowest - 0.001
    with trio.CancelScope() as cs:
@ -135,57 +133,21 @@ async def increment_ohlc_buffer(
                    # write to the buffer
                    shm.push(last)
-            await broadcast(delay_s, shm=lowest_shm)
+            # broadcast the buffer index step to any subscribers for
            # a given sample period.
            subs = sampler.subscribers.get(delay_s, ())
-
+            for stream in subs:
-async def broadcast(
+                try:
-    delay_s: int,
+                    await stream.send({'index': shm._last.value})
-    shm: Optional[ShmArray] = None,
+                except (
-
+                    trio.BrokenResourceError,
-) -> None:
+                    trio.ClosedResourceError
-    '''
+                ):
-    Broadcast the given ``shm: ShmArray``'s buffer index step to any
+                    log.error(
-    subscribers for a given sample period.
+                        f'{stream._ctx.chan.uid} dropped connection'
-
+                    )
-    The sent msg will include the first and last index which slice into
+                    subs.remove(stream)
    the buffer's non-empty data.
    '''
    subs = sampler.subscribers.get(delay_s, ())
    first = last = -1
    if shm is None:
        periods = sampler.ohlcv_shms.keys()
        # if this is an update triggered by a history update there
        # might not actually be any sampling bus setup since there's
        # no "live feed" active yet.
        if periods:
            lowest = min(periods)
            shm = sampler.ohlcv_shms[lowest][0]
            first = shm._first.value
            last = shm._last.value
    for stream in subs:
        try:
            await stream.send({
                'first': first,
                'last': last,
                'index': last,
            })
        except (
            trio.BrokenResourceError,
            trio.ClosedResourceError
        ):
            log.error(
                f'{stream._ctx.chan.uid} dropped connection'
            )
            try:
                subs.remove(stream)
            except ValueError:
                log.warning(
                    f'{stream._ctx.chan.uid} sub already removed!?'
                )
@tractor.context
@ -220,7 +182,7 @@ async def iter_ohlc_periods(
 async def sample_and_broadcast(
-    bus: _FeedsBus,  # noqa
+    bus: '_FeedsBus',  # noqa
    shm: ShmArray,
    quote_stream: trio.abc.ReceiveChannel,
    brokername: str,
@ -299,13 +261,7 @@ async def sample_and_broadcast(
            # end up triggering backpressure which which will
            # eventually block this producer end of the feed and
            # thus other consumers still attached.
-            subs: list[
+            subs = bus._subscribers[broker_symbol.lower()]
                tuple[
                    Union[tractor.MsgStream, trio.MemorySendChannel],
                    tractor.Context,
                    Optional[float],  # tick throttle in Hz
                ]
            ] = bus._subscribers[broker_symbol.lower()]
            # NOTE: by default the broker backend doesn't append
            # it's own "name" into the fqsn schema (but maybe it
@ -314,7 +270,7 @@ async def sample_and_broadcast(
            bsym = f'{broker_symbol}.{brokername}'
            lags: int = 0
-            for (stream, ctx, tick_throttle) in subs:
+            for (stream, tick_throttle) in subs:
                try:
                    with trio.move_on_after(0.2) as cs:
@ -326,41 +282,25 @@ async def sample_and_broadcast(
                                    (bsym, quote)
                                )
                            except trio.WouldBlock:
-                                chan = ctx.chan
+                                ctx = getattr(stream, '_ctx', None)
                                if ctx:
                                    log.warning(
                                        f'Feed overrun {bus.brokername} ->'
-                                        f'{chan.uid} !!!'
+                                        f'{ctx.channel.uid} !!!'
                                    )
                                else:
                                    key = id(stream)
                                    overruns[key] += 1
                                    log.warning(
-                                        f'Feed overrun {broker_symbol}'
+                                        f'Feed overrun {bus.brokername} -> '
                                        '@{bus.brokername} -> '
                                        f'feed @ {tick_throttle} Hz'
                                    )
                                    if overruns[key] > 6:
-                                        # TODO: should we check for the
+                                        log.warning(
-                                        # context being cancelled? this
+                                            f'Dropping consumer {stream}'
-                                        # could happen but the
+                                        )
-                                        # channel-ipc-pipe is still up.
+                                        await stream.aclose()
-                                        if not chan.connected():
+                                        raise trio.BrokenResourceError
                                            log.warning(
                                                'Dropping broken consumer:\n'
                                                f'{broker_symbol}:'
                                                f'{ctx.cid}@{chan.uid}'
                                            )
                                            await stream.aclose()
                                            raise trio.BrokenResourceError
                                        else:
                                            log.warning(
                                                'Feed getting overrun bro!\n'
                                                f'{broker_symbol}:'
                                                f'{ctx.cid}@{chan.uid}'
                                            )
                                            continue
                        else:
                            await stream.send(
                                {bsym: quote}
@ -376,12 +316,11 @@ async def sample_and_broadcast(
                    trio.ClosedResourceError,
                    trio.EndOfChannel,
                ):
-                    chan = ctx.chan
+                    ctx = getattr(stream, '_ctx', None)
                    if ctx:
                        log.warning(
-                            'Dropped `brokerd`-quotes-feed connection:\n'
+                            f'{ctx.chan.uid} dropped  '
-                            f'{broker_symbol}:'
+                            '`brokerd`-quotes-feed connection'
                            f'{ctx.cid}@{chan.uid}'
                        )
                    if tick_throttle:
                        assert stream._closed
@ -394,11 +333,7 @@ async def sample_and_broadcast(
                    try:
                        subs.remove((stream, tick_throttle))
                    except ValueError:
-                        log.error(
+                        log.error(f'{stream} was already removed from subs!?')
                            f'Stream was already removed from subs!?\n'
                            f'{broker_symbol}:'
                            f'{ctx.cid}@{chan.uid}'
                        )
 # TODO: a less naive throttler, here's some snippets:
@ -432,12 +367,7 @@ async def uniform_rate_send(
        if left_to_sleep > 0:
            with trio.move_on_after(left_to_sleep) as cs:
-                try:
+                sym, last_quote = await quote_stream.receive()
                    sym, last_quote = await quote_stream.receive()
                except trio.EndOfChannel:
                    log.exception(f"feed for {stream} ended?")
                    break
                diff = time.time() - last_send
                if not first_quote:
@ -510,7 +440,6 @@ async def uniform_rate_send(
            # if the feed consumer goes down then drop
            # out of this rate limiter
            log.warning(f'{stream} closed')
            await stream.aclose()
            return
        # reset send cycle state
--- a/piker/data/_sharedmem.py
+++ b/piker/data/_sharedmem.py
@ -20,9 +20,9 @@ NumPy compatible shared memory buffers for real-time IPC streaming.
 """
 from __future__ import annotations
 from sys import byteorder
 import time
 from typing import Optional
 from multiprocessing.shared_memory import SharedMemory, _USE_POSIX
 from multiprocessing import resource_tracker as mantracker
 if _USE_POSIX:
    from _posixshmem import shm_unlink
@ -30,7 +30,6 @@ if _USE_POSIX:
 import tractor
 import numpy as np
 from pydantic import BaseModel
 from numpy.lib import recfunctions as rfn
 from ..log import get_logger
 from ._source import base_iohlc_dtype
@ -41,39 +40,32 @@ log = get_logger(__name__)
 # how  much is probably dependent on lifestyle
 _secs_in_day = int(60 * 60 * 24)
-# we try for a buncha times, but only on a run-every-other-day kinda week.
+# we try for 3 times but only on a run-every-other-day kinda week.
-_days_worth = 16
+_default_size = 10 * _secs_in_day
 _default_size = _days_worth * _secs_in_day
 # where to start the new data append index
-_rt_buffer_start = int((_days_worth - 1) * _secs_in_day)
+_rt_buffer_start = int(9*_secs_in_day)
-def cuckoff_mantracker():
+# Tell the "resource tracker" thing to fuck off.
 class ManTracker(mantracker.ResourceTracker):
    def register(self, name, rtype):
        pass
-    from multiprocessing import resource_tracker as mantracker
+    def unregister(self, name, rtype):
        pass
-    # Tell the "resource tracker" thing to fuck off.
+    def ensure_running(self):
-    class ManTracker(mantracker.ResourceTracker):
+        pass
        def register(self, name, rtype):
            pass
        def unregister(self, name, rtype):
            pass
        def ensure_running(self):
            pass
    # "know your land and know your prey"
    # https://www.dailymotion.com/video/x6ozzco
    mantracker._resource_tracker = ManTracker()
    mantracker.register = mantracker._resource_tracker.register
    mantracker.ensure_running = mantracker._resource_tracker.ensure_running
    # ensure_running = mantracker._resource_tracker.ensure_running
    mantracker.unregister = mantracker._resource_tracker.unregister
    mantracker.getfd = mantracker._resource_tracker.getfd
-cuckoff_mantracker()
+# "know your land and know your prey"
 # https://www.dailymotion.com/video/x6ozzco
 mantracker._resource_tracker = ManTracker()
 mantracker.register = mantracker._resource_tracker.register
 mantracker.ensure_running = mantracker._resource_tracker.ensure_running
 ensure_running = mantracker._resource_tracker.ensure_running
 mantracker.unregister = mantracker._resource_tracker.unregister
 mantracker.getfd = mantracker._resource_tracker.getfd
 class SharedInt:
@ -99,12 +91,7 @@ class SharedInt:
        if _USE_POSIX:
            # We manually unlink to bypass all the "resource tracker"
            # nonsense meant for non-SC systems.
-            name = self._shm.name
+            shm_unlink(self._shm.name)
            try:
                shm_unlink(name)
            except FileNotFoundError:
                # might be a teardown race here?
                log.warning(f'Shm for {name} already unlinked?')
 class _Token(BaseModel):
@ -204,11 +191,7 @@ class ShmArray:
        self._post_init: bool = False
        # pushing data does not write the index (aka primary key)
-        dtype = shmarr.dtype
+        self._write_fields = list(shmarr.dtype.fields.keys())[1:]
        if dtype.fields:
            self._write_fields = list(shmarr.dtype.fields.keys())[1:]
        else:
            self._write_fields = None
    # TODO: ringbuf api?
@ -254,48 +237,6 @@ class ShmArray:
        return a
    def ustruct(
        self,
        fields: Optional[list[str]] = None,
        # type that all field values will be cast to
        # in the returned view.
        common_dtype: np.dtype = np.float,
    ) -> np.ndarray:
        array = self._array
        if fields:
            selection = array[fields]
            # fcount = len(fields)
        else:
            selection = array
            # fcount = len(array.dtype.fields)
        # XXX: manual ``.view()`` attempt that also doesn't work.
        # uview = selection.view(
        #     dtype='<f16',
        # ).reshape(-1, 4, order='A')
        # assert len(selection) == len(uview)
        u = rfn.structured_to_unstructured(
            selection,
            # dtype=float,
            copy=True,
        )
        # unstruct = np.ndarray(u.shape, dtype=a.dtype, buffer=shm.buf)
        # array[:] = a[:]
        return u
        # return ShmArray(
        #     shmarr=u,
        #     first=self._first,
        #     last=self._last,
        #     shm=self._shm
        # )
    def last(
        self,
        length: int = 1,
@ -314,7 +255,6 @@ class ShmArray:
        field_map: Optional[dict[str, str]] = None,
        prepend: bool = False,
        update_first: bool = True,
        start: Optional[int] = None,
    ) -> int:
@ -327,9 +267,10 @@ class ShmArray:
        '''
        length = len(data)
        index = start if start is not None else self._last.value
        if prepend:
-            index = (start or self._first.value) - length
+            index = self._first.value - length
            if index < 0:
                raise ValueError(
@ -337,9 +278,6 @@ class ShmArray:
                    f'You have passed {abs(index)} too many datums.'
                )
        else:
            index = start if start is not None else self._last.value
        end = index + length
        if field_map:
@ -357,17 +295,12 @@ class ShmArray:
            # tries to access ``.array`` (which due to the index
            # overlap will be empty). Pretty sure we've fixed it now
            # but leaving this here as a reminder.
-            if prepend and update_first and length:
+            if prepend:
                assert index < self._first.value
-            if (
+            if index < self._first.value:
                index < self._first.value
                and update_first
            ):
                assert prepend, 'prepend=True not passed but index decreased?'
                self._first.value = index
-
+            else:
            elif not prepend:
                self._last.value = end
            self._post_init = True
@ -403,7 +336,6 @@ class ShmArray:
                f"Input array has unknown field(s): {only_in_theirs}"
            )
    # TODO: support "silent" prepends that don't update ._first.value?
    def prepend(
        self,
        data: np.ndarray,
@ -454,11 +386,7 @@ def open_shm_array(
        create=True,
        size=a.nbytes
    )
-    array = np.ndarray(
+    array = np.ndarray(a.shape, dtype=a.dtype, buffer=shm.buf)
        a.shape,
        dtype=a.dtype,
        buffer=shm.buf
    )
    array[:] = a[:]
    array.setflags(write=int(not readonly))
@ -542,26 +470,8 @@ def attach_shm_array(
    if key in _known_tokens:
        assert _Token.from_msg(_known_tokens[key]) == token, "WTF"
    # XXX: ugh, looks like due to the ``shm_open()`` C api we can't
    # actually place files in a subdir, see discussion here:
    # https://stackoverflow.com/a/11103289
    # attach to array buffer and view as per dtype
-    _err: Optional[Exception] = None
+    shm = SharedMemory(name=key)
    for _ in range(3):
        try:
            shm = SharedMemory(
                name=key,
                create=False,
            )
            break
        except OSError as oserr:
            _err = oserr
            time.sleep(0.1)
    else:
        if _err:
            raise _err
    shmarr = np.ndarray(
        (size,),
        dtype=token.dtype,
--- a/piker/data/_source.py
+++ b/piker/data/_source.py
@ -21,7 +21,6 @@ from __future__ import annotations
 from typing import Any
 import decimal
 from bidict import bidict
 import numpy as np
 from pydantic import BaseModel
 # from numba import from_dtype
@ -33,7 +32,7 @@ ohlc_fields = [
    ('high', float),
    ('low', float),
    ('close', float),
-    ('volume', float),
+    ('volume', int),
    ('bar_wap', float),
 ]
@ -48,16 +47,16 @@ base_ohlc_dtype = np.dtype(ohlc_fields)
 # https://github.com/numba/numba/issues/4511
 # numba_ohlc_dtype = from_dtype(base_ohlc_dtype)
-# map time frame "keys" to seconds values
+# map time frame "keys" to minutes values
-tf_in_1s = bidict({
+tf_in_1m = {
-    1: '1s',
+    '1m': 1,
-    60: '1m',
+    '5m':  5,
-    60*5: '5m',
+    '15m': 15,
-    60*15: '15m',
+    '30m':  30,
-    60*30: '30m',
+    '1h': 60,
-    60*60: '1h',
+    '4h': 240,
-    60*60*24: '1d',
+    '1d': 1440,
-})
+}
 def mk_fqsn(
--- a/piker/data/feed.py
+++ b/piker/data/feed.py
@ -22,32 +22,26 @@ This module is enabled for ``brokerd`` daemons.
 """
 from __future__ import annotations
 from dataclasses import dataclass, field
 from datetime import datetime
 from contextlib import asynccontextmanager
 from functools import partial
 from pprint import pformat
 from types import ModuleType
 from typing import (
    Any,
    AsyncIterator, Optional,
    Generator,
    Awaitable,
    TYPE_CHECKING,
    Union,
 )
 import pendulum
 import trio
 from trio.abc import ReceiveChannel
 from trio_typing import TaskStatus
 import trimeter
 import tractor
 from tractor.trionics import maybe_open_context
 from pydantic import BaseModel
 import pendulum
 import numpy as np
 from ..brokers import get_brokermod
-from ..calc import humanize
+from .._cacheables import maybe_open_context
 from ..log import get_logger, get_console_log
 from .._daemon import (
    maybe_spawn_brokerd,
@ -67,19 +61,12 @@ from ._source import (
 from ..ui import _search
 from ._sampling import (
    sampler,
    broadcast,
    increment_ohlc_buffer,
    iter_ohlc_periods,
    sample_and_broadcast,
    uniform_rate_send,
 )
 from ..brokers._util import (
    NoData,
    DataUnavailable,
 )
 if TYPE_CHECKING:
    from .marketstore import Storage
 log = get_logger(__name__)
@ -117,13 +104,7 @@ class _FeedsBus(BaseModel):
    # https://github.com/samuelcolvin/pydantic/issues/2816
    _subscribers: dict[
        str,
-        list[
+        list[tuple[tractor.MsgStream, Optional[float]]]
            tuple[
                Union[tractor.MsgStream, trio.MemorySendChannel],
                tractor.Context,
                Optional[float],  # tick throttle in Hz
            ]
        ]
    ] = {}
    async def start_task(
@ -212,449 +193,20 @@ async def _setup_persistent_brokerd(
        await trio.sleep_forever()
 def diff_history(
    array,
    start_dt,
    end_dt,
    last_tsdb_dt: Optional[datetime] = None
 ) -> np.ndarray:
    to_push = array
    if last_tsdb_dt:
        s_diff = (start_dt - last_tsdb_dt).seconds
        # if we detect a partial frame's worth of data
        # that is new, slice out only that history and
        # write to shm.
        if (
            s_diff < 0
        ):
            if abs(s_diff) < len(array):
                # the + 1 is because ``last_tsdb_dt`` is pulled from
                # the last row entry for the ``'time'`` field retreived
                # from the tsdb.
                to_push = array[abs(s_diff) + 1:]
            else:
                # pass back only the portion of the array that is
                # greater then the last time stamp in the tsdb.
                time = array['time']
                to_push = array[time >= last_tsdb_dt.timestamp()]
            log.info(
                f'Pushing partial frame {to_push.size} to shm'
            )
    return to_push
 async def start_backfill(
    mod: ModuleType,
-    bfqsn: str,
+    fqsn: str,
    shm: ShmArray,
    last_tsdb_dt: Optional[datetime] = None,
    storage: Optional[Storage] = None,
    write_tsdb: bool = True,
    tsdb_is_up: bool = False,
    task_status: TaskStatus[trio.CancelScope] = trio.TASK_STATUS_IGNORED,
 ) -> int:
-    async with mod.open_history_client(bfqsn) as (hist, config):
+    return await mod.backfill_bars(
-
+        fqsn,
-        # get latest query's worth of history all the way
+        shm,
-        # back to what is recorded in the tsdb
+        task_status=task_status,
-        array, start_dt, end_dt = await hist(end_dt=None)
+    )
        times = array['time']
        # sample period step size in seconds
        step_size_s = (
            pendulum.from_timestamp(times[-1])
            - pendulum.from_timestamp(times[-2])
        ).seconds
        # "frame"'s worth of sample period steps in seconds
        frame_size_s = len(array) * step_size_s
        to_push = diff_history(
            array,
            start_dt,
            end_dt,
            last_tsdb_dt=last_tsdb_dt,
        )
        log.info(f'Pushing {to_push.size} to shm!')
        shm.push(to_push)
        for delay_s in sampler.subscribers:
            await broadcast(delay_s)
        # signal that backfilling to tsdb's end datum is complete
        bf_done = trio.Event()
        # let caller unblock and deliver latest history frame
        task_status.started((shm, start_dt, end_dt, bf_done))
        # based on the sample step size, maybe load a certain amount history
        if last_tsdb_dt is None:
            if step_size_s 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 {step_size_s} ' 'seconds.. so ye, dun '
                    'do dat brudder.'
                )
            # when no tsdb "last datum" is provided, we just load
            # some near-term history.
            periods = {
                1: {'days': 1},
                60: {'days': 14},
            }
            if tsdb_is_up:
                # do a decently sized backfill and load it into storage.
                periods = {
                    1: {'days': 6},
                    60: {'years': 2},
                }
            kwargs = periods[step_size_s]
            last_tsdb_dt = start_dt.subtract(**kwargs)
        # configure async query throttling
        erlangs = config.get('erlangs', 1)
        rate = config.get('rate', 1)
        frames = {}
        def iter_dts(start: datetime):
            while True:
                hist_period = pendulum.period(
                    start,
                    last_tsdb_dt,
                )
                dtrange = list(hist_period.range('seconds', frame_size_s))
                log.debug(f'New datetime index:\n{pformat(dtrange)}')
                for end_dt in dtrange:
                    log.info(f'Yielding next frame start {end_dt}')
                    start = yield end_dt
                    # if caller sends a new start date, reset to that
                    if start is not None:
                        log.warning(f'Resetting date range: {start}')
                        break
                else:
                    # from while
                    return
        # pull new history frames until we hit latest
        # already in the tsdb or a max count.
        count = 0
        # NOTE: when gaps are detected in the retreived history (by
        # comparisor of the end - start versus the expected "frame size"
        # in seconds) we need a way to alert the async request code not
        # to continue to query for data "within the gap". This var is
        # set in such cases such that further requests in that period
        # are discarded and further we reset the "datetimem query frame
        # index" in such cases to avoid needless noop requests.
        earliest_end_dt: Optional[datetime] = start_dt
        async def get_ohlc_frame(
            input_end_dt: datetime,
            iter_dts_gen: Generator[datetime],
        ) -> np.ndarray:
            nonlocal count, frames, earliest_end_dt, frame_size_s
            count += 1
            if input_end_dt > earliest_end_dt:
                # if a request comes in for an inter-gap frame we
                # discard it since likely this request is still
                # lingering from before the reset of ``iter_dts()`` via
                # ``.send()`` below.
                log.info(f'Discarding request history ending @ {input_end_dt}')
                # signals to ``trimeter`` loop to discard and
                # ``continue`` in it's schedule loop.
                return None
            try:
                log.info(
                    f'Requesting {step_size_s}s frame ending in {input_end_dt}'
                )
                array, start_dt, end_dt = await hist(end_dt=input_end_dt)
                assert array['time'][0] == start_dt.timestamp()
            except NoData:
                log.warning(
                    f'NO DATA for {frame_size_s}s frame @ {input_end_dt} ?!?'
