Compare commits

..

35 Commits

Author SHA1 Message Date
Tyler Goodlet 7dd72e042d Show full KBI trace for help with CI hangs 2022-07-27 11:38:13 -04:00
Tyler Goodlet ee8ead4d7a Move pydantic-click hang example to new dir, skip in test suite 2022-07-27 11:38:13 -04:00
Tyler Goodlet 8a70a52ff9 Add spaces before values in log msg 2022-07-27 11:38:06 -04:00
Tyler Goodlet 70e4458fb0 Add runtime level msg around channel draining 2022-07-27 11:38:06 -04:00
Tyler Goodlet 70d1c98c10 Always undo SIGINT overrides , cancel detached children
Ensure that even when `pdb` resumption methods are called during a crash
where `trio`'s runtime has already terminated (eg. `Event.set()` will
raise) we always revert our sigint handler to the original. Further
inside the handler if we hit a case where a child is in debug and
(thinks it) has the global pdb lock, if it has no IPC connection to
a parent, simply presume tty sync-coordination is now lost and cancel
the child immediately.
2022-07-27 11:38:06 -04:00
Tyler Goodlet dade6a4b43 Readme formatting tweaks 2022-07-27 11:38:06 -04:00
Tyler Goodlet 7b4049198a Tolerate double `.remove()`s of stream on portal teardowns 2022-07-27 11:37:59 -04:00
Tyler Goodlet 931b20cf35 Always propagate SIGINT when no locking peer found
A hopefully significant fix here is to always avoid suppressing a SIGINT
when the root actor can not detect an active IPC connections (via
a connected channel) to the supposed debug lock holding actor. In that
case it is most likely that the actor has either terminated or has lost
its connection for debugger control and there is no way the root can
verify the lock is in use; thus we choose to allow KBI cancellation.

Drop the (by comment) `try`-`finally` block in
`_hijoack_stdin_for_child()` around the `_acquire_debug_lock()` call
since all that logic should now be handled internal to that locking
manager. Try to catch a weird error around the `.do_longlist()` method
call that seems to sometimes break on py3.10 and latest `pdbpp`.
2022-07-27 11:37:59 -04:00
Tyler Goodlet 11c1582c39 Always call pdb hook even if tty locking fails 2022-07-27 11:37:59 -04:00
Tyler Goodlet d1f347c21f Log cancels with appropriate level 2022-07-27 11:37:59 -04:00
Tyler Goodlet 2800100b21 Just warn on IPC breaks 2022-07-27 11:37:59 -04:00
Tyler Goodlet 1fd4588243 Only warn on `trio.BrokenResourceError`s from `_invoke()` 2022-07-27 11:37:59 -04:00
Tyler Goodlet 7ecc48b3c9 Make example a subpkg for `python -m <mod>` testing 2022-07-27 11:37:59 -04:00
Tyler Goodlet ebefd6e6b4 Add example that triggers bug #302 2022-07-27 11:37:59 -04:00
Tyler Goodlet 67607a4d37 Add back in async gen loop 2022-07-27 11:37:59 -04:00
Tyler Goodlet df16a0c315 Pre-declare disconnected flag 2022-07-27 11:37:59 -04:00
Tyler Goodlet 1163ec5034 Avoid attr error XD 2022-07-27 11:37:59 -04:00
Tyler Goodlet e2169f227d Type annot updates 2022-07-27 11:37:59 -04:00
Tyler Goodlet 4e06b10502 Drop uneeded backframe traceback hide annotation 2022-07-27 11:37:59 -04:00
Tyler Goodlet a5f543eb22 Make `Actor._process_messages()` report disconnects
The method now returns a `bool` which flags whether the transport died
to the caller and allows for reporting a disconnect in the
channel-transport handler task. This is something a user will normally
want to know about on the caller side especially after seeing
a traceback from the peer (if in tree) on console.
2022-07-27 11:37:59 -04:00
Tyler Goodlet d280c26f15 Only cancel/get-result from a ctx if transport is up
There's no point in sending a cancel message to the remote linked task
and especially no reason to block waiting on a result from that task if
the transport layer is detected to be disconnected. We expect that the
transport shouldn't go down at the layer of the message loop
(reconnection logic should be handled in the transport layer itself) so
if we detect the channel is not connected we don't bother requesting
cancels nor waiting on a final result message.

Why?

- if the connection goes down in error the caller side won't have a way
  to know "how long" it should block to wait for a cancel ack or result
  and causes a potential hang that may require an additional ctrl-c from
  the user especially if using the debugger or if the traceback is not
  seen on console.
- obviously there's no point in waiting for messages when there's no
  transport to deliver them XD

Further, add some more detailed cancel logging detailing the task and
actor ids.
2022-07-27 11:37:59 -04:00
Tyler Goodlet 74819821a8 Drop high log level in ctx example 2022-07-27 11:37:59 -04:00
Tyler Goodlet 5dd8adcfb8 Typing fixes, simplify `_set_trace()` 2022-07-27 11:37:59 -04:00
Tyler Goodlet 95ccb27004 Add notes around py3.10 stdlib bug from `pdb++`
There's a bug that's triggered in the stdlib without latest `pdb++`
installed; add a note for that.

Further inside `wait_for_parent_stdin_hijack()` don't `.started()` until
the interactor stream has been opened to avoid races when debugging this
`._debug.py` module (at the least) since we usually don't want the
spawning (parent) task to resume until we know for sure the tty lock has
been acquired. Also, drop the random checkpoint we had inside
`_breakpoint()`, not sure it was actually adding anything useful since
we're (mostly) carefully shielded throughout this func.
2022-07-27 11:37:59 -04:00
Tyler Goodlet 4e6d00918b Add and use a pdb instance factory 2022-07-27 11:37:59 -04:00
Tyler Goodlet a0016bcdc8 A `.open_context()` example that causes a hang!
Finally! I think this may be the root issue we've been seeing in
production in a client project.

No idea yet why this is happening but the fault-causing sequence seems
to be:
- `.open_context()` in a child actor
- enter the debugger via `tractor.breakpoint()`
- continue from that entry via `c` command in REPL
- raise an error just after inside the context task's body

Looking at logging it appears as though the child thinks it has the tty
but no input is accepted on the REPL and a further `ctrl-c` results in
some teardown but also a further hang where both parent and child become
unresponsive..
2022-07-27 11:37:59 -04:00
Tyler Goodlet 9e0bb4f90c Drop all the `@cm.__exit__()` override attempts..
None of it worked (you still will see `.__exit__()` frames on debugger
entry - you'd think this would have been solved by now but, shrug) so
instead wrap the debugger entry-point in a `try:` and put the SIGINT
handler restoration inside `MultiActorPdb` teardown hooks.

This seems to restore the UX as it was prior but with also giving the
desired SIGINT override handler behaviour.
2022-07-27 11:37:59 -04:00
Tyler Goodlet a617631400 Try overriding `_GeneratorContextManager.__exit__()`; didn't work..
Using either of `@pdb.hideframe` or `__tracebackhide__` on stdlib
methods doesn't seem to work either.. This all seems to have something
to do with async generator usage I think ?
2022-07-27 11:37:59 -04:00
Tyler Goodlet 4ea2bc5932 Fix example name typo 2022-07-27 11:37:59 -04:00
Tyler Goodlet a8a2110458 Handle a context cancel? Might be a noop 2022-07-27 11:37:59 -04:00
Tyler Goodlet aad9d7e947 Add a pre-started breakpoint example 2022-07-27 11:37:59 -04:00
Tyler Goodlet aee00e6741 Make `mypy` happy 2022-07-27 11:37:59 -04:00
Tyler Goodlet 688e0b9ebe Refine the handler for child vs. root cases
This gets very close to avoiding any possible hangs to do with tty
locking and SIGINT handling minus a special case that will be detailed
below.

Summary of implementation changes:

- convert `_mk_pdb()` -> `with _open_pdb() as pdb:` which implicitly
  handles the `bdb.BdbQuit` case such that debugger teardown hooks are
  always called.
- rename the handler to `shield_sigint()` and handle a variety of new
  cases:
  * the root is in debug but hasn't been cancelled -> call
    `Actor.cancel_soon()`
  * the root is in debug but *has* been called (`Actor.cancel_soon()`
    already called) -> raise KBI
  * a child is in debug *and* has a task locking the debugger -> ignore
    SIGINT in child *and* the root actor.
- if the debugger instance is provided to the handler at acquire time,
  on SIGINT handling completion re-print the last pdb++ REPL output so
  that the user realizes they are still actively in debug.
- ignore the unlock case where a race condition of "no task" holding the
  lock causes the `RuntimeError` normally associated with the "wrong
  task" doing so (not sure if this is a `trio` bug?).
- change debug logs to runtime level.

Unhandled case(s):

- a child is maybe in debug mode but does not itself have any task using
  the debugger.
    * ToDo: we need a way to decide what to do with
      "intermediate" child actors who themselves either are not in
      `debug_mode=True` but have children who *are* such that a SIGINT
      won't cause cancellation of that child-as-parent-of-another-child
      **iff** any of their children are in in debug mode.
2022-07-27 11:37:59 -04:00
Tyler Goodlet 1e789ecad2 (facepalm) Reraise `BdbQuit` and discard ownerless lock releases 2022-07-27 11:37:59 -04:00
Tyler Goodlet 0503142332 Add WIP while-debugger-active SIGINT ignore handler 2022-07-27 11:37:59 -04:00
75 changed files with 2071 additions and 3821 deletions

View File

@ -26,10 +26,8 @@ jobs:
run: pip install -U . --upgrade-strategy eager -r requirements-test.txt run: pip install -U . --upgrade-strategy eager -r requirements-test.txt
- name: Run MyPy check - name: Run MyPy check
run: mypy tractor/ --ignore-missing-imports --show-traceback run: mypy tractor/ --ignore-missing-imports
# test that we can generate a software distribution and install it
# thus avoid missing file issues after packaging.
sdist-linux: sdist-linux:
name: 'sdist' name: 'sdist'
runs-on: ubuntu-latest runs-on: ubuntu-latest
@ -59,12 +57,8 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
os: [ubuntu-latest] os: [ubuntu-latest]
python: ['3.10'] python: ['3.9', '3.10']
spawn_backend: [ spawn_backend: ['trio', 'mp']
'trio',
'mp_spawn',
'mp_forkserver',
]
steps: steps:
@ -79,53 +73,42 @@ jobs:
- name: Install dependencies - name: Install dependencies
run: pip install -U . -r requirements-test.txt -r requirements-docs.txt --upgrade-strategy eager run: pip install -U . -r requirements-test.txt -r requirements-docs.txt --upgrade-strategy eager
- name: List dependencies
run: pip list
- name: Run tests - name: Run tests
run: pytest tests/ --spawn-backend=${{ matrix.spawn_backend }} -rsx run: pytest tests/ --spawn-backend=${{ matrix.spawn_backend }} -rs
# We skip 3.10 on windows for now due to not having any collabs to # We skip 3.10 on windows for now due to
# debug the CI failures. Anyone wanting to hack and solve them is very # https://github.com/pytest-dev/pytest/issues/8733
# welcome, but our primary user base is not using that OS. # some kinda weird `pyreadline` issue..
# TODO: use job filtering to accomplish instead of repeated # TODO: use job filtering to accomplish instead of repeated
# boilerplate as is above XD: # boilerplate as is above XD:
# - https://docs.github.com/en/actions/learn-github-actions/managing-complex-workflows # - https://docs.github.com/en/actions/learn-github-actions/managing-complex-workflows
# - https://docs.github.com/en/actions/learn-github-actions/managing-complex-workflows#using-a-build-matrix # - https://docs.github.com/en/actions/learn-github-actions/managing-complex-workflows#using-a-build-matrix
# - https://docs.github.com/en/actions/learn-github-actions/workflow-syntax-for-github-actions#jobsjob_idif # - https://docs.github.com/en/actions/learn-github-actions/workflow-syntax-for-github-actions#jobsjob_idif
# testing-windows: testing-windows:
# name: '${{ matrix.os }} Python ${{ matrix.python }} - ${{ matrix.spawn_backend }}' name: '${{ matrix.os }} Python ${{ matrix.python }} - ${{ matrix.spawn_backend }}'
# timeout-minutes: 12 timeout-minutes: 12
# runs-on: ${{ matrix.os }} runs-on: ${{ matrix.os }}
# strategy: strategy:
# fail-fast: false fail-fast: false
# matrix: matrix:
# os: [windows-latest] os: [windows-latest]
# python: ['3.10'] python: ['3.9', '3.10']
# spawn_backend: ['trio', 'mp'] spawn_backend: ['trio', 'mp']
# steps: steps:
# - name: Checkout - name: Checkout
# uses: actions/checkout@v2 uses: actions/checkout@v2
# - name: Setup python - name: Setup python
# uses: actions/setup-python@v2 uses: actions/setup-python@v2
# with: with:
# python-version: '${{ matrix.python }}' python-version: '${{ matrix.python }}'
# - name: Install dependencies - name: Install dependencies
# run: pip install -U . -r requirements-test.txt -r requirements-docs.txt --upgrade-strategy eager run: pip install -U . -r requirements-test.txt -r requirements-docs.txt --upgrade-strategy eager
# # TODO: pretty sure this solves debugger deps-issues on windows, but it needs to - name: Run tests
# # be verified by someone with a native setup. run: pytest tests/ --spawn-backend=${{ matrix.spawn_backend }} -rs --full-trace
# # - name: Force pyreadline3
# # run: pip uninstall pyreadline; pip install -U pyreadline3
# - name: List dependencies
# run: pip list
# - name: Run tests
# run: pytest tests/ --spawn-backend=${{ matrix.spawn_backend }} -rsx

7
305.misc.rst 100644
View File

@ -0,0 +1,7 @@
Add ``tractor.query_actor()`` an addr looker-upper which doesn't deliver
a ``Portal`` instance and instead just a socket address ``tuple``.
Sometimes it's handy to just have a simple way to figure out if
a "service" actor is up, so add this discovery helper for that. We'll
prolly just leave it undocumented for now until we figure out
a longer-term/better discovery system.

142
NEWS.rst
View File

@ -4,148 +4,6 @@ Changelog
.. towncrier release notes start .. towncrier release notes start
tractor 0.1.0a5 (2022-08-03)
============================
This is our final release supporting Python 3.9 since we will be moving
internals to the new `match:` syntax from 3.10 going forward and
further, we have officially dropped usage of the `msgpack` library and
happily adopted `msgspec`.
Features
--------
- `#165 <https://github.com/goodboy/tractor/issues/165>`_: Add SIGINT
protection to our `pdbpp` based debugger subystem such that for
(single-depth) actor trees in debug mode we ignore interrupts in any
actor currently holding the TTY lock thus avoiding clobbering IPC
connections and/or task and process state when working in the REPL.
As a big note currently so called "nested" actor trees (trees with
actors having more then one parent/ancestor) are not fully supported
since we don't yet have a mechanism to relay the debug mode knowledge
"up" the actor tree (for eg. when handling a crash in a leaf actor).
As such currently there is a set of tests and known scenarios which will
result in process cloberring by the zombie repaing machinery and these
have been documented in https://github.com/goodboy/tractor/issues/320.
The implementation details include:
- utilizing a custom SIGINT handler which we apply whenever an actor's
runtime enters the debug machinery, which we also make sure the
stdlib's `pdb` configuration doesn't override (which it does by
default without special instance config).
- litter the runtime with `maybe_wait_for_debugger()` mostly in spots
where the root actor should block before doing embedded nursery
teardown ops which both cancel potential-children-in-deubg as well
as eventually trigger zombie reaping machinery.
- hardening of the TTY locking semantics/API both in terms of IPC
terminations and cancellation and lock release determinism from
sync debugger instance methods.
- factoring of locking infrastructure into a new `._debug.Lock` global
which encapsulates all details of the ``trio`` sync primitives and
task/actor uid management and tracking.
We also add `ctrl-c` cases throughout the test suite though these are
disabled for py3.9 (`pdbpp` UX differences that don't seem worth
compensating for, especially since this will be our last 3.9 supported
release) and there are a slew of marked cases that aren't expected to
work in CI more generally (as mentioned in the "nested" tree note
above) despite seemingly working when run manually on linux.
- `#304 <https://github.com/goodboy/tractor/issues/304>`_: Add a new
``to_asyncio.LinkedTaskChannel.subscribe()`` which gives task-oriented
broadcast functionality semantically equivalent to
``tractor.MsgStream.subscribe()`` this makes it possible for multiple
``trio``-side tasks to consume ``asyncio``-side task msgs in tandem.
Further Improvements to the test suite were added in this patch set
including a new scenario test for a sub-actor managed "service nursery"
(implementing the basics of a "service manager") including use of
*infected asyncio* mode. Further we added a lower level
``test_trioisms.py`` to start to track issues we need to work around in
``trio`` itself which in this case included a bug we were trying to
solve related to https://github.com/python-trio/trio/issues/2258.
Bug Fixes
---------
- `#318 <https://github.com/goodboy/tractor/issues/318>`_: Fix
a previously undetected ``trio``-``asyncio`` task lifetime linking
issue with the ``to_asyncio.open_channel_from()`` api where both sides
where not properly waiting/signalling termination and it was possible
for ``asyncio``-side errors to not propagate due to a race condition.
The implementation fix summary is:
- add state to signal the end of the ``trio`` side task to be
read by the ``asyncio`` side and always cancel any ongoing
task in such cases.
- always wait on the ``asyncio`` task termination from the ``trio``
side on error before maybe raising said error.
- always close the ``trio`` mem chan on exit to ensure the other
side can detect it and follow.
Trivial/Internal Changes
------------------------
- `#248 <https://github.com/goodboy/tractor/issues/248>`_: Adjust the
`tractor._spawn.soft_wait()` strategy to avoid sending an actor cancel
request (via `Portal.cancel_actor()`) if either the child process is
detected as having terminated or the IPC channel is detected to be
closed.
This ensures (even) more deterministic inter-actor cancellation by
avoiding the timeout condition where possible when a whild never
sucessfully spawned, crashed, or became un-contactable over IPC.
- `#295 <https://github.com/goodboy/tractor/issues/295>`_: Add an
experimental ``tractor.msg.NamespacePath`` type for passing Python
objects by "reference" through a ``str``-subtype message and using the
new ``pkgutil.resolve_name()`` for reference loading.
- `#298 <https://github.com/goodboy/tractor/issues/298>`_: Add a new
`tractor.experimental` subpackage for staging new high level APIs and
subystems that we might eventually make built-ins.
- `#300 <https://github.com/goodboy/tractor/issues/300>`_: Update to and
pin latest ``msgpack`` (1.0.3) and ``msgspec`` (0.4.0) both of which
required adjustments for backwards imcompatible API tweaks.
- `#303 <https://github.com/goodboy/tractor/issues/303>`_: Fence off
``multiprocessing`` imports until absolutely necessary in an effort to
avoid "resource tracker" spawning side effects that seem to have
varying degrees of unreliability per Python release. Port to new
``msgspec.DecodeError``.
- `#305 <https://github.com/goodboy/tractor/issues/305>`_: Add
``tractor.query_actor()`` an addr looker-upper which doesn't deliver
a ``Portal`` instance and instead just a socket address ``tuple``.
Sometimes it's handy to just have a simple way to figure out if
a "service" actor is up, so add this discovery helper for that. We'll
prolly just leave it undocumented for now until we figure out
a longer-term/better discovery system.
- `#316 <https://github.com/goodboy/tractor/issues/316>`_: Run windows
CI jobs on python 3.10 after some hacks for ``pdbpp`` dependency
issues.
Issue was to do with the now deprecated `pyreadline` project which
should be changed over to `pyreadline3`.
- `#317 <https://github.com/goodboy/tractor/issues/317>`_: Drop use of
the ``msgpack`` package and instead move fully to the ``msgspec``
codec library.
We've now used ``msgspec`` extensively in production and there's no
reason to not use it as default. Further this change preps us for the up
and coming typed messaging semantics (#196), dialog-unprotocol system
(#297), and caps-based messaging-protocols (#299) planned before our
first beta.
tractor 0.1.0a4 (2021-12-18) tractor 0.1.0a4 (2021-12-18)
============================ ============================

View File

@ -6,14 +6,8 @@
``tractor`` is a `structured concurrent`_, multi-processing_ runtime ``tractor`` is a `structured concurrent`_, multi-processing_ runtime
built on trio_. built on trio_.
Fundamentally, ``tractor`` gives you parallelism via Fundamentally ``tractor`` gives you parallelism via ``trio``-"*actors*":
``trio``-"*actors*": independent Python processes (aka our nurseries_ let you spawn new Python processes which each run a ``trio``
non-shared-memory threads) which maintain structured
concurrency (SC) *end-to-end* inside a *supervision tree*.
Cross-process (and thus cross-host) SC is accomplished through the
combined use of our "actor nurseries_" and an "SC-transitive IPC
protocol" constructed on top of multiple Pythons each running a ``trio``
scheduled runtime - a call to ``trio.run()``. scheduled runtime - a call to ``trio.run()``.
We believe the system adheres to the `3 axioms`_ of an "`actor model`_" We believe the system adheres to the `3 axioms`_ of an "`actor model`_"
@ -29,8 +23,7 @@ Features
- **It's just** a ``trio`` API - **It's just** a ``trio`` API
- *Infinitely nesteable* process trees - *Infinitely nesteable* process trees
- Builtin IPC streaming APIs with task fan-out broadcasting - Builtin IPC streaming APIs with task fan-out broadcasting
- A "native" multi-core debugger REPL using `pdbp`_ (a fork & fix of - A (first ever?) "native" multi-core debugger UX for Python using `pdb++`_
`pdb++`_ thanks to @mdmintz!)
- Support for a swappable, OS specific, process spawning layer - Support for a swappable, OS specific, process spawning layer
- A modular transport stack, allowing for custom serialization (eg. with - A modular transport stack, allowing for custom serialization (eg. with
`msgspec`_), communications protocols, and environment specific IPC `msgspec`_), communications protocols, and environment specific IPC
@ -156,7 +149,7 @@ it **is a bug**.
"Native" multi-process debugging "Native" multi-process debugging
-------------------------------- --------------------------------
Using the magic of `pdbp`_ and our internal IPC, we've Using the magic of `pdb++`_ and our internal IPC, we've
been able to create a native feeling debugging experience for been able to create a native feeling debugging experience for
any (sub-)process in your ``tractor`` tree. any (sub-)process in your ``tractor`` tree.
@ -574,13 +567,6 @@ Help us push toward the future of distributed `Python`.
- Typed capability-based (dialog) protocols ( see `#196 - Typed capability-based (dialog) protocols ( see `#196
<https://github.com/goodboy/tractor/issues/196>`_ with draft work <https://github.com/goodboy/tractor/issues/196>`_ with draft work
started in `#311 <https://github.com/goodboy/tractor/pull/311>`_) started in `#311 <https://github.com/goodboy/tractor/pull/311>`_)
- We **recently disabled CI-testing on windows** and need help getting
it running again! (see `#327
<https://github.com/goodboy/tractor/pull/327>`_). **We do have windows
support** (and have for quite a while) but since no active hacker
exists in the user-base to help test on that OS, for now we're not
actively maintaining testing due to the added hassle and general
latency..
Feel like saying hi? Feel like saying hi?
@ -604,7 +590,6 @@ channel`_!
.. _adherance to: https://www.youtube.com/watch?v=7erJ1DV_Tlo&t=1821s .. _adherance to: https://www.youtube.com/watch?v=7erJ1DV_Tlo&t=1821s
.. _trio gitter channel: https://gitter.im/python-trio/general .. _trio gitter channel: https://gitter.im/python-trio/general
.. _matrix channel: https://matrix.to/#/!tractor:matrix.org .. _matrix channel: https://matrix.to/#/!tractor:matrix.org
.. _pdbp: https://github.com/mdmintz/pdbp
.. _pdb++: https://github.com/pdbpp/pdbpp .. _pdb++: https://github.com/pdbpp/pdbpp
.. _guest mode: https://trio.readthedocs.io/en/stable/reference-lowlevel.html?highlight=guest%20mode#using-guest-mode-to-run-trio-on-top-of-other-event-loops .. _guest mode: https://trio.readthedocs.io/en/stable/reference-lowlevel.html?highlight=guest%20mode#using-guest-mode-to-run-trio-on-top-of-other-event-loops
.. _messages: https://en.wikipedia.org/wiki/Message_passing .. _messages: https://en.wikipedia.org/wiki/Message_passing

View File

@ -396,7 +396,7 @@ tasks spawned via multiple RPC calls to an actor can modify
# a per process cache # a per process cache
_actor_cache: dict[str, bool] = {} _actor_cache: Dict[str, bool] = {}
def ping_endpoints(endpoints: List[str]): def ping_endpoints(endpoints: List[str]):

