piker/piker/ui/_fsp.py

977 lines
28 KiB
Python

# 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/>.
'''
FSP UI and graphics components.
Financial signal processing cluster and real-time graphics management.
'''
from contextlib import asynccontextmanager as acm
from functools import partial
from itertools import cycle
from typing import Optional, AsyncGenerator, Any
import numpy as np
from pydantic import create_model
import tractor
import pyqtgraph as pg
import trio
from trio_typing import TaskStatus
from ._axes import PriceAxis
from .._cacheables import maybe_open_context
from ..calc import humanize
from ..data._sharedmem import (
ShmArray,
_Token,
try_read,
)
from ._chart import (
ChartPlotWidget,
LinkedSplits,
)
from ._forms import (
FieldsForm,
mk_form,
open_form_input_handling,
)
from ..fsp._api import maybe_mk_fsp_shm, Fsp
from ..fsp import cascade
from ..fsp._volume import (
tina_vwap,
dolla_vlm,
flow_rates,
)
from ..log import get_logger
log = get_logger(__name__)
def has_vlm(ohlcv: ShmArray) -> bool:
# make sure that the instrument supports volume history
# (sometimes this is not the case for some commodities and
# derivatives)
vlm = ohlcv.array['volume']
return not bool(np.all(np.isin(vlm, -1)) or np.all(np.isnan(vlm)))
def update_fsp_chart(
chart: ChartPlotWidget,
shm: ShmArray,
graphics_name: str,
array_key: Optional[str],
) -> None:
array = shm.array
last_row = try_read(array)
# guard against unreadable case
if not last_row:
log.warning(f'Read-race on shm array: {graphics_name}@{shm.token}')
return
# update graphics
# NOTE: this does a length check internally which allows it
# staying above the last row check below..
chart.update_curve_from_array(
graphics_name,
array,
array_key=array_key or graphics_name,
)
# XXX: re: ``array_key``: fsp func names must be unique meaning we
# can't have duplicates of the underlying data even if multiple
# sub-charts reference it under different 'named charts'.
# 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)
@acm
async def open_fsp_sidepane(
linked: LinkedSplits,
conf: dict[str, dict[str, str]],
) -> FieldsForm:
schema = {}
assert len(conf) == 1 # for now
# add (single) selection widget
for name, config in conf.items():
schema[name] = {
'label': '**fsp**:',
'type': 'select',
'default_value': [name],
}
# add parameters for selection "options"
params = config.get('params', {})
for name, config in params.items():
default = config['default_value']
kwargs = config.get('widget_kwargs', {})
# add to ORM schema
schema.update({
name: {
'label': f'**{name}**:',
'type': 'edit',
'default_value': default,
'kwargs': kwargs,
},
})
sidepane: FieldsForm = mk_form(
parent=linked.godwidget,
fields_schema=schema,
)
# https://pydantic-docs.helpmanual.io/usage/models/#dynamic-model-creation
FspConfig = create_model(
'FspConfig',
name=name,
**params,
)
sidepane.model = FspConfig()
# just a logger for now until we get fsp configs up and running.
async def settings_change(key: str, value: str) -> bool:
print(f'{key}: {value}')
return True
# TODO:
async with (
open_form_input_handling(
sidepane,
focus_next=linked.godwidget,
on_value_change=settings_change,
)
):
yield sidepane
@acm
async def open_fsp_actor_cluster(
names: list[str] = ['fsp_0', 'fsp_1'],
) -> AsyncGenerator[int, dict[str, tractor.Portal]]:
from tractor._clustering import open_actor_cluster
# profiler = pg.debug.Profiler(
# delayed=False,
# disabled=False
# )
async with open_actor_cluster(
count=2,
names=names,
modules=['piker.fsp._engine'],
) as cluster_map:
# profiler('started fsp cluster')
yield cluster_map
async def run_fsp_ui(
linkedsplits: LinkedSplits,
shm: ShmArray,
started: trio.Event,
target: Fsp,
conf: dict[str, dict],
loglevel: str,
# profiler: pg.debug.Profiler,
# _quote_throttle_rate: int = 58,
) -> None:
'''
Taskf for UI spawning around a ``LinkedSplits`` chart for fsp
related graphics/UX management.