                )
                return None  # discard signal
            except DataUnavailable as duerr:
                # broker is being a bish and we can't pull
                # any more..
                log.warning('backend halted on data deliver !?!?')
                # 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 duerr
            diff = end_dt - start_dt
            frame_time_diff_s = diff.seconds
            expected_frame_size_s = frame_size_s + step_size_s
            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 indicate to the request loop that this gap is
                # expected by both,
                # - resetting the ``iter_dts()`` generator to start at
                #   the new start point delivered in this result
                # - setting the non-locally scoped ``earliest_end_dt``
                #   to this new value so that the request loop doesn't
                #   get tripped up thinking there's an out of order
                #   request-result condition.
                log.warning(
                    f'History frame ending @ {end_dt} appears to have a gap:\n'
                    f'{diff} ~= {frame_time_diff_s} seconds'
                )
                # reset dtrange gen to new start point
                try:
                    next_end = iter_dts_gen.send(start_dt)
                    log.info(
                        f'Reset frame index to start at {start_dt}\n'
                        f'Was at {next_end}'
                    )
                    # NOTE: manually set "earliest end datetime" index-value
                    # to avoid the request loop getting confused about
                    # new frames that are earlier in history - i.e. this
                    # **is not** the case of out-of-order frames from
                    # an async batch request.
                    earliest_end_dt = start_dt
                except StopIteration:
                    # gen already terminated meaning we probably already
                    # exhausted it via frame requests.
                    log.info(
                        "Datetime index already exhausted, can't reset.."
                    )
            to_push = diff_history(
                array,
                start_dt,
                end_dt,
                last_tsdb_dt=last_tsdb_dt,
            )
            ln = len(to_push)
            if ln:
                log.info(f'{ln} bars for {start_dt} -> {end_dt}')
                frames[input_end_dt.timestamp()] = (to_push, start_dt, end_dt)
                return to_push, start_dt, end_dt
            else:
                log.warning(
                    f'{ln} BARS TO PUSH after diff?!: {start_dt} -> {end_dt}'
                )
                return None
        # initial dt index starts at the start of the first query result
        idts = iter_dts(start_dt)
        async with trimeter.amap(
            partial(
                get_ohlc_frame,
                # we close in the ``iter_dt()`` gen in so we can send
                # reset signals as needed for gap dection in the
                # history.
                iter_dts_gen=idts,
            ),
            idts,
            capture_outcome=True,
            include_value=True,
            # better technical names bruv...
            max_at_once=erlangs,
            max_per_second=rate,
        ) as outcomes:
            # Then iterate over the return values, as they become available
            # (i.e., not necessarily in the original order)
            async for input_end_dt, outcome in outcomes:
                try:
                    out = outcome.unwrap()
                    if out is None:
                        # skip signal
                        continue
                    elif isinstance(out, DataUnavailable):
                        # no data available case signal.. so just kill
                        # further requests and basically just stop
                        # trying...
                        break
                except Exception:
                    log.exception('uhh trimeter bail')
                    raise
                else:
                    to_push, start_dt, end_dt = out
                if not len(to_push):
                    # diff returned no new data (i.e. we probablyl hit
                    # the ``last_tsdb_dt`` point).
                    # TODO: raise instead?
                    log.warning(f'No history for range {start_dt} -> {end_dt}')
                    continue
                # pipeline-style pull frames until we need to wait for
                # the next in order to arrive.
                # i = end_dts.index(input_end_dt)
                # print(f'latest end_dt {end_dt} found at index {i}')
                epochs = list(reversed(sorted(frames)))
                for epoch in epochs:
                    start = shm.array['time'][0]
                    last_shm_prepend_dt = pendulum.from_timestamp(start)
                    earliest_frame_queue_dt = pendulum.from_timestamp(epoch)
                    diff = start - epoch
                    if diff < 0:
                        log.warning(
                            'Discarding out of order frame:\n'
                            f'{earliest_frame_queue_dt}'
                        )
                        frames.pop(epoch)
                        continue
                        # await tractor.breakpoint()
                    if diff > step_size_s:
                        if earliest_end_dt < earliest_frame_queue_dt:
                            # XXX: an expected gap was encountered (see
                            # logic in ``get_ohlc_frame()``, so allow
                            # this frame through to the storage layer.
                            log.warning(
                                f'Expected history gap of {diff}s:\n'
                                f'{earliest_frame_queue_dt} <- '
                                f'{earliest_end_dt}'
                            )
                        elif (
                            erlangs > 1
                        ):
                            # we don't yet have the next frame to push
                            # so break back to the async request loop
                            # while we wait for more async frame-results
                            # to arrive.
                            if len(frames) >= erlangs:
                                log.warning(
                                    'Frame count in async-queue is greater '
                                    'then erlangs?\n'
                                    'There seems to be a gap between:\n'
                                    f'{earliest_frame_queue_dt} <- '
                                    f'{last_shm_prepend_dt}\n'
                                    'Conducting manual call for frame ending: '
                                    f'{last_shm_prepend_dt}'
                                )
                                (
                                    to_push,
                                    start_dt,
                                    end_dt,
                                ) = await get_ohlc_frame(
                                    input_end_dt=last_shm_prepend_dt,
                                    iter_dts_gen=idts,
                                )
                                last_epoch = to_push['time'][-1]
                                diff = start - last_epoch
                                if diff > step_size_s:
                                    await tractor.breakpoint()
                                    raise DataUnavailable(
                                        'An awkward frame was found:\n'
                                        f'{start_dt} -> {end_dt}:\n{to_push}'
                                    )
                                else:
                                    frames[last_epoch] = (
                                        to_push, start_dt, end_dt)
                                    break
                            expect_end = pendulum.from_timestamp(start)
                            expect_start = expect_end.subtract(
                                seconds=frame_size_s)
                            log.warning(
                                'waiting on out-of-order history frame:\n'
                                f'{expect_end - expect_start}'
                            )
                            break
                    to_push, start_dt, end_dt = frames.pop(epoch)
                    ln = len(to_push)
                    # bail gracefully on shm allocation overrun/full condition
                    try:
                        shm.push(to_push, prepend=True)
                    except ValueError:
                        log.info(
                            f'Shm buffer overrun on: {start_dt} -> {end_dt}?'
                        )
                        break
                    log.info(
                        f'Shm pushed {ln} frame:\n'
                        f'{start_dt} -> {end_dt}'
                    )
                    # keep track of most recent "prepended" ``start_dt``
                    # both for detecting gaps and ensuring async
                    # frame-result order.
                    earliest_end_dt = start_dt
                    if (
                        storage is not None
                        and write_tsdb
                    ):
                        log.info(
                            f'Writing {ln} frame to storage:\n'
                            f'{start_dt} -> {end_dt}'
                        )
                        await storage.write_ohlcv(
                            f'{bfqsn}.{mod.name}',  # lul..
                            to_push,
                        )
                # TODO: can we only trigger this if the respective
                # history in "in view"?!?
                # 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...
                for delay_s in sampler.subscribers:
                    await broadcast(delay_s)
        bf_done.set()
 async def manage_history(
@ -698,154 +250,115 @@ async def manage_history(
    # for now only do backfilling if no tsdb can be found
    do_legacy_backfill = not is_up and opened
    bfqsn = fqsn.replace('.' + mod.name, '')
    open_history_client = getattr(mod, 'open_history_client', None)
    assert open_history_client
    if is_up and opened and open_history_client:
        log.info('Found existing `marketstored`')
        from . import marketstore
        async with marketstore.open_storage_client(
            fqsn,
        ) as storage:
-            # TODO: this should be used verbatim for the pure
+            tsdb_arrays = await storage.read_ohlcv(fqsn)
            # shm backfiller approach below.
-            # start history anal and load missing new data via backend.
+            if not tsdb_arrays:
-            series, _, last_tsdb_dt = await storage.load(fqsn)
+                do_legacy_backfill = True
            broker, symbol, expiry = unpack_fqsn(fqsn)
            (
                shm,
                latest_start_dt,
                latest_end_dt,
                bf_done,
            ) = await bus.nursery.start(
                partial(
                    start_backfill,
                    mod,
                    bfqsn,
                    shm,
                    last_tsdb_dt=last_tsdb_dt,
                    tsdb_is_up=True,
                    storage=storage,
                )
            )
            # if len(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.
            task_status.started(shm)
            some_data_ready.set()
            await bf_done.wait()
            # do diff against last start frame of history and only fill
            # in from the tsdb an allotment that allows for most recent
            # to be loaded into mem *before* tsdb data.
            if last_tsdb_dt:
                dt_diff_s = (
                    latest_start_dt - last_tsdb_dt
                ).seconds
            else:
-                dt_diff_s = 0
+                log.info(f'Loaded tsdb history {tsdb_arrays}')
-            # await trio.sleep_forever()
+                fastest = list(tsdb_arrays.values())[0]
-            # TODO: see if there's faster multi-field reads:
+                times = fastest['Epoch']
-            # https://numpy.org/doc/stable/user/basics.rec.html#accessing-multiple-fields
+                first, last = times[0], times[-1]
-            # re-index  with a `time` and index field
+                first_tsdb_dt, last_tsdb_dt = map(
-            prepend_start = shm._first.value
+                    pendulum.from_timestamp, [first, last]
            # sanity check on most-recent-data loading
            assert prepend_start > dt_diff_s
            history = list(series.values())
            if history:
                fastest = history[0]
                to_push = fastest[: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=marketstore.ohlc_key_map,
                )
-                # load as much from storage into shm as space will
+                # TODO: this should be used verbatim for the pure
-                # allow according to user's shm size settings.
+                # shm backfiller approach below.
                count = 0
                end = fastest['Epoch'][0]
-                while shm._first.value > 0:
+                def diff_history(
-                    count += 1
+                    array,
-                    series = await storage.read_ohlcv(
+                    start_dt,
-                        fqsn,
+                    end_dt,
                        end=end,
                    )
                    history = list(series.values())
                    fastest = history[0]
                    end = fastest['Epoch'][0]
                    prepend_start -= len(to_push)
                    to_push = fastest[:prepend_start]
                ) -> np.ndarray:
                    s_diff = (last_tsdb_dt - start_dt).seconds
                    # if we detect a partial frame's worth of data
                    # that is new, slice out only that history and
                    # write to shm.
                    if s_diff > 0:
                        assert last_tsdb_dt > start_dt
                        selected = array['time'] > last_tsdb_dt.timestamp()
                        to_push = array[selected]
                        log.info(
                            f'Pushing partial frame {to_push.size} to shm'
                        )
                        return to_push
                    else:
                        return array
                # start history anal and load missing new data via backend.
                async with open_history_client(fqsn) as hist:
                    # get latest query's worth of history all the way
                    # back to what is recorded in the tsdb
                    array, start_dt, end_dt = await hist(end_dt='')
                    to_push = diff_history(array, start_dt, end_dt)
                    shm.push(to_push)
                    # let caller unblock and deliver latest history frame
                    task_status.started(shm)
                    some_data_ready.set()
                    # pull new history frames until we hit latest
                    # already in the tsdb
                    while start_dt > last_tsdb_dt:
                        array, start_dt, end_dt = await hist(end_dt=start_dt)
                        to_push = diff_history(array, start_dt, end_dt)
                        shm.push(to_push, prepend=True)
                    # 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
                    shm.push(
-                        to_push,
+                        fastest[-shm._first.value:],
                        # 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=shm._len - _secs_in_day,
-                        # start=prepend_start,
+                        field_map={
-                        field_map=marketstore.ohlc_key_map,
+                            'Epoch': 'time',
                            'Open': 'open',
                            'High': 'high',
                            'Low': 'low',
                            'Close': 'close',
                            'Volume': 'volume',
                        },
                    )
-                    # manually trigger step update to update charts/fsps
+                    # TODO: write new data to tsdb to be ready to for next
-                    # which need an incremental update.
+                    # read.
                    # 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.
                    for delay_s in sampler.subscribers:
                        await broadcast(delay_s)
                    if count > 6:
                        break
                log.info(f'Loaded {to_push.shape} datums from storage')
                # TODO: write new data to tsdb to be ready to for next read.
    if do_legacy_backfill:
        # do a legacy incremental backfill from the provider.
        log.info('No existing `marketstored` found..')
        bfqsn = fqsn.replace('.' + mod.name, '')
        # start history backfill task ``backfill_bars()`` is
        # a required backend func this must block until shm is
        # filled with first set of ohlc bars
        await bus.nursery.start(
-            partial(
+            start_backfill,
-                start_backfill,
+            mod,
-                mod,
+            bfqsn,
-                bfqsn,
+            shm,
                shm,
            )
        )
        # yield back after client connect with filled shm
@ -1021,7 +534,7 @@ async def open_feed_bus(
    brokername: str,
    symbol: str,  # normally expected to the broker-specific fqsn
    loglevel: str,
-    tick_throttle: Optional[float] = None,
+    tick_throttle:  Optional[float] = None,
    start_stream: bool = True,
 ) -> None:
@ -1125,10 +638,10 @@ async def open_feed_bus(
                recv,
                stream,
            )
-            sub = (send, ctx, tick_throttle)
+            sub = (send, tick_throttle)
        else:
-            sub = (stream, ctx, tick_throttle)
+            sub = (stream, tick_throttle)
        subs = bus._subscribers[bfqsn]
        subs.append(sub)
@ -1211,10 +724,10 @@ class Feed:
    shm: ShmArray
    mod: ModuleType
    first_quotes: dict  # symbol names to first quote dicts
    _portal: tractor.Portal
    stream: trio.abc.ReceiveChannel[dict[str, Any]]
    status: dict[str, Any]
    _portal: tractor.Portal
    stream: trio.abc.ReceiveChannel[dict[str, Any]]
    throttle_rate: Optional[int] = None
    _trade_stream: Optional[AsyncIterator[dict[str, Any]]] = None
@ -1282,7 +795,7 @@ async def install_brokerd_search(
                # a backend module?
                pause_period=getattr(
                    brokermod, '_search_conf', {}
-                ).get('pause_period', 0.0616),
+                    ).get('pause_period', 0.0616),
            ):
                yield
@ -1355,24 +868,9 @@ async def open_feed(
            first_quotes=first_quotes,
            stream=stream,
            _portal=portal,
            status={},
            throttle_rate=tick_throttle,
        )
        # fill out "status info" that the UI can show
        host, port = feed.portal.channel.raddr
        if host == '127.0.0.1':
            host = 'localhost'
        feed.status.update({
            'actor_name': feed.portal.channel.uid[0],
            'host': host,
            'port': port,
            'shm': f'{humanize(feed.shm._shm.size)}',
            'throttle_rate': feed.throttle_rate,
        })
        feed.status.update(init_msg.pop('status', {}))
        for sym, data in init_msg.items():
            si = data['symbol_info']
            fqsn = data['fqsn'] + f'.{brokername}'
--- a/piker/data/marketstore.py
+++ b/piker/data/marketstore.py
@ -23,15 +23,12 @@
 - todo: tick sequence stream-cloning for testing
 '''
 from __future__ import annotations
 from contextlib import asynccontextmanager as acm
 from datetime import datetime
 from pprint import pformat
 from typing import (
    Any,
    Optional,
    Union,
    TYPE_CHECKING,
 )
 import time
 from math import isnan
@ -40,6 +37,7 @@ from bidict import bidict
 import msgpack
 import pyqtgraph as pg
 import numpy as np
 import pandas as pd
 import tractor
 from trio_websocket import open_websocket_url
 from anyio_marketstore import (
@ -47,154 +45,14 @@ from anyio_marketstore import (
    MarketstoreClient,
    Params,
 )
 import pendulum
 import purerpc
 if TYPE_CHECKING:
    import docker
    from ._ahab import DockerContainer
 from .feed import maybe_open_feed
 from ..log import get_logger, get_console_log
 log = get_logger(__name__)
 # container level config
 _config = {
    'grpc_listen_port': 5995,
    'ws_listen_port': 5993,
    'log_level': 'debug',
 }
 _yaml_config = '''
 # piker's ``marketstore`` config.
 # mount this config using:
 # sudo docker run --mount \
 # type=bind,source="$HOME/.config/piker/",target="/etc" -i -p \
 # 5993:5993 alpacamarkets/marketstore:latest
 root_directory: data
 listen_port: {ws_listen_port}
 grpc_listen_port: {grpc_listen_port}
 log_level: {log_level}
 queryable: true
 stop_grace_period: 0
 wal_rotate_interval: 5
 stale_threshold: 5
 enable_add: true
 enable_remove: false
 triggers:
  - module: ondiskagg.so
    on: "*/1Sec/OHLCV"
    config:
        # filter: "nasdaq"
        destinations:
            - 1Min
            - 5Min
            - 15Min
            - 1H
            - 1D
  - module: stream.so
    on: '*/*/*'
    # config:
    #     filter: "nasdaq"
 '''.format(**_config)
 def start_marketstore(
    client: docker.DockerClient,
    **kwargs,
 ) -> tuple[DockerContainer, dict[str, Any]]:
    '''
    Start and supervise a marketstore instance with its config bind-mounted
    in from the piker config directory on the system.
    The equivalent cli cmd to this code is:
        sudo docker run --mount \
        type=bind,source="$HOME/.config/piker/",target="/etc" -i -p \
        5993:5993 alpacamarkets/marketstore:latest
    '''
    import os
    import docker
    from .. import config
    get_console_log('info', name=__name__)
    mktsdir = os.path.join(config._config_dir, 'marketstore')
    # create when dne
    if not os.path.isdir(mktsdir):
        os.mkdir(mktsdir)
    yml_file = os.path.join(mktsdir, 'mkts.yml')
    if not os.path.isfile(yml_file):
        log.warning(
            f'No `marketstore` config exists?: {yml_file}\n'
            'Generating new file from template:\n'
            f'{_yaml_config}\n'
        )
        with open(yml_file, 'w') as yf:
            yf.write(_yaml_config)
    # create a mount from user's local piker config dir into container
    config_dir_mnt = docker.types.Mount(
        target='/etc',
        source=mktsdir,
        type='bind',
    )
    # create a user config subdir where the marketstore
    # backing filesystem database can be persisted.
    persistent_data_dir = os.path.join(
        mktsdir, 'data',
    )
    if not os.path.isdir(persistent_data_dir):
        os.mkdir(persistent_data_dir)
    data_dir_mnt = docker.types.Mount(
        target='/data',
        source=persistent_data_dir,
        type='bind',
    )
    dcntr: DockerContainer = client.containers.run(
        'alpacamarkets/marketstore:latest',
        # do we need this for cmds?
        # '-i',
        # '-p 5993:5993',
        ports={
            '5993/tcp': 5993,  # jsonrpc / ws?
            '5995/tcp': 5995,  # grpc
        },
        mounts=[
            config_dir_mnt,
            data_dir_mnt,
        ],
        detach=True,
        # stop_signal='SIGINT',
        init=True,
        # remove=True,
    )
    return (
        dcntr,
        _config,
        # expected startup and stop msgs
        "launching tcp listener for all services...",
        "exiting...",
    )
 _tick_tbk_ids: tuple[str, str] = ('1Sec', 'TICK')
 _tick_tbk: str = '{}/' + '/'.join(_tick_tbk_ids)
@ -242,22 +100,12 @@ _ohlcv_dt = [
    # ohlcv sampling
    ('Open', 'f4'),
    ('High', 'f4'),
-    ('Low', 'f4'),
+    ('Low', 'i8'),
-    ('Close', 'f4'),
+    ('Close', 'i8'),
    ('Volume', 'f4'),
 ]
 ohlc_key_map = bidict({
    'Epoch': 'time',
    'Open': 'open',
    'High': 'high',
    'Low': 'low',
    'Close': 'close',
    'Volume': 'volume',
 })
 def mk_tbk(keys: tuple[str, str, str]) -> str:
    '''
    Generate a marketstore table key from a tuple.
@ -279,7 +127,7 @@ def quote_to_marketstore_structarray(
    '''
    if last_fill:
        # new fill bby
-        now = int(pendulum.parse(last_fill).timestamp)
+        now = timestamp(last_fill)
    else:
        # this should get inserted upstream by the broker-client to
        # subtract from IPC latency
@ -309,6 +157,15 @@ def quote_to_marketstore_structarray(
    return np.array([tuple(array_input)], dtype=_quote_dt)
 def timestamp(date, **kwargs) -> int:
    '''
    Return marketstore compatible 'Epoch' integer in nanoseconds
    from a date formatted str.
    '''
    return int(pd.Timestamp(date, **kwargs).value)
@acm
 async def get_client(
    host: str = 'localhost',
@ -342,7 +199,6 @@ class MarketStoreError(Exception):
 #                 raise MarketStoreError(err)
 # map of seconds ints to "time frame" accepted keys
 tf_in_1s = bidict({
    1: '1Sec',
    60: '1Min',
@ -384,41 +240,13 @@ class Storage:
    async def write_ticks(self, ticks: list) -> None:
        ...