View File

@ -1,151 +0,0 @@
'''
Complex edge case where during real-time streaming the IPC tranport
channels are wiped out (purposely in this example though it could have
been an outage) and we want to ensure that despite being in debug mode
(or not) the user can sent SIGINT once they notice the hang and the
actor tree will eventually be cancelled without leaving any zombies.
'''
import trio
from tractor import (
open_nursery,
context,
Context,
MsgStream,
)
async def break_channel_silently_then_error(
stream: MsgStream,
):
async for msg in stream:
await stream.send(msg)
# XXX: close the channel right after an error is raised
# purposely breaking the IPC transport to make sure the parent
# doesn't get stuck in debug or hang on the connection join.
# this more or less simulates an infinite msg-receive hang on
# the other end.
await stream._ctx.chan.send(None)
assert 0
async def close_stream_and_error(
stream: MsgStream,
):
async for msg in stream:
await stream.send(msg)
# wipe out channel right before raising
await stream._ctx.chan.send(None)
await stream.aclose()
assert 0
@context
async def recv_and_spawn_net_killers(
ctx: Context,
break_ipc_after: bool | int = False,
) -> None:
'''
Receive stream msgs and spawn some IPC killers mid-stream.
'''
await ctx.started()
async with (
ctx.open_stream() as stream,
trio.open_nursery() as n,
):
async for i in stream:
print(f'child echoing {i}')
await stream.send(i)
if (
break_ipc_after
and i > break_ipc_after
):
'#################################\n'
'Simulating child-side IPC BREAK!\n'
'#################################'
n.start_soon(break_channel_silently_then_error, stream)
n.start_soon(close_stream_and_error, stream)
async def main(
debug_mode: bool = False,
start_method: str = 'trio',
# by default we break the parent IPC first (if configured to break
# at all), but this can be changed so the child does first (even if
# both are set to break).
break_parent_ipc_after: int | bool = False,
break_child_ipc_after: int | bool = False,
) -> None:
async with (
open_nursery(
start_method=start_method,
# NOTE: even debugger is used we shouldn't get
# a hang since it never engages due to broken IPC
debug_mode=debug_mode,
loglevel='warning',
) as an,
):
portal = await an.start_actor(
'chitty_hijo',
enable_modules=[__name__],
)
async with portal.open_context(
recv_and_spawn_net_killers,
break_ipc_after=break_child_ipc_after,
) as (ctx, sent):
async with ctx.open_stream() as stream:
for i in range(1000):
if (
break_parent_ipc_after
and i > break_parent_ipc_after
):
print(
'#################################\n'
'Simulating parent-side IPC BREAK!\n'
'#################################'
)
await stream._ctx.chan.send(None)
# it actually breaks right here in the
# mp_spawn/forkserver backends and thus the zombie
# reaper never even kicks in?
print(f'parent sending {i}')
await stream.send(i)
with trio.move_on_after(2) as cs:
# NOTE: in the parent side IPC failure case this
# will raise an ``EndOfChannel`` after the child
# is killed and sends a stop msg back to it's
# caller/this-parent.
rx = await stream.receive()
print(f"I'm a happy user and echoed to me is {rx}")
if cs.cancelled_caught:
# pretend to be a user seeing no streaming action
# thinking it's a hang, and then hitting ctl-c..
print("YOO i'm a user anddd thingz hangin..")
print(
"YOO i'm mad send side dun but thingz hangin..\n"
'MASHING CTlR-C Ctl-c..'
)
raise KeyboardInterrupt
if __name__ == '__main__':
trio.run(main)

View File

@ -27,17 +27,6 @@ async def main():
# retreive results # retreive results
async with p0.open_stream_from(breakpoint_forever) as stream: async with p0.open_stream_from(breakpoint_forever) as stream:
# triggers the first name error
try:
await p1.run(name_error)
except tractor.RemoteActorError as rae:
assert rae.type is NameError
async for i in stream:
# a second time try the failing subactor and this tie
# let error propagate up to the parent/nursery.
await p1.run(name_error) await p1.run(name_error)

View File

@ -12,31 +12,18 @@ async def breakpoint_forever():
while True: while True:
await tractor.breakpoint() await tractor.breakpoint()
# NOTE: if the test never sent 'q'/'quit' commands
# on the pdb repl, without this checkpoint line the
# repl would spin in this actor forever.
# await trio.sleep(0)
async def spawn_until(depth=0): async def spawn_until(depth=0):
""""A nested nursery that triggers another ``NameError``. """"A nested nursery that triggers another ``NameError``.
""" """
async with tractor.open_nursery() as n: async with tractor.open_nursery() as n:
if depth < 1: if depth < 1:
# await n.run_in_actor('breakpoint_forever', breakpoint_forever)
await n.run_in_actor(breakpoint_forever) await n.run_in_actor(
p = await n.run_in_actor(
name_error, name_error,
name='name_error' name='name_error'
) )
await trio.sleep(0.5)
# rx and propagate error from child
await p.result()
else: else:
# recusrive call to spawn another process branching layer of
# the tree
depth -= 1 depth -= 1
await n.run_in_actor( await n.run_in_actor(
spawn_until, spawn_until,
@ -66,7 +53,6 @@ async def main():
""" """
async with tractor.open_nursery( async with tractor.open_nursery(
debug_mode=True, debug_mode=True,
# loglevel='cancel',
) as n: ) as n:
# spawn both actors # spawn both actors
@ -81,16 +67,8 @@ async def main():
name='spawner1', name='spawner1',
) )
# TODO: test this case as well where the parent don't see
# the sub-actor errors by default and instead expect a user
# ctrl-c to kill the root.
with trio.move_on_after(3):
await trio.sleep_forever()
# gah still an issue here. # gah still an issue here.
await portal.result() await portal.result()
# should never get here
await portal1.result() await portal1.result()

View File

@ -1,40 +0,0 @@
import trio
import tractor
@tractor.context
async def just_sleep(
ctx: tractor.Context,
**kwargs,
) -> None:
'''
Start and sleep.
'''
await ctx.started()
await trio.sleep_forever()
async def main() -> None:
async with tractor.open_nursery(
debug_mode=True,
) as n:
portal = await n.start_actor(
'ctx_child',
# XXX: we don't enable the current module in order
# to trigger `ModuleNotFound`.
enable_modules=[],
)
async with portal.open_context(
just_sleep, # taken from pytest parameterization
) as (ctx, sent):
raise KeyboardInterrupt
if __name__ == '__main__':
trio.run(main)

View File

@ -1,24 +0,0 @@
import os
import sys
import trio
import tractor
async def main() -> None:
async with tractor.open_nursery(debug_mode=True) as an:
assert os.environ['PYTHONBREAKPOINT'] == 'tractor._debug._set_trace'
# TODO: an assert that verifies the hook has indeed been, hooked
# XD
assert sys.breakpointhook is not tractor._debug._set_trace
breakpoint()
# TODO: an assert that verifies the hook is unhooked..
assert sys.breakpointhook
breakpoint()
if __name__ == '__main__':
trio.run(main)

View File

@ -9,7 +9,7 @@ is ``tractor``'s channels.
""" """
from contextlib import asynccontextmanager from contextlib import asynccontextmanager
from typing import Callable from typing import List, Callable
import itertools import itertools
import math import math
import time import time
@ -71,8 +71,8 @@ async def worker_pool(workers=4):
async def _map( async def _map(
worker_func: Callable[[int], bool], worker_func: Callable[[int], bool],
sequence: list[int] sequence: List[int]
) -> list[bool]: ) -> List[bool]:
# define an async (local) task to collect results from workers # define an async (local) task to collect results from workers
async def send_result(func, value, portal): async def send_result(func, value, portal):

View File

@ -0,0 +1,8 @@
Adjust the `tractor._spawn.soft_wait()` strategy to avoid sending an
actor cancel request (via `Portal.cancel_actor()`) if either the child
process is detected as having terminated or the IPC channel is detected
to be closed.
This ensures (even) more deterministic inter-actor cancellation by
avoiding the timeout condition where possible when a whild never
sucessfully spawned, crashed, or became un-contactable over IPC.

View File

@ -0,0 +1,3 @@
Add an experimental ``tractor.msg.NamespacePath`` type for passing Python
objects by "reference" through a ``str``-subtype message and using the
new ``pkgutil.resolve_name()`` for reference loading.

View File

@ -0,0 +1,2 @@
Add a new `tractor.experimental` subpackage for staging new high level
APIs and subystems that we might eventually make built-ins.

View File

@ -0,0 +1,3 @@
Update to and pin latest ``msgpack`` (1.0.3) and ``msgspec`` (0.4.0)
both of which required adjustments for backwards imcompatible API
tweaks.

View File

@ -0,0 +1,4 @@
Fence off ``multiprocessing`` imports until absolutely necessary in an
effort to avoid "resource tracker" spawning side effects that seem to
have varying degrees of unreliability per Python release. Port to new
``msgspec.DecodeError``.

View File

@ -0,0 +1,12 @@
Add a new ``to_asyncio.LinkedTaskChannel.subscribe()`` which gives
task-oriented broadcast functionality semantically equivalent to
``tractor.MsgStream.subscribe()`` this makes it possible for multiple
``trio``-side tasks to consume ``asyncio``-side task msgs in tandem.
Further Improvements to the test suite were added in this patch set
including a new scenario test for a sub-actor managed "service nursery"
(implementing the basics of a "service manager") including use of
*infected asyncio* mode. Further we added a lower level
``test_trioisms.py`` to start to track issues we need to work around in
``trio`` itself which in this case included a bug we were trying to
solve related to https://github.com/python-trio/trio/issues/2258.

View File

@ -0,0 +1,5 @@
Run windows CI jobs on python 3.10 after some
hacks for ``pdbpp`` dependency issues.
Issue was to do with the now deprecated `pyreadline` project which
should be changed over to `pyreadline3`.

View File

@ -0,0 +1,8 @@
Drop use of the ``msgpack`` package and instead move fully to the
``msgspec`` codec library.
We've now used ``msgspec`` extensively in production and there's no
reason to not use it as default. Further this change preps us for the up
and coming typed messaging semantics (#196), dialog-unprotocol system
(#297), and caps-based messaging-protocols (#299) planned before our
first beta.

13
nooz/318.bug.rst 100644
View File

@ -0,0 +1,13 @@
Fix a previously undetected ``trio``-``asyncio`` task lifetime linking
issue with the ``to_asyncio.open_channel_from()`` api where both sides
where not properly waiting/signalling termination and it was possible
for ``asyncio``-side errors to not propagate due to a race condition.
The implementation fix summary is:
- add state to signal the end of the ``trio`` side task to be
read by the ``asyncio`` side and always cancel any ongoing
task in such cases.
- always wait on the ``asyncio`` task termination from the ``trio``
side on error before maybe raising said error.
- always close the ``trio`` mem chan on exit to ensure the other
side can detect it and follow.

View File

@ -1,16 +0,0 @@
Strictly support Python 3.10+, start runtime machinery reorg
Since we want to push forward using the new `match:` syntax for our
internal RPC-msg loops, we officially drop 3.9 support for the next
release which should coincide well with the first release of 3.11.
This patch set also officially removes the ``tractor.run()`` API (which
has been deprecated for some time) as well as starts an initial re-org
of the internal runtime core by:
- renaming ``tractor._actor`` -> ``._runtime``
- moving the ``._runtime.ActorActor._process_messages()`` and
``._async_main()`` to be module level singleton-task-functions since
they are only started once for each connection and actor spawn
respectively; this internal API thus looks more similar to (at the
time of writing) the ``trio``-internals in ``trio._core._run``.
- officially remove ``tractor.run()``, now deprecated for some time.

View File

@ -1,4 +0,0 @@
Only set `._debug.Lock.local_pdb_complete` if has been created.
This can be triggered by a very rare race condition (and thus we have no
working test yet) but it is known to exist in (a) consumer project(s).

View File

@ -1,25 +0,0 @@
Add support for ``trio >= 0.22`` and support for the new Python 3.11
``[Base]ExceptionGroup`` from `pep 654`_ via the backported
`exceptiongroup`_ package and some final fixes to the debug mode
subsystem.
This port ended up driving some (hopefully) final fixes to our debugger
subsystem including the solution to all lingering stdstreams locking
race-conditions and deadlock scenarios. This includes extending the
debugger tests suite as well as cancellation and ``asyncio`` mode cases.
Some of the notable details:
- always reverting to the ``trio`` SIGINT handler when leaving debug
mode.
- bypassing child attempts to acquire the debug lock when detected
to be amdist actor-runtime-cancellation.
- allowing the root actor to cancel local but IPC-stale subactor
requests-tasks for the debug lock when in a "no IPC peers" state.
Further we refined our ``ActorNursery`` semantics to be more similar to
``trio`` in the sense that parent task errors are always packed into the
actor-nursery emitted exception group and adjusted all tests and
examples accordingly.
.. _pep 654: https://peps.python.org/pep-0654/#handling-exception-groups
.. _exceptiongroup: https://github.com/python-trio/exceptiongroup

View File

@ -1,5 +0,0 @@
Establish an explicit "backend spawning" method table; use it from CI
More clearly lays out the current set of (3) backends: ``['trio',
'mp_spawn', 'mp_forkserver']`` and adjusts the ``._spawn.py`` internals
as well as the test suite to accommodate.

View File

@ -1,4 +0,0 @@
Add ``key: Callable[..., Hashable]`` support to ``.trionics.maybe_open_context()``
Gives users finer grained control over cache hit behaviour using
a callable which receives the input ``kwargs: dict``.

View File

@ -1,41 +0,0 @@
Add support for debug-lock blocking using a ``._debug.Lock._blocked:
set[tuple]`` and add ids when no-more IPC connections with the
root actor are detected.
This is an enhancement which (mostly) solves a lingering debugger
locking race case we needed to handle:
- child crashes acquires TTY lock in root and attaches to ``pdb``
- child IPC goes down such that all channels to the root are broken
/ non-functional.
- root is stuck thinking the child is still in debug even though it
can't be contacted and the child actor machinery hasn't been
cancelled by its parent.
- root get's stuck in deadlock with child since it won't send a cancel
request until the child is finished debugging (to avoid clobbering
a child that is actually using the debugger), but the child can't
unlock the debugger bc IPC is down and it can't contact the root.
To avoid this scenario add debug lock blocking list via
`._debug.Lock._blocked: set[tuple]` which holds actor uids for any actor
that is detected by the root as having no transport channel connections
(of which at least one should exist if this sub-actor at some point
acquired the debug lock). The root consequently checks this list for any
actor that tries to (re)acquire the lock and blocks with
a ``ContextCancelled``. Further, when a debug condition is tested in
``._runtime._invoke``, the context's ``._enter_debugger_on_cancel`` is
set to `False` if the actor was put on the block list then all
post-mortem / crash handling will be bypassed for that task.
In theory this approach to block list management may cause problems
where some nested child actor acquires and releases the lock multiple
times and it gets stuck on the block list after the first use? If this
turns out to be an issue we can try changing the strat so blocks are
only added when the root has zero IPC peers left?
Further, this adds a root-locking-task side cancel scope,
``Lock._root_local_task_cs_in_debug``, which can be ``.cancel()``-ed by the root
runtime when a stale lock is detected during the IPC channel testing.
However, right now we're NOT using this since it seems to cause test
failures likely due to causing pre-mature cancellation and maybe needs
a bit more experimenting?

View File

@ -1,19 +0,0 @@
Rework our ``.trionics.BroadcastReceiver`` internals to avoid method
recursion and approach a design and interface closer to ``trio``'s
``MemoryReceiveChannel``.
The details of the internal changes include:
- implementing a ``BroadcastReceiver.receive_nowait()`` and using it
within the async ``.receive()`` thus avoiding recursion from
``.receive()``.
- failing over to an internal ``._receive_from_underlying()`` when the
``_nowait()`` call raises ``trio.WouldBlock``
- adding ``BroadcastState.statistics()`` for debugging and testing both
internals and by users.
- add an internal ``BroadcastReceiver._raise_on_lag: bool`` which can be
set to avoid ``Lagged`` raising for possible use cases where a user
wants to choose between a [cheap or nasty
pattern](https://zguide.zeromq.org/docs/chapter7/#The-Cheap-or-Nasty-Pattern)
the the particular stream (we use this in ``piker``'s dark clearing
engine to avoid fast feeds breaking during HFT periods).

View File

@ -1,11 +0,0 @@
Always ``list``-cast the ``mngrs`` input to
``.trionics.gather_contexts()`` and ensure its size otherwise raise
a ``ValueError``.
Turns out that trying to pass an inline-style generator comprehension
doesn't seem to work inside the ``async with`` expression? Further, in
such a case we can get a hang waiting on the all-entered event
completion when the internal mngrs iteration is a noop. Instead we
always greedily check a size and error on empty input; the lazy
iteration of a generator input is not beneficial anyway since we're
entering all manager instances in concurrent tasks.

View File

@ -1,15 +0,0 @@
Fixes to ensure IPC (channel) breakage doesn't result in hung actor
trees; the zombie reaping and general supervision machinery will always
clean up and terminate.
This includes not only the (mostly minor) fixes to solve these cases but
also a new extensive test suite in `test_advanced_faults.py` with an
accompanying highly configurable example module-script in
`examples/advanced_faults/ipc_failure_during_stream.py`. Tests ensure we
never get hang or zombies despite operating in debug mode and attempt to
simulate all possible IPC transport failure cases for a local-host actor
tree.
Further we simplify `Context.open_stream.__aexit__()` to just call
`MsgStream.aclose()` directly more or less avoiding a pure duplicate
code path.

View File

@ -1,10 +0,0 @@
Always redraw the `pdbpp` prompt on `SIGINT` during REPL use.
There was recent changes todo with Python 3.10 that required us to pin
to a specific commit in `pdbpp` which have recently been fixed minus
this last issue with `SIGINT` shielding: not clobbering or not
showing the `(Pdb++)` prompt on ctlr-c by the user. This repairs all
that by firstly removing the standard KBI intercepting of the std lib's
`pdb.Pdb._cmdloop()` as well as ensuring that only the actor with REPL
control ever reports `SIGINT` handler log msgs and prompt redraws. With
this we move back to using pypi `pdbpp` release.

View File

@ -1,7 +0,0 @@
Drop `trio.Process.aclose()` usage, copy into our spawning code.
The details are laid out in https://github.com/goodboy/tractor/issues/330.
`trio` changed is process running quite some time ago, this just copies
out the small bit we needed (from the old `.aclose()`) for hard kills
where a soft runtime cancel request fails and our "zombie killer"
implementation kicks in.

View File

@ -1,15 +0,0 @@
Switch to using the fork & fix of `pdb++`, `pdbp`:
https://github.com/mdmintz/pdbp
Allows us to sidestep a variety of issues that aren't being maintained
in the upstream project thanks to the hard work of @mdmintz!
We also include some default settings adjustments as per recent
development on the fork:
- sticky mode is still turned on by default but now activates when
a using the `ll` repl command.
- turn off line truncation by default to avoid inter-line gaps when
resizing the terimnal during use.
- when using the backtrace cmd either by `w` or `bt`, the config
automatically switches to non-sticky mode.

View File

@ -1,28 +0,0 @@
[tool.towncrier]
package = "tractor"
filename = "NEWS.rst"
directory = "nooz/"
version = "0.1.0a6"
title_format = "tractor {version} ({project_date})"
template = "nooz/_template.rst"
all_bullets = true
[[tool.towncrier.type]]
directory = "feature"
name = "Features"
showcontent = true
[[tool.towncrier.type]]
directory = "bugfix"
name = "Bug Fixes"
showcontent = true
[[tool.towncrier.type]]
directory = "doc"
name = "Improved Documentation"
showcontent = true
[[tool.towncrier.type]]
directory = "trivial"
name = "Trivial/Internal Changes"
showcontent = true

View File

@ -1,8 +1,7 @@
pytest pytest
pytest-trio pytest-trio
pytest-timeout pdbpp
pdbp mypy<0.920
mypy trio_typing<0.7.0
trio_typing
pexpect pexpect
towncrier towncrier

View File

@ -25,55 +25,47 @@ with open('docs/README.rst', encoding='utf-8') as f:
setup( setup(
name="tractor", name="tractor",
version='0.1.0a6dev0', # alpha zone version='0.1.0a5.dev', # alpha zone
description='structured concurrrent `trio`-"actors"', description='structured concurrrent "actors"',
long_description=readme, long_description=readme,
license='AGPLv3', license='AGPLv3',
author='Tyler Goodlet', author='Tyler Goodlet',
maintainer='Tyler Goodlet', maintainer='Tyler Goodlet',
maintainer_email='goodboy_foss@protonmail.com', maintainer_email='jgbt@protonmail.com',
url='https://github.com/goodboy/tractor', url='https://github.com/goodboy/tractor',
platforms=['linux', 'windows'], platforms=['linux', 'windows'],
packages=[ packages=[
'tractor', 'tractor',
'tractor.experimental', 'tractor.experimental',
'tractor.trionics', 'tractor.trionics',
'tractor.testing',
], ],
install_requires=[ install_requires=[
# trio related # trio related
# proper range spec: 'trio>0.8',
# https://packaging.python.org/en/latest/discussions/install-requires-vs-requirements/#id5
'trio >= 0.22',
'async_generator', 'async_generator',
'trio_typing', 'trio_typing',
'exceptiongroup',
# tooling # tooling
'tricycle', 'tricycle',
'trio_typing', 'trio_typing',
# tooling
'colorlog', 'colorlog',
'wrapt', 'wrapt',
'pdbpp',
# IPC serialization
'msgspec',
# debug mode REPL
'pdbp',
# pip ref docs on these specs:
# https://pip.pypa.io/en/stable/reference/requirement-specifiers/#examples
# and pep:
# https://peps.python.org/pep-0440/#version-specifiers
# windows deps workaround for ``pdbpp`` # windows deps workaround for ``pdbpp``
# https://github.com/pdbpp/pdbpp/issues/498 # https://github.com/pdbpp/pdbpp/issues/498
# https://github.com/pdbpp/fancycompleter/issues/37 # https://github.com/pdbpp/fancycompleter/issues/37
'pyreadline3 ; platform_system == "Windows"', 'pyreadline3 ; platform_system == "Windows"',
# serialization
'msgspec >= "0.4.0"'
], ],
tests_require=['pytest'], tests_require=['pytest'],
python_requires=">=3.10", python_requires=">=3.9",
keywords=[ keywords=[
'trio', 'trio',
'async', 'async',
@ -92,6 +84,7 @@ setup(
"Programming Language :: Python :: Implementation :: CPython", "Programming Language :: Python :: Implementation :: CPython",
"Programming Language :: Python :: 3 :: Only", "Programming Language :: Python :: 3 :: Only",
"Programming Language :: Python :: 3.10", "Programming Language :: Python :: 3.10",
"Programming Language :: Python :: 3.9",
"Intended Audience :: Science/Research", "Intended Audience :: Science/Research",
"Intended Audience :: Developers", "Intended Audience :: Developers",
"Topic :: System :: Distributed Computing", "Topic :: System :: Distributed Computing",

