Commit Graph

3099 Commits (7c863b50e9af55717dce7724facf882dfc86751e)

Author SHA1 Message Date
goodboy 087a34f061
Merge pull request #367 from pikers/livenpaper
`ib`: live & paper accounts together, infra refinements
2022-08-31 18:15:39 -04:00
Tyler Goodlet 653f5c824b Drop empty vnc server script idea for live account 2022-08-31 17:45:02 -04:00
Tyler Goodlet f9217570ab Add intiial `ib` backend readme 2022-08-31 17:38:24 -04:00
Tyler Goodlet 7f224f0342 Doc string typos 2022-08-31 17:22:15 -04:00
Tyler Goodlet 75a5f3795a I guess go back to doing vnc servers on both? 2022-08-31 17:22:15 -04:00
Tyler Goodlet de9f215c83 If more then one `ib` api client is available use next available for search 2022-08-31 17:22:15 -04:00
Tyler Goodlet 848e345364 POC using paper-in-docker gw for symbol search 2022-08-31 17:22:15 -04:00
Tyler Goodlet 38b190e598 Add `ib` `Crypto` contract support 2022-08-31 17:22:15 -04:00
Tyler Goodlet 3a9bc8058f Spawn a live account gateway alongside paper
This is like, super first-draft-y (and ideally we move to offering an
`piker.data._ahab` super for this) but, it's a start at allowing easy
setup of both paper and live `ib-gw` container spawning. We Expect the
user to input creds for the live account manually and the vnc server is
(hackily) only run inside the paper instance which most of the time
seems to make it possible click on the live gui window and input creds
manually.

We also add extra files for the live instance:
- a `dockering/ib/run_x11_vnc_live.sh` which is is a blank script
  that avoids running an `x11vnc` server in the line account cntr.
- a `dockering/ib/jts_live.ini` config which manually sets the live
  gw to use the `4001` port for api connections.

Further config tweaks:
- IBC: drop the api dynamic port override, decrease login display
  timeout to the riskier but likely to be faster 20s.
- `x11vnc` cmd: got back to using `rfbport` instead of `autoport`, drop
  `-logappend` so we see logging on docker console again, drop the
  frame cacheing flags and add in some x-hack disable flags.
2022-08-31 17:22:15 -04:00
Guillermo Rodriguez 739a231afc
Merge pull request #394 from pikers/size_in_shm_token
Store shm array size in token schema, use for loading
2022-08-29 15:15:49 -03:00
Tyler Goodlet 7dfa4c3cde Better comment on the `size`'s purpose/units 2022-08-29 13:56:26 -04:00
Tyler Goodlet 7b653fe4f4 Store shm array size in token schema, use for loading 2022-08-29 13:46:41 -04:00
goodboy 77a687bced
Merge pull request #386 from pikers/paper_tolerance
Paper race tolerance
2022-08-29 13:28:38 -04:00
Tyler Goodlet d5c1cdd91d Configure allocator from pos msg on startup
This fixes a regression added after moving the msg parsing to later in
the order mode startup sequence. The `Allocator` needs to be configured
*to* the initial pos otherwise default settings will show in the UI..

Move the startup config logic from inside `mk_allocator()` to
`PositionTracker.update_from_pp()` and add a flag to allow setting the
`.startup_pp` from the current live one as is needed during initial
load.
2022-08-29 11:39:28 -04:00
Tyler Goodlet 46d3fe88ca Fix sub-slot-remains limiting for -ve sizes
In the short case (-ve size) we had a bug where the last sub-slots worth
of exit size would never be limited to zero once the allocator limit pos
size was hit (i.e. you could keep going more -ve on the pos,
exponentially per slot over the limit). It's a simple fix, just
a `max()` around the `l_sub_pp` var used in the next-step-size calc.

Resolves #392
2022-08-28 13:51:54 -04:00
Tyler Goodlet 5c8c5d8fbf Fix disti-mode paper pps relaying
Turns out we were putting too many brokername suffixes in the symbol
field and thus the order mode msg parser wasn't matching the current
asset to said msgs correctly and pps weren't being shown...

This repairs that plus simplifies the order mode initial pos msg loading
to just delegate into `process_trade_msg()` just as is done for
real-time msg updates.
2022-08-27 15:37:54 -04:00
goodboy 71412310c4
Merge pull request #391 from pikers/json_rpc_generic
Pull jsonrpc machinery out of deribit backend
2022-08-27 15:33:12 -04:00
Guillermo Rodriguez 0c323fdc0b
Minor style changes and warning on unexpected msg 2022-08-27 09:12:02 -03:00
Tyler Goodlet 02f53d0c13 Error on zero-size orders received by paper engine 2022-08-26 10:46:47 -04:00
Tyler Goodlet 8792c97de6 More stringent settings pane input handling
If a setting fails to apply try to log an error msg and revert to the
previous setting by not applying the UI read-update until after the new
`SettingsPane.apply_setting()` call. This prevents crashes when the user
tries to give bad inputs on editable allocator fields.
2022-08-26 10:46:47 -04:00
Tyler Goodlet 980815d075 Avoid handling account as numeric field in settings 2022-08-26 10:46:46 -04:00
Tyler Goodlet 4cedfedc21 Support clearing ticks ('last' & 'trade') fills
Previously we only simulated paper engine fills when the data feed
provide L1 queue-levels matched an execution. This patch add further
support for clear-level matches when there are real live clears on the
data feed that are faster/not synced with the L1 (aka usually during
periods of HFT).

