07e35d3ff5
Merge pull request #217 from pikers/hot_fix_cache_event_is_none
...
Only set event if entry still exists
2021-09-02 12:52:27 -04:00
2227721dac
Only set event if entry still exists
2021-09-02 10:52:21 -04:00
37d94fbb28
Merge pull request #212 from pikers/feed_caching
...
Feed caching
2021-09-01 10:25:49 -04:00
4527d4a677
Allocate an event per context
2021-09-01 10:17:03 -04:00
26cb7aa660
Drop tractor stream shielding use
2021-09-01 09:03:55 -04:00
2df16e11ed
Re-implement client caching using `maybe_open_ctx`
2021-09-01 09:01:25 -04:00
c3682348fe
Use the actor's service nursery instead
...
In order to ensure the lifetime of the feed can in fact be kept open
until the last consumer task has completed we need to maintain
a lifetime which is hierarchically greater then all consumer tasks.
This solution is somewhat hacky but seems to work well: we just use the
`tractor` actor's "service nursery" (the one normally used to invoke rpc
tasks) to launch the task which will start and keep open the target
cached async context manager. To make this more "proper" we may want to
offer a "root nursery" in all piker actors that is exposed through some
singleton api or even introduce a public api for it into `tractor`
directly.
2021-08-31 12:46:47 -04:00
1184a4d88e
Cache sample step streams per actor
2021-08-31 09:28:22 -04:00
bbcce0cab6
Facepalm^2: pass through kwargs
2021-08-30 18:04:19 -04:00
cae7f486e4
Revert "Lol, don't use `maybe_open_feed()` for now, it's totally borked..."
...
Think this was fixed by passing through `**kwargs` in
`maybe_open_feed()`, the shielding for fsp respawns wasn't being
properly passed through..
This reverts commit 2f1455d423
2021-08-30 17:55:10 -04:00
ff322ae7be
Re-impl ctx-mng caching using `trio.Nursery.start()`
...
Maybe i've finally learned my lesson that exit stacks and per task ctx
manager caching is just not trionic.. Use the approach we've taken for
the daemon service manager as well: create a process global nursery for
each unique ctx manager we wish to cache and simply tear it down when
the number of consumers goes to zero.
This seems to resolve all prior issues and gets us error-free cached
feeds!
2021-08-30 17:54:43 -04:00
2f1455d423
Lol, don't use `maybe_open_feed()` for now, it's totally borked...
2021-08-26 17:04:59 -04:00
2a9d24ccac
Remove dead OHLC index consumers from subs list on error
2021-08-26 17:04:59 -04:00
fe0d66e847
Drop removed module import
2021-08-26 17:04:59 -04:00
1e42f58478
Add pause/resume feed api, delegate to msg stream for broadcast api
2021-08-26 17:04:59 -04:00
2f5abaa47a
Add njs token bucket gist url
2021-08-26 17:04:59 -04:00
c8e320849a
Add super basic support for data feed "pausing"
2021-08-26 17:04:59 -04:00
0c9516051b
TO SQUASH cached ctx.
2021-08-26 17:04:59 -04:00
71b50fdae8
Use broadcast chan for order client and avoid chan repacking
2021-08-26 17:04:59 -04:00
954dc6a8b0
Fix missing cache hit bool element of return
2021-08-26 17:04:59 -04:00
310d8f485e
Add disclaimer to old data mod
2021-08-26 17:04:59 -04:00
2202abc9fb
Add (lack of proper) ring buffer note
2021-08-26 17:04:59 -04:00
7d0f47364c
Use `maybe_open_feed()` in ems and fsp daemons
2021-08-26 17:04:59 -04:00
a7d3afc9b1
Add a `maybe_open_feed()` which uses new broadcast chans
...
Try out he new broadcast channels from `tractor` for data feeds
we already have cached. Any time there's a cache hit we load the
cached feed and just slap a broadcast receiver on it for the local
consumer task.
2021-08-26 17:04:59 -04:00
224dbbc4e3
Drop feed refs
2021-08-26 17:04:59 -04:00
7d5add1c3a
Add an njs cache gist link
2021-08-26 17:04:59 -04:00
66f1d91541
Let's abstractify: ->
2021-08-26 17:04:59 -04:00
68ce5b3550
Add lifo cache to new module; drop "utils", bleh
2021-08-26 17:04:59 -04:00
0ce8057823
Move feed cacheing to cache mod; put entry retreival into ctx mng
2021-08-26 17:04:59 -04:00
a0660e553f
Start top level cacheing apis module
2021-08-26 17:04:59 -04:00
146c684f21
Cache `brokerd` feeds for reuse in clearing loop
2021-08-26 17:04:59 -04:00
f03f051e7f
Merge pull request #213 from pikers/brokers_config
...
Brokers config schema bump, ib patches
2021-08-24 10:37:44 -04:00
c21d299193
Drop data/ version of config
2021-08-24 10:32:01 -04:00
89b2089562
Fixup missing ib section handling; drop `.api` subsection
2021-08-24 10:27:25 -04:00
d5394ac677
Fix TWS triggered trades msg packing
2021-08-24 10:26:41 -04:00
12c8d26906
Update brokers.toml schema
2021-08-24 10:25:07 -04:00
a10d20c041
Add ib config section support
2021-07-27 07:09:01 -04:00
96b0102f04
Merge pull request #208 from hengestone/master
...
Fix a few QtGui -> QtWidget issues
2021-07-23 11:17:58 -04:00
108f0da181
Merge pull request #210 from pikers/readme_bump_zone
...
Readme bump to include latest cli symbol syntax
2021-07-23 10:40:48 -04:00
afafe56ec0
Remove List typing
2021-07-21 18:30:50 -07:00
c830183073
More fixes, some reversions
2021-07-21 18:26:24 -07:00
3912b22b41
Fix a few QtGui -> QtWidget issues
2021-07-21 18:26:24 -07:00
1ef0725e74
Readme bump to include latest cli symbol syntax and add binance to provider list
2021-07-21 20:37:40 -04:00
161089de87
Merge pull request #209 from pikers/ci_on_forks
...
Run CI on cross-fork PRs?
2021-07-21 20:24:01 -04:00
cea8272a11
Run CI on cross-fork PRs?
2021-07-21 20:19:28 -04:00
663373cb04
Merge pull request #207 from hengestone/master
...
Fix PyQt5 error
2021-07-21 14:16:29 -04:00
9d8d8fd1a2
Fix PyQt5 error
2021-07-21 10:53:40 -07:00
8a6142632d
Merge pull request #200 from pikers/asyncify_input_modes
...
Asyncify input modes
2021-07-11 12:57:09 -04:00
83ad071cb4
Load provider search engines in tasks instead of exit stack
2021-07-11 11:13:35 -04:00
90588018a6
Label doc tweak
2021-07-11 11:13:35 -04:00
goodboy
Tyler Goodlet
Conrad Steenberg