This is normally spawned/called once for each entry in the fsp
config.
'''
name = target.name
# profiler(f'started UI task for fsp: {name}')
async with (
# side UI for parameters/controls
open_fsp_sidepane(
linkedsplits,
{name: conf},
) as sidepane,
):
await started.wait()
# profiler(f'fsp:{name} attached to fsp ctx-stream')
overlay_with = conf.get('overlay', False)
if overlay_with:
if overlay_with == 'ohlc':
chart = linkedsplits.chart
else:
chart = linkedsplits.subplots[overlay_with]
chart.draw_curve(
name=name,
data=shm.array,
overlay=True,
color='default_light',
array_key=name,
separate_axes=conf.get('separate_axes', False),
**conf.get('chart_kwargs', {})
)
# specially store ref to shm for lookup in display loop
chart._overlays[name] = shm
else:
# create a new sub-chart widget for this fsp
chart = linkedsplits.add_plot(
name=name,
array=shm.array,
array_key=name,
sidepane=sidepane,
# curve by default
ohlc=False,
# settings passed down to ``ChartPlotWidget``
**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
# should **not** be the same sub-chart widget
assert chart.name != linkedsplits.chart.name
array_key = name
# profiler(f'fsp:{name} chart created')
# first UI update, usually from shm pushed history
update_fsp_chart(
chart,
shm,
name,
array_key=array_key,
)
chart.linked.focus()
# TODO: figure out if we can roll our own `FillToThreshold` to
# get brush filled polygons for OS/OB conditions.
# ``pg.FillBetweenItems`` seems to be one technique using
# generic fills between curve types while ``PlotCurveItem`` has
# logic inside ``.paint()`` for ``self.opts['fillLevel']`` which
# might be the best solution?
# graphics = chart.update_from_array(chart.name, array[name])
# graphics.curve.setBrush(50, 50, 200, 100)
# graphics.curve.setFillLevel(50)
# if func_name == 'rsi':
# from ._lines import level_line
# # add moveable over-[sold/bought] lines
# # and labels only for the 70/30 lines
# level_line(chart, 20)
# level_line(chart, 30, orient_v='top')
# level_line(chart, 70, orient_v='bottom')
# level_line(chart, 80, orient_v='top')
chart.view._set_yrange()
# done() # status updates
# profiler(f'fsp:{func_name} starting update loop')
# profiler.finish()
# update chart graphics
# last = time.time()
# XXX: this currently doesn't loop since
# the FSP engine does **not** push updates atm
# since we do graphics update in the main loop
# in ``._display.
# async for value in stream:
# print(value)
# # chart isn't actively shown so just skip render cycle
# if chart.linked.isHidden():
# continue
# else:
# now = time.time()
# period = now - last
# if period <= 1/_quote_throttle_rate:
# # faster then display refresh rate
# print(f'fsp too fast: {1/period}')
# continue
# # run synchronous update
# update_fsp_chart(
# chart,
# shm,
# display_name,
# array_key=func_name,
# )
# # set time of last graphics update
# last = time.time()
class FspAdmin:
'''
Client API for orchestrating FSP actors and displaying
real-time graphics output.
'''
def __init__(
self,
tn: trio.Nursery,
cluster: dict[str, tractor.Portal],
linked: LinkedSplits,
src_shm: ShmArray,
) -> None:
self.tn = tn
self.cluster = cluster
self.linked = linked
self._rr_next_actor = cycle(cluster.items())
self._registry: dict[
tuple,
tuple[tractor.MsgStream, ShmArray]
] = {}
self._flow_registry: dict[_Token, str] = {}
self.src_shm = src_shm
def rr_next_portal(self) -> tractor.Portal:
name, portal = next(self._rr_next_actor)
return portal
async def open_chain(
self,
portal: tractor.Portal,
complete: trio.Event,
started: trio.Event,
dst_shm: ShmArray,
conf: dict,
target: Fsp,
loglevel: str,
) -> None:
'''
Task which opens a remote FSP endpoint in the managed
cluster and sleeps until signalled to exit.