View File

@ -7,91 +7,16 @@ import os
import random import random
import signal import signal
import platform import platform
import pathlib
import time import time
import inspect
from functools import partial, wraps
import pytest import pytest
import trio
import tractor import tractor
# export for tests
from tractor.testing import tractor_test # noqa
pytest_plugins = ['pytester'] pytest_plugins = ['pytester']
def tractor_test(fn):
"""
Use:
@tractor_test
async def test_whatever():
await ...
If fixtures:
- ``arb_addr`` (a socket addr tuple where arbiter is listening)
- ``loglevel`` (logging level passed to tractor internals)
- ``start_method`` (subprocess spawning backend)
are defined in the `pytest` fixture space they will be automatically
injected to tests declaring these funcargs.
"""
@wraps(fn)
def wrapper(
*args,
loglevel=None,
arb_addr=None,
start_method=None,
**kwargs
):
# __tracebackhide__ = True
if 'arb_addr' in inspect.signature(fn).parameters:
# injects test suite fixture value to test as well
# as `run()`
kwargs['arb_addr'] = arb_addr
if 'loglevel' in inspect.signature(fn).parameters:
# allows test suites to define a 'loglevel' fixture
# that activates the internal logging
kwargs['loglevel'] = loglevel
if start_method is None:
if platform.system() == "Windows":
start_method = 'trio'
if 'start_method' in inspect.signature(fn).parameters:
# set of subprocess spawning backends
kwargs['start_method'] = start_method
if kwargs:
# use explicit root actor start
async def _main():
async with tractor.open_root_actor(
# **kwargs,
arbiter_addr=arb_addr,
loglevel=loglevel,
start_method=start_method,
# TODO: only enable when pytest is passed --pdb
# debug_mode=True,
):
await fn(*args, **kwargs)
main = _main
else:
# use implicit root actor start
main = partial(fn, *args, **kwargs)
return trio.run(main)
return wrapper
_arb_addr = '127.0.0.1', random.randint(1000, 9999) _arb_addr = '127.0.0.1', random.randint(1000, 9999)
@ -114,21 +39,14 @@ no_windows = pytest.mark.skipif(
) )
def repodir() -> pathlib.Path: def repodir():
''' """Return the abspath to the repo directory.
Return the abspath to the repo directory. """
dirname = os.path.dirname
''' dirpath = os.path.abspath(
# 2 parents up to step up through tests/<repo_dir> dirname(dirname(os.path.realpath(__file__)))
return pathlib.Path(__file__).parent.parent.absolute() )
return dirpath
def examples_dir() -> pathlib.Path:
'''
Return the abspath to the examples directory as `pathlib.Path`.
'''
return repodir() / 'examples'
def pytest_addoption(parser): def pytest_addoption(parser):
@ -146,6 +64,10 @@ def pytest_addoption(parser):
def pytest_configure(config): def pytest_configure(config):
backend = config.option.spawn_backend backend = config.option.spawn_backend
if backend == 'mp':
tractor._spawn.try_set_start_method('spawn')
elif backend == 'trio':
tractor._spawn.try_set_start_method(backend) tractor._spawn.try_set_start_method(backend)
@ -159,18 +81,15 @@ def loglevel(request):
@pytest.fixture(scope='session') @pytest.fixture(scope='session')
def spawn_backend(request) -> str: def spawn_backend(request):
return request.config.option.spawn_backend return request.config.option.spawn_backend
_ci_env: bool = os.environ.get('CI', False)
@pytest.fixture(scope='session') @pytest.fixture(scope='session')
def ci_env() -> bool: def ci_env() -> bool:
"""Detect CI envoirment. """Detect CI envoirment.
""" """
return _ci_env return os.environ.get('TRAVIS', False) or os.environ.get('CI', False)
@pytest.fixture(scope='session') @pytest.fixture(scope='session')
@ -180,24 +99,24 @@ def arb_addr():
def pytest_generate_tests(metafunc): def pytest_generate_tests(metafunc):
spawn_backend = metafunc.config.option.spawn_backend spawn_backend = metafunc.config.option.spawn_backend
if not spawn_backend: if not spawn_backend:
# XXX some weird windows bug with `pytest`? # XXX some weird windows bug with `pytest`?
spawn_backend = 'trio' spawn_backend = 'mp'
assert spawn_backend in ('mp', 'trio')
# TODO: maybe just use the literal `._spawn.SpawnMethodKey`?
assert spawn_backend in (
'mp_spawn',
'mp_forkserver',
'trio',
)
# NOTE: used to be used to dyanmically parametrize tests for when
# you just passed --spawn-backend=`mp` on the cli, but now we expect
# that cli input to be manually specified, BUT, maybe we'll do
# something like this again in the future?
if 'start_method' in metafunc.fixturenames: if 'start_method' in metafunc.fixturenames:
metafunc.parametrize("start_method", [spawn_backend], scope='module') if spawn_backend == 'mp':
from multiprocessing import get_all_start_methods
methods = get_all_start_methods()
if 'fork' in methods:
# fork not available on windows, so check before
# removing XXX: the fork method is in general
# incompatible with trio's global scheduler state
methods.remove('fork')
elif spawn_backend == 'trio':
methods = ['trio']
metafunc.parametrize("start_method", methods, scope='module')
def sig_prog(proc, sig): def sig_prog(proc, sig):
@ -213,22 +132,16 @@ def sig_prog(proc, sig):
@pytest.fixture @pytest.fixture
def daemon( def daemon(loglevel, testdir, arb_addr):
loglevel: str, """Run a daemon actor as a "remote arbiter".
testdir, """
arb_addr: tuple[str, int],
):
'''
Run a daemon actor as a "remote arbiter".
'''
if loglevel in ('trace', 'debug'): if loglevel in ('trace', 'debug'):
# too much logging will lock up the subproc (smh) # too much logging will lock up the subproc (smh)
loglevel = 'info' loglevel = 'info'
cmdargs = [ cmdargs = [
sys.executable, '-c', sys.executable, '-c',
"import tractor; tractor.run_daemon([], registry_addr={}, loglevel={})" "import tractor; tractor.run_daemon([], arbiter_addr={}, loglevel={})"
.format( .format(
arb_addr, arb_addr,
"'{}'".format(loglevel) if loglevel else None) "'{}'".format(loglevel) if loglevel else None)

View File

@ -1,193 +0,0 @@
'''
Sketchy network blackoutz, ugly byzantine gens, puedes eschuchar la
cancelacion?..
'''
from functools import partial
import pytest
from _pytest.pathlib import import_path
import trio
import tractor
from conftest import (
examples_dir,
)
@pytest.mark.parametrize(
'debug_mode',
[False, True],
ids=['no_debug_mode', 'debug_mode'],
)
@pytest.mark.parametrize(
'ipc_break',
[
# no breaks
{
'break_parent_ipc_after': False,
'break_child_ipc_after': False,
},
# only parent breaks
{
'break_parent_ipc_after': 500,
'break_child_ipc_after': False,
},
# only child breaks
{
'break_parent_ipc_after': False,
'break_child_ipc_after': 500,
},
# both: break parent first
{
'break_parent_ipc_after': 500,
'break_child_ipc_after': 800,
},
# both: break child first
{
'break_parent_ipc_after': 800,
'break_child_ipc_after': 500,
},
],
ids=[
'no_break',
'break_parent',
'break_child',
'break_both_parent_first',
'break_both_child_first',
],
)
def test_ipc_channel_break_during_stream(
debug_mode: bool,
spawn_backend: str,
ipc_break: dict | None,
):
'''
Ensure we can have an IPC channel break its connection during
streaming and it's still possible for the (simulated) user to kill
the actor tree using SIGINT.
We also verify the type of connection error expected in the parent
depending on which side if the IPC breaks first.
'''
if spawn_backend != 'trio':
if debug_mode:
pytest.skip('`debug_mode` only supported on `trio` spawner')
# non-`trio` spawners should never hit the hang condition that
# requires the user to do ctl-c to cancel the actor tree.
expect_final_exc = trio.ClosedResourceError
mod = import_path(
examples_dir() / 'advanced_faults' / 'ipc_failure_during_stream.py',
root=examples_dir(),
)
expect_final_exc = KeyboardInterrupt
# when ONLY the child breaks we expect the parent to get a closed
# resource error on the next `MsgStream.receive()` and then fail out
# and cancel the child from there.
if (
# only child breaks
(
ipc_break['break_child_ipc_after']
and ipc_break['break_parent_ipc_after'] is False
)
# both break but, parent breaks first
or (
ipc_break['break_child_ipc_after'] is not False
and (
ipc_break['break_parent_ipc_after']
> ipc_break['break_child_ipc_after']
)
)
):
expect_final_exc = trio.ClosedResourceError
# when the parent IPC side dies (even if the child's does as well
# but the child fails BEFORE the parent) we expect the channel to be
# sent a stop msg from the child at some point which will signal the
# parent that the stream has been terminated.
# NOTE: when the parent breaks "after" the child you get this same
# case as well, the child breaks the IPC channel with a stop msg
# before any closure takes place.
elif (
# only parent breaks
(
ipc_break['break_parent_ipc_after']
and ipc_break['break_child_ipc_after'] is False
)
# both break but, child breaks first
or (
ipc_break['break_parent_ipc_after'] is not False
and (
ipc_break['break_child_ipc_after']
> ipc_break['break_parent_ipc_after']
)
)
):
expect_final_exc = trio.EndOfChannel
with pytest.raises(expect_final_exc):
trio.run(
partial(
mod.main,
debug_mode=debug_mode,
start_method=spawn_backend,
**ipc_break,
)
)
@tractor.context
async def break_ipc_after_started(
ctx: tractor.Context,
) -> None:
await ctx.started()
async with ctx.open_stream() as stream:
await stream.aclose()
await trio.sleep(0.2)
await ctx.chan.send(None)
print('child broke IPC and terminating')
def test_stream_closed_right_after_ipc_break_and_zombie_lord_engages():
'''
Verify that is a subactor's IPC goes down just after bringing up a stream
the parent can trigger a SIGINT and the child will be reaped out-of-IPC by
the localhost process supervision machinery: aka "zombie lord".
'''
async def main():
async with tractor.open_nursery() as n:
portal = await n.start_actor(
'ipc_breaker',
enable_modules=[__name__],
)
with trio.move_on_after(1):
async with (
portal.open_context(
break_ipc_after_started
) as (ctx, sent),
):
async with ctx.open_stream():
await trio.sleep(0.5)
print('parent waiting on context')
print('parent exited context')
raise KeyboardInterrupt
with pytest.raises(KeyboardInterrupt):
trio.run(main)

View File

@ -5,6 +5,7 @@ Advanced streaming patterns using bidirectional streams and contexts.
from collections import Counter from collections import Counter
import itertools import itertools
import platform import platform
from typing import Set, Dict, List
import trio import trio
import tractor import tractor
@ -14,7 +15,7 @@ def is_win():
return platform.system() == 'Windows' return platform.system() == 'Windows'
_registry: dict[str, set[tractor.MsgStream]] = { _registry: Dict[str, Set[tractor.ReceiveMsgStream]] = {
'even': set(), 'even': set(),
'odd': set(), 'odd': set(),
} }
@ -76,7 +77,7 @@ async def subscribe(
async def consumer( async def consumer(
subs: list[str], subs: List[str],
) -> None: ) -> None:

View File

@ -8,10 +8,6 @@ import platform
import time import time
from itertools import repeat from itertools import repeat
from exceptiongroup import (
BaseExceptionGroup,
ExceptionGroup,
)
import pytest import pytest
import trio import trio
import tractor import tractor
@ -60,49 +56,29 @@ def test_remote_error(arb_addr, args_err):
arbiter_addr=arb_addr, arbiter_addr=arb_addr,
) as nursery: ) as nursery:
# on a remote type error caused by bad input args
# this should raise directly which means we **don't** get
# an exception group outside the nursery since the error
# here and the far end task error are one in the same?
portal = await nursery.run_in_actor( portal = await nursery.run_in_actor(
assert_err, name='errorer', **args assert_err, name='errorer', **args
) )
# get result(s) from main task # get result(s) from main task
try: try:
# this means the root actor will also raise a local
# parent task error and thus an eg will propagate out
# of this actor nursery.
await portal.result() await portal.result()
except tractor.RemoteActorError as err: except tractor.RemoteActorError as err:
assert err.type == errtype assert err.type == errtype
print("Look Maa that actor failed hard, hehh") print("Look Maa that actor failed hard, hehh")
raise raise
# ensure boxed errors
if args:
with pytest.raises(tractor.RemoteActorError) as excinfo: with pytest.raises(tractor.RemoteActorError) as excinfo:
trio.run(main) trio.run(main)
# ensure boxed error is correct
assert excinfo.value.type == errtype assert excinfo.value.type == errtype
else:
# the root task will also error on the `.result()` call
# so we expect an error from there AND the child.
with pytest.raises(BaseExceptionGroup) as excinfo:
trio.run(main)
# ensure boxed errors
for exc in excinfo.value.exceptions:
assert exc.type == errtype
def test_multierror(arb_addr): def test_multierror(arb_addr):
''' """Verify we raise a ``trio.MultiError`` out of a nursery where
Verify we raise a ``BaseExceptionGroup`` out of a nursery where
more then one actor errors. more then one actor errors.
"""
'''
async def main(): async def main():
async with tractor.open_nursery( async with tractor.open_nursery(
arbiter_addr=arb_addr, arbiter_addr=arb_addr,
@ -119,10 +95,10 @@ def test_multierror(arb_addr):
print("Look Maa that first actor failed hard, hehh") print("Look Maa that first actor failed hard, hehh")
raise raise
# here we should get a ``BaseExceptionGroup`` containing exceptions # here we should get a `trio.MultiError` containing exceptions
# from both subactors # from both subactors
with pytest.raises(BaseExceptionGroup): with pytest.raises(trio.MultiError):
trio.run(main) trio.run(main)
@ -131,7 +107,7 @@ def test_multierror(arb_addr):
'num_subactors', range(25, 26), 'num_subactors', range(25, 26),
) )
def test_multierror_fast_nursery(arb_addr, start_method, num_subactors, delay): def test_multierror_fast_nursery(arb_addr, start_method, num_subactors, delay):
"""Verify we raise a ``BaseExceptionGroup`` out of a nursery where """Verify we raise a ``trio.MultiError`` out of a nursery where
more then one actor errors and also with a delay before failure more then one actor errors and also with a delay before failure
to test failure during an ongoing spawning. to test failure during an ongoing spawning.
""" """
@ -147,11 +123,10 @@ def test_multierror_fast_nursery(arb_addr, start_method, num_subactors, delay):
delay=delay delay=delay
) )
# with pytest.raises(trio.MultiError) as exc_info: with pytest.raises(trio.MultiError) as exc_info:
with pytest.raises(BaseExceptionGroup) as exc_info:
trio.run(main) trio.run(main)
assert exc_info.type == ExceptionGroup assert exc_info.type == tractor.MultiError
err = exc_info.value err = exc_info.value
exceptions = err.exceptions exceptions = err.exceptions
@ -239,8 +214,8 @@ async def test_cancel_infinite_streamer(start_method):
[ [
# daemon actors sit idle while single task actors error out # daemon actors sit idle while single task actors error out
(1, tractor.RemoteActorError, AssertionError, (assert_err, {}), None), (1, tractor.RemoteActorError, AssertionError, (assert_err, {}), None),
(2, BaseExceptionGroup, AssertionError, (assert_err, {}), None), (2, tractor.MultiError, AssertionError, (assert_err, {}), None),
(3, BaseExceptionGroup, AssertionError, (assert_err, {}), None), (3, tractor.MultiError, AssertionError, (assert_err, {}), None),
# 1 daemon actor errors out while single task actors sleep forever # 1 daemon actor errors out while single task actors sleep forever
(3, tractor.RemoteActorError, AssertionError, (sleep_forever, {}), (3, tractor.RemoteActorError, AssertionError, (sleep_forever, {}),
@ -251,7 +226,7 @@ async def test_cancel_infinite_streamer(start_method):
(do_nuthin, {}), (assert_err, {'delay': 1}, True)), (do_nuthin, {}), (assert_err, {'delay': 1}, True)),
# daemon complete quickly delay while single task # daemon complete quickly delay while single task
# actors error after brief delay # actors error after brief delay
(3, BaseExceptionGroup, AssertionError, (3, tractor.MultiError, AssertionError,
(assert_err, {'delay': 1}), (do_nuthin, {}, False)), (assert_err, {'delay': 1}), (do_nuthin, {}, False)),
], ],
ids=[ ids=[
@ -318,7 +293,7 @@ async def test_some_cancels_all(num_actors_and_errs, start_method, loglevel):
# should error here with a ``RemoteActorError`` or ``MultiError`` # should error here with a ``RemoteActorError`` or ``MultiError``
except first_err as err: except first_err as err:
if isinstance(err, BaseExceptionGroup): if isinstance(err, tractor.MultiError):
assert len(err.exceptions) == num_actors assert len(err.exceptions) == num_actors
for exc in err.exceptions: for exc in err.exceptions:
if isinstance(exc, tractor.RemoteActorError): if isinstance(exc, tractor.RemoteActorError):
@ -362,7 +337,7 @@ async def spawn_and_error(breadth, depth) -> None:
@tractor_test @tractor_test
async def test_nested_multierrors(loglevel, start_method): async def test_nested_multierrors(loglevel, start_method):
''' '''
Test that failed actor sets are wrapped in `BaseExceptionGroup`s. This Test that failed actor sets are wrapped in `trio.MultiError`s. This
test goes only 2 nurseries deep but we should eventually have tests test goes only 2 nurseries deep but we should eventually have tests
for arbitrary n-depth actor trees. for arbitrary n-depth actor trees.
@ -390,7 +365,7 @@ async def test_nested_multierrors(loglevel, start_method):
breadth=subactor_breadth, breadth=subactor_breadth,
depth=depth, depth=depth,
) )
except BaseExceptionGroup as err: except trio.MultiError as err:
assert len(err.exceptions) == subactor_breadth assert len(err.exceptions) == subactor_breadth
for subexc in err.exceptions: for subexc in err.exceptions:
@ -408,10 +383,10 @@ async def test_nested_multierrors(loglevel, start_method):
assert subexc.type in ( assert subexc.type in (
tractor.RemoteActorError, tractor.RemoteActorError,
trio.Cancelled, trio.Cancelled,
BaseExceptionGroup, trio.MultiError
) )
elif isinstance(subexc, BaseExceptionGroup): elif isinstance(subexc, trio.MultiError):
for subsub in subexc.exceptions: for subsub in subexc.exceptions:
if subsub in (tractor.RemoteActorError,): if subsub in (tractor.RemoteActorError,):
@ -419,7 +394,7 @@ async def test_nested_multierrors(loglevel, start_method):
assert type(subsub) in ( assert type(subsub) in (
trio.Cancelled, trio.Cancelled,
BaseExceptionGroup, trio.MultiError,
) )
else: else:
assert isinstance(subexc, tractor.RemoteActorError) assert isinstance(subexc, tractor.RemoteActorError)
@ -431,13 +406,13 @@ async def test_nested_multierrors(loglevel, start_method):
if is_win(): if is_win():
if isinstance(subexc, tractor.RemoteActorError): if isinstance(subexc, tractor.RemoteActorError):
assert subexc.type in ( assert subexc.type in (
BaseExceptionGroup, trio.MultiError,
tractor.RemoteActorError tractor.RemoteActorError
) )
else: else:
assert isinstance(subexc, BaseExceptionGroup) assert isinstance(subexc, trio.MultiError)
else: else:
assert subexc.type is ExceptionGroup assert subexc.type is trio.MultiError
else: else:
assert subexc.type in ( assert subexc.type in (
tractor.RemoteActorError, tractor.RemoteActorError,
@ -460,7 +435,7 @@ def test_cancel_via_SIGINT(
with trio.fail_after(2): with trio.fail_after(2):
async with tractor.open_nursery() as tn: async with tractor.open_nursery() as tn:
await tn.start_actor('sucka') await tn.start_actor('sucka')
if 'mp' in spawn_backend: if spawn_backend == 'mp':
time.sleep(0.1) time.sleep(0.1)
os.kill(pid, signal.SIGINT) os.kill(pid, signal.SIGINT)
await trio.sleep_forever() await trio.sleep_forever()
@ -499,7 +474,7 @@ def test_cancel_via_SIGINT_other_task(
with trio.fail_after(timeout): with trio.fail_after(timeout):
async with trio.open_nursery() as n: async with trio.open_nursery() as n:
await n.start(spawn_and_sleep_forever) await n.start(spawn_and_sleep_forever)
if 'mp' in spawn_backend: if spawn_backend == 'mp':
time.sleep(0.1) time.sleep(0.1)
os.kill(pid, signal.SIGINT) os.kill(pid, signal.SIGINT)

View File

@ -1,6 +1,5 @@
import itertools import itertools
import pytest
import trio import trio
import tractor import tractor
from tractor import open_actor_cluster from tractor import open_actor_cluster
@ -12,72 +11,26 @@ from conftest import tractor_test
MESSAGE = 'tractoring at full speed' MESSAGE = 'tractoring at full speed'
def test_empty_mngrs_input_raises() -> None:
async def main():
with trio.fail_after(1):
async with (
open_actor_cluster(
modules=[__name__],
# NOTE: ensure we can passthrough runtime opts
loglevel='info',
# debug_mode=True,
) as portals,
gather_contexts(
# NOTE: it's the use of inline-generator syntax
# here that causes the empty input.
mngrs=(
p.open_context(worker) for p in portals.values()
),
),
):
assert 0
with pytest.raises(ValueError):
trio.run(main)
@tractor.context @tractor.context
async def worker( async def worker(ctx: tractor.Context) -> None:
ctx: tractor.Context,
) -> None:
await ctx.started() await ctx.started()
async with ctx.open_stream(backpressure=True) as stream:
async with ctx.open_stream(
backpressure=True,
) as stream:
# TODO: this with the below assert causes a hang bug?
# with trio.move_on_after(1):
async for msg in stream: async for msg in stream:
# do something with msg # do something with msg
print(msg) print(msg)
assert msg == MESSAGE assert msg == MESSAGE
# TODO: does this ever cause a hang
# assert 0
@tractor_test @tractor_test
async def test_streaming_to_actor_cluster() -> None: async def test_streaming_to_actor_cluster() -> None:
async with ( async with (
open_actor_cluster(modules=[__name__]) as portals, open_actor_cluster(modules=[__name__]) as portals,
gather_contexts( gather_contexts(
mngrs=[p.open_context(worker) for p in portals.values()], mngrs=[p.open_context(worker) for p in portals.values()],
) as contexts, ) as contexts,
gather_contexts( gather_contexts(
mngrs=[ctx[0].open_stream() for ctx in contexts], mngrs=[ctx[0].open_stream() for ctx in contexts],
) as streams, ) as streams,
): ):
with trio.move_on_after(1): with trio.move_on_after(1):
for stream in itertools.cycle(streams): for stream in itertools.cycle(streams):

View File

@ -265,7 +265,6 @@ async def test_callee_closes_ctx_after_stream_open():
enable_modules=[__name__], enable_modules=[__name__],
) )
with trio.fail_after(2):
async with portal.open_context( async with portal.open_context(
close_ctx_immediately, close_ctx_immediately,
@ -298,7 +297,6 @@ async def test_callee_closes_ctx_after_stream_open():
# of a stream to the context (at least until a time of # of a stream to the context (at least until a time of
# if/when we decide that's a good idea?) # if/when we decide that's a good idea?)
try: try:
with trio.fail_after(0.5):
async with ctx.open_stream() as stream: async with ctx.open_stream() as stream:
pass pass
except trio.ClosedResourceError: except trio.ClosedResourceError:
@ -571,7 +569,7 @@ def test_one_end_stream_not_opened(overrun_by):
''' '''
overrunner, buf_size_increase, entrypoint = overrun_by overrunner, buf_size_increase, entrypoint = overrun_by
from tractor._runtime import Actor from tractor._actor import Actor
buf_size = buf_size_increase + Actor.msg_buffer_size buf_size = buf_size_increase + Actor.msg_buffer_size
async def main(): async def main():

