richardsheridan
richardsheridan
**What kind of change does this PR introduce?** (check at least one) - [ ] Bugfix - [x] Feature - [ ] Code style update - [ ] Refactor -...
This PR reuses the thread local object in `_threads.py` to stash the `raise_cancel` provided to abort, which basically implements a suggestion by @njsmith in https://github.com/python-trio/trio/issues/961#issuecomment-501597755: >Another idea: in `run_sync_in_thread`, start...
I guess I'm building a reputation for finding trio bugs by writing embarrassingly bad code! Raising an exception in `deliver` argument to `start_thread_soon` will cause the next usage of the...
Fixes #1271. This strategy forgoes the high-level threading API, but gains the ability to ensure that the `cancel_handle` is only closed _as a consequence of_ the wait thread's exit. I've...
Fixes #1271, but also optimizes `WaitForSingleObject` calls. This fulfills a wish I had in #1783 to reduce the number of threads needed during async subprocess usage on Windows. UPDATE: The...
Relevant to #1773, #1767, #824. Maybe even #959 and #1208? This PR leverages Windows message-oriented pipes to create objects of type `Channel[Bytes]` (according to the current state of #1208), supplementing...
Trying to convert a `multiprocessing.Pipe` to a `PipeSendStream`/`PipeReceiveStream` pair as I try to make my own `multiprocessing` wrapper. This is on windows 10, python 3.8.5, trio 0.17. ```python import trio...
This PR provides a new setting, `smoothing_time` applicable for any use of tqdm. This new way to define smoothing is more general/universal; if you know how often you might glance...
## PR Summary Fixes #22847. Not sure if there are knock-on effects, looking forward to code review. ## PR Checklist **Tests and Styling** - [ ] Has pytest style unit...
Revamp of #468. Fixes #467. I was in a position to fix this one because during the python-atomicwrites package fiasco I read somewhere that the author basically said people should...