'''
brokername, sym = self.linked.symbol.front_feed()
ns_path = str(target.ns_path)
async with (
portal.open_context(
# chaining entrypoint
cascade,
# data feed key
brokername=brokername,
symbol=sym,
# mems
src_shm_token=self.src_shm.token,
dst_shm_token=dst_shm.token,
# target
ns_path=ns_path,
loglevel=loglevel,
zero_on_step=conf.get('zero_on_step', False),
shm_registry=[
(token.as_msg(), fsp_name, dst_token.as_msg())
for (token, fsp_name), dst_token
in self._flow_registry.items()
],
) as (ctx, last_index),
ctx.open_stream() as stream,
):
# register output data
self._registry[
(brokername, sym, ns_path)
] = (
stream,
dst_shm,
complete
)
started.set()
# wait for graceful shutdown signal
await complete.wait()
async def start_engine_task(
self,
target: Fsp,
conf: dict[str, dict[str, Any]],
worker_name: Optional[str] = None,
loglevel: str = 'info',
) -> (ShmArray, trio.Event):
fqsn = self.linked.symbol.front_feed()
# allocate an output shm array
key, dst_shm, opened = maybe_mk_fsp_shm(
'.'.join(fqsn),
target=target,
readonly=True,
)
self._flow_registry[
(self.src_shm._token, target.name)
] = dst_shm._token
# if not opened:
# raise RuntimeError(
# f'Already started FSP `{fqsn}:{func_name}`'
# )
portal = self.cluster.get(worker_name) or self.rr_next_portal()
complete = trio.Event()
started = trio.Event()
self.tn.start_soon(
self.open_chain,
portal,
complete,
started,
dst_shm,
conf,
target,
loglevel,
)
return dst_shm, started
async def open_fsp_chart(
self,
target: Fsp,
conf: dict, # yeah probably dumb..
loglevel: str = 'error',
) -> (trio.Event, ChartPlotWidget):
shm, started = await self.start_engine_task(
target,
conf,
loglevel,
)
# init async
self.tn.start_soon(
partial(
run_fsp_ui,
self.linked,
shm,
started,
target,
conf=conf,
loglevel=loglevel,
)
)
return started
@acm
async def open_fsp_admin(
linked: LinkedSplits,
src_shm: ShmArray,
**kwargs,
) -> AsyncGenerator[dict, dict[str, tractor.Portal]]:
async with (
maybe_open_context(
# for now make a cluster per client?
acm_func=open_fsp_actor_cluster,
kwargs=kwargs,
) as (cache_hit, cluster_map),
trio.open_nursery() as tn,
):
if cache_hit:
log.info('re-using existing fsp cluster')
admin = FspAdmin(
tn,
cluster_map,
linked,
src_shm,
)
try:
yield admin
finally:
# terminate all tasks via signals
for key, entry in admin._registry.items():
_, _, event = entry
event.set()
async def open_vlm_displays(
linked: LinkedSplits,
ohlcv: ShmArray,
dvlm: bool = True,
task_status: TaskStatus[ChartPlotWidget] = trio.TASK_STATUS_IGNORED,
) -> ChartPlotWidget:
'''
Volume subchart displays.
Since "volume" is often included directly alongside OHLCV price
data, we don't really need a separate FSP-actor + shm array for it
since it's likely already directly adjacent to OHLC samples from the
data provider.
Further only if volume data is detected (it sometimes isn't provided
eg. forex, certain commodities markets) will volume dependent FSPs
be spawned here.
'''
async with (
open_fsp_sidepane(
linked, {
'vlm': {
'params': {
'price_func': {
'default_value': 'chl3',
# tell target ``Edit`` widget to not allow
# edits for now.
'widget_kwargs': {'readonly': True},
},
},
}
},
) as sidepane,
open_fsp_admin(linked, ohlcv) as admin,
):
# built-in vlm which we plot ASAP since it's
# usually data provided directly with OHLC history.
shm = ohlcv
chart = linked.add_plot(
name='volume',
array=shm.array,
array_key='volume',
sidepane=sidepane,
# curve by default
ohlc=False,
# Draw vertical bars from zero.
# we do this internally ourselves since
# the curve item internals are pretty convoluted.
style='step',
)
# force 0 to always be in view
def maxmin(
names: list[str],
) -> tuple[float, float]:
mx = 0
for name in names:
mxmn = chart.maxmin(name=name)
if mxmn:
mx = max(mxmn[1], mx)
# if mx:
# return 0, mxmn[1]
return 0, mx
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...