View File

@ -1,5 +1,5 @@
""" """
That "native" debug mode better work! That native debug better work!
All these tests can be understood (somewhat) by running the equivalent All these tests can be understood (somewhat) by running the equivalent
`examples/debugging/` scripts manually. `examples/debugging/` scripts manually.
@ -10,25 +10,15 @@ TODO:
- wonder if any of it'll work on OS X? - wonder if any of it'll work on OS X?
""" """
import itertools
from os import path
from typing import Optional
import platform
import pathlib
import sys
import time import time
from os import path
import platform
import pytest import pytest
import pexpect import pexpect
from pexpect.exceptions import (
TIMEOUT,
EOF,
)
from conftest import ( from conftest import repodir
examples_dir,
_ci_env,
)
# TODO: The next great debugger audit could be done by you! # TODO: The next great debugger audit could be done by you!
# - recurrent entry to breakpoint() from single actor *after* and an # - recurrent entry to breakpoint() from single actor *after* and an
@ -47,31 +37,19 @@ if platform.system() == 'Windows':
) )
def examples_dir():
"""Return the abspath to the examples directory.
"""
return path.join(repodir(), 'examples', 'debugging/')
def mk_cmd(ex_name: str) -> str: def mk_cmd(ex_name: str) -> str:
''' """Generate a command suitable to pass to ``pexpect.spawn()``.
Generate a command suitable to pass to ``pexpect.spawn()``. """
return ' '.join(
''' ['python',
script_path: pathlib.Path = examples_dir() / 'debugging' / f'{ex_name}.py' path.join(examples_dir(), f'{ex_name}.py')]
return ' '.join(['python', str(script_path)]) )
# TODO: was trying to this xfail style but some weird bug i see in CI
# that's happening at collect time.. pretty soon gonna dump actions i'm
# thinkin...
# in CI we skip tests which >= depth 1 actor trees due to there
# still being an oustanding issue with relaying the debug-mode-state
# through intermediary parents.
has_nested_actors = pytest.mark.has_nested_actors
# .xfail(
# os.environ.get('CI', False),
# reason=(
# 'This test uses nested actors and fails in CI\n'
# 'The test seems to run fine locally but until we solve the '
# 'following issue this CI test will be xfail:\n'
# 'https://github.com/goodboy/tractor/issues/320'
# )
# )
@pytest.fixture @pytest.fixture
@ -95,83 +73,6 @@ def spawn(
return _spawn return _spawn
PROMPT = r"\(Pdb\+\)"
def expect(
child,
# prompt by default
patt: str = PROMPT,
**kwargs,
) -> None:
'''
Expect wrapper that prints last seen console
data before failing.
'''
try:
child.expect(
patt,
**kwargs,
)
except TIMEOUT:
before = str(child.before.decode())
print(before)
raise
def assert_before(
child,
patts: list[str],
) -> None:
before = str(child.before.decode())
for patt in patts:
try:
assert patt in before
except AssertionError:
print(before)
raise
@pytest.fixture(
params=[False, True],
ids='ctl-c={}'.format,
)
def ctlc(
request,
ci_env: bool,
) -> bool:
use_ctlc = request.param
node = request.node
markers = node.own_markers
for mark in markers:
if mark.name == 'has_nested_actors':
pytest.skip(
f'Test {node} has nested actors and fails with Ctrl-C.\n'
f'The test can sometimes run fine locally but until'
' we solve' 'this issue this CI test will be xfail:\n'
'https://github.com/goodboy/tractor/issues/320'
)
if use_ctlc:
# XXX: disable pygments highlighting for auto-tests
# since some envs (like actions CI) will struggle
# the the added color-char encoding..
from tractor._debug import TractorConfig
TractorConfig.use_pygements = False
yield use_ctlc
@pytest.mark.parametrize( @pytest.mark.parametrize(
'user_in_out', 'user_in_out',
[ [
@ -181,16 +82,14 @@ def ctlc(
ids=lambda item: f'{item[0]} -> {item[1]}', ids=lambda item: f'{item[0]} -> {item[1]}',
) )
def test_root_actor_error(spawn, user_in_out): def test_root_actor_error(spawn, user_in_out):
''' """Demonstrate crash handler entering pdbpp from basic error in root actor.
Demonstrate crash handler entering pdb from basic error in root actor. """
'''
user_input, expect_err_str = user_in_out user_input, expect_err_str = user_in_out
child = spawn('root_actor_error') child = spawn('root_actor_error')
# scan for the prompt # scan for the pdbpp prompt
expect(child, PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
@ -202,7 +101,7 @@ def test_root_actor_error(spawn, user_in_out):
child.sendline(user_input) child.sendline(user_input)
# process should exit # process should exit
expect(child, EOF) child.expect(pexpect.EOF)
assert expect_err_str in str(child.before) assert expect_err_str in str(child.before)
@ -220,8 +119,8 @@ def test_root_actor_bp(spawn, user_in_out):
user_input, expect_err_str = user_in_out user_input, expect_err_str = user_in_out
child = spawn('root_actor_breakpoint') child = spawn('root_actor_breakpoint')
# scan for the prompt # scan for the pdbpp prompt
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
assert 'Error' not in str(child.before) assert 'Error' not in str(child.before)
@ -238,129 +137,56 @@ def test_root_actor_bp(spawn, user_in_out):
assert expect_err_str in str(child.before) assert expect_err_str in str(child.before)
def do_ctlc( def test_root_actor_bp_forever(spawn):
child,
count: int = 3,
delay: float = 0.1,
patt: Optional[str] = None,
# expect repl UX to reprint the prompt after every
# ctrl-c send.
# XXX: no idea but, in CI this never seems to work even on 3.10 so
# needs some further investigation potentially...
expect_prompt: bool = not _ci_env,
) -> None:
# make sure ctl-c sends don't do anything but repeat output
for _ in range(count):
time.sleep(delay)
child.sendcontrol('c')
# TODO: figure out why this makes CI fail..
# if you run this test manually it works just fine..
if expect_prompt:
before = str(child.before.decode())
time.sleep(delay)
child.expect(PROMPT)
time.sleep(delay)
if patt:
# should see the last line on console
assert patt in before
def test_root_actor_bp_forever(
spawn,
ctlc: bool,
):
"Re-enter a breakpoint from the root actor-task." "Re-enter a breakpoint from the root actor-task."
child = spawn('root_actor_breakpoint_forever') child = spawn('root_actor_breakpoint_forever')
# do some "next" commands to demonstrate recurrent breakpoint # do some "next" commands to demonstrate recurrent breakpoint
# entries # entries
for _ in range(10): for _ in range(10):
child.expect(PROMPT)
if ctlc:
do_ctlc(child)
child.sendline('next') child.sendline('next')
child.expect(r"\(Pdb\+\+\)")
# do one continue which should trigger a # do one continue which should trigger a new task to lock the tty
# new task to lock the tty
child.sendline('continue') child.sendline('continue')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
# seems that if we hit ctrl-c too fast the
# sigint guard machinery might not kick in..
time.sleep(0.001)
if ctlc:
do_ctlc(child)
# XXX: this previously caused a bug! # XXX: this previously caused a bug!
child.sendline('n') child.sendline('n')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
child.sendline('n') child.sendline('n')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
# quit out of the loop
child.sendline('q')
child.expect(pexpect.EOF)
@pytest.mark.parametrize( def test_subactor_error(spawn):
'do_next', "Single subactor raising an error"
(True, False),
ids='do_next={}'.format,
)
def test_subactor_error(
spawn,
ctlc: bool,
do_next: bool,
):
'''
Single subactor raising an error
'''
child = spawn('subactor_error') child = spawn('subactor_error')
# scan for the prompt # scan for the pdbpp prompt
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "Attaching to pdb in crashed actor: ('name_error'" in before assert "Attaching to pdb in crashed actor: ('name_error'" in before
if do_next: # send user command
child.sendline('n') # (in this case it's the same for 'continue' vs. 'quit')
else:
# make sure ctl-c sends don't do anything but repeat output
if ctlc:
do_ctlc(
child,
)
# send user command and (in this case it's the same for 'continue'
# vs. 'quit') the debugger should enter a second time in the nursery
# creating actor
child.sendline('continue') child.sendline('continue')
child.expect(PROMPT) # the debugger should enter a second time in the nursery
# creating actor
child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
# root actor gets debugger engaged # root actor gets debugger engaged
assert "Attaching to pdb in crashed actor: ('root'" in before assert "Attaching to pdb in crashed actor: ('root'" in before
# error is a remote error propagated from the subactor # error is a remote error propagated from the subactor
assert "RemoteActorError: ('name_error'" in before assert "RemoteActorError: ('name_error'" in before
# another round
if ctlc:
do_ctlc(child)
child.sendline('c') child.sendline('c')
child.expect('\r\n') child.expect('\r\n')
@ -368,16 +194,13 @@ def test_subactor_error(
child.expect(pexpect.EOF) child.expect(pexpect.EOF)
def test_subactor_breakpoint( def test_subactor_breakpoint(spawn):
spawn,
ctlc: bool,
):
"Single subactor with an infinite breakpoint loop" "Single subactor with an infinite breakpoint loop"
child = spawn('subactor_breakpoint') child = spawn('subactor_breakpoint')
# scan for the prompt # scan for the pdbpp prompt
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "Attaching pdb to actor: ('breakpoint_forever'" in before assert "Attaching pdb to actor: ('breakpoint_forever'" in before
@ -386,34 +209,25 @@ def test_subactor_breakpoint(
# entries # entries
for _ in range(10): for _ in range(10):
child.sendline('next') child.sendline('next')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
if ctlc:
do_ctlc(child)
# now run some "continues" to show re-entries # now run some "continues" to show re-entries
for _ in range(5): for _ in range(5):
child.sendline('continue') child.sendline('continue')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "Attaching pdb to actor: ('breakpoint_forever'" in before assert "Attaching pdb to actor: ('breakpoint_forever'" in before
if ctlc:
do_ctlc(child)
# finally quit the loop # finally quit the loop
child.sendline('q') child.sendline('q')
# child process should exit but parent will capture pdb.BdbQuit # child process should exit but parent will capture pdb.BdbQuit
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "RemoteActorError: ('breakpoint_forever'" in before assert "RemoteActorError: ('breakpoint_forever'" in before
assert 'bdb.BdbQuit' in before assert 'bdb.BdbQuit' in before
if ctlc:
do_ctlc(child)
# quit the parent # quit the parent
child.sendline('c') child.sendline('c')
@ -425,159 +239,110 @@ def test_subactor_breakpoint(
assert 'bdb.BdbQuit' in before assert 'bdb.BdbQuit' in before
@has_nested_actors def test_multi_subactors(spawn):
def test_multi_subactors( """
spawn, Multiple subactors, both erroring and breakpointing as well as
ctlc: bool, a nested subactor erroring.
): """
'''
Multiple subactors, both erroring and
breakpointing as well as a nested subactor erroring.
'''
child = spawn(r'multi_subactors') child = spawn(r'multi_subactors')
# scan for the prompt # scan for the pdbpp prompt
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "Attaching pdb to actor: ('breakpoint_forever'" in before assert "Attaching pdb to actor: ('breakpoint_forever'" in before
if ctlc:
do_ctlc(child)
# do some "next" commands to demonstrate recurrent breakpoint # do some "next" commands to demonstrate recurrent breakpoint
# entries # entries
for _ in range(10): for _ in range(10):
child.sendline('next') child.sendline('next')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
if ctlc:
do_ctlc(child)
# continue to next error # continue to next error
child.sendline('c') child.sendline('c')
# first name_error failure # first name_error failure
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "Attaching to pdb in crashed actor: ('name_error'" in before assert "Attaching to pdb in crashed actor: ('name_error'" in before
assert "NameError" in before assert "NameError" in before
if ctlc:
do_ctlc(child)
# continue again # continue again
child.sendline('c') child.sendline('c')
# 2nd name_error failure # 2nd name_error failure
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode())
# TODO: will we ever get the race where this crash will show up? assert "Attaching to pdb in crashed actor: ('name_error_1'" in before
# blocklist strat now prevents this crash assert "NameError" in before
# assert_before(child, [
# "Attaching to pdb in crashed actor: ('name_error_1'",
# "NameError",
# ])
if ctlc:
do_ctlc(child)
# breakpoint loop should re-engage # breakpoint loop should re-engage
child.sendline('c') child.sendline('c')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "Attaching pdb to actor: ('breakpoint_forever'" in before assert "Attaching pdb to actor: ('breakpoint_forever'" in before
if ctlc:
do_ctlc(child)
# wait for spawn error to show up # wait for spawn error to show up
spawn_err = "Attaching to pdb in crashed actor: ('spawn_error'" spawn_err = "Attaching to pdb in crashed actor: ('spawn_error'"
start = time.time() while spawn_err not in before:
while (
spawn_err not in before
and (time.time() - start) < 3 # timeout eventually
):
child.sendline('c') child.sendline('c')
time.sleep(0.1) time.sleep(0.1)
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
if ctlc:
do_ctlc(child)
# 2nd depth nursery should trigger # 2nd depth nursery should trigger
# (XXX: this below if guard is technically a hack that makes the # child.sendline('c')
# nested case seem to work locally on linux but ideally in the long # child.expect(r"\(Pdb\+\+\)")
# run this can be dropped.) # before = str(child.before.decode())
if not ctlc: assert spawn_err in before
assert_before(child, [ assert "RemoteActorError: ('name_error_1'" in before
spawn_err,
"RemoteActorError: ('name_error_1'",
])
# now run some "continues" to show re-entries # now run some "continues" to show re-entries
for _ in range(5): for _ in range(5):
child.sendline('c') child.sendline('c')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
# quit the loop and expect parent to attach # quit the loop and expect parent to attach
child.sendline('q') child.sendline('q')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert_before(child, [
# debugger attaches to root # debugger attaches to root
"Attaching to pdb in crashed actor: ('root'", assert "Attaching to pdb in crashed actor: ('root'" in before
# expect a multierror with exceptions for each sub-actor # expect a multierror with exceptions for each sub-actor
"RemoteActorError: ('breakpoint_forever'", assert "RemoteActorError: ('breakpoint_forever'" in before
"RemoteActorError: ('name_error'", assert "RemoteActorError: ('name_error'" in before
"RemoteActorError: ('spawn_error'", assert "RemoteActorError: ('spawn_error'" in before
"RemoteActorError: ('name_error_1'", assert "RemoteActorError: ('name_error_1'" in before
'bdb.BdbQuit', assert 'bdb.BdbQuit' in before
])
if ctlc:
do_ctlc(child)
# process should exit # process should exit
child.sendline('c') child.sendline('c')
child.expect(pexpect.EOF) child.expect(pexpect.EOF)
# repeat of previous multierror for final output # repeat of previous multierror for final output
assert_before(child, [ before = str(child.before.decode())
"RemoteActorError: ('breakpoint_forever'", assert "RemoteActorError: ('breakpoint_forever'" in before
"RemoteActorError: ('name_error'", assert "RemoteActorError: ('name_error'" in before
"RemoteActorError: ('spawn_error'", assert "RemoteActorError: ('spawn_error'" in before
"RemoteActorError: ('name_error_1'", assert "RemoteActorError: ('name_error_1'" in before
'bdb.BdbQuit', assert 'bdb.BdbQuit' in before
])
def test_multi_daemon_subactors( def test_multi_daemon_subactors(spawn, loglevel):
spawn, """Multiple daemon subactors, both erroring and breakpointing within a
loglevel: str,
ctlc: bool
):
'''
Multiple daemon subactors, both erroring and breakpointing within a
stream. stream.
"""
'''
child = spawn('multi_daemon_subactors') child = spawn('multi_daemon_subactors')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
# there can be a race for which subactor will acquire # there is a race for which subactor will acquire
# the root's tty lock first so anticipate either crash # the root's tty lock first
# message on the first entry.
bp_forever_msg = "Attaching pdb to actor: ('bp_forever'"
name_error_msg = "NameError: name 'doggypants' is not defined"
before = str(child.before.decode()) before = str(child.before.decode())
bp_forever_msg = "Attaching pdb to actor: ('bp_forever'"
name_error_msg = "NameError"
if bp_forever_msg in before: if bp_forever_msg in before:
next_msg = name_error_msg next_msg = name_error_msg
@ -587,9 +352,6 @@ def test_multi_daemon_subactors(
else: else:
raise ValueError("Neither log msg was found !?") raise ValueError("Neither log msg was found !?")
if ctlc:
do_ctlc(child)
# NOTE: previously since we did not have clobber prevention # NOTE: previously since we did not have clobber prevention
# in the root actor this final resume could result in the debugger # in the root actor this final resume could result in the debugger
# tearing down since both child actors would be cancelled and it was # tearing down since both child actors would be cancelled and it was
@ -598,8 +360,10 @@ def test_multi_daemon_subactors(
# second entry by `bp_forever`. # second entry by `bp_forever`.
child.sendline('c') child.sendline('c')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
assert_before(child, [next_msg]) before = str(child.before.decode())
assert next_msg in before
# XXX: hooray the root clobbering the child here was fixed! # XXX: hooray the root clobbering the child here was fixed!
# IMO, this demonstrates the true power of SC system design. # IMO, this demonstrates the true power of SC system design.
@ -607,7 +371,7 @@ def test_multi_daemon_subactors(
# now the root actor won't clobber the bp_forever child # now the root actor won't clobber the bp_forever child
# during it's first access to the debug lock, but will instead # during it's first access to the debug lock, but will instead
# wait for the lock to release, by the edge triggered # wait for the lock to release, by the edge triggered
# ``_debug.Lock.no_remote_has_tty`` event before sending cancel messages # ``_debug._no_remote_has_tty`` event before sending cancel messages
# (via portals) to its underlings B) # (via portals) to its underlings B)
# at some point here there should have been some warning msg from # at some point here there should have been some warning msg from
@ -615,61 +379,32 @@ def test_multi_daemon_subactors(
# it seems unreliable in testing here to gnab it: # it seems unreliable in testing here to gnab it:
# assert "in use by child ('bp_forever'," in before # assert "in use by child ('bp_forever'," in before
if ctlc:
do_ctlc(child)
# expect another breakpoint actor entry
child.sendline('c')
child.expect(PROMPT)
try:
assert_before(child, [bp_forever_msg])
except AssertionError:
assert_before(child, [name_error_msg])
else:
if ctlc:
do_ctlc(child)
# should crash with the 2nd name error (simulates
# a retry) and then the root eventually (boxed) errors
# after 1 or more further bp actor entries.
child.sendline('c')
child.expect(PROMPT)
assert_before(child, [name_error_msg])
# wait for final error in root # wait for final error in root
# where it crashs with boxed error
while True: while True:
try:
child.sendline('c') child.sendline('c')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
assert_before( before = str(child.before.decode())
child, try:
[bp_forever_msg]
) # root error should be packed as remote error
except AssertionError: assert "_exceptions.RemoteActorError: ('name_error'" in before
break break
assert_before( except AssertionError:
child, assert bp_forever_msg in before
[
# boxed error raised in root task
"Attaching to pdb in crashed actor: ('root'",
"_exceptions.RemoteActorError: ('name_error'",
]
)
try:
child.sendline('c') child.sendline('c')
child.expect(pexpect.EOF) child.expect(pexpect.EOF)
except pexpect.exceptions.TIMEOUT:
# Failed to exit using continue..?
child.sendline('q')
child.expect(pexpect.EOF)
@has_nested_actors
def test_multi_subactors_root_errors( def test_multi_subactors_root_errors(spawn):
spawn,
ctlc: bool
):
''' '''
Multiple subactors, both erroring and breakpointing as well as Multiple subactors, both erroring and breakpointing as well as
a nested subactor erroring. a nested subactor erroring.
@ -677,87 +412,50 @@ def test_multi_subactors_root_errors(
''' '''
child = spawn('multi_subactor_root_errors') child = spawn('multi_subactor_root_errors')
# scan for the prompt # scan for the pdbpp prompt
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
# at most one subactor should attach before the root is cancelled # at most one subactor should attach before the root is cancelled
before = str(child.before.decode()) before = str(child.before.decode())
assert "NameError: name 'doggypants' is not defined" in before assert "NameError: name 'doggypants' is not defined" in before
if ctlc:
do_ctlc(child)
# continue again to catch 2nd name error from # continue again to catch 2nd name error from
# actor 'name_error_1' (which is 2nd depth). # actor 'name_error_1' (which is 2nd depth).
child.sendline('c') child.sendline('c')
child.expect(r"\(Pdb\+\+\)")
# due to block list strat from #337, this will no longer
# propagate before the root errors and cancels the spawner sub-tree.
child.expect(PROMPT)
# only if the blocking condition doesn't kick in fast enough
before = str(child.before.decode()) before = str(child.before.decode())
if "Debug lock blocked for ['name_error_1'" not in before: assert "Attaching to pdb in crashed actor: ('name_error_1'" in before
assert "NameError" in before
assert_before(child, [
"Attaching to pdb in crashed actor: ('name_error_1'",
"NameError",
])
if ctlc:
do_ctlc(child)
child.sendline('c') child.sendline('c')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
# check if the spawner crashed or was blocked from debug
# and if this intermediary attached check the boxed error
before = str(child.before.decode()) before = str(child.before.decode())
if "Attaching to pdb in crashed actor: ('spawn_error'" in before: assert "Attaching to pdb in crashed actor: ('spawn_error'" in before
# boxed error from previous step
assert_before(child, [ assert "RemoteActorError: ('name_error_1'" in before
# boxed error from spawner's child assert "NameError" in before
"RemoteActorError: ('name_error_1'",
"NameError",
])
if ctlc:
do_ctlc(child)
child.sendline('c') child.sendline('c')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode())
assert "Attaching to pdb in crashed actor: ('root'" in before
# boxed error from first level failure
assert "RemoteActorError: ('name_error'" in before
assert "NameError" in before
# expect a root actor crash # warnings assert we probably don't need
assert_before(child, [ # assert "Cancelling nursery in ('spawn_error'," in before
"RemoteActorError: ('name_error'",
"NameError",
# error from root actor and root task that created top level nursery
"Attaching to pdb in crashed actor: ('root'",
"AssertionError",
])
# continue again
child.sendline('c') child.sendline('c')
child.expect(pexpect.EOF) child.expect(pexpect.EOF)
assert_before(child, [ before = str(child.before.decode())
# "Attaching to pdb in crashed actor: ('root'", # error from root actor and root task that created top level nursery
# boxed error from previous step assert "AssertionError" in before
"RemoteActorError: ('name_error'",
"NameError",
"AssertionError",
'assert 0',
])
@has_nested_actors def test_multi_nested_subactors_error_through_nurseries(spawn):
def test_multi_nested_subactors_error_through_nurseries(
spawn,
# TODO: address debugger issue for nested tree:
# https://github.com/goodboy/tractor/issues/320
# ctlc: bool,
):
"""Verify deeply nested actors that error trigger debugger entries """Verify deeply nested actors that error trigger debugger entries
at each actor nurserly (level) all the way up the tree. at each actor nurserly (level) all the way up the tree.
@ -772,70 +470,55 @@ def test_multi_nested_subactors_error_through_nurseries(
timed_out_early: bool = False timed_out_early: bool = False
for send_char in itertools.cycle(['c', 'q']): for i in range(12):
try: try:
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
child.sendline(send_char) child.sendline('c')
time.sleep(0.01) time.sleep(0.1)
except EOF: except pexpect.exceptions.EOF:
# race conditions on how fast the continue is sent?
print(f"Failed early on {i}?")
timed_out_early = True
break break
else:
child.expect(pexpect.EOF)
assert_before(child, [ if not timed_out_early:
before = str(child.before.decode())
# boxed source errors assert "NameError" in before
"NameError: name 'doggypants' is not defined",
"tractor._exceptions.RemoteActorError: ('name_error'",
"bdb.BdbQuit",
# first level subtrees
"tractor._exceptions.RemoteActorError: ('spawner0'",
# "tractor._exceptions.RemoteActorError: ('spawner1'",
# propagation of errors up through nested subtrees
"tractor._exceptions.RemoteActorError: ('spawn_until_0'",
"tractor._exceptions.RemoteActorError: ('spawn_until_1'",
"tractor._exceptions.RemoteActorError: ('spawn_until_2'",
])
@pytest.mark.timeout(15)
@has_nested_actors
def test_root_nursery_cancels_before_child_releases_tty_lock( def test_root_nursery_cancels_before_child_releases_tty_lock(
spawn, spawn,
start_method, start_method
ctlc: bool,
): ):
''' """Test that when the root sends a cancel message before a nested
Test that when the root sends a cancel message before a nested child child has unblocked (which can happen when it has the tty lock and
has unblocked (which can happen when it has the tty lock and is is engaged in pdb) it is indeed cancelled after exiting the debugger.
engaged in pdb) it is indeed cancelled after exiting the debugger. """
'''
timed_out_early = False timed_out_early = False
child = spawn('root_cancelled_but_child_is_in_tty_lock') child = spawn('root_cancelled_but_child_is_in_tty_lock')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "NameError: name 'doggypants' is not defined" in before assert "NameError: name 'doggypants' is not defined" in before
assert "tractor._exceptions.RemoteActorError: ('name_error'" not in before assert "tractor._exceptions.RemoteActorError: ('name_error'" not in before
time.sleep(0.5) time.sleep(0.5)
if ctlc:
do_ctlc(child)
child.sendline('c') child.sendline('c')
for i in range(4): for i in range(4):
time.sleep(0.5) time.sleep(0.5)
try: try:
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
except ( except (
EOF, pexpect.exceptions.EOF,
TIMEOUT, pexpect.exceptions.TIMEOUT,
): ):
# races all over.. # races all over..
@ -850,37 +533,26 @@ def test_root_nursery_cancels_before_child_releases_tty_lock(
before = str(child.before.decode()) before = str(child.before.decode())
assert "NameError: name 'doggypants' is not defined" in before assert "NameError: name 'doggypants' is not defined" in before
if ctlc:
do_ctlc(child)
child.sendline('c') child.sendline('c')
time.sleep(0.1)
for i in range(3): while True:
try: try:
child.expect(pexpect.EOF, timeout=0.5) child.expect(pexpect.EOF)
break break
except TIMEOUT: except pexpect.exceptions.TIMEOUT:
child.sendline('c') child.sendline('c')
time.sleep(0.1)
print('child was able to grab tty lock again?') print('child was able to grab tty lock again?')
else:
print('giving up on child releasing, sending `quit` cmd')
child.sendline('q')
expect(child, EOF)
if not timed_out_early: if not timed_out_early:
before = str(child.before.decode()) before = str(child.before.decode())
assert_before(child, [ assert "tractor._exceptions.RemoteActorError: ('spawner0'" in before
"tractor._exceptions.RemoteActorError: ('spawner0'", assert "tractor._exceptions.RemoteActorError: ('name_error'" in before
"tractor._exceptions.RemoteActorError: ('name_error'", assert "NameError: name 'doggypants' is not defined" in before
"NameError: name 'doggypants' is not defined",
])
def test_root_cancels_child_context_during_startup( def test_root_cancels_child_context_during_startup(
spawn, spawn,
ctlc: bool,
): ):
'''Verify a fast fail in the root doesn't lock up the child reaping '''Verify a fast fail in the root doesn't lock up the child reaping
and all while using the new context api. and all while using the new context api.
@ -888,33 +560,26 @@ def test_root_cancels_child_context_during_startup(
''' '''
child = spawn('fast_error_in_root_after_spawn') child = spawn('fast_error_in_root_after_spawn')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
before = str(child.before.decode()) before = str(child.before.decode())
assert "AssertionError" in before assert "AssertionError" in before
if ctlc:
do_ctlc(child)
child.sendline('c') child.sendline('c')
child.expect(pexpect.EOF) child.expect(pexpect.EOF)
def test_different_debug_mode_per_actor( def test_different_debug_mode_per_actor(
spawn, spawn,
ctlc: bool,
): ):
child = spawn('per_actor_debug') child = spawn('per_actor_debug')
child.expect(PROMPT) child.expect(r"\(Pdb\+\+\)")
# only one actor should enter the debugger # only one actor should enter the debugger
before = str(child.before.decode()) before = str(child.before.decode())
assert "Attaching to pdb in crashed actor: ('debugged_boi'" in before assert "Attaching to pdb in crashed actor: ('debugged_boi'" in before
assert "RuntimeError" in before assert "RuntimeError" in before
if ctlc:
do_ctlc(child)
child.sendline('c') child.sendline('c')
child.expect(pexpect.EOF) child.expect(pexpect.EOF)