-    async def load(
+    async def write_ohlcv(self, ohlcv: np.ndarray) -> None:
-        self,
+        ...
        fqsn: str,
    ) -> tuple[
        dict[int, np.ndarray],  # timeframe (in secs) to series
        Optional[datetime],  # first dt
        Optional[datetime],  # last dt
    ]:
        first_tsdb_dt, last_tsdb_dt = None, None
        tsdb_arrays = await self.read_ohlcv(
            fqsn,
            # on first load we don't need to pull the max
            # history per request size worth.
            limit=3000,
        )
        log.info(f'Loaded tsdb history {tsdb_arrays}')
        if tsdb_arrays:
            fastest = list(tsdb_arrays.values())[0]
            times = fastest['Epoch']
            first, last = times[0], times[-1]
            first_tsdb_dt, last_tsdb_dt = map(
                pendulum.from_timestamp, [first, last]
            )
        return tsdb_arrays, first_tsdb_dt, last_tsdb_dt
    async def read_ohlcv(
        self,
        fqsn: str,
        timeframe: Optional[Union[int, str]] = None,
        end: Optional[int] = None,
        limit: int = int(800e3),
    ) -> tuple[
        MarketstoreClient,
@ -436,12 +264,11 @@ class Storage:
            symbols=fqsn,
            timeframe=tfstr,
            attrgroup='OHLCV',
            end=end,
            # limit_from_start=True,
            # TODO: figure the max limit here given the
            # ``purepc`` msg size limit of purerpc: 33554432
-            limit=limit,
+            limit=int(800e3),
        )
        if timeframe is None:
@ -492,90 +319,6 @@ class Storage:
        return await client.destroy(tbk=key)
    async def write_ohlcv(
        self,
        fqsn: str,
        ohlcv: np.ndarray,
        append_and_duplicate: bool = True,
        limit: int = int(800e3),
    ) -> None:
        # build mkts schema compat array for writing
        mkts_dt = np.dtype(_ohlcv_dt)
        mkts_array = np.zeros(
            len(ohlcv),
            dtype=mkts_dt,
        )
        # copy from shm array (yes it's this easy):
        # https://numpy.org/doc/stable/user/basics.rec.html#assignment-from-other-structured-arrays
        mkts_array[:] = ohlcv[[
            'time',
            'open',
            'high',
            'low',
            'close',
            'volume',
        ]]
        m, r = divmod(len(mkts_array), limit)
        for i in range(m, 1):
            to_push = mkts_array[i-1:i*limit]
            # write to db
            resp = await self.client.write(
                to_push,
                tbk=f'{fqsn}/1Sec/OHLCV',
                # NOTE: will will append duplicates
                # for the same timestamp-index.
                # TODO: pre deduplicate?
                isvariablelength=append_and_duplicate,
            )
            log.info(
                f'Wrote {mkts_array.size} datums to tsdb\n'
            )
            for resp in resp.responses:
                err = resp.error
                if err:
                    raise MarketStoreError(err)
        if r:
            to_push = mkts_array[m*limit:]
            # write to db
            resp = await self.client.write(
                to_push,
                tbk=f'{fqsn}/1Sec/OHLCV',
                # NOTE: will will append duplicates
                # for the same timestamp-index.
                # TODO: pre deduplicate?
                isvariablelength=append_and_duplicate,
            )
            log.info(
                f'Wrote {mkts_array.size} datums to tsdb\n'
            )
            for resp in resp.responses:
                err = resp.error
                if err:
                    raise MarketStoreError(err)
    # XXX: currently the only way to do this is through the CLI:
    # sudo ./marketstore connect --dir ~/.config/piker/data
    # >> \show mnq.globex.20220617.ib/1Sec/OHLCV 2022-05-15
    # and this seems to block and use up mem..
    # >> \trim mnq.globex.20220617.ib/1Sec/OHLCV 2022-05-15
    # relevant source code for this is here:
    # https://github.com/alpacahq/marketstore/blob/master/cmd/connect/session/trim.go#L14
    # def delete_range(self, start_dt, end_dt) -> None:
    #     ...
@acm
 async def open_storage_client(
@ -596,7 +339,7 @@ async def open_storage_client(
 async def tsdb_history_update(
-    fqsn: Optional[str] = None,
+    fqsn: str,
 ) -> list[str]:
@ -642,51 +385,80 @@ async def tsdb_history_update(
    ):
        profiler(f'opened feed for {fqsn}')
-        to_append = feed.shm.array
+        symbol = feed.symbols.get(fqsn)
-        to_prepend = None
+        if symbol:
-
+            fqsn = symbol.front_fqsn()
        if fqsn:
            symbol = feed.symbols.get(fqsn)
            if symbol:
                fqsn = symbol.front_fqsn()
            # diff db history with shm and only write the missing portions
            ohlcv = feed.shm.array
            # TODO: use pg profiler
            tsdb_arrays = await storage.read_ohlcv(fqsn)
            # hist diffing
            if tsdb_arrays:
                for secs in (1, 60):
                    ts = tsdb_arrays.get(secs)
                    if ts is not None and len(ts):
                        # these aren't currently used but can be referenced from
                        # within the embedded ipython shell below.
                        to_append = ohlcv[ohlcv['time'] > ts['Epoch'][-1]]
                        to_prepend = ohlcv[ohlcv['time'] < ts['Epoch'][0]]
            profiler('Finished db arrays diffs')
        syms = await storage.client.list_symbols()
        log.info(f'Existing tsdb symbol set:\n{pformat(syms)}')
        profiler(f'listed symbols {syms}')
-        # TODO: ask if user wants to write history for detected
+        # diff db history with shm and only write the missing portions
-        # available shm buffers?
+        ohlcv = feed.shm.array
        # TODO: use pg profiler
        tsdb_arrays = await storage.read_ohlcv(fqsn)
        to_append = feed.shm.array
        to_prepend = None
        # hist diffing
        if tsdb_arrays:
            onesec = tsdb_arrays[1]
            to_append = ohlcv[ohlcv['time'] > onesec['Epoch'][-1]]
            to_prepend = ohlcv[ohlcv['time'] < onesec['Epoch'][0]]
        profiler('Finished db arrays diffs')
        for array in [to_append, to_prepend]:
            if array is None:
                continue
            log.info(
                f'Writing datums {array.size} -> to tsdb from shm\n'
            )
            # build mkts schema compat array for writing
            mkts_dt = np.dtype(_ohlcv_dt)
            mkts_array = np.zeros(
                len(array),
                dtype=mkts_dt,
            )
            # copy from shm array (yes it's this easy):
            # https://numpy.org/doc/stable/user/basics.rec.html#assignment-from-other-structured-arrays
            mkts_array[:] = array[[
                'time',
                'open',
                'high',
                'low',
                'close',
                'volume',
            ]]
            # write to db
            resp = await storage.client.write(
                mkts_array,
                tbk=f'{fqsn}/1Sec/OHLCV',
                # NOTE: will will append duplicates
                # for the same timestamp-index.
                # TODO: pre deduplicate?
                isvariablelength=True,
            )
            log.info(
                f'Wrote {to_append.size} datums to tsdb\n'
            )
            profiler('Finished db writes')
            for resp in resp.responses:
                err = resp.error
                if err:
                    raise MarketStoreError(err)
        from tractor.trionics import ipython_embed
        await ipython_embed()
        # for array in [to_append, to_prepend]:
        #     if array is None:
        #         continue
        #     log.info(
        #         f'Writing datums {array.size} -> to tsdb from shm\n'
        #     )
        #     await storage.write_ohlcv(fqsn, array)
        # profiler('Finished db writes')
 async def ingest_quote_stream(
    symbols: list[str],
--- a/piker/fsp/_engine.py
+++ b/piker/fsp/_engine.py
@ -361,7 +361,7 @@ async def cascade(
                ) -> tuple[TaskTracker, int]:
                    # TODO: adopt an incremental update engine/approach
                    # where possible here eventually!
-                    log.debug(f're-syncing fsp {func_name} to source')
+                    log.warning(f're-syncing fsp {func_name} to source')
                    tracker.cs.cancel()
                    await tracker.complete.wait()
                    tracker, index = await n.start(fsp_target)
@ -369,12 +369,7 @@ async def cascade(
                    # always trigger UI refresh after history update,
                    # see ``piker.ui._fsp.FspAdmin.open_chain()`` and
                    # ``piker.ui._display.trigger_update()``.
-                    await client_stream.send({
+                    await client_stream.send('update')
                        'fsp_update': {
                            'key': dst_shm_token,
                            'first': dst._first.value,
                            'last': dst._last.value,
                    }})
                    return tracker, index
                def is_synced(
--- a/piker/fsp/_momo.py
+++ b/piker/fsp/_momo.py
@ -167,7 +167,6 @@ def _wma(
    assert length == len(weights)
    # lol, for long sequences this is nutso slow and expensive..
    return np.convolve(signal, weights, 'valid')
--- a/piker/fsp/_volume.py
+++ b/piker/fsp/_volume.py
@ -309,7 +309,7 @@ async def flow_rates(
        if period > 1:
            trade_rate_wma = _wma(
-                dvlm_shm.array['trade_count'][-period:],
+                dvlm_shm.array['trade_count'],
                period,
                weights=weights,
            )
@ -332,7 +332,7 @@ async def flow_rates(
        if period > 1:
            dark_trade_rate_wma = _wma(
-                dvlm_shm.array['dark_trade_count'][-period:],
+                dvlm_shm.array['dark_trade_count'],
                period,
                weights=weights,
            )
--- a/piker/log.py
+++ b/piker/log.py
@ -25,13 +25,10 @@ from pygments import highlight, lexers, formatters
 # Makes it so we only see the full module name when using ``__name__``
 # without the extra "piker." prefix.
-_proj_name: str = 'piker'
+_proj_name = 'piker'
-def get_logger(
+def get_logger(name: str = None) -> logging.Logger:
    name: str = None,
 ) -> logging.Logger:
    '''Return the package log or a sub-log for `name` if provided.
    '''
    return tractor.log.get_logger(name=name, _root_name=_proj_name)
--- a/piker/ui/_axes.py
+++ b/piker/ui/_axes.py
@ -19,10 +19,10 @@ Chart axes graphics and behavior.
 """
 from functools import lru_cache
-from typing import Optional, Callable
+from typing import List, Tuple, Optional, Callable
 from math import floor
-import numpy as np
+import pandas as pd
 import pyqtgraph as pg
 from PyQt5 import QtCore, QtGui, QtWidgets
 from PyQt5.QtCore import QPointF
@ -103,7 +103,7 @@ class Axis(pg.AxisItem):
    def size_to_values(self) -> None:
        pass
-    def txt_offsets(self) -> tuple[int, int]:
+    def txt_offsets(self) -> Tuple[int, int]:
        return tuple(self.style['tickTextOffset'])
@ -218,14 +218,13 @@ class DynamicDateAxis(Axis):
    def _indexes_to_timestrs(
        self,
-        indexes: list[int],
+        indexes: List[int],
-    ) -> list[str]:
+    ) -> List[str]:
        chart = self.linkedsplits.chart
-        flow = chart._flows[chart.name]
+        bars = chart._arrays[chart.name]
-        shm = flow.shm
+        shm = self.linkedsplits.chart._shm
        bars = shm.array
        first = shm._first.value
        bars_len = len(bars)
@ -242,17 +241,10 @@ class DynamicDateAxis(Axis):
        )]
        # TODO: **don't** have this hard coded shift to EST
-        # delay = times[-1] - times[-2]
+        dts = pd.to_datetime(epochs, unit='s')  # - 4*pd.offsets.Hour()
        dts = np.array(epochs, dtype='datetime64[s]')
-        # see units listing:
+        delay = times[-1] - times[-2]
-        # https://numpy.org/devdocs/reference/arrays.datetime.html#datetime-units
+        return dts.strftime(self.tick_tpl[delay])
        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])
    def tickStrings(
        self,
@ -438,7 +430,7 @@ class XAxisLabel(AxisLabel):
        | QtCore.Qt.AlignCenter
    )
-    def size_hint(self) -> tuple[float, float]:
+    def size_hint(self) -> Tuple[float, float]:
        # size to parent axis height
        return self._parent.height(), None
@ -452,11 +444,11 @@ class XAxisLabel(AxisLabel):
        timestrs = self._parent._indexes_to_timestrs([int(value)])
-        if not len(timestrs):
+        if not timestrs.any():
            return
        pad = 1*' '
-        self.label_str = pad + str(timestrs[0]) + pad
+        self.label_str = pad + timestrs[0] + pad
        _, y_offset = self._parent.txt_offsets()
@ -517,7 +509,7 @@ class YAxisLabel(AxisLabel):
        if getattr(self._parent, 'txt_offsets', False):
            self.x_offset, y_offset = self._parent.txt_offsets()
-    def size_hint(self) -> tuple[float, float]:
+    def size_hint(self) -> Tuple[float, float]:
        # size to parent axis width(-ish)
        wsh = self._dpifont.boundingRect(' ').height() / 2
        return (
--- a/piker/ui/_chart.py
+++ b/piker/ui/_chart.py
@ -34,7 +34,9 @@ from PyQt5.QtWidgets import (
    QVBoxLayout,
    QSplitter,
 )
 import msgspec
 import numpy as np
 # from pydantic import BaseModel
 import pyqtgraph as pg
 import trio
@ -47,13 +49,9 @@ from ._cursor import (
    Cursor,
    ContentsLabel,
 )
 from ..data._sharedmem import ShmArray
 from ._l1 import L1Labels
 from ._ohlc import BarItems
-from ._curve import (
+from ._curve import FastAppendCurve
    Curve,
    StepCurve,
 )
 from ._style import (
    hcolor,
    CHART_MARGINS,
@ -62,12 +60,15 @@ from ._style import (
 )
 from ..data.feed import Feed
 from ..data._source import Symbol
 from ..data._sharedmem import (
    ShmArray,
    # _Token,
 )
 from ..log import get_logger
 from ._interaction import ChartView
 from ._forms import FieldsForm
 from .._profile import pg_profile_enabled, ms_slower_then
 from ._overlay import PlotItemOverlay
 from ._flows import Flow
 if TYPE_CHECKING:
    from ._display import DisplayState
@ -368,11 +369,11 @@ class LinkedSplits(QWidget):
        self._symbol: Symbol = None
-    def graphics_cycle(self, **kwargs) -> None:
+    def graphics_cycle(self) -> None:
        from . import _display
        ds = self.display_state
        if ds:
-            return _display.graphics_update_cycle(ds, **kwargs)
+            return _display.graphics_update_cycle(ds)
    @property
    def symbol(self) -> Symbol:
@ -418,7 +419,7 @@ class LinkedSplits(QWidget):
        self,
        symbol: Symbol,
-        shm: ShmArray,
+        array: np.ndarray,
        sidepane: FieldsForm,
        style: str = 'bar',
@ -443,7 +444,7 @@ class LinkedSplits(QWidget):
        self.chart = self.add_plot(
            name=symbol.key,
-            shm=shm,
+            array=array,
            style=style,
            _is_main=True,
@ -471,7 +472,7 @@ class LinkedSplits(QWidget):
        self,
        name: str,
-        shm: ShmArray,
+        array: np.ndarray,
        array_key: Optional[str] = None,
        style: str = 'line',
@ -515,6 +516,7 @@ class LinkedSplits(QWidget):
            name=name,
            data_key=array_key or name,
            array=array,
            parent=qframe,
            linkedsplits=self,
            axisItems=axes,
@ -578,7 +580,7 @@ class LinkedSplits(QWidget):
            graphics, data_key = cpw.draw_ohlc(
                name,
-                shm,
+                array,
                array_key=array_key
            )
            self.cursor.contents_labels.add_label(
@ -592,7 +594,7 @@ class LinkedSplits(QWidget):
            add_label = True
            graphics, data_key = cpw.draw_curve(
                name,
-                shm,
+                array,
                array_key=array_key,
                color='default_light',
            )
@ -601,7 +603,7 @@ class LinkedSplits(QWidget):
            add_label = True
            graphics, data_key = cpw.draw_curve(
                name,
-                shm,
+                array,
                array_key=array_key,
                step_mode=True,
                color='davies',
@ -689,6 +691,7 @@ class ChartPlotWidget(pg.PlotWidget):
        # the "data view" we generate graphics from
        name: str,
        array: np.ndarray,
        data_key: str,
        linkedsplits: LinkedSplits,
@ -741,6 +744,14 @@ class ChartPlotWidget(pg.PlotWidget):
        self._max_l1_line_len: float = 0
        # self.setViewportMargins(0, 0, 0, 0)
        # self._ohlc = array  # readonly view of ohlc data
        # TODO: move to Aggr above XD
        # readonly view of data arrays
        self._arrays = {
            self.data_key: array,
        }
        self._graphics = {}  # registry of underlying graphics
        # registry of overlay curve names
        self._flows: dict[str, Flow] = {}
@ -756,6 +767,7 @@ class ChartPlotWidget(pg.PlotWidget):
        # show background grid
        self.showGrid(x=False, y=True, alpha=0.3)
        self.default_view()
        self.cv.enable_auto_yrange()
        self.pi_overlay: PlotItemOverlay = PlotItemOverlay(self.plotItem)
@ -804,8 +816,14 @@ class ChartPlotWidget(pg.PlotWidget):
        Return a range tuple for the bars present in view.
        '''
-        main_flow = self._flows[self.name]
+        l, r = self.view_range()
-        ifirst, l, lbar, rbar, r, ilast = main_flow.datums_range()
+        array = self._arrays[self.name]
        start, stop = self._xrange = (
            array[0]['index'],
            array[-1]['index'],
        )
        lbar = max(l, start)
        rbar = min(r, stop)
        return l, lbar, rbar, r
    def curve_width_pxs(
@ -859,51 +877,40 @@ class ChartPlotWidget(pg.PlotWidget):
    def default_view(
        self,
-        bars_from_y: int = 3000,
+        steps_on_screen: Optional[int] = None
    ) -> None:
        '''
        Set the view box to the "default" startup view of the scene.
        '''
-        flow = self._flows.get(self.name)
+        try:
-        if not flow:
+            index = self._arrays[self.name]['index']
-            log.warning(f'`Flow` for {self.name} not loaded yet?')
+        except IndexError:
            log.warning(f'array for {self.name} not loaded yet?')
            return
        index = flow.shm.array['index']
        xfirst, xlast = index[0], index[-1]
        l, lbar, rbar, r = self.bars_range()
-        view = self.view
+
        marker_pos, l1_len = self.pre_l1_xs()
        end = xlast + l1_len + 1
        if (
            rbar < 0
            or l < xfirst
            or l < 0
            or (rbar - lbar) < 6
        ):
-            # TODO: set fixed bars count on screen that approx includes as
+            # set fixed bars count on screen that approx includes as
            # many bars as possible before a downsample line is shown.
-            begin = xlast - bars_from_y
+            begin = xlast - round(6116 / 6)
            view.setXRange(
                min=begin,
                max=xlast,
                padding=0,
            )
            # re-get range
            l, lbar, rbar, r = self.bars_range()
-        # we get the L1 spread label "length" in view coords
+        else:
-        # terms now that we've scaled either by user control
+            begin = end - (r - l)
        # or to the default set of bars as per the immediate block
        # above.
        marker_pos, l1_len = self.pre_l1_xs()
        end = xlast + l1_len + 1
        begin = end - (r - l)
        # for debugging
        # print(
-        #     # f'bars range: {brange}\n'
+        #     f'bars range: {brange}\n'
        #     f'xlast: {xlast}\n'
        #     f'marker pos: {marker_pos}\n'
        #     f'l1 len: {l1_len}\n'
@ -915,13 +922,14 @@ class ChartPlotWidget(pg.PlotWidget):
        if self._static_yrange == 'axis':
            self._static_yrange = None
        view = self.view
        view.setXRange(
            min=begin,
            max=end,
            padding=0,
        )
        self.view.maybe_downsample_graphics()
        view._set_yrange()
        self.view.maybe_downsample_graphics()
        try:
            self.linked.graphics_cycle()
        except IndexError:
@ -952,7 +960,7 @@ class ChartPlotWidget(pg.PlotWidget):
    def draw_ohlc(
        self,
        name: str,
-        shm: ShmArray,
+        data: np.ndarray,
        array_key: Optional[str] = None,
@ -972,12 +980,15 @@ class ChartPlotWidget(pg.PlotWidget):
        # the np array buffer to be drawn on next render cycle
        self.plotItem.addItem(graphics)
        # draw after to allow self.scene() to work...
        graphics.draw_from_data(data)
        data_key = array_key or name
        self._graphics[data_key] = graphics
        self._flows[data_key] = Flow(
            name=name,
            plot=self.plotItem,
            _shm=shm,
            is_ohlc=True,
            graphics=graphics,
        )
@ -1047,21 +1058,20 @@ class ChartPlotWidget(pg.PlotWidget):
        self,
        name: str,
-        shm: ShmArray,
+        data: np.ndarray,
        array_key: Optional[str] = None,
        overlay: bool = False,
        color: Optional[str] = None,
        add_label: bool = True,
        pi: Optional[pg.PlotItem] = None,
        step_mode: bool = False,
        **pdi_kwargs,
    ) -> (pg.PlotDataItem, str):
        '''
        Draw a "curve" (line plot graphics) for the provided data in
-        the input shm array ``shm``.
+        the input array ``data``.