# show volume units value on LHS (for dinkus)
# 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)
# should **not** be the same sub-chart widget
assert chart.name != linked.chart.name
# sticky only on sub-charts atm
last_val_sticky = chart._ysticks[chart.name]
# read from last calculated value
value = shm.array['volume'][-1]
last_val_sticky.update_from_data(-1, value)
vlm_curve = chart.update_curve_from_array(
'volume',
shm.array,
)
# size view to data once at outset
chart.view._set_yrange()
# add axis title
axis = chart.getAxis('right')
axis.set_title(' vlm')
if dvlm:
tasks_ready = []
# spawn and overlay $ vlm on the same subchart
dvlm_shm, started = await admin.start_engine_task(
dolla_vlm,
{ # fsp engine conf
'func_name': 'dolla_vlm',
'zero_on_step': True,
'params': {
'price_func': {
'default_value': 'chl3',
},
},
},
# loglevel,
)
tasks_ready.append(started)
# FIXME: we should error on starting the same fsp right
# since it might collide with existing shm.. or wait we
# had this before??
# dolla_vlm,
tasks_ready.append(started)
# profiler(f'created shm for fsp actor: {display_name}')
# wait for all engine tasks to startup
async with trio.open_nursery() as n:
for event in tasks_ready:
n.start_soon(event.wait)
###################
# dolla vlm overlay
###################
dvlm_pi = chart.overlay_plotitem(
'dolla_vlm',
index=0, # place axis on inside (nearest to chart)
axis_title=' $vlm',
axis_side='right',
axis_kwargs={
'typical_max_str': ' 100.0 M ',
'formatter': partial(
humanize,
digits=2,
),
},
)
# add custom auto range handler
dvlm_pi.vb._maxmin = partial(
maxmin,
# keep both regular and dark vlm in view
names=[
'dolla_vlm',
'dark_vlm',
'dvlm_rate',
'dark_dvlm_rate',
],
)
curve, _ = chart.draw_curve(
name='dolla_vlm',
data=dvlm_shm.array,
array_key='dolla_vlm',
overlay=dvlm_pi,
step_mode=True,
)
# TODO: is there a way to "sync" the dual axes such that only
# one curve is needed?
# hide the original vlm curve since the $vlm one is now
# displayed and the curves are effectively the same minus
# liquidity events (well at least on low OHLC periods - 1s).
vlm_curve.hide()
# TODO: we need a better API to do this..
# specially store ref to shm for lookup in display loop
# since only a placeholder of `None` is entered in
# ``.draw_curve()``.
chart._overlays['dolla_vlm'] = dvlm_shm
################
# dark vlm curve
################
# darker theme hue (obvsly)
dark_vlm_color = 'charcoal'
curve, _ = chart.draw_curve(
name='dark_vlm',
data=dvlm_shm.array,
array_key='dark_vlm',
overlay=dvlm_pi,
color=dark_vlm_color,
step_mode=True,
# **conf.get('chart_kwargs', {})
)
chart._overlays['dark_vlm'] = dvlm_shm
# spawn flow rates fsp **ONLY AFTER** the 'dolla_vlm' fsp is
# up since this one depends on it.
fr_shm, started = await admin.start_engine_task(
flow_rates,
{ # fsp engine conf
'func_name': 'flow_rates',
},
# loglevel,
)
await started.wait()