View File

@ -12,17 +12,17 @@ import shutil
import pytest import pytest
from conftest import ( from conftest import repodir
examples_dir,
)
def examples_dir():
"""Return the abspath to the examples directory.
"""
return os.path.join(repodir(), 'examples')
@pytest.fixture @pytest.fixture
def run_example_in_subproc( def run_example_in_subproc(loglevel, testdir, arb_addr):
loglevel: str,
testdir,
arb_addr: tuple[str, int],
):
@contextmanager @contextmanager
def run(script_code): def run(script_code):
@ -32,8 +32,8 @@ def run_example_in_subproc(
# on windows we need to create a special __main__.py which will # on windows we need to create a special __main__.py which will
# be executed with ``python -m <modulename>`` on windows.. # be executed with ``python -m <modulename>`` on windows..
shutil.copyfile( shutil.copyfile(
examples_dir() / '__main__.py', os.path.join(examples_dir(), '__main__.py'),
str(testdir / '__main__.py'), os.path.join(str(testdir), '__main__.py')
) )
# drop the ``if __name__ == '__main__'`` guard onwards from # drop the ``if __name__ == '__main__'`` guard onwards from
@ -88,7 +88,6 @@ def run_example_in_subproc(
and f[0] != '_' and f[0] != '_'
and 'debugging' not in p[0] and 'debugging' not in p[0]
and 'integration' not in p[0] and 'integration' not in p[0]
and 'advanced_faults' not in p[0]
], ],
ids=lambda t: t[1], ids=lambda t: t[1],
@ -117,19 +116,9 @@ def test_example(run_example_in_subproc, example_script):
# print(f'STDOUT: {out}') # print(f'STDOUT: {out}')
# if we get some gnarly output let's aggregate and raise # if we get some gnarly output let's aggregate and raise
if err:
errmsg = err.decode() errmsg = err.decode()
errlines = errmsg.splitlines() errlines = errmsg.splitlines()
last_error = errlines[-1] if err and 'Error' in errlines[-1]:
if (
'Error' in last_error
# XXX: currently we print this to console, but maybe
# shouldn't eventually once we figure out what's
# a better way to be explicit about aio side
# cancels?
and 'asyncio.exceptions.CancelledError' not in last_error
):
raise Exception(errmsg) raise Exception(errmsg)
assert proc.returncode == 0 assert proc.returncode == 0

View File

@ -8,7 +8,6 @@ import builtins
import itertools import itertools
import importlib import importlib
from exceptiongroup import BaseExceptionGroup
import pytest import pytest
import trio import trio
import tractor import tractor
@ -171,11 +170,11 @@ async def trio_ctx(
# message. # message.
with trio.fail_after(2): with trio.fail_after(2):
async with ( async with (
trio.open_nursery() as n,
tractor.to_asyncio.open_channel_from( tractor.to_asyncio.open_channel_from(
sleep_and_err, sleep_and_err,
) as (first, chan), ) as (first, chan),
trio.open_nursery() as n,
): ):
assert first == 'start' assert first == 'start'
@ -204,7 +203,6 @@ def test_context_spawns_aio_task_that_errors(
''' '''
async def main(): async def main():
with trio.fail_after(2):
async with tractor.open_nursery() as n: async with tractor.open_nursery() as n:
p = await n.start_actor( p = await n.start_actor(
'aio_daemon', 'aio_daemon',
@ -410,12 +408,11 @@ def test_trio_error_cancels_intertask_chan(arb_addr):
# should trigger remote actor error # should trigger remote actor error
await portal.result() await portal.result()
with pytest.raises(BaseExceptionGroup) as excinfo: with pytest.raises(RemoteActorError) as excinfo:
trio.run(main) trio.run(main)
# ensure boxed errors # ensure boxed error is correct
for exc in excinfo.value.exceptions: assert excinfo.value.type == Exception
assert exc.type == Exception
def test_trio_closes_early_and_channel_exits(arb_addr): def test_trio_closes_early_and_channel_exits(arb_addr):
@ -444,12 +441,11 @@ def test_aio_errors_and_channel_propagates_and_closes(arb_addr):
# should trigger remote actor error # should trigger remote actor error
await portal.result() await portal.result()
with pytest.raises(BaseExceptionGroup) as excinfo: with pytest.raises(RemoteActorError) as excinfo:
trio.run(main) trio.run(main)
# ensure boxed errors # ensure boxed error is correct
for exc in excinfo.value.exceptions: assert excinfo.value.type == Exception
assert exc.type == Exception
@tractor.context @tractor.context

View File

@ -11,18 +11,25 @@ from conftest import tractor_test
@pytest.mark.trio @pytest.mark.trio
async def test_no_runtime(): async def test_no_arbitter():
"""An arbitter must be established before any nurseries """An arbitter must be established before any nurseries
can be created. can be created.
(In other words ``tractor.open_root_actor()`` must be engaged at (In other words ``tractor.open_root_actor()`` must be engaged at
some point?) some point?)
""" """
with pytest.raises(RuntimeError) : with pytest.raises(RuntimeError):
async with tractor.find_actor('doggy'): with tractor.open_nursery():
pass pass
def test_no_main():
"""An async function **must** be passed to ``tractor.run()``.
"""
with pytest.raises(TypeError):
tractor.run(None)
@tractor_test @tractor_test
async def test_self_is_registered(arb_addr): async def test_self_is_registered(arb_addr):
"Verify waiting on the arbiter to register itself using the standard api." "Verify waiting on the arbiter to register itself using the standard api."

View File

@ -4,10 +4,9 @@ from itertools import cycle
import pytest import pytest
import trio import trio
import tractor import tractor
from tractor.testing import tractor_test
from tractor.experimental import msgpub from tractor.experimental import msgpub
from conftest import tractor_test
def test_type_checks(): def test_type_checks():

View File

@ -1,73 +0,0 @@
"""
Verifying internal runtime state and undocumented extras.
"""
import os
import pytest
import trio
import tractor
from conftest import tractor_test
_file_path: str = ''
def unlink_file():
print('Removing tmp file!')
os.remove(_file_path)
async def crash_and_clean_tmpdir(
tmp_file_path: str,
error: bool = True,
):
global _file_path
_file_path = tmp_file_path
actor = tractor.current_actor()
actor.lifetime_stack.callback(unlink_file)
assert os.path.isfile(tmp_file_path)
await trio.sleep(0.1)
if error:
assert 0
else:
actor.cancel_soon()
@pytest.mark.parametrize(
'error_in_child',
[True, False],
)
@tractor_test
async def test_lifetime_stack_wipes_tmpfile(
tmp_path,
error_in_child: bool,
):
child_tmp_file = tmp_path / "child.txt"
child_tmp_file.touch()
assert child_tmp_file.exists()
path = str(child_tmp_file)
try:
with trio.move_on_after(0.5):
async with tractor.open_nursery() as n:
await ( # inlined portal
await n.run_in_actor(
crash_and_clean_tmpdir,
tmp_file_path=path,
error=error_in_child,
)
).result()
except (
tractor.RemoteActorError,
tractor.BaseExceptionGroup,
):
pass
# tmp file should have been wiped by
# teardown stack.
assert not child_tmp_file.exists()

View File

@ -1,8 +1,7 @@
""" """
Spawning basics Spawning basics
""" """
from typing import Optional from typing import Dict, Tuple, Optional
import pytest import pytest
import trio import trio
@ -15,8 +14,8 @@ data_to_pass_down = {'doggy': 10, 'kitty': 4}
async def spawn( async def spawn(
is_arbiter: bool, is_arbiter: bool,
data: dict, data: Dict,
arb_addr: tuple[str, int], arb_addr: Tuple[str, int],
): ):
namespaces = [__name__] namespaces = [__name__]
@ -142,7 +141,7 @@ def test_loglevel_propagated_to_subactor(
capfd, capfd,
arb_addr, arb_addr,
): ):
if start_method == 'mp_forkserver': if start_method == 'forkserver':
pytest.skip( pytest.skip(
"a bug with `capfd` seems to make forkserver capture not work?") "a bug with `capfd` seems to make forkserver capture not work?")
@ -151,13 +150,13 @@ def test_loglevel_propagated_to_subactor(
async def main(): async def main():
async with tractor.open_nursery( async with tractor.open_nursery(
name='arbiter', name='arbiter',
loglevel=level,
start_method=start_method, start_method=start_method,
arbiter_addr=arb_addr, arbiter_addr=arb_addr,
) as tn: ) as tn:
await tn.run_in_actor( await tn.run_in_actor(
check_loglevel, check_loglevel,
loglevel=level,
level=level, level=level,
) )

View File

@ -7,10 +7,9 @@ import platform
import trio import trio
import tractor import tractor
from tractor.testing import tractor_test
import pytest import pytest
from conftest import tractor_test
def test_must_define_ctx(): def test_must_define_ctx():
@ -251,7 +250,7 @@ def test_a_quadruple_example(time_quad_ex, ci_env, spawn_backend):
results, diff = time_quad_ex results, diff = time_quad_ex
assert results assert results
this_fast = 6 if platform.system() in ('Windows', 'Darwin') else 3 this_fast = 6 if platform.system() in ('Windows', 'Darwin') else 2.666
assert diff < this_fast assert diff < this_fast

View File

@ -6,16 +6,13 @@ from contextlib import asynccontextmanager
from functools import partial from functools import partial
from itertools import cycle from itertools import cycle
import time import time
from typing import Optional from typing import Optional, List, Tuple
import pytest import pytest
import trio import trio
from trio.lowlevel import current_task from trio.lowlevel import current_task
import tractor import tractor
from tractor.trionics import ( from tractor.trionics import broadcast_receiver, Lagged
broadcast_receiver,
Lagged,
)
@tractor.context @tractor.context
@ -40,7 +37,7 @@ async def echo_sequences(
async def ensure_sequence( async def ensure_sequence(
stream: tractor.MsgStream, stream: tractor.ReceiveMsgStream,
sequence: list, sequence: list,
delay: Optional[float] = None, delay: Optional[float] = None,
@ -65,8 +62,8 @@ async def ensure_sequence(
@asynccontextmanager @asynccontextmanager
async def open_sequence_streamer( async def open_sequence_streamer(
sequence: list[int], sequence: List[int],
arb_addr: tuple[str, int], arb_addr: Tuple[str, int],
start_method: str, start_method: str,
) -> tractor.MsgStream: ) -> tractor.MsgStream:
@ -214,8 +211,7 @@ def test_faster_task_to_recv_is_cancelled_by_slower(
arb_addr, arb_addr,
start_method, start_method,
): ):
''' '''Ensure that if a faster task consuming from a stream is cancelled
Ensure that if a faster task consuming from a stream is cancelled
the slower task can continue to receive all expected values. the slower task can continue to receive all expected values.
''' '''
@ -464,51 +460,3 @@ def test_first_recver_is_cancelled():
assert value == 1 assert value == 1
trio.run(main) trio.run(main)
def test_no_raise_on_lag():
'''
Run a simple 2-task broadcast where one task is slow but configured
so that it does not raise `Lagged` on overruns using
`raise_on_lasg=False` and verify that the task does not raise.
'''
size = 100
tx, rx = trio.open_memory_channel(size)
brx = broadcast_receiver(rx, size)
async def slow():
async with brx.subscribe(
raise_on_lag=False,
) as br:
async for msg in br:
print(f'slow task got: {msg}')
await trio.sleep(0.1)
async def fast():
async with brx.subscribe() as br:
async for msg in br:
print(f'fast task got: {msg}')
async def main():
async with (
tractor.open_root_actor(
# NOTE: so we see the warning msg emitted by the bcaster
# internals when the no raise flag is set.
loglevel='warning',
),
trio.open_nursery() as n,
):
n.start_soon(slow)
n.start_soon(fast)
for i in range(1000):
await tx.send(i)
# simulate user nailing ctl-c after realizing
# there's a lag in the slow task.
await trio.sleep(1)
raise KeyboardInterrupt
with pytest.raises(KeyboardInterrupt):
trio.run(main)

7
towncrier.toml 100644
View File

@ -0,0 +1,7 @@
[tool.towncrier]
package = "tractor"
filename = "NEWS.rst"
directory = "nooz/"
title_format = "tractor {version} ({project_date})"
version = "0.1.0a4"
template = "nooz/_template.rst"

View File

@ -18,12 +18,13 @@
tractor: structured concurrent "actors". tractor: structured concurrent "actors".
""" """
from exceptiongroup import BaseExceptionGroup from trio import MultiError
from ._clustering import open_actor_cluster from ._clustering import open_actor_cluster
from ._ipc import Channel from ._ipc import Channel
from ._streaming import ( from ._streaming import (
Context, Context,
ReceiveMsgStream,
MsgStream, MsgStream,
stream, stream,
context, context,
@ -35,37 +36,27 @@ from ._discovery import (
query_actor, query_actor,
) )
from ._supervise import open_nursery from ._supervise import open_nursery
from ._state import ( from ._state import current_actor, is_root_process
current_actor,
is_root_process,
)
from ._exceptions import ( from ._exceptions import (
RemoteActorError, RemoteActorError,
ModuleNotExposed, ModuleNotExposed,
ContextCancelled, ContextCancelled,
) )
from ._debug import ( from ._debug import breakpoint, post_mortem
breakpoint,
post_mortem,
)
from . import msg from . import msg
from ._root import ( from ._root import run, run_daemon, open_root_actor
run_daemon,
open_root_actor,
)
from ._portal import Portal from ._portal import Portal
from ._runtime import Actor
__all__ = [ __all__ = [
'Actor',
'Channel', 'Channel',
'Context', 'Context',
'ContextCancelled', 'ContextCancelled',
'ModuleNotExposed', 'ModuleNotExposed',
'MsgStream', 'MsgStream',
'BaseExceptionGroup', 'MultiError',
'Portal', 'Portal',
'ReceiveMsgStream',
'RemoteActorError', 'RemoteActorError',
'breakpoint', 'breakpoint',
'context', 'context',
@ -79,6 +70,7 @@ __all__ = [
'open_root_actor', 'open_root_actor',
'post_mortem', 'post_mortem',
'query_actor', 'query_actor',
'run',
'run_daemon', 'run_daemon',
'stream', 'stream',
'to_asyncio', 'to_asyncio',

File diff suppressed because it is too large Load Diff

View File

@ -24,7 +24,7 @@ import argparse
from ast import literal_eval from ast import literal_eval
from ._runtime import Actor from ._actor import Actor
from ._entry import _trio_main from ._entry import _trio_main

View File

@ -32,12 +32,9 @@ import tractor
async def open_actor_cluster( async def open_actor_cluster(
modules: list[str], modules: list[str],
count: int = cpu_count(), count: int = cpu_count(),
names: list[str] | None = None, names: Optional[list[str]] = None,
start_method: Optional[str] = None,
hard_kill: bool = False, hard_kill: bool = False,
# passed through verbatim to ``open_root_actor()``
**runtime_kwargs,
) -> AsyncGenerator[ ) -> AsyncGenerator[
dict[str, tractor.Portal], dict[str, tractor.Portal],
None, None,
@ -52,9 +49,7 @@ async def open_actor_cluster(
raise ValueError( raise ValueError(
'Number of names is {len(names)} but count it {count}') 'Number of names is {len(names)} but count it {count}')
async with tractor.open_nursery( async with tractor.open_nursery(start_method=start_method) as an:
**runtime_kwargs,
) as an:
async with trio.open_nursery() as n: async with trio.open_nursery() as n:
uid = tractor.current_actor().uid uid = tractor.current_actor().uid

File diff suppressed because it is too large Load Diff

View File

@ -18,11 +18,7 @@
Actor discovery API. Actor discovery API.
""" """
from typing import ( from typing import Tuple, Optional, Union, AsyncGenerator
Optional,
Union,
AsyncGenerator,
)
from contextlib import asynccontextmanager as acm from contextlib import asynccontextmanager as acm
from ._ipc import _connect_chan, Channel from ._ipc import _connect_chan, Channel
@ -108,7 +104,7 @@ async def query_actor(
@acm @acm
async def find_actor( async def find_actor(
name: str, name: str,
arbiter_sockaddr: tuple[str, int] | None = None arbiter_sockaddr: Tuple[str, int] = None
) -> AsyncGenerator[Optional[Portal], None]: ) -> AsyncGenerator[Optional[Portal], None]:
''' '''
@ -134,7 +130,7 @@ async def find_actor(
@acm @acm
async def wait_for_actor( async def wait_for_actor(
name: str, name: str,
arbiter_sockaddr: tuple[str, int] | None = None arbiter_sockaddr: Tuple[str, int] = None
) -> AsyncGenerator[Portal, None]: ) -> AsyncGenerator[Portal, None]:
"""Wait on an actor to register with the arbiter. """Wait on an actor to register with the arbiter.

View File

@ -18,28 +18,15 @@
Sub-process entry points. Sub-process entry points.
""" """
from __future__ import annotations
from functools import partial from functools import partial
from typing import ( from typing import Tuple, Any
Any, import signal
TYPE_CHECKING,
)
import trio # type: ignore import trio # type: ignore
from .log import ( from .log import get_console_log, get_logger
get_console_log,
get_logger,
)
from . import _state from . import _state
from .to_asyncio import run_as_asyncio_guest from .to_asyncio import run_as_asyncio_guest
from ._runtime import (
async_main,
Actor,
)
if TYPE_CHECKING:
from ._spawn import SpawnMethodKey
log = get_logger(__name__) log = get_logger(__name__)
@ -47,11 +34,11 @@ log = get_logger(__name__)
def _mp_main( def _mp_main(
actor: Actor, # type: ignore actor: 'Actor', # type: ignore
accept_addr: tuple[str, int], accept_addr: Tuple[str, int],
forkserver_info: tuple[Any, Any, Any, Any, Any], forkserver_info: Tuple[Any, Any, Any, Any, Any],
start_method: SpawnMethodKey, start_method: str,
parent_addr: tuple[str, int] | None = None, parent_addr: Tuple[str, int] = None,
infect_asyncio: bool = False, infect_asyncio: bool = False,
) -> None: ) -> None:
@ -76,8 +63,7 @@ def _mp_main(
log.debug(f"parent_addr is {parent_addr}") log.debug(f"parent_addr is {parent_addr}")
trio_main = partial( trio_main = partial(
async_main, actor._async_main,
actor,
accept_addr, accept_addr,
parent_addr=parent_addr parent_addr=parent_addr
) )
@ -96,9 +82,9 @@ def _mp_main(
def _trio_main( def _trio_main(
actor: Actor, # type: ignore actor: 'Actor', # type: ignore
*, *,
parent_addr: tuple[str, int] | None = None, parent_addr: Tuple[str, int] = None,
infect_asyncio: bool = False, infect_asyncio: bool = False,
) -> None: ) -> None:
@ -120,8 +106,7 @@ def _trio_main(
log.debug(f"parent_addr is {parent_addr}") log.debug(f"parent_addr is {parent_addr}")
trio_main = partial( trio_main = partial(
async_main, actor._async_main,
actor,
parent_addr=parent_addr parent_addr=parent_addr
) )

View File

@ -18,16 +18,11 @@
Our classy exception set. Our classy exception set.
""" """
from typing import ( from typing import Dict, Any, Optional, Type
Any,
Optional,
Type,
)
import importlib import importlib
import builtins import builtins
import traceback import traceback
import exceptiongroup as eg
import trio import trio
@ -53,6 +48,9 @@ class RemoteActorError(Exception):
self.type = suberror_type self.type = suberror_type
self.msgdata = msgdata self.msgdata = msgdata
# TODO: a trio.MultiError.catch like context manager
# for catching underlying remote errors of a particular type
class InternalActorError(RemoteActorError): class InternalActorError(RemoteActorError):
"""Remote internal ``tractor`` error indicating """Remote internal ``tractor`` error indicating
@ -97,7 +95,7 @@ def pack_error(
exc: BaseException, exc: BaseException,
tb=None, tb=None,
) -> dict[str, Any]: ) -> Dict[str, Any]:
"""Create an "error message" for tranmission over """Create an "error message" for tranmission over
a channel (aka the wire). a channel (aka the wire).
""" """
@ -116,17 +114,15 @@ def pack_error(
def unpack_error( def unpack_error(
msg: dict[str, Any], msg: Dict[str, Any],
chan=None, chan=None,
err_type=RemoteActorError err_type=RemoteActorError
) -> Exception: ) -> Exception:
''' """Unpack an 'error' message from the wire
Unpack an 'error' message from the wire
into a local ``RemoteActorError``. into a local ``RemoteActorError``.
''' """
__tracebackhide__ = True
error = msg['error'] error = msg['error']
tb_str = error.get('tb_str', '') tb_str = error.get('tb_str', '')
@ -139,12 +135,7 @@ def unpack_error(
suberror_type = trio.Cancelled suberror_type = trio.Cancelled
else: # try to lookup a suitable local error type else: # try to lookup a suitable local error type
for ns in [ for ns in [builtins, _this_mod, trio]:
builtins,
_this_mod,
eg,
trio,
]:
try: try:
suberror_type = getattr(ns, type_name) suberror_type = getattr(ns, type_name)
break break
@ -163,15 +154,12 @@ def unpack_error(
def is_multi_cancelled(exc: BaseException) -> bool: def is_multi_cancelled(exc: BaseException) -> bool:
''' """Predicate to determine if a ``trio.MultiError`` contains only
Predicate to determine if a possible ``eg.BaseExceptionGroup`` contains ``trio.Cancelled`` sub-exceptions (and is likely the result of
only ``trio.Cancelled`` sub-exceptions (and is likely the result of
cancelling a collection of subtasks. cancelling a collection of subtasks.
''' """
if isinstance(exc, eg.BaseExceptionGroup): return not trio.MultiError.filter(
return exc.subgroup( lambda exc: exc if not isinstance(exc, trio.Cancelled) else None,
lambda exc: isinstance(exc, trio.Cancelled) exc,
) is not None )
return False

View File

@ -341,7 +341,7 @@ class Channel:
async def connect( async def connect(
self, self,
destaddr: tuple[Any, ...] | None = None, destaddr: tuple[Any, ...] = None,
**kwargs **kwargs
) -> MsgTransport: ) -> MsgTransport:

View File

@ -35,7 +35,6 @@ import warnings
import trio import trio
from async_generator import asynccontextmanager from async_generator import asynccontextmanager
from .trionics import maybe_open_nursery
from ._state import current_actor from ._state import current_actor
from ._ipc import Channel from ._ipc import Channel
from .log import get_logger from .log import get_logger
@ -45,27 +44,43 @@ from ._exceptions import (
NoResult, NoResult,
ContextCancelled, ContextCancelled,
) )
from ._streaming import ( from ._streaming import Context, ReceiveMsgStream
Context,
MsgStream,
)
log = get_logger(__name__) log = get_logger(__name__)
@asynccontextmanager
async def maybe_open_nursery(
nursery: trio.Nursery = None,
shield: bool = False,
) -> AsyncGenerator[trio.Nursery, Any]:
'''
Create a new nursery if None provided.
Blocks on exit as expected if no input nursery is provided.
'''
if nursery is not None:
yield nursery
else:
async with trio.open_nursery() as nursery:
nursery.cancel_scope.shield = shield
yield nursery
def _unwrap_msg( def _unwrap_msg(
msg: dict[str, Any], msg: dict[str, Any],
channel: Channel channel: Channel
) -> Any: ) -> Any:
__tracebackhide__ = True
try: try:
return msg['return'] return msg['return']
except KeyError: except KeyError:
# internal error should never get here # internal error should never get here
assert msg.get('cid'), "Received internal error at portal?" assert msg.get('cid'), "Received internal error at portal?"
raise unpack_error(msg, channel) from None raise unpack_error(msg, channel)
class MessagingError(Exception): class MessagingError(Exception):
@ -104,7 +119,7 @@ class Portal:
# it is expected that ``result()`` will be awaited at some # it is expected that ``result()`` will be awaited at some
# point. # point.
self._expect_result: Optional[Context] = None self._expect_result: Optional[Context] = None
self._streams: set[MsgStream] = set() self._streams: set[ReceiveMsgStream] = set()
self.actor = current_actor() self.actor = current_actor()
async def _submit_for_result( async def _submit_for_result(
@ -139,7 +154,6 @@ class Portal:
Return the result(s) from the remote actor's "main" task. Return the result(s) from the remote actor's "main" task.
''' '''
# __tracebackhide__ = True
# Check for non-rpc errors slapped on the # Check for non-rpc errors slapped on the
# channel for which we always raise # channel for which we always raise
exc = self.channel._exc exc = self.channel._exc
@ -189,7 +203,7 @@ class Portal:
async def cancel_actor( async def cancel_actor(
self, self,
timeout: float | None = None, timeout: float = None,
) -> bool: ) -> bool:
''' '''
@ -319,7 +333,7 @@ class Portal:
async_gen_func: Callable, # typing: ignore async_gen_func: Callable, # typing: ignore
**kwargs, **kwargs,
) -> AsyncGenerator[MsgStream, None]: ) -> AsyncGenerator[ReceiveMsgStream, None]:
if not inspect.isasyncgenfunction(async_gen_func): if not inspect.isasyncgenfunction(async_gen_func):
if not ( if not (
@ -344,7 +358,7 @@ class Portal:
try: try:
# deliver receive only stream # deliver receive only stream
async with MsgStream( async with ReceiveMsgStream(
ctx, ctx._recv_chan, ctx, ctx._recv_chan,
) as rchan: ) as rchan:
self._streams.add(rchan) self._streams.add(rchan)
@ -464,6 +478,7 @@ class Portal:
# sure it's worth being pedantic: # sure it's worth being pedantic:
# Exception, # Exception,
# trio.Cancelled, # trio.Cancelled,
# trio.MultiError,
# KeyboardInterrupt, # KeyboardInterrupt,
) as err: ) as err:
@ -500,10 +515,6 @@ class Portal:
f'actor: {uid}' f'actor: {uid}'
) )
result = await ctx.result() result = await ctx.result()
log.runtime(
f'Context {fn_name} returned '
f'value from callee `{result}`'
)
# though it should be impossible for any tasks # though it should be impossible for any tasks
# operating *in* this scope to have survived # operating *in* this scope to have survived
@ -525,22 +536,14 @@ class Portal:
f'task:{cid}\n' f'task:{cid}\n'
f'actor:{uid}' f'actor:{uid}'
) )
# XXX: (MEGA IMPORTANT) if this is a root opened process we else:
# wait for any immediate child in debug before popping the log.runtime(
# context from the runtime msg loop otherwise inside f'Context {fn_name} returned '
# ``Actor._push_result()`` the msg will be discarded and in f'value from callee `{result}`'
# the case where that msg is global debugger unlock (via )
# a "stop" msg for a stream), this can result in a deadlock
# where the root is waiting on the lock to clear but the
# child has already cleared it and clobbered IPC.
from ._debug import maybe_wait_for_debugger
await maybe_wait_for_debugger()
# remove the context from runtime tracking # remove the context from runtime tracking
self.actor._contexts.pop( self.actor._contexts.pop((self.channel.uid, ctx.cid))
(self.channel.uid, ctx.cid),
None,
)
@dataclass @dataclass
@ -597,11 +600,9 @@ async def open_portal(
msg_loop_cs: Optional[trio.CancelScope] = None msg_loop_cs: Optional[trio.CancelScope] = None
if start_msg_loop: if start_msg_loop:
from ._runtime import process_messages
msg_loop_cs = await nursery.start( msg_loop_cs = await nursery.start(
partial( partial(
process_messages, actor._process_messages,
actor,
channel, channel,
# if the local task is cancelled we want to keep # if the local task is cancelled we want to keep
# the msg loop running until our block ends # the msg loop running until our block ends