The solution was to simply iterate the interleaved paper book entries on
both sides for said tick types and instead yield side-specific predicate
per entry.
2022-08-26 10:46:46 -04:00
Tyler Goodlet fe3d0c6fdd Handle too-fast-edits with `defaultdict[str, bidict[str, tuple]]`
Not entirely sure why this all of a sudden became a problem but it seems
price changes on order edits were sometimes resulting in key errors when
modifying paper book entries quickly. This changes the implementation to
not care about matching the last price when keying/popping old orders
and use `bidict`s to more easily pop cleared orders in the paper loop.
2022-08-26 10:46:46 -04:00
Tyler Goodlet 9200e8da57 Raw-dog-pop cancelled paper entries; old price dun matter 2022-08-26 10:46:46 -04:00
Tyler Goodlet 430d065da6 Handle paper-engine too-fast clearing race cases
When the paper engine is used it seems we can definitely hit races where
order ack msgs arrive close enough to status messages that `trio`
schedules the status processing before the acks. In such cases we want
to be tolerant and not crash but instead warn that we got an
unknown/out-of-order msg.
2022-08-26 10:46:46 -04:00
Tyler Goodlet ecd93cb05a Pass symbol with broker suffix to `.submit_limit()`; fix clearing 2022-08-26 10:46:46 -04:00
Guillermo Rodriguez 4facd161a9
Pull jsonrpc machinery out of deribit backend into piker.data._web_bs module and make it generic 2022-08-25 14:08:09 -03:00
goodboy c5447fda06
Merge pull request #390 from pikers/actually_enable_modules
Oneliner enable rpc modules on runtime open
2022-08-25 13:06:53 -04:00
Guillermo Rodriguez 0447612b34
Oneliner enable rpc modules on runtime open 2022-08-25 11:47:40 -03:00
goodboy b5499b8225
Merge pull request #331 from pikers/deribit
Deribit backend & minimal broker check tool
2022-08-25 10:08:29 -04:00
Guillermo Rodriguez 00aabddfe8
Fix link 2022-08-25 09:22:15 -03:00
Guillermo Rodriguez 43fb720877
Do multiline imports 2022-08-25 09:20:41 -03:00
Guillermo Rodriguez 9626dbd7ac
Simplify rpc machinery, and switch refs to Dict and List to builtins, make brokercheck call public broker methods and get their results again 2022-08-25 09:18:52 -03:00
Guillermo Rodriguez f286c79a03
Woops enable backfill_bars in module __init__.py 2022-08-24 19:41:04 -03:00
Guillermo Rodriguez accb0eee6c
Add brokercheck guard on deribit.get_client && drop method running in brokercheck 2022-08-24 19:32:54 -03:00
Guillermo Rodriguez e97dd1cbdb
Stop using as much closures
Use a custom tractor branch that fixes a `maybe_open_context` re entrant related bug
2022-08-24 18:09:35 -03:00
Guillermo Rodriguez 34fb497eb4
Add aiter api to NoBsWs and rework cryptofeed relay to not be OOPy 2022-08-24 18:09:35 -03:00
Guillermo Rodriguez 6669ba6590
Switch back to using async for and dont install signal handlers on cryptofeed 2022-08-24 18:09:35 -03:00
Guillermo Rodriguez cb8099bb8c
Add README.rst and brokers.toml section in config example 2022-08-24 18:09:35 -03:00
Guillermo Rodriguez 80a1a58bfc
Refactor cryptofeed relay api and move it to client
Added submit_limit and submit_cancel
Cache syms correctly
Lowercase search results
2022-08-24 18:09:32 -03:00
Guillermo Rodriguez d60f222bb7
Add get_balances, and get_assets rpc to deribit.api.Client
Improve symbol_info search results
Expect cancellation on cryptofeeds asyncio task
Fix the no trades on instrument bug that we had on startup
2022-08-24 18:08:45 -03:00
Guillermo Rodriguez 2c2e43d8ac
Add comments and update cryptofeed fork url in requirements 2022-08-24 18:08:31 -03:00
Guillermo Rodriguez 212b3d620d
Tweaks on Client init to make api credentials optional 2022-08-24 18:08:29 -03:00
Guillermo Rodriguez 92090b01b8
Begin jsonrpc over ws refactor 2022-08-24 18:06:00 -03:00
Guillermo Rodriguez 9073fbc317
drop pydantic to match master 2022-08-23 15:18:45 -03:00
Guillermo Rodriguez f55f56a29f
Refactored deribit backend into new multi file format 2022-08-23 15:18:45 -03:00
Guillermo Rodriguez 28e025d02e
Finally get a chart going! lots of fixes to streaming machinery and custom cryptofeed fork with fixes 2022-08-23 15:18:43 -03:00
Guillermo Rodriguez e558e5837e
Introduce piker protocol in stream_messages 2022-08-23 15:17:18 -03:00
Guillermo Rodriguez a0b415095a
Brokermod check output fixed and tweaks to deribit Client.bars function 2022-08-23 15:17:18 -03:00
Guillermo Rodriguez 6df181c233
Add brokercheck test and got deribit to dump l1 and trades to console 2022-08-23 15:17:18 -03:00