# curve, _ = chart.draw_curve(
# name='1m_vlm_rate',
# data=fr_shm.array,
# array_key='1m_vlm_rate',
# overlay=fr_pi,
# color='jet',
# style='solid',
# )
# chart._overlays['1m_vlm_rate'] = fr_shm
# use slightly less light (then bracket) gray
# for volume from "main exchange".
vlm_color = 'i3'
curve, _ = chart.draw_curve(
name='dvlm_rate',
data=fr_shm.array,
array_key='dvlm_rate',
overlay=dvlm_pi,
color=vlm_color,
style='solid',
)
chart._overlays['dvlm_rate'] = fr_shm
curve, _ = chart.draw_curve(
name='dark_dvlm_rate',
data=fr_shm.array,
array_key='dark_dvlm_rate',
overlay=dvlm_pi,
color=dark_vlm_color,
style='solid',
)
chart._overlays['dark_dvlm_rate'] = fr_shm
# vlm rate overlay
####################
# (needs separate axis since trade counts are likely
# different scale then vlm)
# vlmrate_pi = chart.overlay_plotitem(
# 'vlm_rates',
# index=0, # place axis on inside (nearest to chart)
# # NOTE: we might want to suffix with a \w
# # on lhs and prefix for the rhs axis labels?
# axis_title=' vlm/m',
# axis_side='left',
# axis_kwargs={
# 'typical_max_str': ' 100.0 M ',
# 'formatter': partial(
# humanize,
# digits=2,
# ),
# 'text_color': vlm_color,
# },
# )
# # add custom auto range handler
# vlmrate.vb._maxmin = partial(
# maxmin,
# # keep both regular and dark vlm in view
# names=[
# # '1m_vlm_rate',
# ],
# )
####################
# Trade rate overlay
####################
tr_pi = chart.overlay_plotitem(
'trade_rates',
index=1, # place axis on inside (nearest to chart)
# TODO: dynamically update period (and thus this axis?)
# title from user input.
axis_title='clears/P',
axis_side='left',
axis_kwargs={
'typical_max_str': ' 10.0 M ',
'formatter': partial(
humanize,
digits=2,
),
'text_color': vlm_color,
},
)
fields = [
'trade_rate',
'dark_trade_rate',
# '1m_trade_rate',
]
# add custom auto range handler
tr_pi.vb._maxmin = partial(
maxmin,
# keep both regular and dark vlm in view
names=fields,
)
for field in fields:
if 'dark' in field:
color = dark_vlm_color
else:
color = vlm_color
curve, _ = chart.draw_curve(
name=field,
data=fr_shm.array,
array_key=field,
overlay=tr_pi,
color=color,
# dashed line to represent "individual trades" being
# more "granular" B)
style='dash',
)
chart._overlays[field] = fr_shm
for pi in (dvlm_pi, tr_pi):
for name, axis_info in pi.axes.items():
# lol this sux XD
axis = axis_info['item']
if isinstance(axis, PriceAxis):
axis.size_to_values()
# built-in vlm fsps
for target, conf in {
tina_vwap: {
'overlay': 'ohlc', # overlays with OHLCV (main) chart
'anchor': 'session',
},
}.items():
started = await admin.open_fsp_chart(
target,
conf,
)
async def start_fsp_displays(
linked: LinkedSplits,
ohlcv: ShmArray,
group_status_key: str,
loglevel: str,
) -> None:
'''
Create fsp charts from a config input attached to a local actor
compute cluster.
Pass target entrypoint and historical data via ``ShmArray``.
'''
linked.focus()
# TODO: eventually we'll support some kind of n-compose syntax
fsp_conf = {
# 'rsi': {
# 'func_name': 'rsi', # literal python func ref lookup name
# # map of parameters to place on the fsp sidepane widget
# # which should map to dynamic inputs available to the
# # fsp function at runtime.
# 'params': {
# 'period': {
# 'default_value': 14,
# 'widget_kwargs': {'readonly': True},
# },
# },
# # ``ChartPlotWidget`` options passthrough
# 'chart_kwargs': {
# 'static_yrange': (0, 100),
# },
# },
}
profiler = pg.debug.Profiler(
delayed=False,
disabled=False
)
async with (
# NOTE: this admin internally opens an actor cluster
open_fsp_admin(linked, ohlcv) as admin,
):
statuses = []
for target, conf in fsp_conf.items():
started = await admin.open_fsp_chart(
target,
conf,
)
done = linked.window().status_bar.open_status(
f'loading fsp, {target}..',
group_key=group_status_key,
)
statuses.append((started, done))
for fsp_loaded, status_cb in statuses:
await fsp_loaded.wait()
profiler(f'attached to fsp portal: {target}')
status_cb()
# blocks on nursery until all fsp actors complete