View File

@ -22,21 +22,14 @@ from contextlib import asynccontextmanager
from functools import partial from functools import partial
import importlib import importlib
import logging import logging
import signal
import sys
import os import os
from typing import Tuple, Optional, List, Any
import typing import typing
import warnings import warnings
from exceptiongroup import BaseExceptionGroup
import trio import trio
from ._runtime import ( from ._actor import Actor, Arbiter
Actor,
Arbiter,
async_main,
)
from . import _debug from . import _debug
from . import _spawn from . import _spawn
from . import _state from . import _state
@ -56,45 +49,37 @@ logger = log.get_logger('tractor')
@asynccontextmanager @asynccontextmanager
async def open_root_actor( async def open_root_actor(
*,
# defaults are above # defaults are above
arbiter_addr: tuple[str, int] | None = None, arbiter_addr: Optional[Tuple[str, int]] = (
_default_arbiter_host,
_default_arbiter_port,
),
# defaults are above name: Optional[str] = 'root',
registry_addr: tuple[str, int] | None = None,
name: str | None = 'root',
# either the `multiprocessing` start method: # either the `multiprocessing` start method:
# https://docs.python.org/3/library/multiprocessing.html#contexts-and-start-methods # https://docs.python.org/3/library/multiprocessing.html#contexts-and-start-methods
# OR `trio` (the new default). # OR `trio` (the new default).
start_method: _spawn.SpawnMethodKey | None = None, start_method: Optional[str] = None,
# enables the multi-process debugger support # enables the multi-process debugger support
debug_mode: bool = False, debug_mode: bool = False,
# internal logging # internal logging
loglevel: str | None = None, loglevel: Optional[str] = None,
enable_modules: list | None = None, enable_modules: Optional[List] = None,
rpc_module_paths: list | None = None, rpc_module_paths: Optional[List] = None,
) -> typing.Any: ) -> typing.Any:
''' """Async entry point for ``tractor``.
Runtime init entry point for ``tractor``.
''' """
# Override the global debugger hook to make it play nice with # Override the global debugger hook to make it play nice with
# ``trio``, see much discussion in: # ``trio``, see:
# https://github.com/python-trio/trio/issues/1155#issuecomment-742964018 # https://github.com/python-trio/trio/issues/1155#issuecomment-742964018
builtin_bp_handler = sys.breakpointhook
orig_bp_path: str | None = os.environ.get('PYTHONBREAKPOINT', None)
os.environ['PYTHONBREAKPOINT'] = 'tractor._debug._set_trace' os.environ['PYTHONBREAKPOINT'] = 'tractor._debug._set_trace'
# attempt to retreive ``trio``'s sigint handler and stash it
# on our debugger lock state.
_debug.Lock._trio_handler = signal.getsignal(signal.SIGINT)
# mark top most level process as root actor # mark top most level process as root actor
_state._runtime_vars['_is_root'] = True _state._runtime_vars['_is_root'] = True
@ -113,24 +98,18 @@ async def open_root_actor(
if start_method is not None: if start_method is not None:
_spawn.try_set_start_method(start_method) _spawn.try_set_start_method(start_method)
if arbiter_addr is not None: arbiter_addr = (host, port) = arbiter_addr or (
warnings.warn(
'`arbiter_addr` is now deprecated and has been renamed to'
'`registry_addr`.\nUse that instead..',
DeprecationWarning,
stacklevel=2,
)
registry_addr = (host, port) = (
registry_addr
or arbiter_addr
or (
_default_arbiter_host, _default_arbiter_host,
_default_arbiter_port, _default_arbiter_port,
) )
)
loglevel = (loglevel or log._default_loglevel).upper() if loglevel is None:
loglevel = log.get_loglevel()
else:
log._default_loglevel = loglevel
log.get_console_log(loglevel)
assert loglevel
if debug_mode and _spawn._spawn_method == 'trio': if debug_mode and _spawn._spawn_method == 'trio':
_state._runtime_vars['_debug_mode'] = True _state._runtime_vars['_debug_mode'] = True
@ -145,7 +124,7 @@ async def open_root_actor(
logging.getLevelName( logging.getLevelName(
# lul, need the upper case for the -> int map? # lul, need the upper case for the -> int map?
# sweet "dynamic function behaviour" stdlib... # sweet "dynamic function behaviour" stdlib...
loglevel, loglevel.upper()
) > logging.getLevelName('PDB') ) > logging.getLevelName('PDB')
): ):
loglevel = 'PDB' loglevel = 'PDB'
@ -155,25 +134,20 @@ async def open_root_actor(
"Debug mode is only supported for the `trio` backend!" "Debug mode is only supported for the `trio` backend!"
) )
log.get_console_log(loglevel) # make a temporary connection to see if an arbiter exists
try:
# make a temporary connection to see if an arbiter exists,
# if one can't be made quickly we assume none exists.
arbiter_found = False arbiter_found = False
try:
# TODO: this connect-and-bail forces us to have to carefully # TODO: this connect-and-bail forces us to have to carefully
# rewrap TCP 104-connection-reset errors as EOF so as to avoid # rewrap TCP 104-connection-reset errors as EOF so as to avoid
# propagating cancel-causing errors to the channel-msg loop # propagating cancel-causing errors to the channel-msg loop
# machinery. Likely it would be better to eventually have # machinery. Likely it would be better to eventually have
# a "discovery" protocol with basic handshake instead. # a "discovery" protocol with basic handshake instead.
with trio.move_on_after(1):
async with _connect_chan(host, port): async with _connect_chan(host, port):
arbiter_found = True arbiter_found = True
except OSError: except OSError:
# TODO: make this a "discovery" log level? logger.warning(f"No actor could be found @ {host}:{port}")
logger.warning(f"No actor registry found @ {host}:{port}")
# create a local actor and start up its main routine/task # create a local actor and start up its main routine/task
if arbiter_found: if arbiter_found:
@ -183,7 +157,7 @@ async def open_root_actor(
actor = Actor( actor = Actor(
name or 'anonymous', name or 'anonymous',
arbiter_addr=registry_addr, arbiter_addr=arbiter_addr,
loglevel=loglevel, loglevel=loglevel,
enable_modules=enable_modules, enable_modules=enable_modules,
) )
@ -199,7 +173,7 @@ async def open_root_actor(
actor = Arbiter( actor = Arbiter(
name or 'arbiter', name or 'arbiter',
arbiter_addr=registry_addr, arbiter_addr=arbiter_addr,
loglevel=loglevel, loglevel=loglevel,
enable_modules=enable_modules, enable_modules=enable_modules,
) )
@ -215,14 +189,13 @@ async def open_root_actor(
# start the actor runtime in a new task # start the actor runtime in a new task
async with trio.open_nursery() as nursery: async with trio.open_nursery() as nursery:
# ``_runtime.async_main()`` creates an internal nursery and # ``Actor._async_main()`` creates an internal nursery and
# thus blocks here until the entire underlying actor tree has # thus blocks here until the entire underlying actor tree has
# terminated thereby conducting structured concurrency. # terminated thereby conducting structured concurrency.
await nursery.start( await nursery.start(
partial( partial(
async_main, actor._async_main,
actor,
accept_addr=(host, port), accept_addr=(host, port),
parent_addr=None parent_addr=None
) )
@ -230,10 +203,7 @@ async def open_root_actor(
try: try:
yield actor yield actor
except ( except (Exception, trio.MultiError) as err:
Exception,
BaseExceptionGroup,
) as err:
entered = await _debug._maybe_enter_pm(err) entered = await _debug._maybe_enter_pm(err)
@ -246,8 +216,7 @@ async def open_root_actor(
finally: finally:
# NOTE: not sure if we'll ever need this but it's # NOTE: not sure if we'll ever need this but it's
# possibly better for even more determinism? # possibly better for even more determinism?
# logger.cancel( # logger.cancel(f'Waiting on {len(nurseries)} nurseries in root..')
# f'Waiting on {len(nurseries)} nurseries in root..')
# nurseries = actor._actoruid2nursery.values() # nurseries = actor._actoruid2nursery.values()
# async with trio.open_nursery() as tempn: # async with trio.open_nursery() as tempn:
# for an in nurseries: # for an in nurseries:
@ -257,40 +226,64 @@ async def open_root_actor(
await actor.cancel() await actor.cancel()
finally: finally:
_state._current_actor = None _state._current_actor = None
# restore breakpoint hook state
sys.breakpointhook = builtin_bp_handler
if orig_bp_path is not None:
os.environ['PYTHONBREAKPOINT'] = orig_bp_path
else:
# clear env back to having no entry
os.environ.pop('PYTHONBREAKPOINT')
logger.runtime("Root actor terminated") logger.runtime("Root actor terminated")
def run_daemon( def run(
enable_modules: list[str],
# target
async_fn: typing.Callable[..., typing.Awaitable],
*args,
# runtime kwargs # runtime kwargs
name: str | None = 'root', name: Optional[str] = 'root',
registry_addr: tuple[str, int] = ( arbiter_addr: Tuple[str, int] = (
_default_arbiter_host, _default_arbiter_host,
_default_arbiter_port, _default_arbiter_port,
), ),
start_method: str | None = None, start_method: Optional[str] = None,
debug_mode: bool = False, debug_mode: bool = False,
**kwargs **kwargs,
) -> Any:
"""Run a trio-actor async function in process.
This is tractor's main entry and the start point for any async actor.
"""
async def _main():
async with open_root_actor(
arbiter_addr=arbiter_addr,
name=name,
start_method=start_method,
debug_mode=debug_mode,
**kwargs,
):
return await async_fn(*args)
warnings.warn(
"`tractor.run()` is now deprecated. `tractor` now"
" implicitly starts the root actor on first actor nursery"
" use. If you want to start the root actor manually, use"
" `tractor.open_root_actor()`.",
DeprecationWarning,
stacklevel=2,
)
return trio.run(_main)
def run_daemon(
enable_modules: list[str],
**kwargs
) -> None: ) -> None:
''' '''
Spawn daemon actor which will respond to RPC; the main task simply Spawn daemon actor which will respond to RPC.
starts the runtime and then sleeps forever.
This is a very minimal convenience wrapper around starting This is a convenience wrapper around
a "run-until-cancelled" root actor which can be started with a set ``tractor.run(trio.sleep(float('inf')))`` such that the first actor spawned
of enabled modules for RPC request handling. is meant to run forever responding to RPC requests.
''' '''
kwargs['enable_modules'] = list(enable_modules) kwargs['enable_modules'] = list(enable_modules)
@ -298,15 +291,4 @@ def run_daemon(
for path in enable_modules: for path in enable_modules:
importlib.import_module(path) importlib.import_module(path)
async def _main(): return run(partial(trio.sleep, float('inf')), **kwargs)
async with open_root_actor(
registry_addr=registry_addr,
name=name,
start_method=start_method,
debug_mode=debug_mode,
**kwargs,
):
return await trio.sleep_forever()
return trio.run(_main)

View File

@ -22,15 +22,10 @@ from __future__ import annotations
import sys import sys
import platform import platform
from typing import ( from typing import (
Any, Any, Optional, Callable, TypeVar, TYPE_CHECKING
Awaitable,
Literal,
Callable,
TypeVar,
TYPE_CHECKING,
) )
from collections.abc import Awaitable
from exceptiongroup import BaseExceptionGroup
import trio import trio
from trio_typing import TaskStatus from trio_typing import TaskStatus
@ -44,28 +39,23 @@ from ._state import (
is_root_process, is_root_process,
debug_mode, debug_mode,
) )
from .log import get_logger from .log import get_logger
from ._portal import Portal from ._portal import Portal
from ._runtime import Actor from ._actor import Actor
from ._entry import _mp_main from ._entry import _mp_main
from ._exceptions import ActorFailure from ._exceptions import ActorFailure
if TYPE_CHECKING: if TYPE_CHECKING:
from ._supervise import ActorNursery
import multiprocessing as mp import multiprocessing as mp
ProcessType = TypeVar('ProcessType', mp.Process, trio.Process) ProcessType = TypeVar('ProcessType', mp.Process, trio.Process)
log = get_logger('tractor') log = get_logger('tractor')
# placeholder for an mp start context if so using that backend # placeholder for an mp start context if so using that backend
_ctx: mp.context.BaseContext | None = None _ctx: Optional[mp.context.BaseContext] = None
SpawnMethodKey = Literal[ _spawn_method: str = "trio"
'trio', # supported on all platforms
'mp_spawn',
'mp_forkserver', # posix only
]
_spawn_method: SpawnMethodKey = 'trio'
if platform.system() == 'Windows': if platform.system() == 'Windows':
@ -82,10 +72,7 @@ else:
await trio.lowlevel.wait_readable(proc.sentinel) await trio.lowlevel.wait_readable(proc.sentinel)
def try_set_start_method( def try_set_start_method(name: str) -> Optional[mp.context.BaseContext]:
key: SpawnMethodKey
) -> mp.context.BaseContext | None:
''' '''
Attempt to set the method for process starting, aka the "actor Attempt to set the method for process starting, aka the "actor
spawning backend". spawning backend".
@ -100,30 +87,28 @@ def try_set_start_method(
global _ctx global _ctx
global _spawn_method global _spawn_method
mp_methods = mp.get_all_start_methods() methods = mp.get_all_start_methods()
if 'fork' in mp_methods: if 'fork' in methods:
# forking is incompatible with ``trio``s global task tree # forking is incompatible with ``trio``s global task tree
mp_methods.remove('fork') methods.remove('fork')
match key: # supported on all platforms
case 'mp_forkserver': methods += ['trio']
if name not in methods:
raise ValueError(
f"Spawn method `{name}` is invalid please choose one of {methods}"
)
elif name == 'forkserver':
from . import _forkserver_override from . import _forkserver_override
_forkserver_override.override_stdlib() _forkserver_override.override_stdlib()
_ctx = mp.get_context('forkserver') _ctx = mp.get_context(name)
elif name == 'trio':
case 'mp_spawn':
_ctx = mp.get_context('spawn')
case 'trio':
_ctx = None _ctx = None
else:
_ctx = mp.get_context(name)
case _: _spawn_method = name
raise ValueError(
f'Spawn method `{key}` is invalid!\n'
f'Please choose one of {SpawnMethodKey}'
)
_spawn_method = key
return _ctx return _ctx
@ -139,7 +124,6 @@ async def exhaust_portal(
If the main task is an async generator do our best to consume If the main task is an async generator do our best to consume
what's left of it. what's left of it.
''' '''
__tracebackhide__ = True
try: try:
log.debug(f"Waiting on final result from {actor.uid}") log.debug(f"Waiting on final result from {actor.uid}")
@ -147,11 +131,8 @@ async def exhaust_portal(
# always be established and shutdown using a context manager api # always be established and shutdown using a context manager api
final = await portal.result() final = await portal.result()
except ( except (Exception, trio.MultiError) as err:
Exception, # we reraise in the parent task via a ``trio.MultiError``
BaseExceptionGroup,
) as err:
# we reraise in the parent task via a ``BaseExceptionGroup``
return err return err
except trio.Cancelled as err: except trio.Cancelled as err:
# lol, of course we need this too ;P # lol, of course we need this too ;P
@ -179,7 +160,7 @@ async def cancel_on_completion(
''' '''
# if this call errors we store the exception for later # if this call errors we store the exception for later
# in ``errors`` which will be reraised inside # in ``errors`` which will be reraised inside
# an exception group and we still send out a cancel request # a MultiError and we still send out a cancel request
result = await exhaust_portal(portal, actor) result = await exhaust_portal(portal, actor)
if isinstance(result, Exception): if isinstance(result, Exception):
errors[actor.uid] = result errors[actor.uid] = result
@ -199,37 +180,16 @@ async def cancel_on_completion(
async def do_hard_kill( async def do_hard_kill(
proc: trio.Process, proc: trio.Process,
terminate_after: int = 3, terminate_after: int = 3,
) -> None: ) -> None:
# NOTE: this timeout used to do nothing since we were shielding # NOTE: this timeout used to do nothing since we were shielding
# the ``.wait()`` inside ``new_proc()`` which will pretty much # the ``.wait()`` inside ``new_proc()`` which will pretty much
# never release until the process exits, now it acts as # never release until the process exits, now it acts as
# a hard-kill time ultimatum. # a hard-kill time ultimatum.
log.debug(f"Terminating {proc}")
with trio.move_on_after(terminate_after) as cs: with trio.move_on_after(terminate_after) as cs:
# NOTE: code below was copied verbatim from the now deprecated # NOTE: This ``__aexit__()`` shields internally.
# (in 0.20.0) ``trio._subrocess.Process.aclose()``, orig doc async with proc: # calls ``trio.Process.aclose()``
# string: log.debug(f"Terminating {proc}")
#
# Close any pipes we have to the process (both input and output)
# and wait for it to exit. If cancelled, kills the process and
# waits for it to finish exiting before propagating the
# cancellation.
with trio.CancelScope(shield=True):
if proc.stdin is not None:
await proc.stdin.aclose()
if proc.stdout is not None:
await proc.stdout.aclose()
if proc.stderr is not None:
await proc.stderr.aclose()
try:
await proc.wait()
finally:
if proc.returncode is None:
proc.kill()
with trio.CancelScope(shield=True):
await proc.wait()
if cs.cancelled_caught: if cs.cancelled_caught:
# XXX: should pretty much never get here unless we have # XXX: should pretty much never get here unless we have
@ -280,17 +240,16 @@ async def soft_wait(
if proc.poll() is None: # type: ignore if proc.poll() is None: # type: ignore
log.warning( log.warning(
'Actor still alive after cancel request:\n' f'Process still alive after cancel request:\n{uid}')
f'{uid}'
)
n.cancel_scope.cancel() n.cancel_scope.cancel()
raise raise
async def new_proc( async def new_proc(
name: str, name: str,
actor_nursery: ActorNursery, actor_nursery: 'ActorNursery', # type: ignore # noqa
subactor: Actor, subactor: Actor,
errors: dict[tuple[str, str], Exception], errors: dict[tuple[str, str], Exception],
@ -304,41 +263,6 @@ async def new_proc(
infect_asyncio: bool = False, infect_asyncio: bool = False,
task_status: TaskStatus[Portal] = trio.TASK_STATUS_IGNORED task_status: TaskStatus[Portal] = trio.TASK_STATUS_IGNORED
) -> None:
# lookup backend spawning target
target = _methods[_spawn_method]
# mark the new actor with the global spawn method
subactor._spawn_method = _spawn_method
await target(
name,
actor_nursery,
subactor,
errors,
bind_addr,
parent_addr,
_runtime_vars, # run time vars
infect_asyncio=infect_asyncio,
task_status=task_status,
)
async def trio_proc(
name: str,
actor_nursery: ActorNursery,
subactor: Actor,
errors: dict[tuple[str, str], Exception],
# passed through to actor main
bind_addr: tuple[str, int],
parent_addr: tuple[str, int],
_runtime_vars: dict[str, Any], # serialized and sent to _child
*,
infect_asyncio: bool = False,
task_status: TaskStatus[Portal] = trio.TASK_STATUS_IGNORED
) -> None: ) -> None:
''' '''
Create a new ``Process`` using a "spawn method" as (configured using Create a new ``Process`` using a "spawn method" as (configured using
@ -348,6 +272,11 @@ async def trio_proc(
here is to be considered the core supervision strategy. here is to be considered the core supervision strategy.
''' '''
# mark the new actor with the global spawn method
subactor._spawn_method = _spawn_method
uid = subactor.uid
if _spawn_method == 'trio':
spawn_cmd = [ spawn_cmd = [
sys.executable, sys.executable,
"-m", "-m",
@ -375,11 +304,10 @@ async def trio_proc(
spawn_cmd.append("--asyncio") spawn_cmd.append("--asyncio")
cancelled_during_spawn: bool = False cancelled_during_spawn: bool = False
proc: trio.Process | None = None proc: Optional[trio.Process] = None
try: try:
try: try:
# TODO: needs ``trio_typing`` patch? proc = await trio.open_process(spawn_cmd)
proc = await trio.lowlevel.open_process(spawn_cmd)
log.runtime(f"Started {proc}") log.runtime(f"Started {proc}")
@ -400,21 +328,15 @@ async def trio_proc(
await maybe_wait_for_debugger() await maybe_wait_for_debugger()
elif proc is not None: elif proc is not None:
async with acquire_debug_lock(subactor.uid): async with acquire_debug_lock(uid):
# soft wait on the proc to terminate # soft wait on the proc to terminate
with trio.move_on_after(0.5): with trio.move_on_after(0.5):
await proc.wait() await proc.wait()
raise raise
# a sub-proc ref **must** exist now
assert proc
portal = Portal(chan) portal = Portal(chan)
actor_nursery._children[subactor.uid] = ( actor_nursery._children[subactor.uid] = (
subactor, subactor, proc, portal)
proc,
portal,
)
# send additional init params # send additional init params
await chan.send({ await chan.send({
@ -463,28 +385,21 @@ async def trio_proc(
nursery.cancel_scope.cancel() nursery.cancel_scope.cancel()
finally: finally:
# XXX NOTE XXX: The "hard" reap since no actor zombies are # The "hard" reap since no actor zombies are allowed!
# allowed! Do this **after** cancellation/teardown to avoid # XXX: do this **after** cancellation/tearfown to avoid
# killing the process too early. # killing the process too early.
if proc: if proc:
log.cancel(f'Hard reap sequence starting for {subactor.uid}') log.cancel(f'Hard reap sequence starting for {uid}')
with trio.CancelScope(shield=True): with trio.CancelScope(shield=True):
# don't clobber an ongoing pdb # don't clobber an ongoing pdb
if cancelled_during_spawn: if cancelled_during_spawn:
# Try again to avoid TTY clobbering. # Try again to avoid TTY clobbering.
async with acquire_debug_lock(subactor.uid): async with acquire_debug_lock(uid):
with trio.move_on_after(0.5): with trio.move_on_after(0.5):
await proc.wait() await proc.wait()
if is_root_process(): if is_root_process():
# TODO: solve the following issue where we need
# to do a similar wait like this but in an
# "intermediary" parent actor that itself isn't
# in debug but has a child that is, and we need
# to hold off on relaying SIGINT until that child
# is complete.
# https://github.com/goodboy/tractor/issues/320
await maybe_wait_for_debugger( await maybe_wait_for_debugger(
child_in_debug=_runtime_vars.get( child_in_debug=_runtime_vars.get(
'_debug_mode', False), '_debug_mode', False),
@ -503,10 +418,28 @@ async def trio_proc(
# subactor # subactor
actor_nursery._children.pop(subactor.uid) actor_nursery._children.pop(subactor.uid)
else:
# `multiprocessing`
# async with trio.open_nursery() as nursery:
await mp_new_proc(
name=name,
actor_nursery=actor_nursery,
subactor=subactor,
errors=errors,
# passed through to actor main
bind_addr=bind_addr,
parent_addr=parent_addr,
_runtime_vars=_runtime_vars,
infect_asyncio=infect_asyncio,
task_status=task_status,
)
async def mp_new_proc(
async def mp_proc(
name: str, name: str,
actor_nursery: ActorNursery, # type: ignore # noqa actor_nursery: 'ActorNursery', # type: ignore # noqa
subactor: Actor, subactor: Actor,
errors: dict[tuple[str, str], Exception], errors: dict[tuple[str, str], Exception],
# passed through to actor main # passed through to actor main
@ -531,7 +464,6 @@ async def mp_proc(
assert _ctx assert _ctx
start_method = _ctx.get_start_method() start_method = _ctx.get_start_method()
if start_method == 'forkserver': if start_method == 'forkserver':
from multiprocessing import forkserver # type: ignore from multiprocessing import forkserver # type: ignore
# XXX do our hackery on the stdlib to avoid multiple # XXX do our hackery on the stdlib to avoid multiple
# forkservers (one at each subproc layer). # forkservers (one at each subproc layer).
@ -544,24 +476,23 @@ async def mp_proc(
# forkserver.set_forkserver_preload(enable_modules) # forkserver.set_forkserver_preload(enable_modules)
forkserver.ensure_running() forkserver.ensure_running()
fs_info = ( fs_info = (
fs._forkserver_address, # type: ignore # noqa fs._forkserver_address,
fs._forkserver_alive_fd, # type: ignore # noqa fs._forkserver_alive_fd,
getattr(fs, '_forkserver_pid', None), getattr(fs, '_forkserver_pid', None),
getattr( getattr(
resource_tracker._resource_tracker, '_pid', None), resource_tracker._resource_tracker, '_pid', None),
resource_tracker._resource_tracker._fd, resource_tracker._resource_tracker._fd,
) )
else: # request to forkerserver to fork a new child else:
assert curr_actor._forkserver_info assert curr_actor._forkserver_info
fs_info = ( fs_info = (
fs._forkserver_address, # type: ignore # noqa fs._forkserver_address,
fs._forkserver_alive_fd, # type: ignore # noqa fs._forkserver_alive_fd,
fs._forkserver_pid, # type: ignore # noqa fs._forkserver_pid,
resource_tracker._resource_tracker._pid, resource_tracker._resource_tracker._pid,
resource_tracker._resource_tracker._fd, resource_tracker._resource_tracker._fd,
) = curr_actor._forkserver_info ) = curr_actor._forkserver_info
else: else:
# spawn method
fs_info = (None, None, None, None, None) fs_info = (None, None, None, None, None)
proc: mp.Process = _ctx.Process( # type: ignore proc: mp.Process = _ctx.Process( # type: ignore
@ -570,7 +501,7 @@ async def mp_proc(
subactor, subactor,
bind_addr, bind_addr,
fs_info, fs_info,
_spawn_method, start_method,
parent_addr, parent_addr,
infect_asyncio, infect_asyncio,
), ),
@ -664,16 +595,4 @@ async def mp_proc(
log.debug(f"Joined {proc}") log.debug(f"Joined {proc}")
# pop child entry to indicate we are no longer managing subactor # pop child entry to indicate we are no longer managing subactor
actor_nursery._children.pop(subactor.uid) subactor, proc, portal = actor_nursery._children.pop(subactor.uid)
# TODO: prolly report to ``mypy`` how this causes all sorts of
# false errors..
# subactor, proc, portal = actor_nursery._children.pop(subactor.uid)
# proc spawning backend target map
_methods: dict[SpawnMethodKey, Callable] = {
'trio': trio_proc,
'mp_spawn': mp_proc,
'mp_forkserver': mp_proc,
}