        '''
        color = color or self.pen_color or 'default_light'
@ -1071,26 +1081,40 @@ class ChartPlotWidget(pg.PlotWidget):
        data_key = array_key or name
-        curve_type = {
+        # yah, we wrote our own B)
-            None: Curve,
+        curve = FastAppendCurve(
-            'step': StepCurve,
+            y=data[data_key],
-            # TODO:
+            x=data['index'],
-            # 'bars': BarsItems
+            # antialias=True,
        }['step' if step_mode else None]
        curve = curve_type(
            name=name,
            # XXX: pretty sure this is just more overhead
            # on data reads and makes graphics rendering no faster
            # clipToView=True,
            **pdi_kwargs,
        )
        # XXX: see explanation for different caching modes:
        # https://stackoverflow.com/a/39410081
        # seems to only be useful if we don't re-generate the entire
        # QPainterPath every time
        # curve.setCacheMode(QtWidgets.QGraphicsItem.DeviceCoordinateCache)
        # don't ever use this - it's a colossal nightmare of artefacts
        # and is disastrous for performance.
        # curve.setCacheMode(QtWidgets.QGraphicsItem.ItemCoordinateCache)
        # register curve graphics and backing array for name
        self._graphics[name] = curve
        self._arrays[data_key] = data
        pi = pi or self.plotItem
        self._flows[data_key] = Flow(
            name=name,
            plot=pi,
            _shm=shm,
            is_ohlc=False,
            # register curve graphics with this flow
            graphics=curve,
        )
@ -1151,11 +1175,16 @@ class ChartPlotWidget(pg.PlotWidget):
        )
        return last
-    def update_graphics_from_flow(
+    def update_graphics_from_array(
        self,
        graphics_name: str,
        array: Optional[np.ndarray] = None,
        array_key: Optional[str] = None,
        use_vr: bool = True,
        render: bool = True,
        **kwargs,
    ) -> pg.GraphicsObject:
@ -1163,11 +1192,63 @@ class ChartPlotWidget(pg.PlotWidget):
        Update the named internal graphics from ``array``.
        '''
-        flow = self._flows[array_key or graphics_name]
+        if array is not None:
-        return flow.update_graphics(
+            assert len(array)
-            array_key=array_key,
+
-            **kwargs,
+        data_key = array_key or graphics_name
-        )
+        if graphics_name not in self._flows:
            data_key = self.name
        if array is not None:
            # write array to internal graphics table
            self._arrays[data_key] = array
        else:
            array = self._arrays[data_key]
        # array key and graphics "name" might be different..
        graphics = self._graphics[graphics_name]
        # compute "in-view" indices
        l, lbar, rbar, r = self.bars_range()
        indexes = array['index']
        ifirst = indexes[0]
        ilast = indexes[-1]
        lbar_i = max(l, ifirst) - ifirst
        rbar_i = min(r, ilast) - ifirst
        # TODO: we could do it this way as well no?
        # to_draw = array[lbar - ifirst:(rbar - ifirst) + 1]
        in_view = array[lbar_i: rbar_i + 1]
        if (
            not in_view.size
            or not render
        ):
            return graphics
        if isinstance(graphics, BarItems):
            graphics.update_from_array(
                array,
                in_view,
                view_range=(lbar_i, rbar_i) if use_vr else None,
                **kwargs,
            )
        else:
            graphics.update_from_array(
                x=array['index'],
                y=array[data_key],
                x_iv=in_view['index'],
                y_iv=in_view[data_key],
                view_range=(lbar_i, rbar_i) if use_vr else None,
                **kwargs
            )
        return graphics
    # def _label_h(self, yhigh: float, ylow: float) -> float:
    #     # compute contents label "height" in view terms
@ -1214,7 +1295,7 @@ class ChartPlotWidget(pg.PlotWidget):
        # TODO: this should go onto some sort of
        # data-view thinger..right?
-        ohlc = self._flows[self.name].shm.array
+        ohlc = self._shm.array
        # XXX: not sure why the time is so off here
        # looks like we're gonna have to do some fixing..
@ -1244,9 +1325,7 @@ class ChartPlotWidget(pg.PlotWidget):
    def maxmin(
        self,
        name: Optional[str] = None,
-        bars_range: Optional[tuple[
+        bars_range: Optional[tuple[int, int, int, int]] = None,
            int, int, int, int, int, int
        ]] = None,
    ) -> tuple[float, float]:
        '''
@ -1255,14 +1334,16 @@ class ChartPlotWidget(pg.PlotWidget):
        If ``bars_range`` is provided use that range.
        '''
        # print(f'Chart[{self.name}].maxmin()')
        profiler = pg.debug.Profiler(
-            msg=f'`{str(self)}.maxmin(name={name})`: `{self.name}`',
+            msg=f'`{str(self)}.maxmin()` loop cycle for: `{self.name}`',
            disabled=not pg_profile_enabled(),
-            ms_threshold=ms_slower_then,
+            gt=ms_slower_then,
            delayed=True,
        )
        l, lbar, rbar, r = bars_range or self.bars_range()
        profiler(f'{self.name} got bars range')
        # TODO: here we should instead look up the ``Flow.shm.array``
        # and read directly from shm to avoid copying to memory first
        # and then reading it again here.
@ -1272,26 +1353,112 @@ class ChartPlotWidget(pg.PlotWidget):
            flow is None
        ):
            log.error(f"flow {flow_key} doesn't exist in chart {self.name} !?")
-            key = res = 0, 0
+            res = 0, 0
        else:
            (
                first,
                l,
                lbar,
                rbar,
                r,
                last,
            ) = bars_range or flow.datums_range()
            profiler(f'{self.name} got bars range')
            key = round(lbar), round(rbar)
            res = flow.maxmin(*key)
            profiler(f'yrange mxmn: {key} -> {res}')
            if res == (None, None):
                log.error(
                    f"{flow_key} no mxmn for bars_range => {key} !?"
                )
                res = 0, 0
        profiler(f'yrange mxmn: {key} -> {res}')
        return res
 # class FlowsTable(pydantic.BaseModel):
 #     '''
 #     Data-AGGRegate: high level API onto multiple (categorized)
 #     ``Flow``s with high level processing routines for
 #     multi-graphics computations and display.
 #     '''
 #     flows: dict[str, np.ndarray] = {}
 class Flow(msgspec.Struct):  # , frozen=True):
    '''
    (FinancialSignal-)Flow compound type which wraps a real-time
    graphics (curve) and its backing data stream together for high level
    access and control.
    The intention is for this type to eventually be capable of shm-passing
    of incrementally updated graphics stream data between actors.
    '''
    name: str
    plot: pg.PlotItem
    is_ohlc: bool = False
    graphics: pg.GraphicsObject
    # TODO: hackery to be able to set a shm later
    # but whilst also allowing this type to hashable,
    # likely will require serializable token that is used to attach
    # to the underlying shm ref after startup?
    _shm: Optional[ShmArray] = None  # currently, may be filled in "later"
    # cache of y-range values per x-range input.
    _mxmns: dict[tuple[int, int], tuple[float, float]] = {}
    @property
    def shm(self) -> ShmArray:
        return self._shm
    @shm.setter
    def shm(self, shm: ShmArray) -> ShmArray:
        self._shm = shm
    def maxmin(
        self,
        lbar,
        rbar,
    ) -> tuple[float, float]:
        '''
        Compute the cached max and min y-range values for a given
        x-range determined by ``lbar`` and ``rbar``.
        '''
        rkey = (lbar, rbar)
        cached_result = self._mxmns.get(rkey)
        if cached_result:
            return cached_result
        shm = self.shm
        if shm is None:
            mxmn = None
        else:  # new block for profiling?..
            arr = shm.array
            # build relative indexes into shm array
            # TODO: should we just add/use a method
            # on the shm to do this?
            ifirst = arr[0]['index']
            slice_view = arr[
                lbar - ifirst:
                (rbar - ifirst) + 1
            ]
            if not slice_view.size:
                mxmn = None
            else:
                if self.is_ohlc:
                    ylow = np.min(slice_view['low'])
                    yhigh = np.max(slice_view['high'])
                else:
                    view = slice_view[self.name]
                    ylow = np.min(view)
                    yhigh = np.max(view)
                mxmn = ylow, yhigh
            if mxmn is not None:
                # cache new mxmn result
                self._mxmns[rkey] = mxmn
            return mxmn
--- a/piker/ui/_compression.py
+++ b/piker/ui/_compression.py
@ -138,20 +138,50 @@ def ohlc_flatten(
    return x, flat
 def ohlc_to_m4_line(
    ohlc: np.ndarray,
    px_width: int,
    downsample: bool = False,
    uppx: Optional[float] = None,
    pretrace: bool = False,
 ) -> tuple[np.ndarray, np.ndarray]:
    '''
    Convert an OHLC struct-array to a m4 downsampled 1-d array.
    '''
    xpts, flat = ohlc_flatten(
        ohlc,
        use_mxmn=pretrace,
    )
    if downsample:
        bins, x, y = ds_m4(
            xpts,
            flat,
            px_width=px_width,
            uppx=uppx,
            log_scale=bool(uppx)
        )
        x = np.broadcast_to(x[:, None], y.shape)
        x = (x + np.array([-0.43, 0, 0, 0.43])).flatten()
        y = y.flatten()
        return x, y
    else:
        return xpts, flat
 def ds_m4(
    x: np.ndarray,
    y: np.ndarray,
    # units-per-pixel-x(dimension)
    uppx: float,
-    # XXX: troll zone / easter egg..
+    # this is the width of the data in view
-    # want to mess with ur pal, pass in the actual
+    # in display-device-local pixel units.
-    # pixel width here instead of uppx-proper (i.e. pass
+    px_width: int,
-    # in our ``pg.GraphicsObject`` derivative's ``.px_width()``
+    uppx: Optional[float] = None,
-    # gto mega-trip-out ur bud). Hint, it used to be implemented
+    log_scale: bool = True,
    # (wrongly) using "pixel width", so check the git history ;)
    xrange: Optional[float] = None,
 ) -> tuple[int, np.ndarray, np.ndarray]:
    '''
@ -178,49 +208,52 @@ def ds_m4(
    # "i didn't show it in the sample code, but it's accounted for
    # in the start and end indices and number of bins"
-    # should never get called unless actually needed
+    # optionally log-scale down the "supposed pxs on screen"
-    assert uppx > 1
+    # as the units-per-px (uppx) get's large.
    if log_scale:
        assert uppx, 'You must provide a `uppx` value to use log scaling!'
        # scaler = 2**7 / (1 + math.log(uppx, 2))
        scaler = round(
            max(
                # NOTE: found that a 16x px width brought greater
                # detail, likely due to dpi scaling?
                # px_width=px_width * 16,
                2**7 / (1 + math.log(uppx, 2)),
                1
            )
        )
        px_width *= scaler
    assert px_width > 1  # width of screen in pxs?
    # NOTE: if we didn't pre-slice the data to downsample
    # you could in theory pass these as the slicing params,
    # do we care though since we can always just pre-slice the
    # input?
    x_start = x[0]  # x value start/lowest in domain
-
+    x_end = x[-1]  # x end value/highest in domain
    if xrange is None:
        x_end = x[-1]  # x end value/highest in domain
        xrange = (x_end - x_start)
    # XXX: always round up on the input pixels
-    # lnx = len(x)
+    px_width = math.ceil(px_width)
    # uppx *= max(4 / (1 + math.log(uppx, 2)), 1)
-    pxw = math.ceil(xrange / uppx)
+    x_range = x_end - x_start
-    # scale up the frame "width" directly with uppx
+    # ratio of indexed x-value to width of raster in pixels.
-    w = uppx
+    # this is more or less, uppx: units-per-pixel.
    w = x_range / float(px_width)
    # ensure we make more then enough
    # frames (windows) for the output pixel
-    frames = pxw
+    frames = px_width
    # if we have more and then exact integer's
    # (uniform quotient output) worth of datum-domain-points
    # per windows-frame, add one more window to ensure
    # we have room for all output down-samples.
-    pts_per_pixel, r = divmod(xrange, frames)
+    pts_per_pixel, r = divmod(len(x), frames)
    if r:
        # while r:
        frames += 1
        pts_per_pixel, r = divmod(xrange, frames)
    # print(
    #     f'uppx: {uppx}\n'
    #     f'xrange: {xrange}\n'
    #     f'pxw: {pxw}\n'
    #     f'frames: {frames}\n'
    # )
    assert frames >= (xrange / uppx)
    # call into ``numba``
    nb, i_win, y_out = _m4(
--- a/piker/ui/_cursor.py
+++ b/piker/ui/_cursor.py
@ -43,8 +43,8 @@ log = get_logger(__name__)
 # latency (in terms of perceived lag in cross hair) so really be sure
 # there's an improvement if you want to change it!
-_mouse_rate_limit = 60  # TODO; should we calc current screen refresh rate?
+_mouse_rate_limit = 120  # TODO; should we calc current screen refresh rate?
-_debounce_delay = 0
+_debounce_delay = 1 / 40
 _ch_label_opac = 1
@ -98,30 +98,25 @@ class LineDot(pg.CurvePoint):
        ev: QtCore.QEvent,
    ) -> bool:
-
+        if not isinstance(
-        if (
+            ev, QtCore.QDynamicPropertyChangeEvent
-            not isinstance(ev, QtCore.QDynamicPropertyChangeEvent)
+        ) or self.curve() is None:
            or self.curve() is None
        ):
            return False
        # TODO: get rid of this ``.getData()`` and
        # make a more pythonic api to retreive backing
        # numpy arrays...
-        # (x, y) = self.curve().getData()
+        (x, y) = self.curve().getData()
-        # index = self.property('index')
+        index = self.property('index')
-        # # first = self._plot._arrays['ohlc'][0]['index']
+        # first = self._plot._arrays['ohlc'][0]['index']
-        # # first = x[0]
+        # first = x[0]
-        # # i = index - first
+        # i = index - first
-        # if index:
+        if index:
-        #     i = round(index - x[0])
+            i = round(index - x[0])
-        #     if i > 0 and i < len(y):
+            if i > 0 and i < len(y):
-        #         newPos = (index, y[i])
+                newPos = (index, y[i])
-        #         QtWidgets.QGraphicsItem.setPos(
+                QtWidgets.QGraphicsItem.setPos(self, *newPos)
-        #             self,
+                return True
        #             *newPos,
        #         )
        #         return True
        return False
@ -196,9 +191,6 @@ class ContentsLabel(pg.LabelItem):
        self.setText(
            "<b>i</b>:{index}<br/>"
            # NB: these fields must be indexed in the correct order via
            # the slice syntax below.
            "<b>epoch</b>:{}<br/>"
            "<b>O</b>:{}<br/>"
            "<b>H</b>:{}<br/>"
            "<b>L</b>:{}<br/>"
@ -206,15 +198,7 @@ class ContentsLabel(pg.LabelItem):
            "<b>V</b>:{}<br/>"
            "<b>wap</b>:{}".format(
                *array[index - first][
-                    [
+                    ['open', 'high', 'low', 'close', 'volume', 'bar_wap']
                        'time',
                        'open',
                        'high',
                        'low',
                        'close',
                        'volume',
                        'bar_wap',
                    ]
                ],
                name=name,
                index=index,
@ -259,13 +243,13 @@ class ContentsLabels:
    def update_labels(
        self,
        index: int,
        # array_name: str,
    ) -> None:
        # for name, (label, update) in self._labels.items():
        for chart, name, label, update in self._labels:
-            flow = chart._flows[name]
+            array = chart._arrays[name]
            array = flow.shm.array
            if not (
                index >= 0
                and index < array[-1]['index']
@ -274,6 +258,8 @@ class ContentsLabels:
                print('WTF out of range?')
                continue
            # array = chart._arrays[name]
            # call provided update func with data point
            try:
                label.show()
@ -475,12 +461,9 @@ class Cursor(pg.GraphicsObject):
    ) -> LineDot:
        # if this plot contains curves add line dot "cursors" to denote
        # the current sample under the mouse
        main_flow = plot._flows[plot.name]
        # read out last index
        i = main_flow.shm.array[-1]['index']
        cursor = LineDot(
            curve,
-            index=i,
+            index=plot._arrays[plot.name][-1]['index'],
            plot=plot
        )
        plot.addItem(cursor)
--- a/piker/ui/_curve.py
+++ b/piker/ui/_curve.py
@ -18,37 +18,83 @@
 Fast, smooth, sexy curves.
 """
-from contextlib import contextmanager as cm
+from typing import Optional
 from typing import Optional, Callable
 import numpy as np
 import pyqtgraph as pg
-from PyQt5 import QtWidgets
+from PyQt5 import QtGui, QtWidgets
 from PyQt5.QtWidgets import QGraphicsItem
 from PyQt5.QtCore import (
    Qt,
    QLineF,
    QSizeF,
    QRectF,
    # QRect,
    QPointF,
 )
-from PyQt5.QtGui import (
+
    QPainter,
    QPainterPath,
 )
 from .._profile import pg_profile_enabled, ms_slower_then
 from ._style import hcolor
-# from ._compression import (
+from ._compression import (
-#     # ohlc_to_m4_line,
+    # ohlc_to_m4_line,
-#     ds_m4,
+    ds_m4,
-# )
+)
 from ..log import get_logger
 log = get_logger(__name__)
 def step_path_arrays_from_1d(
    x: np.ndarray,
    y: np.ndarray,
    include_endpoints: bool = False,
 ) -> (np.ndarray, np.ndarray):
    '''
    Generate a "step mode" curve aligned with OHLC style bars
    such that each segment spans each bar (aka "centered" style).
    '''
    y_out = y.copy()
    x_out = x.copy()
    x2 = np.empty(
        # the data + 2 endpoints on either end for
        # "termination of the path".
        (len(x) + 1, 2),
        # we want to align with OHLC or other sampling style
        # bars likely so we need fractinal values
        dtype=float,
    )
    x2[0] = x[0] - 0.5
    x2[1] = x[0] + 0.5
    x2[1:] = x[:, np.newaxis] + 0.5
    # flatten to 1-d
    x_out = x2.reshape(x2.size)
    # we create a 1d with 2 extra indexes to
    # hold the start and (current) end value for the steps
    # on either end
    y2 = np.empty((len(y), 2), dtype=y.dtype)
    y2[:] = y[:, np.newaxis]
    y_out = np.empty(
        2*len(y) + 2,
        dtype=y.dtype
    )
    # flatten and set 0 endpoints
    y_out[1:-1] = y2.reshape(y2.size)
    y_out[0] = 0
    y_out[-1] = 0
    if not include_endpoints:
        return x_out[:-1], y_out[:-1]
    else:
        return x_out, y_out
 _line_styles: dict[str, int] = {
    'solid': Qt.PenStyle.SolidLine,
    'dash': Qt.PenStyle.DashLine,
@ -57,43 +103,24 @@ _line_styles: dict[str, int] = {
 }
-class Curve(pg.GraphicsObject):
+class FastAppendCurve(pg.GraphicsObject):
    '''
-    A faster, simpler, append friendly version of
+    A faster, append friendly version of ``pyqtgraph.PlotCurveItem``
-    ``pyqtgraph.PlotCurveItem`` built for highly customizable real-time
+    built for real-time data updates.
    updates.
-    This type is a much stripped down version of a ``pyqtgraph`` style
+    The main difference is avoiding regeneration of the entire
-    "graphics object" in the sense that the internal lower level
+    historical path where possible and instead only updating the "new"
-    graphics which are drawn in the ``.paint()`` method are actually
+    segment(s) via a ``numpy`` array diff calc. Further the "last"
-    rendered outside of this class entirely and instead are assigned as
+    graphic segment is drawn independently such that near-term (high
-    state (instance vars) here and then drawn during a Qt graphics
+    frequency) discrete-time-sampled style updates don't trigger a full
-    cycle.
+    path redraw.
    The main motivation for this more modular, composed design is that
    lower level graphics data can be rendered in different threads and
    then read and drawn in this main thread without having to worry
    about dealing with Qt's concurrency primitives. See
    ``piker.ui._flows.Renderer`` for details and logic related to lower
    level path generation and incremental update. The main differences in
    the path generation code include:
    - avoiding regeneration of the entire historical path where possible
      and instead only updating the "new" segment(s) via a ``numpy``
      array diff calc.
    - here, the "last" graphics datum-segment is drawn independently
      such that near-term (high frequency) discrete-time-sampled style
      updates don't trigger a full path redraw.
