fix(core): support for Python 3.14 #33461
Conversation
CodSpeed Performance ReportMerging #33461 will degrade performances by 12.27%Comparing
|
| Mode | Benchmark | BASE |
HEAD |
Change | |
|---|---|---|---|---|---|
| 👁 | WallTime | test_import_time[ChatPromptTemplate] |
572.6 ms | 652.6 ms | -12.27% |
| 👁 | WallTime | test_import_time[LangChainTracer] |
412.3 ms | 461.9 ms | -10.73% |
| 👁 | WallTime | test_import_time[PydanticOutputParser] |
505.2 ms | 574.5 ms | -12.05% |
| 👁 | WallTime | test_import_time[RunnableLambda] |
473 ms | 534.9 ms | -11.57% |
Footnotes
-
21 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩
| The coroutine with the context. | ||
| """ | ||
| if asyncio_accepts_context(): | ||
| if sys.version_info >= (3, 11): |
There was a problem hiding this comment.
The signature we get in py314 is asyncio.create_task(coro, **kwargs) so we can't check the presence of context.
There's very little chance that support for context in create_task will ever be removed so I think we can just do a version check here.
There was a problem hiding this comment.
Is it possible to fix the implementation of the function: asyncio_accepts_context instead of not using it?
|
|
||
| loop = asyncio.get_event_loop() | ||
| try: | ||
| loop = asyncio.get_event_loop() |
There was a problem hiding this comment.
asyncio.get_event_loop raises an exception in Python 3.14 if there's no running loop (was a warning in previous versions)
cbornet
force-pushed
the
python-3.14
branch
3 times, most recently
from
c05fe27 to
42fcb96
Compare
| The coroutine with the context. | ||
| """ | ||
| if asyncio_accepts_context(): | ||
| if sys.version_info >= (3, 11): |
There was a problem hiding this comment.
Is it possible to fix the implementation of the function: asyncio_accepts_context instead of not using it?
| f_or_c: Literal["F", "C"] | ||
| forecast: str | ||
|
|
||
| _FORECAST_MODELS_TYPES |= type[ForecastV1] |
There was a problem hiding this comment.
_FORECAST_MODELS_TYPES -- which type is this? something looks off. If it's meant to be a set or a list could we use the dedicated APIs for adding an element instead of relying on the operator shorthand (|=) to maximize for readability?
There was a problem hiding this comment.
_FORECAST_MODELS_TYPES is a typing.Union of types. Used for type checking. I don’t think there is an API to do this. But there’s maybe a simpler way to write it. I’ll have a look.
There was a problem hiding this comment.
I tried to simplify. PTAL.
|
|
||
|
|
||
| @pytest.mark.skipif( | ||
| sys.version_info >= (3, 14), reason="Pydantic v1 not supported with Python 3.14+" |
There was a problem hiding this comment.
Nit: could be helpful to paraphrase the reason a bit (wording is confusing w/ respect to whether this is pydantic 1 or pydantic 2's v1 namespace)
You could say The pydantic.v1 namespace is not supported with python 3.14+ or something like that (to make it clear that it's specifically v1 within pydantic 2)
cbornet
force-pushed
the
python-3.14
branch
2 times, most recently
from
b845d61 to
29c5034
Compare
github-actions
bot
added
infra
|
@eyurtsev in the end, I added the upgrade to pydantic 2.12 to this PR. |
4 participants
asyncio.create_taskasyncio.get_event_loop()raises an exception if there's no running loop