View File

@ -18,10 +18,8 @@
Per process state Per process state
""" """
from typing import ( from typing import Optional, Dict, Any
Optional, from collections.abc import Mapping
Any,
)
import trio import trio
@ -29,7 +27,7 @@ from ._exceptions import NoRuntime
_current_actor: Optional['Actor'] = None # type: ignore # noqa _current_actor: Optional['Actor'] = None # type: ignore # noqa
_runtime_vars: dict[str, Any] = { _runtime_vars: Dict[str, Any] = {
'_debug_mode': False, '_debug_mode': False,
'_is_root': False, '_is_root': False,
'_root_mailbox': (None, None) '_root_mailbox': (None, None)
@ -45,6 +43,30 @@ def current_actor(err_on_no_runtime: bool = True) -> 'Actor': # type: ignore #
return _current_actor return _current_actor
_conc_name_getters = {
'task': trio.lowlevel.current_task,
'actor': current_actor
}
class ActorContextInfo(Mapping):
"Dyanmic lookup for local actor and task names"
_context_keys = ('task', 'actor')
def __len__(self):
return len(self._context_keys)
def __iter__(self):
return iter(self._context_keys)
def __getitem__(self, key: str) -> str:
try:
return _conc_name_getters[key]().name # type: ignore
except RuntimeError:
# no local actor/task context initialized yet
return f'no {key} context'
def is_main_process() -> bool: def is_main_process() -> bool:
"""Bool determining if this actor is running in the top-most process. """Bool determining if this actor is running in the top-most process.
""" """

View File

@ -23,10 +23,8 @@ import inspect
from contextlib import asynccontextmanager from contextlib import asynccontextmanager
from dataclasses import dataclass from dataclasses import dataclass
from typing import ( from typing import (
Any, Any, Optional, Callable,
Optional, AsyncGenerator, Dict,
Callable,
AsyncGenerator,
AsyncIterator AsyncIterator
) )
@ -50,13 +48,12 @@ log = get_logger(__name__)
# - use __slots__ on ``Context``? # - use __slots__ on ``Context``?
class MsgStream(trio.abc.Channel): class ReceiveMsgStream(trio.abc.ReceiveChannel):
''' '''
A bidirectional message stream for receiving logically sequenced A IPC message stream for receiving logically sequenced values over
values over an inter-actor IPC ``Channel``. an inter-actor ``Channel``. This is the type returned to a local
task which entered either ``Portal.open_stream_from()`` or
This is the type returned to a local task which entered either ``Context.open_stream()``.
``Portal.open_stream_from()`` or ``Context.open_stream()``.
Termination rules: Termination rules:
@ -98,9 +95,6 @@ class MsgStream(trio.abc.Channel):
if self._eoc: if self._eoc:
raise trio.EndOfChannel raise trio.EndOfChannel
if self._closed:
raise trio.ClosedResourceError('This stream was closed')
try: try:
msg = await self._rx_chan.receive() msg = await self._rx_chan.receive()
return msg['yield'] return msg['yield']
@ -114,9 +108,6 @@ class MsgStream(trio.abc.Channel):
# - 'error' # - 'error'
# possibly just handle msg['stop'] here! # possibly just handle msg['stop'] here!
if self._closed:
raise trio.ClosedResourceError('This stream was closed')
if msg.get('stop') or self._eoc: if msg.get('stop') or self._eoc:
log.debug(f"{self} was stopped at remote end") log.debug(f"{self} was stopped at remote end")
@ -196,6 +187,7 @@ class MsgStream(trio.abc.Channel):
return return
self._eoc = True self._eoc = True
self._closed = True
# NOTE: this is super subtle IPC messaging stuff: # NOTE: this is super subtle IPC messaging stuff:
# Relay stop iteration to far end **iff** we're # Relay stop iteration to far end **iff** we're
@ -212,8 +204,12 @@ class MsgStream(trio.abc.Channel):
# In the bidirectional case, `Context.open_stream()` will create # In the bidirectional case, `Context.open_stream()` will create
# the `Actor._cids2qs` entry from a call to # the `Actor._cids2qs` entry from a call to
# `Actor.get_context()` and will call us here to send the stop # `Actor.get_context()` and will send the stop message in
# msg in ``__aexit__()`` on teardown. # ``__aexit__()`` on teardown so it **does not** need to be
# called here.
if not self._ctx._portal:
# Only for 2 way streams can we can send stop from the
# caller side.
try: try:
# NOTE: if this call is cancelled we expect this end to # NOTE: if this call is cancelled we expect this end to
# handle as though the stop was never sent (though if it # handle as though the stop was never sent (though if it
@ -230,14 +226,7 @@ class MsgStream(trio.abc.Channel):
# the underlying channel may already have been pulled # the underlying channel may already have been pulled
# in which case our stop message is meaningless since # in which case our stop message is meaningless since
# it can't traverse the transport. # it can't traverse the transport.
ctx = self._ctx log.debug(f'Channel for {self} was already closed')
log.warning(
f'Stream was already destroyed?\n'
f'actor: {ctx.chan.uid}\n'
f'ctx id: {ctx.cid}'
)
self._closed = True
# Do we close the local mem chan ``self._rx_chan`` ??!? # Do we close the local mem chan ``self._rx_chan`` ??!?
@ -280,8 +269,7 @@ class MsgStream(trio.abc.Channel):
self, self,
) -> AsyncIterator[BroadcastReceiver]: ) -> AsyncIterator[BroadcastReceiver]:
''' '''Allocate and return a ``BroadcastReceiver`` which delegates
Allocate and return a ``BroadcastReceiver`` which delegates
to this message stream. to this message stream.
This allows multiple local tasks to receive each their own copy This allows multiple local tasks to receive each their own copy
@ -318,15 +306,15 @@ class MsgStream(trio.abc.Channel):
async with self._broadcaster.subscribe() as bstream: async with self._broadcaster.subscribe() as bstream:
assert bstream.key != self._broadcaster.key assert bstream.key != self._broadcaster.key
assert bstream._recv == self._broadcaster._recv assert bstream._recv == self._broadcaster._recv
# NOTE: we patch on a `.send()` to the bcaster so that the
# caller can still conduct 2-way streaming using this
# ``bstream`` handle transparently as though it was the msg
# stream instance.
bstream.send = self.send # type: ignore
yield bstream yield bstream
class MsgStream(ReceiveMsgStream, trio.abc.Channel):
'''
Bidirectional message stream for use within an inter-actor actor
``Context```.
'''
async def send( async def send(
self, self,
data: Any data: Any
@ -381,8 +369,6 @@ class Context:
# status flags # status flags
_cancel_called: bool = False _cancel_called: bool = False
_cancel_msg: Optional[str] = None
_enter_debugger_on_cancel: bool = True
_started_called: bool = False _started_called: bool = False
_started_received: bool = False _started_received: bool = False
_stream_opened: bool = False _stream_opened: bool = False
@ -407,7 +393,7 @@ class Context:
async def _maybe_raise_from_remote_msg( async def _maybe_raise_from_remote_msg(
self, self,
msg: dict[str, Any], msg: Dict[str, Any],
) -> None: ) -> None:
''' '''
@ -464,11 +450,7 @@ class Context:
if not self._scope_nursery._closed: # type: ignore if not self._scope_nursery._closed: # type: ignore
self._scope_nursery.start_soon(raiser) self._scope_nursery.start_soon(raiser)
async def cancel( async def cancel(self) -> None:
self,
msg: Optional[str] = None,
) -> None:
''' '''
Cancel this inter-actor-task context. Cancel this inter-actor-task context.
@ -477,8 +459,6 @@ class Context:
''' '''
side = 'caller' if self._portal else 'callee' side = 'caller' if self._portal else 'callee'
if msg:
assert side == 'callee', 'Only callee side can provide cancel msg'
log.cancel(f'Cancelling {side} side of context to {self.chan.uid}') log.cancel(f'Cancelling {side} side of context to {self.chan.uid}')
@ -515,10 +495,8 @@ class Context:
log.cancel( log.cancel(
"Timed out on cancelling remote task " "Timed out on cancelling remote task "
f"{cid} for {self._portal.channel.uid}") f"{cid} for {self._portal.channel.uid}")
# callee side remote task
else: else:
self._cancel_msg = msg # callee side remote task
# TODO: should we have an explicit cancel message # TODO: should we have an explicit cancel message
# or is relaying the local `trio.Cancelled` as an # or is relaying the local `trio.Cancelled` as an
@ -603,23 +581,23 @@ class Context:
async with MsgStream( async with MsgStream(
ctx=self, ctx=self,
rx_chan=ctx._recv_chan, rx_chan=ctx._recv_chan,
) as stream: ) as rchan:
if self._portal: if self._portal:
self._portal._streams.add(stream) self._portal._streams.add(rchan)
try: try:
self._stream_opened = True self._stream_opened = True
# XXX: do we need this? # ensure we aren't cancelled before delivering
# ensure we aren't cancelled before yielding the stream # the stream
# await trio.lowlevel.checkpoint() # await trio.lowlevel.checkpoint()
yield stream yield rchan
# NOTE: Make the stream "one-shot use". On exit, signal # XXX: Make the stream "one-shot use". On exit, signal
# ``trio.EndOfChannel``/``StopAsyncIteration`` to the # ``trio.EndOfChannel``/``StopAsyncIteration`` to the
# far end. # far end.
await stream.aclose() await self.send_stop()
finally: finally:
if self._portal: if self._portal:

View File

@ -18,23 +18,19 @@
``trio`` inspired apis and helpers ``trio`` inspired apis and helpers
""" """
from contextlib import asynccontextmanager as acm
from functools import partial from functools import partial
import inspect import inspect
from typing import ( from typing import Tuple, List, Dict, Optional, TYPE_CHECKING
Optional,
TYPE_CHECKING,
)
import typing import typing
import warnings import warnings
from exceptiongroup import BaseExceptionGroup
import trio import trio
from async_generator import asynccontextmanager
from ._debug import maybe_wait_for_debugger from ._debug import maybe_wait_for_debugger
from ._state import current_actor, is_main_process from ._state import current_actor, is_main_process
from .log import get_logger, get_loglevel from .log import get_logger, get_loglevel
from ._runtime import Actor from ._actor import Actor
from ._portal import Portal from ._portal import Portal
from ._exceptions import is_multi_cancelled from ._exceptions import is_multi_cancelled
from ._root import open_root_actor from ._root import open_root_actor
@ -47,7 +43,7 @@ if TYPE_CHECKING:
log = get_logger(__name__) log = get_logger(__name__)
_default_bind_addr: tuple[str, int] = ('127.0.0.1', 0) _default_bind_addr: Tuple[str, int] = ('127.0.0.1', 0)
class ActorNursery: class ActorNursery:
@ -83,19 +79,15 @@ class ActorNursery:
actor: Actor, actor: Actor,
ria_nursery: trio.Nursery, ria_nursery: trio.Nursery,
da_nursery: trio.Nursery, da_nursery: trio.Nursery,
errors: dict[tuple[str, str], BaseException], errors: Dict[Tuple[str, str], Exception],
) -> None: ) -> None:
# self.supervisor = supervisor # TODO # self.supervisor = supervisor # TODO
self._actor: Actor = actor self._actor: Actor = actor
self._ria_nursery = ria_nursery self._ria_nursery = ria_nursery
self._da_nursery = da_nursery self._da_nursery = da_nursery
self._children: dict[ self._children: Dict[
tuple[str, str], Tuple[str, str],
tuple[ Tuple[Actor, mp.Process, Optional[Portal]]
Actor,
trio.Process | mp.Process,
Optional[Portal],
]
] = {} ] = {}
# portals spawned with ``run_in_actor()`` are # portals spawned with ``run_in_actor()`` are
# cancelled when their "main" result arrives # cancelled when their "main" result arrives
@ -110,12 +102,12 @@ class ActorNursery:
self, self,
name: str, name: str,
*, *,
bind_addr: tuple[str, int] = _default_bind_addr, bind_addr: Tuple[str, int] = _default_bind_addr,
rpc_module_paths: list[str] | None = None, rpc_module_paths: List[str] = None,
enable_modules: list[str] | None = None, enable_modules: List[str] = None,
loglevel: str | None = None, # set log level per subactor loglevel: str = None, # set log level per subactor
nursery: trio.Nursery | None = None, nursery: trio.Nursery = None,
debug_mode: Optional[bool] | None = None, debug_mode: Optional[bool] = None,
infect_asyncio: bool = False, infect_asyncio: bool = False,
) -> Portal: ) -> Portal:
''' '''
@ -181,10 +173,10 @@ class ActorNursery:
*, *,
name: Optional[str] = None, name: Optional[str] = None,
bind_addr: tuple[str, int] = _default_bind_addr, bind_addr: Tuple[str, int] = _default_bind_addr,
rpc_module_paths: list[str] | None = None, rpc_module_paths: Optional[List[str]] = None,
enable_modules: list[str] | None = None, enable_modules: List[str] = None,
loglevel: str | None = None, # set log level per subactor loglevel: str = None, # set log level per subactor
infect_asyncio: bool = False, infect_asyncio: bool = False,
**kwargs, # explicit args to ``fn`` **kwargs, # explicit args to ``fn``
@ -295,17 +287,13 @@ class ActorNursery:
self._join_procs.set() self._join_procs.set()
@acm @asynccontextmanager
async def _open_and_supervise_one_cancels_all_nursery( async def _open_and_supervise_one_cancels_all_nursery(
actor: Actor, actor: Actor,
) -> typing.AsyncGenerator[ActorNursery, None]: ) -> typing.AsyncGenerator[ActorNursery, None]:
# TODO: yay or nay?
__tracebackhide__ = True
# the collection of errors retreived from spawned sub-actors # the collection of errors retreived from spawned sub-actors
errors: dict[tuple[str, str], BaseException] = {} errors: Dict[Tuple[str, str], Exception] = {}
# This is the outermost level "deamon actor" nursery. It is awaited # This is the outermost level "deamon actor" nursery. It is awaited
# **after** the below inner "run in actor nursery". This allows for # **after** the below inner "run in actor nursery". This allows for
@ -338,17 +326,19 @@ async def _open_and_supervise_one_cancels_all_nursery(
# after we yield upwards # after we yield upwards
yield anursery yield anursery
# When we didn't error in the caller's scope,
# signal all process-monitor-tasks to conduct
# the "hard join phase".
log.runtime( log.runtime(
f"Waiting on subactors {anursery._children} " f"Waiting on subactors {anursery._children} "
"to complete" "to complete"
) )
# Last bit before first nursery block ends in the case
# where we didn't error in the caller's scope
# signal all process monitor tasks to conduct
# hard join phase.
anursery._join_procs.set() anursery._join_procs.set()
except BaseException as inner_err: except BaseException as err:
errors[actor.uid] = inner_err
# If we error in the root but the debugger is # If we error in the root but the debugger is
# engaged we don't want to prematurely kill (and # engaged we don't want to prematurely kill (and
@ -365,18 +355,19 @@ async def _open_and_supervise_one_cancels_all_nursery(
# worry more are coming). # worry more are coming).
anursery._join_procs.set() anursery._join_procs.set()
try:
# XXX: hypothetically an error could be # XXX: hypothetically an error could be
# raised and then a cancel signal shows up # raised and then a cancel signal shows up
# slightly after in which case the `else:` # slightly after in which case the `else:`
# block here might not complete? For now, # block here might not complete? For now,
# shield both. # shield both.
with trio.CancelScope(shield=True): with trio.CancelScope(shield=True):
etype = type(inner_err) etype = type(err)
if etype in ( if etype in (
trio.Cancelled, trio.Cancelled,
KeyboardInterrupt KeyboardInterrupt
) or ( ) or (
is_multi_cancelled(inner_err) is_multi_cancelled(err)
): ):
log.cancel( log.cancel(
f"Nursery for {current_actor().uid} " f"Nursery for {current_actor().uid} "
@ -384,23 +375,29 @@ async def _open_and_supervise_one_cancels_all_nursery(
else: else:
log.exception( log.exception(
f"Nursery for {current_actor().uid} " f"Nursery for {current_actor().uid} "
f"errored with") f"errored with {err}, ")
# cancel all subactors # cancel all subactors
await anursery.cancel() await anursery.cancel()
except trio.MultiError as merr:
# If we receive additional errors while waiting on
# remaining subactors that were cancelled,
# aggregate those errors with the original error
# that triggered this teardown.
if err not in merr.exceptions:
raise trio.MultiError(merr.exceptions + [err])
else:
raise
# ria_nursery scope end # ria_nursery scope end
# TODO: this is the handler around the ``.run_in_actor()`` # XXX: do we need a `trio.Cancelled` catch here as well?
# nursery. Ideally we can drop this entirely in the future as # this is the catch around the ``.run_in_actor()`` nursery
# the whole ``.run_in_actor()`` API should be built "on top of"
# this lower level spawn-request-cancel "daemon actor" API where
# a local in-actor task nursery is used with one-to-one task
# + `await Portal.run()` calls and the results/errors are
# handled directly (inline) and errors by the local nursery.
except ( except (
Exception, Exception,
BaseExceptionGroup, trio.MultiError,
trio.Cancelled trio.Cancelled
) as err: ) as err:
@ -432,20 +429,18 @@ async def _open_and_supervise_one_cancels_all_nursery(
with trio.CancelScope(shield=True): with trio.CancelScope(shield=True):
await anursery.cancel() await anursery.cancel()
# use `BaseExceptionGroup` as needed # use `MultiError` as needed
if len(errors) > 1: if len(errors) > 1:
raise BaseExceptionGroup( raise trio.MultiError(tuple(errors.values()))
'tractor.ActorNursery errored with',
tuple(errors.values()),
)
else: else:
raise list(errors.values())[0] raise list(errors.values())[0]
# da_nursery scope end - nursery checkpoint # ria_nursery scope end - nursery checkpoint
# final exit
# after nursery exit
@acm @asynccontextmanager
async def open_nursery( async def open_nursery(
**kwargs, **kwargs,

View File

@ -26,10 +26,7 @@ support provided by ``tractor.Context.open_stream()`` and friends.
from __future__ import annotations from __future__ import annotations
import inspect import inspect
import typing import typing
from typing import ( from typing import Dict, Any, Set, Callable, List, Tuple
Any,
Callable,
)
from functools import partial from functools import partial
from async_generator import aclosing from async_generator import aclosing
@ -47,8 +44,8 @@ log = get_logger('messaging')
async def fan_out_to_ctxs( async def fan_out_to_ctxs(
pub_async_gen_func: typing.Callable, # it's an async gen ... gd mypy pub_async_gen_func: typing.Callable, # it's an async gen ... gd mypy
topics2ctxs: dict[str, list], topics2ctxs: Dict[str, list],
packetizer: typing.Callable | None = None, packetizer: typing.Callable = None,
) -> None: ) -> None:
''' '''
Request and fan out quotes to each subscribed actor channel. Request and fan out quotes to each subscribed actor channel.
@ -64,7 +61,7 @@ async def fan_out_to_ctxs(
async for published in pub_gen: async for published in pub_gen:
ctx_payloads: list[tuple[Context, Any]] = [] ctx_payloads: List[Tuple[Context, Any]] = []
for topic, data in published.items(): for topic, data in published.items():
log.debug(f"publishing {topic, data}") log.debug(f"publishing {topic, data}")
@ -106,8 +103,8 @@ async def fan_out_to_ctxs(
def modify_subs( def modify_subs(
topics2ctxs: dict[str, list[Context]], topics2ctxs: Dict[str, List[Context]],
topics: set[str], topics: Set[str],
ctx: Context, ctx: Context,
) -> None: ) -> None:
@ -139,20 +136,20 @@ def modify_subs(
topics2ctxs.pop(topic) topics2ctxs.pop(topic)
_pub_state: dict[str, dict] = {} _pub_state: Dict[str, dict] = {}
_pubtask2lock: dict[str, trio.StrictFIFOLock] = {} _pubtask2lock: Dict[str, trio.StrictFIFOLock] = {}
def pub( def pub(
wrapped: typing.Callable | None = None, wrapped: typing.Callable = None,
*, *,
tasks: set[str] = set(), tasks: Set[str] = set(),
): ):
"""Publisher async generator decorator. """Publisher async generator decorator.
A publisher can be called multiple times from different actors but A publisher can be called multiple times from different actors but
will only spawn a finite set of internal tasks to stream values to will only spawn a finite set of internal tasks to stream values to
each caller. The ``tasks: set[str]`` argument to the decorator each caller. The ``tasks: Set[str]`` argument to the decorator
specifies the names of the mutex set of publisher tasks. When the specifies the names of the mutex set of publisher tasks. When the
publisher function is called, an argument ``task_name`` must be publisher function is called, an argument ``task_name`` must be
passed to specify which task (of the set named in ``tasks``) should passed to specify which task (of the set named in ``tasks``) should
@ -161,9 +158,9 @@ def pub(
necessary. necessary.
Values yielded from the decorated async generator must be Values yielded from the decorated async generator must be
``dict[str, dict[str, Any]]`` where the fist level key is the topic ``Dict[str, Dict[str, Any]]`` where the fist level key is the topic
string and determines which subscription the packet will be string and determines which subscription the packet will be
delivered to and the value is a packet ``dict[str, Any]`` by default delivered to and the value is a packet ``Dict[str, Any]`` by default
of the form: of the form:
.. ::python .. ::python
@ -189,7 +186,7 @@ def pub(
The publisher must be called passing in the following arguments: The publisher must be called passing in the following arguments:
- ``topics: set[str]`` the topic sequence or "subscriptions" - ``topics: Set[str]`` the topic sequence or "subscriptions"
- ``task_name: str`` the task to use (if ``tasks`` was passed) - ``task_name: str`` the task to use (if ``tasks`` was passed)
- ``ctx: Context`` the tractor context (only needed if calling the - ``ctx: Context`` the tractor context (only needed if calling the
pub func without a nursery, otherwise this is provided implicitly) pub func without a nursery, otherwise this is provided implicitly)
@ -234,7 +231,7 @@ def pub(
if wrapped is None: if wrapped is None:
return partial(pub, tasks=tasks) return partial(pub, tasks=tasks)
task2lock: dict[str, trio.StrictFIFOLock] = {} task2lock: Dict[str, trio.StrictFIFOLock] = {}
for name in tasks: for name in tasks:
task2lock[name] = trio.StrictFIFOLock() task2lock[name] = trio.StrictFIFOLock()
@ -246,11 +243,11 @@ def pub(
# `wrapt` docs # `wrapt` docs
async def _execute( async def _execute(
ctx: Context, ctx: Context,
topics: set[str], topics: Set[str],
*args, *args,
# *, # *,
task_name: str | None = None, # default: only one task allocated task_name: str = None, # default: only one task allocated
packetizer: Callable | None = None, packetizer: Callable = None,
**kwargs, **kwargs,
): ):
if task_name is None: if task_name is None:

View File

@ -18,14 +18,12 @@
Log like a forester! Log like a forester!
""" """
from collections.abc import Mapping
import sys import sys
import logging import logging
import colorlog # type: ignore import colorlog # type: ignore
from typing import Optional
import trio from ._state import ActorContextInfo
from ._state import current_actor
_proj_name: str = 'tractor' _proj_name: str = 'tractor'
@ -38,8 +36,7 @@ LOG_FORMAT = (
# "{bold_white}{log_color}{asctime}{reset}" # "{bold_white}{log_color}{asctime}{reset}"
"{log_color}{asctime}{reset}" "{log_color}{asctime}{reset}"
" {bold_white}{thin_white}({reset}" " {bold_white}{thin_white}({reset}"
"{thin_white}{actor_name}[{actor_uid}], " "{thin_white}{actor}, {process}, {task}){reset}{bold_white}{thin_white})"
"{process}, {task}){reset}{bold_white}{thin_white})"
" {reset}{log_color}[{reset}{bold_log_color}{levelname}{reset}{log_color}]" " {reset}{log_color}[{reset}{bold_log_color}{levelname}{reset}{log_color}]"
" {log_color}{name}" " {log_color}{name}"
" {thin_white}{filename}{log_color}:{reset}{thin_white}{lineno}{log_color}" " {thin_white}{filename}{log_color}:{reset}{thin_white}{lineno}{log_color}"
@ -139,40 +136,9 @@ class StackLevelAdapter(logging.LoggerAdapter):
) )
_conc_name_getters = {
'task': lambda: trio.lowlevel.current_task().name,
'actor': lambda: current_actor(),
'actor_name': lambda: current_actor().name,
'actor_uid': lambda: current_actor().uid[1][:6],
}
class ActorContextInfo(Mapping):
"Dyanmic lookup for local actor and task names"
_context_keys = (
'task',
'actor',
'actor_name',
'actor_uid',
)
def __len__(self):
return len(self._context_keys)
def __iter__(self):
return iter(self._context_keys)
def __getitem__(self, key: str) -> str:
try:
return _conc_name_getters[key]()
except RuntimeError:
# no local actor/task context initialized yet
return f'no {key} context'
def get_logger( def get_logger(
name: str | None = None, name: str = None,
_root_name: str = _proj_name, _root_name: str = _proj_name,
) -> StackLevelAdapter: ) -> StackLevelAdapter:
@ -207,7 +173,7 @@ def get_logger(
def get_console_log( def get_console_log(
level: str | None = None, level: str = None,
**kwargs, **kwargs,
) -> logging.LoggerAdapter: ) -> logging.LoggerAdapter:
'''Get the package logger and enable a handler which writes to stderr. '''Get the package logger and enable a handler which writes to stderr.