    '''
    # sub-type customization methods
    sub_br: Optional[Callable] = None
    sub_paint: Optional[Callable] = None
    declare_paintables: Optional[Callable] = None
    def __init__(
        self,
        x: np.ndarray,
        y: np.ndarray,
        *args,
        step_mode: bool = False,
@ -107,25 +134,27 @@ class Curve(pg.GraphicsObject):
    ) -> None:
        self._name = name
        # brutaaalll, see comments within..
-        self.yData = None
+        self._y = self.yData = y
-        self.xData = None
+        self._x = self.xData = x
-        # self._last_cap: int = 0
+        self._name = name
-        self.path: Optional[QPainterPath] = None
+        self.path: Optional[QtGui.QPainterPath] = None
        # additional path used for appends which tries to avoid
        # triggering an update/redraw of the presumably larger
        # historical ``.path`` above.
        self.use_fpath = use_fpath
-        self.fast_path: Optional[QPainterPath] = None
+        self.fast_path: Optional[QtGui.QPainterPath] = None
        # TODO: we can probably just dispense with the parent since
        # we're basically only using the pen setting now...
        super().__init__(*args, **kwargs)
        # self._xrange: tuple[int, int] = self.dataBounds(ax=0)
        self._xrange: Optional[tuple[int, int]] = None
        # self._last_draw = time.time()
        self._in_ds: bool = False
        self._last_uppx: float = 0
        # all history of curve is drawn in single px thickness
        pen = pg.mkPen(hcolor(color))
        pen.setStyle(_line_styles[style])
@ -139,43 +168,29 @@ class Curve(pg.GraphicsObject):
        # self.last_step_pen = pg.mkPen(hcolor(color), width=2)
        self.last_step_pen = pg.mkPen(pen, width=2)
-        # self._last_line: Optional[QLineF] = None
+        self._last_line: Optional[QLineF] = None
-        self._last_line = QLineF()
+        self._last_step_rect: Optional[QRectF] = None
        self._last_w: float = 1
        # flat-top style histogram-like discrete curve
-        # self._step_mode: bool = step_mode
+        self._step_mode: bool = step_mode
        # self._fill = True
        self._brush = pg.functions.mkBrush(hcolor(fill_color or color))
        # NOTE: this setting seems to mostly prevent redraws on mouse
        # interaction which is a huge boon for avg interaction latency.
        # TODO: one question still remaining is if this makes trasform
        # interactions slower (such as zooming) and if so maybe if/when
        # we implement a "history" mode for the view we disable this in
        # that mode?
-        # don't enable caching by default for the case where the
+        if step_mode:
-        # only thing drawn is the "last" line segment which can
+            # don't enable caching by default for the case where the
-        # have a weird artifact where it won't be fully drawn to its
+            # only thing drawn is the "last" line segment which can
-        # endpoint (something we saw on trade rate curves)
+            # have a weird artifact where it won't be fully drawn to its
-        self.setCacheMode(QGraphicsItem.DeviceCoordinateCache)
+            # endpoint (something we saw on trade rate curves)
            self.setCacheMode(
                QGraphicsItem.DeviceCoordinateCache
            )
-        # XXX: see explanation for different caching modes:
+        self.update()
        # https://stackoverflow.com/a/39410081
        # seems to only be useful if we don't re-generate the entire
        # QPainterPath every time
        # curve.setCacheMode(QtWidgets.QGraphicsItem.DeviceCoordinateCache)
        # don't ever use this - it's a colossal nightmare of artefacts
        # and is disastrous for performance.
        # curve.setCacheMode(QtWidgets.QGraphicsItem.ItemCoordinateCache)
        # allow sub-type customization
        declare = self.declare_paintables
        if declare:
            declare()
    # TODO: probably stick this in a new parent
    # type which will contain our own version of
@ -199,6 +214,9 @@ class Curve(pg.GraphicsObject):
        vr = self.viewRect()
        l, r = int(vr.left()), int(vr.right())
        if not self._xrange:
            return 0
        start, stop = self._xrange
        lbar = max(l, start)
        rbar = min(r, stop)
@ -207,10 +225,352 @@ class Curve(pg.GraphicsObject):
            QLineF(lbar, 0, rbar, 0)
        ).length()
    def downsample(
        self,
        x,
        y,
        px_width,
        uppx,
    ) -> tuple[np.ndarray, np.ndarray]:
        # downsample whenever more then 1 pixels per datum can be shown.
        # always refresh data bounds until we get diffing
        # working properly, see above..
        bins, x, y = ds_m4(
            x,
            y,
            px_width=px_width,
            uppx=uppx,
            log_scale=bool(uppx)
        )
        x = np.broadcast_to(x[:, None], y.shape)
        # x = (x + np.array([-0.43, 0, 0, 0.43])).flatten()
        x = (x + np.array([-0.5, 0, 0, 0.5])).flatten()
        y = y.flatten()
        # presumably?
        self._in_ds = True
        return x, y
    def update_from_array(
        self,
        # full array input history
        x: np.ndarray,
        y: np.ndarray,
        # pre-sliced array data that's "in view"
        x_iv: np.ndarray,
        y_iv: np.ndarray,
        view_range: Optional[tuple[int, int]] = None,
        profiler: Optional[pg.debug.Profiler] = None,
    ) -> QtGui.QPainterPath:
        '''
        Update curve from input 2-d data.
        Compare with a cached "x-range" state and (pre/a)ppend based on
        a length diff.
        '''
        profiler = profiler or pg.debug.Profiler(
            msg=f'FastAppendCurve.update_from_array(): `{self._name}`',
            disabled=not pg_profile_enabled(),
            gt=ms_slower_then,
        )
        # flip_cache = False
        if self._xrange:
            istart, istop = self._xrange
        else:
            self._xrange = istart, istop = x[0], x[-1]
        # print(f"xrange: {self._xrange}")
        # XXX: lol brutal, the internals of `CurvePoint` (inherited by
        # our `LineDot`) required ``.getData()`` to work..
        self.xData = x
        self.yData = y
        self._x, self._y = x, y
        if view_range:
            profiler(f'view range slice {view_range}')
        # downsampling incremental state checking
        uppx = self.x_uppx()
        px_width = self.px_width()
        uppx_diff = (uppx - self._last_uppx)
        should_ds = False
        should_redraw = False
        # if a view range is passed, plan to draw the
        # source ouput that's "in view" of the chart.
        if view_range and not self._in_ds:
            # print(f'{self._name} vr: {view_range}')
            # by default we only pull data up to the last (current) index
            x_out, y_out = x_iv[:-1], y_iv[:-1]
            # step mode: draw flat top discrete "step"
            # over the index space for each datum.
            if self._step_mode:
                # TODO: numba this bish
                x_out, y_out = step_path_arrays_from_1d(
                    x_out,
                    y_out
                )
                profiler('generated step arrays')
            should_redraw = True
            profiler('sliced in-view array history')
            # x_last = x_iv[-1]
            # y_last = y_iv[-1]
            self._last_vr = view_range
            # self.disable_cache()
            # flip_cache = True
        else:
            self._xrange = x[0], x[-1]
        x_last = x[-1]
        y_last = y[-1]
        # check for downsampling conditions
        if (
            # std m4 downsample conditions
            px_width
            and uppx_diff >= 4
            or uppx_diff <= -3
            or self._step_mode and abs(uppx_diff) >= 4
        ):
            log.info(
                f'{self._name} sampler change: {self._last_uppx} -> {uppx}'
            )
            self._last_uppx = uppx
            should_ds = True
        elif (
            uppx <= 2
            and self._in_ds
        ):
            # we should de-downsample back to our original
            # source data so we clear our path data in prep
            # to generate a new one from original source data.
            should_redraw = True
            should_ds = False
        # compute the length diffs between the first/last index entry in
        # the input data and the last indexes we have on record from the
        # last time we updated the curve index.
        prepend_length = int(istart - x[0])
        append_length = int(x[-1] - istop)
        # no_path_yet = self.path is None
        if (
            self.path is None
            or should_redraw
            or should_ds
            or prepend_length > 0
        ):
            if (
                not view_range
                or self._in_ds
            ):
                # by default we only pull data up to the last (current) index
                x_out, y_out = x[:-1], y[:-1]
                # step mode: draw flat top discrete "step"
                # over the index space for each datum.
                if self._step_mode:
                    x_out, y_out = step_path_arrays_from_1d(
                        x_out,
                        y_out,
                    )
                    # TODO: numba this bish
                    profiler('generated step arrays')
            if should_redraw:
                profiler('path reversion to non-ds')
                if self.path:
                    self.path.clear()
                if self.fast_path:
                    self.fast_path.clear()
            if should_redraw and not should_ds:
                if self._in_ds:
                    log.info(f'DEDOWN -> {self._name}')
                self._in_ds = False
            elif should_ds and px_width:
                x_out, y_out = self.downsample(
                    x_out,
                    y_out,
                    px_width,
                    uppx,
                )
                profiler(f'FULL PATH downsample redraw={should_ds}')
                self._in_ds = True
            self.path = pg.functions.arrayToQPath(
                x_out,
                y_out,
                connect='all',
                finiteCheck=False,
                path=self.path,
            )
            profiler('generated fresh path')
            # profiler(f'DRAW PATH IN VIEW -> {self._name}')
            # reserve mem allocs see:
            # - https://doc.qt.io/qt-5/qpainterpath.html#reserve
            # - https://doc.qt.io/qt-5/qpainterpath.html#capacity
            # - https://doc.qt.io/qt-5/qpainterpath.html#clear
            # XXX: right now this is based on had hoc checks on a
            # hidpi 3840x2160 4k monitor but we should optimize for
            # the target display(s) on the sys.
            # if no_path_yet:
            #     self.path.reserve(int(500e3))
        # TODO: get this piecewise prepend working - right now it's
        # giving heck on vwap...
        # elif prepend_length:
        #     breakpoint()
        #     prepend_path = pg.functions.arrayToQPath(
        #         x[0:prepend_length],
        #         y[0:prepend_length],
        #         connect='all'
        #     )
        #     # swap prepend path in "front"
        #     old_path = self.path
        #     self.path = prepend_path
        #     # self.path.moveTo(new_x[0], new_y[0])
        #     self.path.connectPath(old_path)
        elif (
            append_length > 0
            and not view_range
        ):
            new_x = x[-append_length - 2:-1]
            new_y = y[-append_length - 2:-1]
            if self._step_mode:
                new_x, new_y = step_path_arrays_from_1d(
                    new_x,
                    new_y,
                )
                # [1:] since we don't need the vertical line normally at
                # the beginning of the step curve taking the first (x,
                # y) poing down to the x-axis **because** this is an
                # appended path graphic.
                new_x = new_x[1:]
                new_y = new_y[1:]
            profiler('diffed append arrays')
            if should_ds:
                new_x, new_y = self.downsample(
                    new_x,
                    new_y,
                    **should_ds,
                )
                profiler(f'fast path downsample redraw={should_ds}')
            append_path = pg.functions.arrayToQPath(
                new_x,
                new_y,
                connect='all',
                finiteCheck=False,
                path=self.fast_path,
            )
            if self.use_fpath:
                # an attempt at trying to make append-updates faster..
                if self.fast_path is None:
                    self.fast_path = append_path
                    self.fast_path.reserve(int(6e3))
                else:
                    self.fast_path.connectPath(append_path)
                    size = self.fast_path.capacity()
                    profiler(f'connected fast path w size: {size}')
                    # print(f"append_path br: {append_path.boundingRect()}")
                    # self.path.moveTo(new_x[0], new_y[0])
                    # path.connectPath(append_path)
                    # XXX: lol this causes a hang..
                    # self.path = self.path.simplified()
            else:
                size = self.path.capacity()
                profiler(f'connected history path w size: {size}')
                self.path.connectPath(append_path)
            # other merging ideas:
            # https://stackoverflow.com/questions/8936225/how-to-merge-qpainterpaths
            # path.addPath(append_path)
            # path.closeSubpath()
            # TODO: try out new work from `pyqtgraph` main which
            # should repair horrid perf:
            # https://github.com/pyqtgraph/pyqtgraph/pull/2032
            # ok, nope still horrible XD
            # if self._fill:
            #     # XXX: super slow set "union" op
            #     self.path = self.path.united(append_path).simplified()
            # self.disable_cache()
            # flip_cache = True
        # draw the "current" step graphic segment so it lines up with
        # the "middle" of the current (OHLC) sample.
        if self._step_mode:
            self._last_line = QLineF(
                x_last - 0.5, 0,
                x_last + 0.5, 0,
            )
            self._last_step_rect = QRectF(
                x_last - 0.5, 0,
                x_last + 0.5, y_last
            )
            # print(
            #     f"path br: {self.path.boundingRect()}",
            #     f"fast path br: {self.fast_path.boundingRect()}",
            #     f"last rect br: {self._last_step_rect}",
            # )
        else:
            self._last_line = QLineF(
                x[-2], y[-2],
                x[-1], y_last
            )
        profiler('draw last segment')
        # trigger redraw of path
        # do update before reverting to cache mode
        # self.prepareGeometryChange()
        self.update()
        profiler('.update()')
        # if flip_cache:
        #     # XXX: seems to be needed to avoid artifacts (see above).
        #     self.setCacheMode(QGraphicsItem.DeviceCoordinateCache)
    # XXX: lol brutal, the internals of `CurvePoint` (inherited by
    # our `LineDot`) required ``.getData()`` to work..
    def getData(self):
-        return self.xData, self.yData
+        return self._x, self._y
    # TODO: drop the above after ``Cursor`` re-work
    def get_arrays(self) -> tuple[np.ndarray, np.ndarray]:
        return self._x, self._y
    def clear(self):
        '''
@ -233,18 +593,25 @@ class Curve(pg.GraphicsObject):
                # self.fast_path.clear()
                self.fast_path = None
-    @cm
+        # self.disable_cache()
-    def reset_cache(self) -> None:
+        # self.setCacheMode(QGraphicsItem.DeviceCoordinateCache)
    def disable_cache(self) -> None:
        '''
        Disable the use of the pixel coordinate cache and trigger a geo event.
        '''
        # XXX: pretty annoying but, without this there's little
        # artefacts on the append updates to the curve...
        self.setCacheMode(QtWidgets.QGraphicsItem.NoCache)
-        yield
+        self.prepareGeometryChange()
        self.setCacheMode(QGraphicsItem.DeviceCoordinateCache)
    def boundingRect(self):
        '''
        Compute and then cache our rect.
        '''
        if self.path is None:
-            return QPainterPath().boundingRect()
+            return QtGui.QPainterPath().boundingRect()
        else:
            # dynamically override this method after initial
            # path is created to avoid requiring the above None check
@ -256,7 +623,6 @@ class Curve(pg.GraphicsObject):
        Post init ``.boundingRect()```.
        '''
        # hb = self.path.boundingRect()
        hb = self.path.controlPointRect()
        hb_size = hb.size()
@ -264,60 +630,17 @@ class Curve(pg.GraphicsObject):
        if fp:
            fhb = fp.controlPointRect()
            hb_size = fhb.size() + hb_size
        # print(f'hb_size: {hb_size}')
-        # if self._last_step_rect:
+        w = hb_size.width() + 1
-        #     hb_size += self._last_step_rect.size()
+        h = hb_size.height() + 1
        # if self._line:
        #     br = self._last_step_rect.bottomRight()
        # tl = QPointF(
        #     # self._vr[0],
        #     # hb.topLeft().y(),
        #     # 0,
        #     # hb_size.height() + 1
        # )
        #     br = self._last_step_rect.bottomRight()
        w = hb_size.width()
        h = hb_size.height()
        sbr = self.sub_br
        if sbr:
            w, h = self.sub_br(w, h)
        else:
            # assume plain line graphic and use
            # default unit step in each direction.
            # only on a plane line do we include
            # and extra index step's worth of width
            # since in the step case the end of the curve
            # actually terminates earlier so we don't need
            # this for the last step.
            w += self._last_w
            # ll = self._last_line
            h += 1  # ll.y2() - ll.y1()
        # br = QPointF(
        #     self._vr[-1],
        #     # tl.x() + w,
        #     tl.y() + h,
        # )
        br = QRectF(
            # top left
            # hb.topLeft()
            # tl,
            QPointF(hb.topLeft()),
            # br,
            # total size
            # QSizeF(hb_size)
            # hb_size,
            QSizeF(w, h)
        )
        # print(f'bounding rect: {br}')
@ -325,36 +648,40 @@ class Curve(pg.GraphicsObject):
    def paint(
        self,
-        p: QPainter,
+        p: QtGui.QPainter,
        opt: QtWidgets.QStyleOptionGraphicsItem,
        w: QtWidgets.QWidget
    ) -> None:
        profiler = pg.debug.Profiler(
-            msg=f'Curve.paint(): `{self._name}`',
+            msg=f'FastAppendCurve.paint(): `{self._name}`',
            disabled=not pg_profile_enabled(),
-            ms_threshold=ms_slower_then,
+            gt=ms_slower_then,
        )
-        sub_paint = self.sub_paint
+        if (
-        if sub_paint:
+            self._step_mode
-            sub_paint(p, profiler)
+            and self._last_step_rect
        ):
            brush = self._brush
-        p.setPen(self.last_step_pen)
+            # p.drawLines(*tuple(filter(bool, self._last_step_lines)))
-        p.drawLine(self._last_line)
+            # p.drawRect(self._last_step_rect)
-        profiler('.drawLine()')
+            p.fillRect(self._last_step_rect, brush)
-        p.setPen(self._pen)
+            profiler('.fillRect()')
        if self._last_line:
            p.setPen(self.last_step_pen)
            p.drawLine(self._last_line)
            profiler('.drawLine()')
            p.setPen(self._pen)
        path = self.path
        # cap = path.capacity()
        # if cap != self._last_cap:
        #     print(f'NEW CAPACITY: {self._last_cap} -> {cap}')
        #     self._last_cap = cap
        if path:
            p.drawPath(path)
-            profiler(f'.drawPath(path): {path.capacity()}')
+            profiler('.drawPath(path)')
        fp = self.fast_path
        if fp:
@ -368,117 +695,3 @@ class Curve(pg.GraphicsObject):
        # if self._fill:
        #     brush = self.opts['brush']
        #     p.fillPath(self.path, brush)
    def draw_last_datum(
        self,
        path: QPainterPath,
        src_data: np.ndarray,
        render_data: np.ndarray,
        reset: bool,
        array_key: str,
    ) -> None:
        # default line draw last call
        # with self.reset_cache():
        x = render_data['index']
        y = render_data[array_key]
        # draw the "current" step graphic segment so it
        # lines up with the "middle" of the current
        # (OHLC) sample.
        self._last_line = QLineF(
            x[-2], y[-2],
            x[-1], y[-1],
        )
        return x, y
 # TODO: this should probably be a "downsampled" curve type
 # that draws a bar-style (but for the px column) last graphics
 # element such that the current datum in view can be shown
 # (via it's max / min) even when highly zoomed out.
 class FlattenedOHLC(Curve):
    def draw_last_datum(
        self,
        path: QPainterPath,
        src_data: np.ndarray,
        render_data: np.ndarray,
        reset: bool,
        array_key: str,
    ) -> None:
        lasts = src_data[-2:]
        x = lasts['index']
        y = lasts['close']
        # draw the "current" step graphic segment so it
        # lines up with the "middle" of the current
        # (OHLC) sample.
        self._last_line = QLineF(
            x[-2], y[-2],
            x[-1], y[-1]
        )
        return x, y
 class StepCurve(Curve):
    def declare_paintables(
        self,
    ) -> None:
        self._last_step_rect = QRectF()
    def draw_last_datum(
        self,
        path: QPainterPath,
        src_data: np.ndarray,
        render_data: np.ndarray,
        reset: bool,
        array_key: str,
        w: float = 0.5,
    ) -> None:
        # TODO: remove this and instead place all step curve
        # updating into pre-path data render callbacks.
        # full input data
        x = src_data['index']
        y = src_data[array_key]
        x_last = x[-1]
        y_last = y[-1]
        # lol, commenting this makes step curves
        # all "black" for me :eyeroll:..
        self._last_line = QLineF(
            x_last - w, 0,
            x_last + w, 0,
        )
        self._last_step_rect = QRectF(
            x_last - w, 0,
            x_last + w, y_last,
        )
        return x, y
    def sub_paint(
        self,
        p: QPainter,
        profiler: pg.debug.Profiler,
    ) -> None:
        # p.drawLines(*tuple(filter(bool, self._last_step_lines)))
        # p.drawRect(self._last_step_rect)
        p.fillRect(self._last_step_rect, self._brush)
        profiler('.fillRect()')
    def sub_br(
        self,
        path_w: float,
        path_h: float,
    ) -> (float, float):
        # passthrough
        return path_w, path_h
--- a/piker/ui/_display.py
+++ b/piker/ui/_display.py
@ -29,10 +29,9 @@ from typing import Optional, Any, Callable
 import numpy as np
 import tractor
 import trio
 import pendulum
 import pyqtgraph as pg
-# from .. import brokers
+from .. import brokers
 from ..data.feed import open_feed
 from ._axes import YAxisLabel
 from ._chart import (
@ -48,22 +47,21 @@ from ._fsp import (
    open_vlm_displays,
 )
 from ..data._sharedmem import ShmArray
 from ..data._source import tf_in_1s
 from ._forms import (
    FieldsForm,
    mk_order_pane_layout,
 )
 from .order_mode import open_order_mode
-from .._profile import (
+# from .._profile import (
-    pg_profile_enabled,
+#     pg_profile_enabled,
-    ms_slower_then,
+#     ms_slower_then,
-)
+# )
 from ..log import get_logger
 log = get_logger(__name__)
 # TODO: load this from a config.toml!
-_quote_throttle_rate: int = 22  # Hz
+_quote_throttle_rate: int = 12  # Hz
 # a working tick-type-classes template
@ -96,19 +94,28 @@ def chart_maxmin(
    Compute max and min datums "in view" for range limits.