View File

@ -24,7 +24,7 @@ Built-in messaging patterns, types, APIs and helpers.
# ``pkgutil.resolve_name()`` internally uses # ``pkgutil.resolve_name()`` internally uses
# ``importlib.import_module()`` which can be filtered by inserting # ``importlib.import_module()`` which can be filtered by inserting
# a ``MetaPathFinder`` into ``sys.meta_path`` (which we could do before # a ``MetaPathFinder`` into ``sys.meta_path`` (which we could do before
# entering the ``_runtime.process_messages()`` loop). # entering the ``Actor._process_messages()`` loop).
# - https://github.com/python/cpython/blob/main/Lib/pkgutil.py#L645 # - https://github.com/python/cpython/blob/main/Lib/pkgutil.py#L645
# - https://stackoverflow.com/questions/1350466/preventing-python-code-from-importing-certain-modules # - https://stackoverflow.com/questions/1350466/preventing-python-code-from-importing-certain-modules
# - https://stackoverflow.com/a/63320902 # - https://stackoverflow.com/a/63320902

View File

@ -0,0 +1,17 @@
# tractor: structured concurrent "actors".
# Copyright 2018-eternity Tyler Goodlet.
# 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/>.
from ._tractor_test import tractor_test

View File

@ -0,0 +1,104 @@
# tractor: structured concurrent "actors".
# Copyright 2018-eternity Tyler Goodlet.
# 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/>.
import inspect
import platform
from functools import partial, wraps
import trio
import tractor
__all__ = ['tractor_test']
def tractor_test(fn):
"""
Use:
@tractor_test
async def test_whatever():
await ...
If fixtures:
- ``arb_addr`` (a socket addr tuple where arbiter is listening)
- ``loglevel`` (logging level passed to tractor internals)
- ``start_method`` (subprocess spawning backend)
are defined in the `pytest` fixture space they will be automatically
injected to tests declaring these funcargs.
"""
@wraps(fn)
def wrapper(
*args,
loglevel=None,
arb_addr=None,
start_method=None,
**kwargs
):
# __tracebackhide__ = True
if 'arb_addr' in inspect.signature(fn).parameters:
# injects test suite fixture value to test as well
# as `run()`
kwargs['arb_addr'] = arb_addr
if 'loglevel' in inspect.signature(fn).parameters:
# allows test suites to define a 'loglevel' fixture
# that activates the internal logging
kwargs['loglevel'] = loglevel
if start_method is None:
if platform.system() == "Windows":
start_method = 'spawn'
else:
start_method = 'trio'
if 'start_method' in inspect.signature(fn).parameters:
# set of subprocess spawning backends
kwargs['start_method'] = start_method
if kwargs:
# use explicit root actor start
async def _main():
async with tractor.open_root_actor(
# **kwargs,
arbiter_addr=arb_addr,
loglevel=loglevel,
start_method=start_method,
# TODO: only enable when pytest is passed --pdb
# debug_mode=True,
) as actor:
await fn(*args, **kwargs)
main = _main
else:
# use implicit root actor start
main = partial(fn, *args, **kwargs)
return trio.run(main)
# arbiter_addr=arb_addr,
# loglevel=loglevel,
# start_method=start_method,
# )
return wrapper

View File

@ -466,11 +466,11 @@ async def open_channel_from(
): ):
# sync to a "started()"-like first delivered value from the # sync to a "started()"-like first delivered value from the
# ``asyncio`` task. # ``asyncio`` task.
try:
with chan._trio_cs:
first = await chan.receive() first = await chan.receive()
# deliver stream handle upward # deliver stream handle upward
try:
with chan._trio_cs:
yield first, chan yield first, chan
finally: finally:
chan._trio_exited = True chan._trio_exited = True
@ -491,18 +491,16 @@ def run_as_asyncio_guest(
SC semantics. SC semantics.
''' '''
# Uh, oh. # Uh, oh. :o
#
# :o
# It looks like your event loop has caught a case of the ``trio``s. # It looks like your event loop has caught a case of the ``trio``s.
# :() # :()
# Don't worry, we've heard you'll barely notice. You might # Don't worry, we've heard you'll barely notice. You might hallucinate
# hallucinate a few more propagating errors and feel like your # a few more propagating errors and feel like your digestion has
# digestion has slowed but if anything get's too bad your parents # slowed but if anything get's too bad your parents will know about
# will know about it. # it.
# :) # :)

View File

@ -21,7 +21,6 @@ Sugary patterns for trio + tractor designs.
from ._mngrs import ( from ._mngrs import (
gather_contexts, gather_contexts,
maybe_open_context, maybe_open_context,
maybe_open_nursery,
) )
from ._broadcast import ( from ._broadcast import (
broadcast_receiver, broadcast_receiver,
@ -36,5 +35,4 @@ __all__ = [
'BroadcastReceiver', 'BroadcastReceiver',
'Lagged', 'Lagged',
'maybe_open_context', 'maybe_open_context',
'maybe_open_nursery',
] ]

View File

@ -23,6 +23,7 @@ from __future__ import annotations
from abc import abstractmethod from abc import abstractmethod
from collections import deque from collections import deque
from contextlib import asynccontextmanager from contextlib import asynccontextmanager
from dataclasses import dataclass
from functools import partial from functools import partial
from operator import ne from operator import ne
from typing import Optional, Callable, Awaitable, Any, AsyncIterator, Protocol from typing import Optional, Callable, Awaitable, Any, AsyncIterator, Protocol
@ -32,10 +33,7 @@ import trio
from trio._core._run import Task from trio._core._run import Task
from trio.abc import ReceiveChannel from trio.abc import ReceiveChannel
from trio.lowlevel import current_task from trio.lowlevel import current_task
from msgspec import Struct
from tractor.log import get_logger
log = get_logger(__name__)
# A regular invariant generic type # A regular invariant generic type
T = TypeVar("T") T = TypeVar("T")
@ -88,7 +86,8 @@ class Lagged(trio.TooSlowError):
''' '''
class BroadcastState(Struct): @dataclass
class BroadcastState:
''' '''
Common state to all receivers of a broadcast. Common state to all receivers of a broadcast.
@ -111,35 +110,7 @@ class BroadcastState(Struct):
eoc: bool = False eoc: bool = False
# If the broadcaster was cancelled, we might as well track it # If the broadcaster was cancelled, we might as well track it
cancelled: dict[int, Task] = {} cancelled: bool = False
def statistics(self) -> dict[str, Any]:
'''
Return broadcast receiver group "statistics" like many of
``trio``'s internal task-sync primitives.
'''
key: int | None
ev: trio.Event | None
subs = self.subs
if self.recv_ready is not None:
key, ev = self.recv_ready
else:
key = ev = None
qlens: dict[int, int] = {}
for tid, sz in subs.items():
qlens[tid] = sz if sz != -1 else 0
return {
'open_consumers': len(subs),
'queued_len_by_task': qlens,
'max_buffer_size': self.maxlen,
'tasks_waiting': ev.statistics().tasks_waiting if ev else 0,
'tasks_cancelled': self.cancelled,
'next_value_receiver_id': key,
}
class BroadcastReceiver(ReceiveChannel): class BroadcastReceiver(ReceiveChannel):
@ -157,40 +128,23 @@ class BroadcastReceiver(ReceiveChannel):
rx_chan: AsyncReceiver, rx_chan: AsyncReceiver,
state: BroadcastState, state: BroadcastState,
receive_afunc: Optional[Callable[[], Awaitable[Any]]] = None, receive_afunc: Optional[Callable[[], Awaitable[Any]]] = None,
raise_on_lag: bool = True,
) -> None: ) -> None:
# register the original underlying (clone) # register the original underlying (clone)
self.key = id(self) self.key = id(self)
self._state = state self._state = state
# each consumer has an int count which indicates
# which index contains the next value that the task has not yet
# consumed and thus should read. In the "up-to-date" case the
# consumer task must wait for a new value from the underlying
# receiver and we use ``-1`` as the sentinel for this state.
state.subs[self.key] = -1 state.subs[self.key] = -1
# underlying for this receiver # underlying for this receiver
self._rx = rx_chan self._rx = rx_chan
self._recv = receive_afunc or rx_chan.receive self._recv = receive_afunc or rx_chan.receive
self._closed: bool = False self._closed: bool = False
self._raise_on_lag = raise_on_lag
def receive_nowait( async def receive(self) -> ReceiveType:
self,
_key: int | None = None,
_state: BroadcastState | None = None,
) -> Any: key = self.key
''' state = self._state
Sync version of `.receive()` which does all the low level work
of receiving from the underlying/wrapped receive channel.
'''
key = _key or self.key
state = _state or self._state
# TODO: ideally we can make some way to "lock out" the # TODO: ideally we can make some way to "lock out" the
# underlying receive channel in some way such that if some task # underlying receive channel in some way such that if some task
@ -223,47 +177,32 @@ class BroadcastReceiver(ReceiveChannel):
# return this value." # return this value."
# https://docs.rs/tokio/1.11.0/tokio/sync/broadcast/index.html#lagging # https://docs.rs/tokio/1.11.0/tokio/sync/broadcast/index.html#lagging
mxln = state.maxlen
lost = seq - mxln
# decrement to the last value and expect # decrement to the last value and expect
# consumer to either handle the ``Lagged`` and come back # consumer to either handle the ``Lagged`` and come back
# or bail out on its own (thus un-subscribing) # or bail out on its own (thus un-subscribing)
state.subs[key] = mxln - 1 state.subs[key] = state.maxlen - 1
# this task was overrun by the producer side # this task was overrun by the producer side
task: Task = current_task() task: Task = current_task()
msg = f'Task `{task.name}` overrun and dropped `{lost}` values' raise Lagged(f'Task {task.name} was overrun')
if self._raise_on_lag:
raise Lagged(msg)
else:
log.warning(msg)
return self.receive_nowait(_key, _state)
state.subs[key] -= 1 state.subs[key] -= 1
return value return value
raise trio.WouldBlock # current task already has the latest value **and** is the
# first task to begin waiting for a new one
async def _receive_from_underlying( if state.recv_ready is None:
self,
key: int,
state: BroadcastState,
) -> ReceiveType:
if self._closed: if self._closed:
raise trio.ClosedResourceError raise trio.ClosedResourceError
event = trio.Event() event = trio.Event()
assert state.recv_ready is None
state.recv_ready = key, event state.recv_ready = key, event
try:
# if we're cancelled here it should be # if we're cancelled here it should be
# fine to bail without affecting any other consumers # fine to bail without affecting any other consumers
# right? # right?
try:
value = await self._recv() value = await self._recv()
# items with lower indices are "newer" # items with lower indices are "newer"
@ -281,6 +220,7 @@ class BroadcastReceiver(ReceiveChannel):
# already retreived the last value # already retreived the last value
# XXX: which of these impls is fastest? # XXX: which of these impls is fastest?
# subs = state.subs.copy() # subs = state.subs.copy()
# subs.pop(key) # subs.pop(key)
@ -311,85 +251,54 @@ class BroadcastReceiver(ReceiveChannel):
# consumers will be awoken with a sequence of -1 # consumers will be awoken with a sequence of -1
# and will potentially try to rewait the underlying # and will potentially try to rewait the underlying
# receiver instead of just cancelling immediately. # receiver instead of just cancelling immediately.
self._state.cancelled[key] = current_task() self._state.cancelled = True
if event.statistics().tasks_waiting: if event.statistics().tasks_waiting:
event.set() event.set()
raise raise
finally: finally:
# Reset receiver waiter task event for next blocking condition. # Reset receiver waiter task event for next blocking condition.
# this MUST be reset even if the above ``.recv()`` call # this MUST be reset even if the above ``.recv()`` call
# was cancelled to avoid the next consumer from blocking on # was cancelled to avoid the next consumer from blocking on
# an event that won't be set! # an event that won't be set!
state.recv_ready = None state.recv_ready = None
async def receive(self) -> ReceiveType:
key = self.key
state = self._state
try:
return self.receive_nowait(
_key=key,
_state=state,
)
except trio.WouldBlock:
pass
# current task already has the latest value **and** is the
# first task to begin waiting for a new one so we begin blocking
# until rescheduled with the a new value from the underlying.
if state.recv_ready is None:
return await self._receive_from_underlying(key, state)
# This task is all caught up and ready to receive the latest # This task is all caught up and ready to receive the latest
# value, so queue/schedule it to be woken on the next internal # value, so queue sched it on the internal event.
# event.
else: else:
while state.recv_ready is not None: seq = state.subs[key]
# seq = state.subs[key] assert seq == -1 # sanity
# assert seq == -1 # sanity
_, ev = state.recv_ready _, ev = state.recv_ready
await ev.wait() await ev.wait()
try:
return self.receive_nowait(
_key=key,
_state=state,
)
except trio.WouldBlock:
if self._closed:
raise trio.ClosedResourceError
subs = state.subs # NOTE: if we ever would like the behaviour where if the
if ( # first task to recv on the underlying is cancelled but it
len(subs) == 1 # still DOES trigger the ``.recv_ready``, event we'll likely need
and key in subs # this logic:
# or cancelled
):
# XXX: we are the last and only user of this BR so
# likely it makes sense to unwind back to the
# underlying?
# import tractor
# await tractor.breakpoint()
log.warning(
f'Only one sub left for {self}?\n'
'We can probably unwind from breceiver?'
)
if seq > -1:
# stuff from above..
seq = state.subs[key]
value = state.queue[seq]
state.subs[key] -= 1
return value
elif seq == -1:
# XXX: In the case where the first task to allocate the # XXX: In the case where the first task to allocate the
# ``.recv_ready`` event is cancelled we will be woken # ``.recv_ready`` event is cancelled we will be woken with
# with a non-incremented sequence number (the ``-1`` # a non-incremented sequence number and thus will read the
# sentinel) and thus will read the oldest value if we # oldest value if we use that. Instead we need to detect if
# use that. Instead we need to detect if we have not # we have not been incremented and then receive again.
# been incremented and then receive again. return await self.receive()
# return await self.receive()
return await self._receive_from_underlying(key, state) else:
raise ValueError(f'Invalid sequence {seq}!?')
@asynccontextmanager @asynccontextmanager
async def subscribe( async def subscribe(
self, self,
raise_on_lag: bool = True,
) -> AsyncIterator[BroadcastReceiver]: ) -> AsyncIterator[BroadcastReceiver]:
''' '''
Subscribe for values from this broadcast receiver. Subscribe for values from this broadcast receiver.
@ -407,7 +316,6 @@ class BroadcastReceiver(ReceiveChannel):
rx_chan=self._rx, rx_chan=self._rx,
state=state, state=state,
receive_afunc=self._recv, receive_afunc=self._recv,
raise_on_lag=raise_on_lag,
) )
# assert clone in state.subs # assert clone in state.subs
assert br.key in state.subs assert br.key in state.subs
@ -444,8 +352,7 @@ def broadcast_receiver(
recv_chan: AsyncReceiver, recv_chan: AsyncReceiver,
max_buffer_size: int, max_buffer_size: int,
receive_afunc: Optional[Callable[[], Awaitable[Any]]] = None, **kwargs,
raise_on_lag: bool = True,
) -> BroadcastReceiver: ) -> BroadcastReceiver:
@ -456,6 +363,5 @@ def broadcast_receiver(
maxlen=max_buffer_size, maxlen=max_buffer_size,
subs={}, subs={},
), ),
receive_afunc=receive_afunc, **kwargs,
raise_on_lag=raise_on_lag,
) )

View File

@ -19,7 +19,6 @@ Async context manager primitives with hard ``trio``-aware semantics
''' '''
from contextlib import asynccontextmanager as acm from contextlib import asynccontextmanager as acm
import inspect
from typing import ( from typing import (
Any, Any,
AsyncContextManager, AsyncContextManager,
@ -35,8 +34,8 @@ from typing import (
import trio import trio
from trio_typing import TaskStatus from trio_typing import TaskStatus
from .._state import current_actor
from ..log import get_logger from ..log import get_logger
from .._state import current_actor
log = get_logger(__name__) log = get_logger(__name__)
@ -45,25 +44,6 @@ log = get_logger(__name__)
T = TypeVar("T") T = TypeVar("T")
@acm
async def maybe_open_nursery(
nursery: trio.Nursery | None = None,
shield: bool = False,
) -> AsyncGenerator[trio.Nursery, Any]:
'''
Create a new nursery if None provided.
Blocks on exit as expected if no input nursery is provided.
'''
if nursery is not None:
yield nursery
else:
async with trio.open_nursery() as nursery:
nursery.cancel_scope.shield = shield
yield nursery
async def _enter_and_wait( async def _enter_and_wait(
mngr: AsyncContextManager[T], mngr: AsyncContextManager[T],
@ -101,7 +81,7 @@ async def gather_contexts(
This function is somewhat similar to common usage of This function is somewhat similar to common usage of
``contextlib.AsyncExitStack.enter_async_context()`` (in a loop) in ``contextlib.AsyncExitStack.enter_async_context()`` (in a loop) in
combo with ``asyncio.gather()`` except the managers are concurrently combo with ``asyncio.gather()`` except the managers are concurrently
entered and exited, and cancellation just works. entered and exited cancellation just works.
''' '''
unwrapped: dict[int, Optional[T]] = {}.fromkeys(id(mngr) for mngr in mngrs) unwrapped: dict[int, Optional[T]] = {}.fromkeys(id(mngr) for mngr in mngrs)
@ -109,17 +89,6 @@ async def gather_contexts(
all_entered = trio.Event() all_entered = trio.Event()
parent_exit = trio.Event() parent_exit = trio.Event()
# XXX: ensure greedy sequence of manager instances
# since a lazy inline generator doesn't seem to work
# with `async with` syntax.
mngrs = list(mngrs)
if not mngrs:
raise ValueError(
'input mngrs is empty?\n'
'Did try to use inline generator syntax?'
)
async with trio.open_nursery() as n: async with trio.open_nursery() as n:
for mngr in mngrs: for mngr in mngrs:
n.start_soon( n.start_soon(
@ -133,12 +102,10 @@ async def gather_contexts(
# deliver control once all managers have started up # deliver control once all managers have started up
await all_entered.wait() await all_entered.wait()
try:
yield tuple(unwrapped.values()) yield tuple(unwrapped.values())
finally:
# NOTE: this is ABSOLUTELY REQUIRED to avoid # we don't need a try/finally since cancellation will be triggered
# the following wacky bug: # by the surrounding nursery on error.
# <tractorbugurlhere>
parent_exit.set() parent_exit.set()
@ -152,15 +119,13 @@ class _Cache:
a kept-alive-while-in-use async resource. a kept-alive-while-in-use async resource.
''' '''
service_n: Optional[trio.Nursery] = None lock = trio.Lock()
locks: dict[Hashable, trio.Lock] = {}
users: int = 0 users: int = 0
values: dict[Any, Any] = {} values: dict[Any, Any] = {}
resources: dict[ resources: dict[
Hashable, Hashable,
tuple[trio.Nursery, trio.Event] tuple[trio.Nursery, trio.Event]
] = {} ] = {}
# nurseries: dict[int, trio.Nursery] = {}
no_more_users: Optional[trio.Event] = None no_more_users: Optional[trio.Event] = None
@classmethod @classmethod
@ -191,7 +156,7 @@ async def maybe_open_context(
# XXX: used as cache key after conversion to tuple # XXX: used as cache key after conversion to tuple
# and all embedded values must also be hashable # and all embedded values must also be hashable
kwargs: dict = {}, kwargs: dict = {},
key: Hashable | Callable[..., Hashable] = None, key: Hashable = None,
) -> AsyncIterator[tuple[bool, T]]: ) -> AsyncIterator[tuple[bool, T]]:
''' '''
@ -200,69 +165,51 @@ async def maybe_open_context(
_Cached instance on a _Cache hit. _Cached instance on a _Cache hit.
''' '''
fid = id(acm_func) # lock resource acquisition around task racing / ``trio``'s
# scheduler protocol
await _Cache.lock.acquire()
if inspect.isfunction(key): ctx_key = (id(acm_func), key or tuple(kwargs.items()))
ctx_key = (fid, key(**kwargs)) value = None
else:
ctx_key = (fid, key or tuple(kwargs.items()))
# yielded output
yielded: Any = None
# Lock resource acquisition around task racing / ``trio``'s
# scheduler protocol.
# NOTE: the lock is target context manager func specific in order
# to allow re-entrant use cases where one `maybe_open_context()`
# wrapped factor may want to call into another.
lock = _Cache.locks.setdefault(fid, trio.Lock())
await lock.acquire()
# XXX: one singleton nursery per actor and we want to
# have it not be closed until all consumers have exited (which is
# currently difficult to implement any other way besides using our
# pre-allocated runtime instance..)
service_n: trio.Nursery = current_actor()._service_n
# TODO: is there any way to allocate
# a 'stays-open-till-last-task-finshed nursery?
# service_n: trio.Nursery
# async with maybe_open_nursery(_Cache.service_n) as service_n:
# _Cache.service_n = service_n
try: try:
# **critical section** that should prevent other tasks from # **critical section** that should prevent other tasks from
# checking the _Cache until complete otherwise the scheduler # checking the _Cache until complete otherwise the scheduler
# may switch and by accident we create more then one resource. # may switch and by accident we create more then one resource.
yielded = _Cache.values[ctx_key] value = _Cache.values[ctx_key]
except KeyError: except KeyError:
log.info(f'Allocating new {acm_func} for {ctx_key}') log.info(f'Allocating new {acm_func} for {ctx_key}')
mngr = acm_func(**kwargs) mngr = acm_func(**kwargs)
# TODO: avoid pulling from ``tractor`` internals and
# instead offer a "root nursery" in piker actors?
service_n = current_actor()._service_n
# TODO: does this need to be a tractor "root nursery"?
resources = _Cache.resources resources = _Cache.resources
assert not resources.get(ctx_key), f'Resource exists? {ctx_key}' assert not resources.get(ctx_key), f'Resource exists? {ctx_key}'
resources[ctx_key] = (service_n, trio.Event()) ln, _ = resources[ctx_key] = (service_n, trio.Event())
# sync up to the mngr's yielded value value = await ln.start(
yielded = await service_n.start(
_Cache.run_ctx, _Cache.run_ctx,
mngr, mngr,
ctx_key, ctx_key,
) )
_Cache.users += 1 _Cache.users += 1
lock.release() _Cache.lock.release()
yield False, yielded yield False, value
else: else:
log.info(f'Reusing _Cached resource for {ctx_key}') log.info(f'Reusing _Cached resource for {ctx_key}')
_Cache.users += 1 _Cache.users += 1
lock.release() _Cache.lock.release()
yield True, yielded yield True, value
finally: finally:
_Cache.users -= 1 _Cache.users -= 1
if yielded is not None: if value is not None:
# if no more consumers, teardown the client # if no more consumers, teardown the client
if _Cache.users <= 0: if _Cache.users <= 0:
log.info(f'De-allocating resource for {ctx_key}') log.info(f'De-allocating resource for {ctx_key}')
@ -274,5 +221,3 @@ async def maybe_open_context(
if entry: if entry:
_, no_more_users = entry _, no_more_users = entry
no_more_users.set() no_more_users.set()
_Cache.locks.pop(fid)