    '''
-    last_bars_range = chart.bars_range()
+    array = ohlcv_shm.array
-    out = chart.maxmin()
+    ifirst = array[0]['index']
-    if out is None:
+    last_bars_range = chart.bars_range()
    l, lbar, rbar, r = last_bars_range
    in_view = array[lbar - ifirst:rbar - ifirst + 1]
    if not in_view.size:
        log.warning('Resetting chart to data')
        chart.default_view()
        return (last_bars_range, 0, 0, 0)
-    mn, mx = out
+    mx, mn = (
        np.nanmax(in_view['high']),
        np.nanmin(in_view['low'],)
    )
    mx_vlm_in_view = 0
    if vlm_chart:
-        out = vlm_chart.maxmin()
+        mx_vlm_in_view = np.max(
-        if out:
+            in_view['volume']
-            _, mx_vlm_in_view = out
+        )
    return (
        last_bars_range,
@ -263,7 +270,6 @@ async def graphics_update_loop(
        'vars': {
            'tick_margin': tick_margin,
            'i_last': i_last,
            'i_last_append': i_last,
            'last_mx_vlm': last_mx_vlm,
            'last_mx': last_mx,
            'last_mn': last_mn,
@ -309,8 +315,6 @@ async def graphics_update_loop(
 def graphics_update_cycle(
    ds: DisplayState,
    wap_in_history: bool = False,
    trigger_all: bool = False,  # flag used by prepend history updates
    prepend_update_index: Optional[int] = None,
 ) -> None:
    # TODO: eventually optimize this whole graphics stack with ``numba``
@ -320,12 +324,9 @@ def graphics_update_cycle(
    profiler = pg.debug.Profiler(
        msg=f'Graphics loop cycle for: `{chart.name}`',
-        delayed=True,
+        disabled=True,  # not pg_profile_enabled(),
-        disabled=not pg_profile_enabled(),
+        gt=1/12 * 1e3,
-        # disabled=True,
+        # gt=ms_slower_then,
        ms_threshold=ms_slower_then,
        # ms_threshold=1/12 * 1e3,
    )
    # unpack multi-referenced components
@ -336,12 +337,12 @@ def graphics_update_cycle(
    vars = ds.vars
    tick_margin = vars['tick_margin']
-    update_uppx = 16
+    update_uppx = 6
    for sym, quote in ds.quotes.items():
        # compute the first available graphic's x-units-per-pixel
-        uppx = vlm_chart.view.x_uppx()
+        xpx = vlm_chart.view.x_uppx()
        # NOTE: vlm may be written by the ``brokerd`` backend
        # event though a tick sample is not emitted.
@ -360,71 +361,88 @@ def graphics_update_cycle(
        i_diff = i_step - vars['i_last']
        vars['i_last'] = i_step
        append_diff = i_step - vars['i_last_append']
        # update the "last datum" (aka extending the flow graphic with
        # new data) only if the number of unit steps is >= the number of
        # such unit steps per pixel (aka uppx). Iow, if the zoom level
        # is such that a datum(s) update to graphics wouldn't span
        # to a new pixel, we don't update yet.
        do_append = (append_diff >= uppx)
        if do_append:
            vars['i_last_append'] = i_step
        do_rt_update = uppx < update_uppx
        # print(
        #     f'append_diff:{append_diff}\n'
        #     f'uppx:{uppx}\n'
        #     f'do_append: {do_append}'
        # )
        # TODO: we should only run mxmn when we know
        # an update is due via ``do_append`` above.
        (
            brange,
            mx_in_view,
            mn_in_view,
            mx_vlm_in_view,
        ) = ds.maxmin()
        l, lbar, rbar, r = brange
        mx = mx_in_view + tick_margin
        mn = mn_in_view - tick_margin
-
+        profiler('maxmin call')
-        profiler('`ds.maxmin()` call')
+        liv = r > i_step  # the last datum is in view
        liv = r >= i_step  # the last datum is in view
        if (
            prepend_update_index is not None
            and lbar > prepend_update_index
        ):
            # on a history update (usually from the FSP subsys)
            # if the segment of history that is being prepended
            # isn't in view there is no reason to do a graphics
            # update.
            log.debug('Skipping prepend graphics cycle: frame not in view')
            return
        # don't real-time "shift" the curve to the
-        # left unless we get one of the following:
+        # left under the following conditions:
        if (
-            (
+            i_diff > 0  # no new sample step
-                # i_diff > 0  # no new sample step
+            and xpx < update_uppx  # chart is zoomed out very far
-                do_append
+            and liv
                # and uppx < 4  # chart is zoomed out very far
                and liv
            )
            or trigger_all
        ):
            # TODO: we should track and compute whether the last
            # pixel in a curve should show new data based on uppx
            # and then iff update curves and shift?
            chart.increment_view(steps=i_diff)
-            if vlm_chart:
+        if vlm_chart:
-                vlm_chart.increment_view(steps=i_diff)
+            # always update y-label
            ds.vlm_sticky.update_from_data(
                *array[-1][['index', 'volume']]
            )
-            profiler('view incremented')
+            if (
                # if zoomed out alot don't update the last "bar"
                (xpx < update_uppx or i_diff > 0)
                and liv
            ):
                # TODO: make it so this doesn't have to be called
                # once the $vlm is up?
                vlm_chart.update_graphics_from_array(
                    'volume',
                    array,
                    # UGGGh, see ``maxmin()`` impl in `._fsp` for
                    # the overlayed plotitems... we need a better
                    # bay to invoke a maxmin per overlay..
                    render=False,
                    # XXX: ^^^^ THIS IS SUPER IMPORTANT! ^^^^
                    # without this, since we disable the
                    # 'volume' (units) chart after the $vlm starts
                    # up we need to be sure to enable this
                    # auto-ranging otherwise there will be no handler
                    # connected to update accompanying overlay
                    # graphics..
                )
                last_mx_vlm = vars['last_mx_vlm']
                if (
                    mx_vlm_in_view != last_mx_vlm
                    or mx_vlm_in_view > last_mx_vlm
                ):
                    yrange = (0, mx_vlm_in_view * 1.375)
                    vlm_chart.view._set_yrange(
                        yrange=yrange,
                    )
                    # print(f'mx vlm: {last_mx_vlm} -> {mx_vlm_in_view}')
                    vars['last_mx_vlm'] = mx_vlm_in_view
                for curve_name, flow in vlm_chart._flows.items():
                    update_fsp_chart(
                        vlm_chart,
                        flow,
                        curve_name,
                        array_key=curve_name,
                    )
                    # is this even doing anything?
                    # (pretty sure it's the real-time
                    # resizing from last quote?)
                    fvb = flow.plot.vb
                    fvb._set_yrange(
                        autoscale_linked_plots=False,
                        name=curve_name,
                    )
        ticks_frame = quote.get('ticks', ())
@ -471,20 +489,14 @@ def graphics_update_cycle(
        # update ohlc sampled price bars
        if (
-            do_rt_update
+            xpx < update_uppx
-            or do_append
+            or i_diff > 0
            or trigger_all
        ):
-            chart.update_graphics_from_flow(
+            chart.update_graphics_from_array(
                chart.name,
-                # do_append=uppx < update_uppx,
+                array,
                do_append=do_append,
            )
        # NOTE: we always update the "last" datum
        # since the current range should at least be updated
        # to it's max/min on the last pixel.
        # iterate in FIFO order per tick-frame
        for typ, tick in lasts.items():
@ -495,9 +507,8 @@ def graphics_update_cycle(
            # tick frames to determine the y-range for chart
            # auto-scaling.
            # TODO: we need a streaming minmax algo here, see def above.
-            if liv:
+            mx = max(price + tick_margin, mx)
-                mx = max(price + tick_margin, mx)
+            mn = min(price - tick_margin, mn)
                mn = min(price - tick_margin, mn)
            if typ in clear_types:
@ -520,8 +531,9 @@ def graphics_update_cycle(
                if wap_in_history:
                    # update vwap overlay line
-                    chart.update_graphics_from_flow(
+                    chart.update_graphics_from_array(
                        'bar_wap',
                        array,
                    )
            # L1 book label-line updates
@ -537,7 +549,7 @@ def graphics_update_cycle(
                if (
                    label is not None
-                    and liv
+                    # and liv
                ):
                    label.update_fields(
                        {'level': price, 'size': size}
@ -551,7 +563,7 @@ def graphics_update_cycle(
                typ in _tick_groups['asks']
                # TODO: instead we could check if the price is in the
                # y-view-range?
-                and liv
+                # and liv
            ):
                l1.ask_label.update_fields({'level': price, 'size': size})
@ -559,7 +571,7 @@ def graphics_update_cycle(
                typ in _tick_groups['bids']
                # TODO: instead we could check if the price is in the
                # y-view-range?
-                and liv
+                # and liv
            ):
                l1.bid_label.update_fields({'level': price, 'size': size})
@ -574,7 +586,6 @@ def graphics_update_cycle(
                main_vb._ic is None
                or not main_vb._ic.is_set()
            ):
                # print(f'updating range due to mxmn')
                main_vb._set_yrange(
                    # TODO: we should probably scale
                    # the view margin based on the size
@ -585,120 +596,21 @@ def graphics_update_cycle(
                    yrange=(mn, mx),
                )
-        # XXX: update this every draw cycle to make L1-always-in-view work.
+            vars['last_mx'], vars['last_mn'] = mx, mn
        vars['last_mx'], vars['last_mn'] = mx, mn
        # run synchronous update on all linked flows
        # TODO: should the "main" (aka source) flow be special?
        for curve_name, flow in chart._flows.items():
-            # update any overlayed fsp flows
+            # TODO: should the "main" (aka source) flow be special?
-            if curve_name != chart.data_key:
+            if curve_name == chart.data_key:
-                update_fsp_chart(
+                continue
                    chart,
                    flow,
                    curve_name,
                    array_key=curve_name,
                )
-            # even if we're downsampled bigly
+            update_fsp_chart(
-            # draw the last datum in the final
+                chart,
-            # px column to give the user the mx/mn
+                flow,
-            # range of that set.
+                curve_name,
-            if (
+                array_key=curve_name,
                not do_append
                # and not do_rt_update
                and liv
            ):
                flow.draw_last(
                    array_key=curve_name,
                    only_last_uppx=True,
                )
        # volume chart logic..
        # TODO: can we unify this with the above loop?
        if vlm_chart:
            # always update y-label
            ds.vlm_sticky.update_from_data(
                *array[-1][['index', 'volume']]
            )
            if (
                (
                    do_rt_update
                    or do_append
                    and liv
                )
                or trigger_all
            ):
                # TODO: make it so this doesn't have to be called
                # once the $vlm is up?
                vlm_chart.update_graphics_from_flow(
                    'volume',
                    # UGGGh, see ``maxmin()`` impl in `._fsp` for
                    # the overlayed plotitems... we need a better
                    # bay to invoke a maxmin per overlay..
                    render=False,
                    # XXX: ^^^^ THIS IS SUPER IMPORTANT! ^^^^
                    # without this, since we disable the
                    # 'volume' (units) chart after the $vlm starts
                    # up we need to be sure to enable this
                    # auto-ranging otherwise there will be no handler
                    # connected to update accompanying overlay
                    # graphics..
                )
                profiler('`vlm_chart.update_graphics_from_flow()`')
                if (
                    mx_vlm_in_view != vars['last_mx_vlm']
                ):
                    yrange = (0, mx_vlm_in_view * 1.375)
                    vlm_chart.view._set_yrange(
                        yrange=yrange,
                    )
                    profiler('`vlm_chart.view._set_yrange()`')
                    # print(f'mx vlm: {last_mx_vlm} -> {mx_vlm_in_view}')
                    vars['last_mx_vlm'] = mx_vlm_in_view
            for curve_name, flow in vlm_chart._flows.items():
                if (
                    curve_name != 'volume' and
                    flow.render and (
                        liv and
                        do_rt_update or do_append
                    )
                ):
                    update_fsp_chart(
                        vlm_chart,
                        flow,
                        curve_name,
                        array_key=curve_name,
                        # do_append=uppx < update_uppx,
                        do_append=do_append,
                    )
                    # is this even doing anything?
                    # (pretty sure it's the real-time
                    # resizing from last quote?)
                    fvb = flow.plot.vb
                    fvb._set_yrange(
                        name=curve_name,
                    )
                elif (
                    curve_name != 'volume'
                    and not do_append
                    and liv
                    and uppx >= 1
                    # even if we're downsampled bigly
                    # draw the last datum in the final
                    # px column to give the user the mx/mn
                    # range of that set.
                ):
                    # always update the last datum-element
                    # graphic for all flows
                    # print(f'drawing last {flow.name}')
                    flow.draw_last(array_key=curve_name)
 async def display_symbol_data(
    godwidget: GodWidget,
@ -723,7 +635,7 @@ async def display_symbol_data(
    )
    # historical data fetch
-    # brokermod = brokers.get_brokermod(provider)
+    brokermod = brokers.get_brokermod(provider)
    # ohlc_status_done = sbar.open_status(
    #     'retreiving OHLC history.. ',
@ -746,17 +658,11 @@ async def display_symbol_data(
        symbol = feed.symbols[sym]
        fqsn = symbol.front_fqsn()
        times = bars['time']
        end = pendulum.from_timestamp(times[-1])
        start = pendulum.from_timestamp(times[times != times[-1]][-1])
        step_size_s = (end - start).seconds
        tf_key = tf_in_1s[step_size_s]
        # load in symbol's ohlc data
        godwidget.window.setWindowTitle(
            f'{fqsn} '
            f'tick:{symbol.tick_size} '
-            f'step:{tf_key} '
+            f'step:1s '
        )
        linked = godwidget.linkedsplits
@ -772,31 +678,32 @@ async def display_symbol_data(
        # create main OHLC chart
        chart = linked.plot_ohlc_main(
            symbol,
-            ohlcv,
+            bars,
            sidepane=pp_pane,
        )
        chart.default_view()
        chart._feeds[symbol.key] = feed
        chart.setFocus()
        # plot historical vwap if available
        wap_in_history = False
-        # XXX: FOR SOME REASON THIS IS CAUSING HANGZ!?!
+        if brokermod._show_wap_in_history:
        # if brokermod._show_wap_in_history:
-        #     if 'bar_wap' in bars.dtype.fields:
+            if 'bar_wap' in bars.dtype.fields:
-        #         wap_in_history = True
+                wap_in_history = True
-        #         chart.draw_curve(
+                chart.draw_curve(
-        #             name='bar_wap',
+                    name='bar_wap',
-        #             shm=ohlcv,
+                    data=bars,
-        #             color='default_light',
+                    add_label=False,
-        #             add_label=False,
+                )
        #         )
        # size view to data once at outset
        chart.cv._set_yrange()
        # TODO: a data view api that makes this less shit
        chart._shm = ohlcv
        chart._flows[chart.data_key].shm = ohlcv
        # NOTE: we must immediately tell Qt to show the OHLC chart
        # to avoid a race where the subplots get added/shown to
        # the linked set *before* the main price chart!
@ -859,5 +766,6 @@ async def display_symbol_data(
                sbar._status_groups[loading_sym_key][1]()
                # let the app run.. bby
                chart.default_view()
                # linked.graphics_cycle()
                await trio.sleep_forever()
--- a/piker/ui/_editors.py
+++ b/piker/ui/_editors.py
@ -343,7 +343,7 @@ class SelectRect(QtGui.QGraphicsRectItem):
        nbars = ixmx - ixmn + 1
        chart = self._chart
-        data = chart._flows[chart.name].shm.array[ixmn:ixmx]
+        data = chart._arrays[chart.name][ixmn:ixmx]
        if len(data):
            std = data['close'].std()
--- a/piker/ui/_exec.py
+++ b/piker/ui/_exec.py
@ -49,6 +49,10 @@ from . import _style
 log = get_logger(__name__)
 # pyqtgraph global config
 # might as well enable this for now?
 pg.useOpenGL = True
 pg.enableExperimental = True
 # engage core tweaks that give us better response
 # latency then the average pg user
 _do_overrides()
--- a/piker/ui/_feedstatus.py
+++ b/piker/ui/_feedstatus.py
@ -1,83 +0,0 @@
 # piker: trading gear for hackers
 # Copyright (C) Tyler Goodlet (in stewardship for 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/>.
 """
 Feed status and controls widget(s) for embedding in a UI-pane.
 """
 from __future__ import annotations
 from textwrap import dedent
 from typing import TYPE_CHECKING
 # from PyQt5.QtCore import Qt
 from ._style import _font, _font_small
 # from ..calc import humanize
 from ._label import FormatLabel
 if TYPE_CHECKING:
    from ._chart import ChartPlotWidget
    from ..data.feed import Feed
    from ._forms import FieldsForm
 def mk_feed_label(
    form: FieldsForm,
    feed: Feed,
    chart: ChartPlotWidget,
 ) -> FormatLabel:
    '''
    Generate a label from feed meta-data to be displayed
    in a UI sidepane.
    TODO: eventually buttons for changing settings over
    a feed control protocol.
    '''
    status = feed.status
    assert status
    msg = dedent("""
        actor: **{actor_name}**\n
        |_ @**{host}:{port}**\n
    """)
    for key, val in status.items():
        if key in ('host', 'port', 'actor_name'):
            continue
        msg += f'\n|_ {key}: **{{{key}}}**\n'
    feed_label = FormatLabel(
        fmt_str=msg,
        # |_ streams: **{symbols}**\n
        font=_font.font,
        font_size=_font_small.px_size,
        font_color='default_lightest',
    )
    # form.vbox.setAlignment(feed_label, Qt.AlignBottom)
    # form.vbox.setAlignment(Qt.AlignBottom)
    _ = chart.height() - (
        form.height() +
        form.fill_bar.height()
        # feed_label.height()
    )
    feed_label.format(**feed.status)
    return feed_label
--- a/piker/ui/_flows.py
+++ b/piker/ui/_flows.py
--- a/piker/ui/_forms.py
+++ b/piker/ui/_forms.py
@ -750,12 +750,12 @@ def mk_order_pane_layout(
        parent=parent,
        fields_schema={
            'account': {
-                'label': '**accnt**:',
+                'label': '**account**:',
                'type': 'select',
                'default_value': ['paper'],
            },
            'size_unit': {
-                'label': '**alloc**:',
+                'label': '**allocate**:',
                'type': 'select',
                'default_value': [
                    '$ size',
--- a/piker/ui/_fsp.py
+++ b/piker/ui/_fsp.py
@ -75,7 +75,6 @@ def update_fsp_chart(
    flow,
    graphics_name: str,
    array_key: Optional[str],
    **kwargs,
 ) -> None:
@ -94,10 +93,10 @@ def update_fsp_chart(
    # update graphics
    # NOTE: this does a length check internally which allows it
    # staying above the last row check below..
-    chart.update_graphics_from_flow(
+    chart.update_graphics_from_array(
        graphics_name,
        array,
        array_key=array_key or graphics_name,
        **kwargs,
    )
    # XXX: re: ``array_key``: fsp func names must be unique meaning we
@ -107,6 +106,9 @@ def update_fsp_chart(
    # read from last calculated value and update any label
    last_val_sticky = chart._ysticks.get(graphics_name)
    if last_val_sticky:
        # array = shm.array[array_key]
        # if len(array):
        #     value = array[-1]
        last = last_row[array_key]
        last_val_sticky.update_from_data(-1, last)
@ -244,18 +246,20 @@ async def run_fsp_ui(
            chart.draw_curve(
                name=name,
-                shm=shm,
+                data=shm.array,
                overlay=True,
                color='default_light',
                array_key=name,
                **conf.get('chart_kwargs', {})
            )
            # specially store ref to shm for lookup in display loop
            chart._flows[name].shm = shm
        else:
            # create a new sub-chart widget for this fsp
            chart = linkedsplits.add_plot(
                name=name,
-                shm=shm,
+                array=shm.array,
                array_key=name,
                sidepane=sidepane,
@ -267,6 +271,12 @@ async def run_fsp_ui(
                **conf.get('chart_kwargs', {})
            )
            # XXX: ONLY for sub-chart fsps, overlays have their
            # data looked up from the chart's internal array set.
            # TODO: we must get a data view api going STAT!!
            chart._shm = shm
            chart._flows[chart.data_key].shm = shm
            # should **not** be the same sub-chart widget
            assert chart.name != linkedsplits.chart.name
@ -435,16 +445,12 @@ class FspAdmin:
            # wait for graceful shutdown signal
            async with stream.subscribe() as stream:
                async for msg in stream:
-                    info = msg.get('fsp_update')
+                    if msg == 'update':
                    if info:
                        # if the chart isn't hidden try to update
                        # the data on screen.
                        if not self.linked.isHidden():
-                            log.debug(f'Re-syncing graphics for fsp: {ns_path}')
+                            log.info(f'Re-syncing graphics for fsp: {ns_path}')
-                            self.linked.graphics_cycle(
+                            self.linked.graphics_cycle()
                                trigger_all=True,
                                prepend_update_index=info['first'],
                            )
                    else:
                        log.info(f'recved unexpected fsp engine msg: {msg}')
@ -620,7 +626,7 @@ async def open_vlm_displays(
        shm = ohlcv
        chart = linked.add_plot(
            name='volume',
-            shm=shm,
+            array=shm.array,
            array_key='volume',
            sidepane=sidepane,
@ -633,9 +639,10 @@ async def open_vlm_displays(
            # the curve item internals are pretty convoluted.
            style='step',
        )
        chart._flows['volume'].shm = ohlcv
        # force 0 to always be in view
-        def multi_maxmin(
+        def maxmin(
            names: list[str],
        ) -> tuple[float, float]:
@ -651,7 +658,7 @@ async def open_vlm_displays(
            return 0, mx
-        chart.view.maxmin = partial(multi_maxmin, names=['volume'])
+        chart.view.maxmin = partial(maxmin, names=['volume'])
        # TODO: fix the x-axis label issue where if you put
        # the axis on the left it's totally not lined up...
@ -659,6 +666,11 @@ async def open_vlm_displays(
        # chart.hideAxis('right')
        # chart.showAxis('left')
        # XXX: ONLY for sub-chart fsps, overlays have their
        # data looked up from the chart's internal array set.
        # TODO: we must get a data view api going STAT!!
        chart._shm = shm
        # send back new chart to caller
        task_status.started(chart)
@ -673,9 +685,9 @@ async def open_vlm_displays(
        last_val_sticky.update_from_data(-1, value)
-        vlm_curve = chart.update_graphics_from_flow(
+        vlm_curve = chart.update_graphics_from_array(
            'volume',
-            # shm.array,
+            shm.array,
        )
        # size view to data once at outset
@ -741,20 +753,19 @@ async def open_vlm_displays(
               'dolla_vlm',
               'dark_vlm',
            ]
-            # dvlm_rate_fields = [
+            dvlm_rate_fields = [
-            #     'dvlm_rate',
+                'dvlm_rate',
-            #     'dark_dvlm_rate',
+                'dark_dvlm_rate',
-            # ]
+            ]
            trade_rate_fields = [
                'trade_rate',
                'dark_trade_rate',
            ]
            group_mxmn = partial(
-                multi_maxmin,
+                maxmin,
                # keep both regular and dark vlm in view
-                names=fields,
+                names=fields + dvlm_rate_fields,
                # names=fields + dvlm_rate_fields,
            )
            # add custom auto range handler
@ -784,8 +795,9 @@ async def open_vlm_displays(
                        color = 'bracket'
                    curve, _ = chart.draw_curve(
                        # name='dolla_vlm',
                        name=name,
-                        shm=shm,
+                        data=shm.array,
                        array_key=name,
                        overlay=pi,
                        color=color,
@ -800,6 +812,7 @@ async def open_vlm_displays(
                    # ``.draw_curve()``.
                    flow = chart._flows[name]
                    assert flow.plot is pi
                    flow.shm = shm
            chart_curves(
                fields,
@ -821,11 +834,11 @@ async def open_vlm_displays(
            )
            await started.wait()
-            # chart_curves(
+            chart_curves(
-            #     dvlm_rate_fields,
+                dvlm_rate_fields,
-            #     dvlm_pi,
+                dvlm_pi,
-            #     fr_shm,
+                fr_shm,
-            # )
+            )
            # TODO: is there a way to "sync" the dual axes such that only
            # one curve is needed?
@ -834,9 +847,7 @@ async def open_vlm_displays(
            # liquidity events (well at least on low OHLC periods - 1s).
            vlm_curve.hide()
            chart.removeItem(vlm_curve)
-            vflow = chart._flows['volume']
+            chart._flows.pop('volume')
            vflow.render = False
            # avoid range sorting on volume once disabled
            chart.view.disable_auto_yrange()
@ -863,7 +874,7 @@ async def open_vlm_displays(
            )
            # add custom auto range handler
            tr_pi.vb.maxmin = partial(
-                multi_maxmin,
+                maxmin,
                # keep both regular and dark vlm in view
                names=trade_rate_fields,
            )
@ -891,10 +902,10 @@ async def open_vlm_displays(
        # built-in vlm fsps
        for target, conf in {
-            # tina_vwap: {
+            tina_vwap: {
-            #     'overlay': 'ohlc',  # overlays with OHLCV (main) chart
+                'overlay': 'ohlc',  # overlays with OHLCV (main) chart
-            #     'anchor': 'session',
+                'anchor': 'session',
-            # },
+            },
        }.items():
            started = await admin.open_fsp_chart(
                target,
--- a/piker/ui/_interaction.py
+++ b/piker/ui/_interaction.py
@ -20,6 +20,7 @@ Chart view box primitives
 """
 from __future__ import annotations
 from contextlib import asynccontextmanager
 # import itertools
 import time
 from typing import Optional, Callable
@ -34,9 +35,10 @@ import trio
 from ..log import get_logger
 from .._profile import pg_profile_enabled, ms_slower_then
-# from ._style import _min_points_to_show
+from ._style import _min_points_to_show
 from ._editors import SelectRect
 from . import _event
 from ._ohlc import BarItems
 log = get_logger(__name__)
@ -484,18 +486,15 @@ class ChartView(ViewBox):
        # don't zoom more then the min points setting
        l, lbar, rbar, r = chart.bars_range()
-        # vl = r - l
+        vl = r - l
-        # if ev.delta() > 0 and vl <= _min_points_to_show:
+        if ev.delta() > 0 and vl <= _min_points_to_show:
-        #     log.debug("Max zoom bruh...")
+            log.debug("Max zoom bruh...")
-        #     return
+            return
-        # if (
+        if ev.delta() < 0 and vl >= len(chart._arrays[chart.name]) + 666:
-        #     ev.delta() < 0
+            log.debug("Min zoom bruh...")
-        #     and vl >= len(chart._flows[chart.name].shm.array) + 666
+            return
        # ):
        #     log.debug("Min zoom bruh...")
        #     return
        # actual scaling factor
        s = 1.015 ** (ev.delta() * -1 / 20)  # self.state['wheelScaleFactor'])
@ -569,23 +568,11 @@ class ChartView(ViewBox):
            self._resetTarget()
            self.scaleBy(s, focal)
            # XXX: the order of the next 2 lines i'm pretty sure
            # matters, we want the resize to trigger before the graphics
            # update, but i gotta feelin that because this one is signal
            # based (and thus not necessarily sync invoked right away)
            # that calling the resize method manually might work better.
            self.sigRangeChangedManually.emit(mask)
-            # XXX: without this is seems as though sometimes
+            # self._ic.set()
-            # when zooming in from far out (and maybe vice versa?)
+            # self._ic = None
-            # the signal isn't being fired enough since if you pan
+            # self.chart.resume_all_feeds()
            # just after you'll see further downsampling code run
            # (pretty noticeable on the OHLC ds curve) but with this
            # that never seems to happen? Only question is how much this
            # "double work" is causing latency when these missing event
            # fires don't happen?
            self.maybe_downsample_graphics()
            ev.accept()
@ -747,8 +734,9 @@ class ChartView(ViewBox):
        # flag to prevent triggering sibling charts from the same linked
        # set from recursion errors.
-        autoscale_linked_plots: bool = False,
+        autoscale_linked_plots: bool = True,
        name: Optional[str] = None,
        # autoscale_overlays: bool = False,
    ) -> None:
        '''
@ -759,12 +747,9 @@ class ChartView(ViewBox):
        data set.
        '''
        name = self.name
        # print(f'YRANGE ON {name}')
        profiler = pg.debug.Profiler(
            msg=f'`ChartView._set_yrange()`: `{name}`',
            disabled=not pg_profile_enabled(),
-            ms_threshold=ms_slower_then,
+            gt=ms_slower_then,
            delayed=True,
        )
        set_range = True
@ -790,22 +775,45 @@ class ChartView(ViewBox):
        elif yrange is not None:
            ylow, yhigh = yrange
        # calculate max, min y values in viewable x-range from data.
        # Make sure min bars/datums on screen is adhered.
        else:
            br = bars_range or chart.bars_range()
            profiler(f'got bars range: {br}')
            # TODO: maybe should be a method on the
            # chart widget/item?
            # if False:
            # if autoscale_linked_plots:
            #     # avoid recursion by sibling plots
            #     linked = self.linkedsplits
            #     plots = list(linked.subplots.copy().values())
            #     main = linked.chart
            #     if main:
            #         plots.append(main)
            #     for chart in plots:
            #         if chart and not chart._static_yrange:
            #             chart.cv._set_yrange(
            #                 bars_range=br,
            #                 autoscale_linked_plots=False,
            #             )
            #     profiler('autoscaled linked plots')
        if set_range:
            # XXX: only compute the mxmn range
            # if none is provided as input!
            if not yrange:
-                # flow = chart._flows[name]
+                # XXX: only compute the mxmn range
                # if none is provided as input!
                yrange = self._maxmin()
                if yrange is None:
-                    log.warning(f'No yrange provided for {name}!?')
+                    log.warning(f'No yrange provided for {self.name}!?')
                    print(f"WTF NO YRANGE {name}")
                    return
            ylow, yhigh = yrange
-            profiler(f'callback ._maxmin(): {yrange}')
+            profiler(f'maxmin(): {yrange}')
            # view margins: stay within a % of the "true range"
            diff = yhigh - ylow
@ -822,8 +830,6 @@ class ChartView(ViewBox):
            self.setYRange(ylow, yhigh)
            profiler(f'set limits: {(ylow, yhigh)}')
        profiler.finish()
    def enable_auto_yrange(
        self,
        src_vb: Optional[ChartView] = None,
@ -837,9 +843,17 @@ class ChartView(ViewBox):
        if src_vb is None:
            src_vb = self
        # such that when a linked chart changes its range
        # this local view is also automatically changed and
        # resized to data.
        src_vb.sigXRangeChanged.connect(self._set_yrange)
        # splitter(s) resizing
        src_vb.sigResized.connect(self._set_yrange)
        # mouse wheel doesn't emit XRangeChanged
        src_vb.sigRangeChangedManually.connect(self._set_yrange)
        # TODO: a smarter way to avoid calling this needlessly?
        # 2 things i can think of:
        # - register downsample-able graphics specially and only
@ -850,16 +864,15 @@ class ChartView(ViewBox):
            self.maybe_downsample_graphics
        )
-        # mouse wheel doesn't emit XRangeChanged
+    def disable_auto_yrange(
-        src_vb.sigRangeChangedManually.connect(self._set_yrange)
+        self,
    ) -> None:
-        # src_vb.sigXRangeChanged.connect(self._set_yrange)
+        # self._chart._static_yrange = 'axis'
        # src_vb.sigXRangeChanged.connect(
        #     self.maybe_downsample_graphics
        # )
    def disable_auto_yrange(self) -> None:
        self.sigXRangeChanged.disconnect(
            self._set_yrange,
        )
        self.sigResized.disconnect(
            self._set_yrange,
        )
@ -870,11 +883,6 @@ class ChartView(ViewBox):
            self._set_yrange,
        )
        # self.sigXRangeChanged.disconnect(self._set_yrange)
        # self.sigXRangeChanged.disconnect(
        #     self.maybe_downsample_graphics
        # )
    def x_uppx(self) -> float:
        '''
        Return the "number of x units" within a single
@ -882,7 +890,7 @@ class ChartView(ViewBox):
        graphics items which are our children.
        '''
-        graphics = [f.graphics for f in self._chart._flows.values()]
+        graphics = list(self._chart._graphics.values())
        if not graphics:
            return 0
@ -893,21 +901,25 @@ class ChartView(ViewBox):
        else:
            return 0
-    def maybe_downsample_graphics(
+    def maybe_downsample_graphics(self):
-        self,
+
-        autoscale_overlays: bool = True,
+        uppx = self.x_uppx()
-    ):
+        if (
            # we probably want to drop this once we are "drawing in
            # view" for downsampled flows..
            uppx and uppx > 16
            and self._ic is not None
        ):
            # don't bother updating since we're zoomed out bigly and
            # in a pan-interaction, in which case we shouldn't be
            # doing view-range based rendering (at least not yet).
            # print(f'{uppx} exiting early!')
            return
        profiler = pg.debug.Profiler(
            msg=f'ChartView.maybe_downsample_graphics() for {self.name}',
            disabled=not pg_profile_enabled(),
-
+            gt=3,
-            # XXX: important to avoid not seeing underlying
+            delayed=True,
            # ``.update_graphics_from_flow()`` nested profiling likely
            # due to the way delaying works and garbage collection of
            # the profiler in the delegated method calls.
            ms_threshold=6,
            # ms_threshold=ms_slower_then,
        )
        # TODO: a faster single-loop-iterator way of doing this XD
@ -916,32 +928,19 @@ class ChartView(ViewBox):
        plots = linked.subplots | {chart.name: chart}
        for chart_name, chart in plots.items():
            for name, flow in chart._flows.items():
                graphics = flow.graphics
-                if (
+                use_vr = False
-                    not flow.render
+                if isinstance(graphics, BarItems):
-
+                    use_vr = True
                    # XXX: super important to be aware of this.
                    # or not flow.graphics.isVisible()
                ):
                    continue
                # pass in no array which will read and render from the last
                # passed array (normally provided by the display loop.)
-                chart.update_graphics_from_flow(
+                chart.update_graphics_from_array(
                    name,
-                    use_vr=True,
+                    use_vr=use_vr,
                    profiler=profiler,
                )
                profiler(f'range change updated {chart_name}:{name}')
-                # for each overlay on this chart auto-scale the
+        profiler.finish()
                # y-range to max-min values.
                if autoscale_overlays:
                    overlay = chart.pi_overlay
                    if overlay:
                        for pi in overlay.overlays:
                            pi.vb._set_yrange(
                                # TODO: get the range once up front...
                                # bars_range=br,
                            )
                    profiler('autoscaled linked plots')
                profiler(f'<{chart_name}>.update_graphics_from_flow({name})')
--- a/piker/ui/_ohlc.py
+++ b/piker/ui/_ohlc.py
@ -25,13 +25,17 @@ from typing import (
 import numpy as np
 import pyqtgraph as pg
 from numba import njit, float64, int64  # , optional
 from PyQt5 import QtCore, QtGui, QtWidgets
 from PyQt5.QtCore import QLineF, QPointF
-from PyQt5.QtGui import QPainterPath
+# from numba import types as ntypes
 # from ..data._source import numba_ohlc_dtype
 from .._profile import pg_profile_enabled, ms_slower_then
 from ._style import hcolor
 from ..log import get_logger
 from ._curve import FastAppendCurve
 from ._compression import ohlc_flatten
 if TYPE_CHECKING:
    from ._chart import LinkedSplits
@ -42,8 +46,7 @@ log = get_logger(__name__)
 def bar_from_ohlc_row(
    row: np.ndarray,
-    # 0.5 is no overlap between arms, 1.0 is full overlap
+    w: float
    w: float = 0.43
 ) -> tuple[QLineF]:
    '''
@ -81,11 +84,128 @@ def bar_from_ohlc_row(
    return [hl, o, c]
@njit(
    # TODO: for now need to construct this manually for readonly arrays, see
    # https://github.com/numba/numba/issues/4511
    # ntypes.tuple((float64[:], float64[:], float64[:]))(
    #     numba_ohlc_dtype[::1],  # contiguous
    #     int64,
    #     optional(float64),
    # ),
    nogil=True
 )
 def path_arrays_from_ohlc(
    data: np.ndarray,
    start: int64,
    bar_gap: float64 = 0.43,
 ) -> np.ndarray:
    '''
    Generate an array of lines objects from input ohlc data.
    '''
    size = int(data.shape[0] * 6)
    x = np.zeros(
        # data,
        shape=size,
        dtype=float64,
    )
    y, c = x.copy(), x.copy()
    # TODO: report bug for assert @
    # /home/goodboy/repos/piker/env/lib/python3.8/site-packages/numba/core/typing/builtins.py:991
    for i, q in enumerate(data[start:], start):
        # TODO: ask numba why this doesn't work..
        # open, high, low, close, index = q[
        #     ['open', 'high', 'low', 'close', 'index']]
        open = q['open']
        high = q['high']
        low = q['low']
        close = q['close']
        index = float64(q['index'])
        istart = i * 6
        istop = istart + 6
        # x,y detail the 6 points which connect all vertexes of a ohlc bar
        x[istart:istop] = (
            index - bar_gap,
            index,
            index,
            index,
            index,
            index + bar_gap,
        )
        y[istart:istop] = (
            open,
            open,
            low,
            high,
            close,
            close,
        )
        # specifies that the first edge is never connected to the
        # prior bars last edge thus providing a small "gap"/"space"
        # between bars determined by ``bar_gap``.
        c[istart:istop] = (1, 1, 1, 1, 1, 0)
    return x, y, c
 def gen_qpath(
    data: np.ndarray,
    start: int,  # XXX: do we need this?
    w: float,
    path: Optional[QtGui.QPainterPath] = None,
 ) -> QtGui.QPainterPath:
    path_was_none = path is None
    profiler = pg.debug.Profiler(
        msg='gen_qpath ohlc',
        disabled=not pg_profile_enabled(),
        gt=ms_slower_then,
    )
    x, y, c = path_arrays_from_ohlc(
        data,
        start,
        bar_gap=w,
    )
    profiler("generate stream with numba")
    # TODO: numba the internals of this!
    path = pg.functions.arrayToQPath(
        x,
        y,
        connect=c,
        path=path,
    )
    # avoid mem allocs if possible
    if path_was_none:
        path.reserve(path.capacity())
    profiler("generate path with arrayToQPath")
    return path
 class BarItems(pg.GraphicsObject):
    '''
    "Price range" bars graphics rendered from a OHLC sampled sequence.
    '''
    sigPlotChanged = QtCore.pyqtSignal(object)
    # 0.5 is no overlap between arms, 1.0 is full overlap
    w: float = 0.43
    def __init__(
        self,
        linked: LinkedSplits,
@ -105,13 +225,388 @@ class BarItems(pg.GraphicsObject):
        self.last_bar_pen = pg.mkPen(hcolor(last_bar_color), width=2)
        self._name = name
        self._ds_line_xy: Optional[
            tuple[np.ndarray, np.ndarray]
        ] = None
        # NOTE: this prevents redraws on mouse interaction which is
        # a huge boon for avg interaction latency.
        # TODO: one question still remaining is if this makes trasform
        # interactions slower (such as zooming) and if so maybe if/when
        # we implement a "history" mode for the view we disable this in
        # that mode?
        self.setCacheMode(QtWidgets.QGraphicsItem.DeviceCoordinateCache)
-        self.path = QPainterPath()
+
        self._pi = plotitem
        self.path = QtGui.QPainterPath()
        self.fast_path = QtGui.QPainterPath()
        self._xrange: tuple[int, int]
        self._yrange: tuple[float, float]
        self._vrange = None
        # TODO: don't render the full backing array each time
        # self._path_data = None
        self._last_bar_lines: Optional[tuple[QLineF, ...]] = None
        # track the current length of drawable lines within the larger array
        self.start_index: int = 0
        self.stop_index: int = 0
        # downsampler-line state
        self._in_ds: bool = False
        self._ds_line: Optional[FastAppendCurve] = None
        self._dsi: tuple[int, int] = 0, 0
        self._xs_in_px: float = 0
    def draw_from_data(
        self,
        ohlc: np.ndarray,
        start: int = 0,
    ) -> QtGui.QPainterPath:
        '''
        Draw OHLC datum graphics from a ``np.ndarray``.
        This routine is usually only called to draw the initial history.
        '''
        hist, last = ohlc[:-1], ohlc[-1]
        self.path = gen_qpath(hist, start, self.w)
        # save graphics for later reference and keep track
        # of current internal "last index"
        # self.start_index = len(ohlc)
        index = ohlc['index']
        self._xrange = (index[0], index[-1])
        self._yrange = (
            np.nanmax(ohlc['high']),
            np.nanmin(ohlc['low']),
        )
        # up to last to avoid double draw of last bar
        self._last_bar_lines = bar_from_ohlc_row(last, self.w)
        x, y = self._ds_line_xy = ohlc_flatten(ohlc)
        # TODO: figuring out the most optimial size for the ideal
        # curve-path by,
        # - calcing the display's max px width `.screen()`
        # - drawing a curve and figuring out it's capacity:
        #   https://doc.qt.io/qt-5/qpainterpath.html#capacity
        # - reserving that cap for each curve-mapped-to-shm with
        # - leveraging clearing when needed to redraw the entire
        #   curve that does not release mem allocs:
        #   https://doc.qt.io/qt-5/qpainterpath.html#clear
        curve = FastAppendCurve(
            y=y,
            x=x,
            name='OHLC',
            color=self._color,
        )
        curve.hide()
        self._pi.addItem(curve)
        self._ds_line = curve
        self._ds_xrange = (index[0], index[-1])
        # trigger render
        # https://doc.qt.io/qt-5/qgraphicsitem.html#update
        self.update()
        return self.path
    def x_uppx(self) -> int:
-        # we expect the downsample curve report this.
+        if self._ds_line:
-        return 0
+            return self._ds_line.x_uppx()
        else:
            return 0
    def update_from_array(
        self,
        # full array input history
        ohlc: np.ndarray,
        # pre-sliced array data that's "in view"
        ohlc_iv: np.ndarray,
        view_range: Optional[tuple[int, int]] = None,
        profiler: Optional[pg.debug.Profiler] = None,
    ) -> None:
        '''
        Update the last datum's bar graphic from input data array.
        This routine should be interface compatible with
        ``pg.PlotCurveItem.setData()``. Normally this method in
        ``pyqtgraph`` seems to update all the data passed to the
        graphics object, and then update/rerender, but here we're
        assuming the prior graphics havent changed (OHLC history rarely
        does) so this "should" be simpler and faster.
        This routine should be made (transitively) as fast as possible.
        '''
        profiler = profiler or pg.debug.Profiler(
            disabled=not pg_profile_enabled(),
            gt=ms_slower_then,
            delayed=True,
        )
        # index = self.start_index
        istart, istop = self._xrange
        ds_istart, ds_istop = self._ds_xrange
        index = ohlc['index']
        first_index, last_index = index[0], index[-1]
        # length = len(ohlc)
        # prepend_length = istart - first_index
        # append_length = last_index - istop
        # ds_prepend_length = ds_istart - first_index
        # ds_append_length = last_index - ds_istop
        flip_cache = False
        x_gt = 16
        if self._ds_line:
            uppx = self._ds_line.x_uppx()
        else:
            uppx = 0
        should_line = self._in_ds
        if (
            self._in_ds
            and uppx < x_gt
        ):
            should_line = False
        elif (
            not self._in_ds
            and uppx >= x_gt
        ):
            should_line = True
        profiler('ds logic complete')
        if should_line:
            # update the line graphic
            # x, y = self._ds_line_xy = ohlc_flatten(ohlc_iv)
            x, y = self._ds_line_xy = ohlc_flatten(ohlc)
            x_iv, y_iv = self._ds_line_xy = ohlc_flatten(ohlc_iv)
            profiler('flattening bars to line')
            # TODO: we should be diffing the amount of new data which
            # needs to be downsampled. Ideally we actually are just
            # doing all the ds-ing in sibling actors so that the data
            # can just be read and rendered to graphics on events of our
            # choice.
            # diff = do_diff(ohlc, new_bit)
            curve = self._ds_line
            curve.update_from_array(
                x=x,
                y=y,
                x_iv=x_iv,
                y_iv=y_iv,
                view_range=None,  # hack
                profiler=profiler,
            )
            profiler('updated ds line')
            if not self._in_ds:
                # hide bars and show line
                self.hide()
                # XXX: is this actually any faster?
                # self._pi.removeItem(self)
                # TODO: a `.ui()` log level?
                log.info(
                    f'downsampling to line graphic {self._name}'
                )
                # self._pi.addItem(curve)
                curve.show()
                curve.update()
                self._in_ds = True
            # stop here since we don't need to update bars path any more
            # as we delegate to the downsample line with updates.
            profiler.finish()
            # print('terminating early')
            return
        else:
            # we should be in bars mode
            if self._in_ds:
                # flip back to bars graphics and hide the downsample line.
                log.info(f'showing bars graphic {self._name}')
                curve = self._ds_line
                curve.hide()
                # self._pi.removeItem(curve)
                # XXX: is this actually any faster?
                # self._pi.addItem(self)
                self.show()
                self._in_ds = False
            # generate in_view path
            self.path = gen_qpath(
                ohlc_iv,
                0,
                self.w,
                # path=self.path,
            )
            # TODO: to make the downsampling faster
            # - allow mapping only a range of lines thus only drawing as
            #   many bars as exactly specified.
            # - move ohlc "flattening" to a shmarr
            # - maybe move all this embedded logic to a higher
            #   level type?
            # if prepend_length:
            #     # new history was added and we need to render a new path
            #     prepend_bars = ohlc[:prepend_length]
            # if ds_prepend_length:
            #     ds_prepend_bars = ohlc[:ds_prepend_length]
            #     pre_x, pre_y = ohlc_flatten(ds_prepend_bars)
            #     fx = np.concatenate((pre_x, fx))
            #     fy = np.concatenate((pre_y, fy))
            #     profiler('ds line prepend diff complete')
            # if append_length:
            #     # generate new graphics to match provided array
            #     # path appending logic:
            #     # we need to get the previous "current bar(s)" for the time step
            #     # and convert it to a sub-path to append to the historical set
            #     # new_bars = ohlc[istop - 1:istop + append_length - 1]
            #     append_bars = ohlc[-append_length - 1:-1]
            #     # print(f'ohlc bars to append size: {append_bars.size}\n')
            # if ds_append_length:
            #     ds_append_bars = ohlc[-ds_append_length - 1:-1]
            #     post_x, post_y = ohlc_flatten(ds_append_bars)
            #     print(
            #         f'ds curve to append sizes: {(post_x.size, post_y.size)}'
            #     )
            #     fx = np.concatenate((fx, post_x))
            #     fy = np.concatenate((fy, post_y))
            #     profiler('ds line append diff complete')
            profiler('array diffs complete')
            # does this work?
            last = ohlc[-1]
            # fy[-1] = last['close']
            # # incremental update and cache line datums
            # self._ds_line_xy = fx, fy
            # maybe downsample to line
            # ds = self.maybe_downsample()
            # if ds:
            #     # if we downsample to a line don't bother with
            #     # any more path generation / updates
            #     self._ds_xrange = first_index, last_index
            #     profiler('downsampled to line')
            #     return
            # print(in_view.size)
            # if self.path:
            #     self.path = path
            #     self.path.reserve(path.capacity())
            #     self.path.swap(path)
            # path updates
            # if prepend_length:
            #     # XXX: SOMETHING IS MAYBE FISHY HERE what with the old_path
            #     # y value not matching the first value from
            #     # ohlc[prepend_length + 1] ???
            #     prepend_path = gen_qpath(prepend_bars, 0, self.w)
            #     old_path = self.path
            #     self.path = prepend_path
            #     self.path.addPath(old_path)
            #     profiler('path PREPEND')
            # if append_length:
            #     append_path = gen_qpath(append_bars, 0, self.w)
            #     self.path.moveTo(
            #         float(istop - self.w),
            #         float(append_bars[0]['open'])
            #     )
            #     self.path.addPath(append_path)
            #     profiler('path APPEND')
            #     fp = self.fast_path
            #     if fp is None:
            #         self.fast_path = append_path
            #     else:
            #         fp.moveTo(
            #             float(istop - self.w), float(new_bars[0]['open'])
            #         )
            #         fp.addPath(append_path)
            #     self.setCacheMode(QtWidgets.QGraphicsItem.NoCache)
            #     flip_cache = True
            self._xrange = first_index, last_index
            # trigger redraw despite caching
            self.prepareGeometryChange()
            # generate new lines objects for updatable "current bar"
            self._last_bar_lines = bar_from_ohlc_row(last, self.w)
            # last bar update
            i, o, h, l, last, v = last[
                ['index', 'open', 'high', 'low', 'close', 'volume']
            ]
            # assert i == self.start_index - 1
            # assert i == last_index
            body, larm, rarm = self._last_bar_lines
            # XXX: is there a faster way to modify this?
            rarm.setLine(rarm.x1(), last, rarm.x2(), last)
            # writer is responsible for changing open on "first" volume of bar
            larm.setLine(larm.x1(), o, larm.x2(), o)
            if l != h:  # noqa
                if body is None:
                    body = self._last_bar_lines[0] = QLineF(i, l, i, h)
                else:
                    # update body
                    body.setLine(i, l, i, h)
                # XXX: pretty sure this is causing an issue where the bar has
                # a large upward move right before the next sample and the body
                # is getting set to None since the next bar is flat but the shm
                # array index update wasn't read by the time this code runs. Iow
                # we're doing this removal of the body for a bar index that is
                # now out of date / from some previous sample. It's weird
                # though because i've seen it do this to bars i - 3 back?
            profiler('last bar set')
            self.update()
            profiler('.update()')
            if flip_cache:
                self.setCacheMode(QtWidgets.QGraphicsItem.DeviceCoordinateCache)
            profiler.finish()
    def boundingRect(self):
        # Qt docs: https://doc.qt.io/qt-5/qgraphicsitem.html#boundingRect
@ -135,6 +630,16 @@ class BarItems(pg.GraphicsObject):
            hb.bottomRight(),
        )
        # fp = self.fast_path
        # if fp:
        #     fhb = fp.controlPointRect()
        #     print((hb_tl, hb_br))
        #     print(fhb)
        #     hb_tl, hb_br = (
        #         fhb.topLeft() + hb.topLeft(),
        #         fhb.bottomRight() + hb.bottomRight(),
        #     )
        # need to include last bar height or BR will be off
        mx_y = hb_br.y()
        mn_y = hb_tl.y()
@ -170,9 +675,12 @@ class BarItems(pg.GraphicsObject):
    ) -> None:
        if self._in_ds:
            return
        profiler = pg.debug.Profiler(
            disabled=not pg_profile_enabled(),
-            ms_threshold=ms_slower_then,
+            gt=ms_slower_then,
        )
        # p.setCompositionMode(0)
@ -184,67 +692,13 @@ class BarItems(pg.GraphicsObject):
        # lead to any perf gains other then when zoomed in to less bars
        # in view.
        p.setPen(self.last_bar_pen)
-        if self._last_bar_lines:
+        p.drawLines(*tuple(filter(bool, self._last_bar_lines)))
-            p.drawLines(*tuple(filter(bool, self._last_bar_lines)))
+        profiler('draw last bar')
            profiler('draw last bar')
        p.setPen(self.bars_pen)
        p.drawPath(self.path)
        profiler(f'draw history path: {self.path.capacity()}')
-    def draw_last_datum(
+        # if self.fast_path:
-        self,
+        #     p.drawPath(self.fast_path)
-        path: QPainterPath,
+        #     profiler('draw fast path')
        src_data: np.ndarray,
        render_data: np.ndarray,
        reset: bool,
        array_key: str,
        fields: list[str] = [
            'index',
            'open',
            'high',
            'low',
            'close',
        ],
    ) -> None:
        # relevant fields
        ohlc = src_data[fields]
        last_row = ohlc[-1:]
        # individual values
        last_row = i, o, h, l, last = ohlc[-1]
        # generate new lines objects for updatable "current bar"
        self._last_bar_lines = bar_from_ohlc_row(last_row)
        # assert i == graphics.start_index - 1
        # assert i == last_index
        body, larm, rarm = self._last_bar_lines
        # XXX: is there a faster way to modify this?
        rarm.setLine(rarm.x1(), last, rarm.x2(), last)
        # writer is responsible for changing open on "first" volume of bar
        larm.setLine(larm.x1(), o, larm.x2(), o)
        if l != h:  # noqa
            if body is None:
                body = self._last_bar_lines[0] = QLineF(i, l, i, h)
            else:
                # update body
                body.setLine(i, l, i, h)
            # XXX: pretty sure this is causing an issue where the
            # bar has a large upward move right before the next
            # sample and the body is getting set to None since the
            # next bar is flat but the shm array index update wasn't
            # read by the time this code runs. Iow we're doing this
            # removal of the body for a bar index that is now out of
            # date / from some previous sample. It's weird though
            # because i've seen it do this to bars i - 3 back?
        return ohlc['index'], ohlc['close']
--- a/piker/ui/_pathops.py
+++ b/piker/ui/_pathops.py
@ -1,236 +0,0 @@
 # piker: trading gear for hackers
 # Copyright (C) 2018-present  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/>.
 """
 Super fast ``QPainterPath`` generation related operator routines.
 """
 from __future__ import annotations
 from typing import (
    # Optional,
    TYPE_CHECKING,
 )
 import numpy as np
 from numpy.lib import recfunctions as rfn
 from numba import njit, float64, int64  # , optional
 # import pyqtgraph as pg
 from PyQt5 import QtGui
 # from PyQt5.QtCore import QLineF, QPointF
 from ..data._sharedmem import (
    ShmArray,
 )
 # from .._profile import pg_profile_enabled, ms_slower_then
 from ._compression import (
    ds_m4,
 )
 if TYPE_CHECKING:
    from ._flows import Renderer
 def xy_downsample(
    x,
    y,
    uppx,
    x_spacer: float = 0.5,
 ) -> tuple[np.ndarray, np.ndarray]:
    # downsample whenever more then 1 pixels per datum can be shown.
    # always refresh data bounds until we get diffing
    # working properly, see above..
    bins, x, y = ds_m4(
        x,
        y,
        uppx,
    )
    # flatten output to 1d arrays suitable for path-graphics generation.
    x = np.broadcast_to(x[:, None], y.shape)
    x = (x + np.array(
        [-x_spacer, 0, 0, x_spacer]
    )).flatten()
    y = y.flatten()
    return x, y
@njit(
    # TODO: for now need to construct this manually for readonly arrays, see
    # https://github.com/numba/numba/issues/4511
    # ntypes.tuple((float64[:], float64[:], float64[:]))(
    #     numba_ohlc_dtype[::1],  # contiguous
    #     int64,
    #     optional(float64),
    # ),
    nogil=True
 )
 def path_arrays_from_ohlc(
    data: np.ndarray,
    start: int64,
    bar_gap: float64 = 0.43,
 ) -> np.ndarray:
    '''
    Generate an array of lines objects from input ohlc data.
    '''
    size = int(data.shape[0] * 6)
    x = np.zeros(
        # data,
        shape=size,
        dtype=float64,
    )
    y, c = x.copy(), x.copy()
    # TODO: report bug for assert @
    # /home/goodboy/repos/piker/env/lib/python3.8/site-packages/numba/core/typing/builtins.py:991
    for i, q in enumerate(data[start:], start):
        # TODO: ask numba why this doesn't work..
        # open, high, low, close, index = q[
        #     ['open', 'high', 'low', 'close', 'index']]
        open = q['open']
        high = q['high']
        low = q['low']
        close = q['close']
        index = float64(q['index'])
        istart = i * 6
        istop = istart + 6
        # x,y detail the 6 points which connect all vertexes of a ohlc bar
        x[istart:istop] = (
            index - bar_gap,
            index,
            index,
            index,
            index,
            index + bar_gap,
        )
        y[istart:istop] = (
            open,
            open,
            low,
            high,
            close,
            close,
        )
        # specifies that the first edge is never connected to the
        # prior bars last edge thus providing a small "gap"/"space"
        # between bars determined by ``bar_gap``.
        c[istart:istop] = (1, 1, 1, 1, 1, 0)
    return x, y, c
 def gen_ohlc_qpath(
    r: Renderer,
    data: np.ndarray,
    array_key: str,  # we ignore this
    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 = 0.43,
 ) -> QtGui.QPainterPath:
    '''
    More or less direct proxy to ``path_arrays_from_ohlc()``
    but with closed in kwargs for line spacing.
    '''
    x, y, c = path_arrays_from_ohlc(
        data,
        start,
        bar_gap=w,
    )
    return x, y, c
 def ohlc_to_line(
    ohlc_shm: ShmArray,
    data_field: str,
    fields: list[str] = ['open', 'high', 'low', 'close']
 ) -> tuple[
    np.ndarray,
    np.ndarray,
 ]:
    '''
    Convert an input struct-array holding OHLC samples into a pair of
    flattened x, y arrays with the same size (datums wise) as the source
    data.
    '''
    y_out = ohlc_shm.ustruct(fields)
    first = ohlc_shm._first.value
    last = ohlc_shm._last.value
    # write pushed data to flattened copy
    y_out[first:last] = rfn.structured_to_unstructured(
        ohlc_shm.array[fields]
    )
    # generate an flat-interpolated x-domain
    x_out = (
        np.broadcast_to(
            ohlc_shm._array['index'][:, None],
            (
                ohlc_shm._array.size,
                # 4,  # only ohlc
                y_out.shape[1],
            ),
        ) + np.array([-0.5, 0, 0, 0.5])
    )
    assert y_out.any()
    return (
        x_out,
        y_out,
    )
 def to_step_format(
    shm: ShmArray,
    data_field: str,
    index_field: str = 'index',
 ) -> tuple[int, np.ndarray, np.ndarray]:
    '''
    Convert an input 1d shm array to a "step array" format
    for use by path graphics generation.
    '''
    i = shm._array['index'].copy()
    out = shm._array[data_field].copy()
    x_out = np.broadcast_to(
        i[:, None],
        (i.size, 2),
    ) + np.array([-0.5, 0.5])
    y_out = np.empty((len(out), 2), dtype=out.dtype)
    y_out[:] = out[:, np.newaxis]
    # start y at origin level
    y_out[0, 0] = 0
    return x_out, y_out
--- a/piker/ui/_window.py
+++ b/piker/ui/_window.py
@ -287,6 +287,7 @@ class MainWindow(QtGui.QMainWindow):
            app = QtGui.QApplication.instance()
            geo = self.current_screen().geometry()
            h, w = geo.height(), geo.width()
            self.setMaximumSize(w, h)
            # use approx 1/3 of the area of the screen by default
            self._size = round(w * .666), round(h * .666)
--- a/piker/ui/order_mode.py
+++ b/piker/ui/order_mode.py
@ -30,7 +30,6 @@ import uuid
 from pydantic import BaseModel
 import tractor
 import trio
 from PyQt5.QtCore import Qt
 from .. import config
 from ..clearing._client import open_ems, OrderBook
@ -38,7 +37,6 @@ from ..clearing._allocate import (
    mk_allocator,
    Position,
 )
 from ._style import _font
 from ..data._source import Symbol
 from ..data.feed import Feed
 from ..log import get_logger
@ -48,8 +46,7 @@ from ._position import (
    PositionTracker,
    SettingsPane,
 )
-from ._forms import FieldsForm
+from ._label import FormatLabel
 # from ._label import FormatLabel
 from ._window import MultiStatus
 from ..clearing._messages import Order, BrokerdPosition
 from ._forms import open_form_input_handling
@ -642,21 +639,63 @@ async def open_order_mode(
                pp_tracker.hide_info()
        # setup order mode sidepane widgets
-        form: FieldsForm = chart.sidepane
+        form = chart.sidepane
-        form.vbox.setSpacing(
+        vbox = form.vbox
        from textwrap import dedent
        from PyQt5.QtCore import Qt
        from ._style import _font, _font_small
        from ..calc import humanize
        feed_label = FormatLabel(
            fmt_str=dedent("""
            actor: **{actor_name}**\n
            |_ @**{host}:{port}**\n
            |_ throttle_hz: **{throttle_rate}**\n
            |_ streams: **{symbols}**\n
            |_ shm: **{shm}**\n
            """),
            font=_font.font,
            font_size=_font_small.px_size,
            font_color='default_lightest',
        )
        form.feed_label = feed_label
        # add feed info label to top
        vbox.insertWidget(
            0,
            feed_label,
            alignment=Qt.AlignBottom,
        )
        # vbox.setAlignment(feed_label, Qt.AlignBottom)
        # vbox.setAlignment(Qt.AlignBottom)
        _ = chart.height() - (
            form.height() +
            form.fill_bar.height()
            # feed_label.height()
        )
        vbox.setSpacing(
            int((1 + 5/8)*_font.px_size)
        )
-        from ._feedstatus import mk_feed_label
+        # fill in brokerd feed info
-
+        host, port = feed.portal.channel.raddr
-        feed_label = mk_feed_label(
+        if host == '127.0.0.1':
-            form,
+            host = 'localhost'
-            feed,
+        mpshm = feed.shm._shm
-            chart,
+        shmstr = f'{humanize(mpshm.size)}'
        form.feed_label.format(
            actor_name=feed.portal.channel.uid[0],
            host=host,
            port=port,
            symbols=len(feed.symbols),
            shm=shmstr,
            throttle_rate=feed.throttle_rate,
        )
        # XXX: we set this because?
        form.feed_label = feed_label
        order_pane = SettingsPane(
            form=form,
            # XXX: ugh, so hideous...
@ -667,11 +706,6 @@ async def open_order_mode(
        )
        order_pane.set_accounts(list(trackers.keys()))
        form.vbox.addWidget(
            feed_label,
            alignment=Qt.AlignBottom,
        )
        # update pp icons
        for name, tracker in trackers.items():
            order_pane.update_account_icons({name: tracker.live_pp})
@ -873,9 +907,7 @@ async def process_trades_and_update_ui(
                mode.lines.remove_line(uuid=oid)
        # each clearing tick is responded individually
-        elif resp in (
+        elif resp in ('broker_filled',):
            'broker_filled',
        ):
            known_order = book._sent_orders.get(oid)
            if not known_order:
--- a/requirements.txt
+++ b/requirements.txt
@ -7,15 +7,3 @@
 # pin this to a dev branch that we have more control over especially
 # as more graphics stuff gets hashed out.
 -e git+https://github.com/pikers/pyqtgraph.git@piker_pin#egg=pyqtgraph
 # our async client for ``marketstore`` (the tsdb)
 -e git+https://github.com/pikers/anyio-marketstore.git@master#egg=anyio-marketstore
 # ``trimeter`` for asysnc history fetching
 -e git+https://github.com/python-trio/trimeter.git@master#egg=trimeter
 # ``asyncvnc`` for sending interactions to ib-gw inside docker
 -e git+https://github.com/pikers/asyncvnc.git@vid_passthrough#egg=asyncvnc
--- a/setup.py
+++ b/setup.py
@ -51,55 +51,42 @@ setup(
        # async
        'trio',
        'trio-websocket',
        # 'tractor',  # from github currently
        'msgspec',  # performant IPC messaging
        'async_generator',
        # from github currently (see requirements.txt)
        # 'trimeter',  # not released yet..
        # 'tractor',
        # asyncvnc,
        # brokers
        'asks==2.4.8',
        'ib_insync',
        # numerics
-        'pendulum', # easier datetimes
+        'arrow',  # better datetimes
        'bidict',  # 2 way map
        'cython',
        'numpy',
        'numba',
        'pandas',
        # UI
        'PyQt5',
-        # 'pyqtgraph',  from our fork see reqs.txt
+        'pyqtgraph',
-        'qdarkstyle >= 3.0.2',  # themeing
+        'qdarkstyle >= 3.0.2',
-        'fuzzywuzzy[speedup]',  # fuzzy search
+        # fuzzy search
        'fuzzywuzzy[speedup]',
        # tsdbs
-        # anyio-marketstore  # from gh see reqs.txt
+        'pymarketstore',
    ],
    extras_require={
        'tsdb': [
            'docker',
        ],
    },
    tests_require=['pytest'],
-    python_requires=">=3.10",
+    python_requires=">=3.9",  # literally for ``datetime.datetime.fromisoformat``...
-    keywords=[
+    keywords=["async", "trading", "finance", "quant", "charting"],
        "async",
        "trading",
        "finance",
        "quant",
        "charting",
    ],
    classifiers=[
        'Development Status :: 3 - Alpha',
        'License :: OSI Approved :: ',
        'Operating System :: POSIX :: Linux',
        "Programming Language :: Python :: Implementation :: CPython",
        "Programming Language :: Python :: 3 :: Only",
        "Programming Language :: Python :: 3.9",
        "Programming Language :: Python :: 3.10",
        'Intended Audience :: Financial and Insurance Industry',
        'Intended Audience :: Science/Research',