Export test fixes #2887
Export test fixes #2887
Conversation
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## master #2887 +/- ##
==========================================
- Coverage 99.54% 99.54% -0.01%
==========================================
Files 115 115
Lines 17660 17657 -3
Branches 3165 3164 -1
==========================================
- Hits 17580 17577 -3
Misses 52 52
Partials 28 28
|
A5rocks
force-pushed
the
export-test-fixes
branch
from
ebc2fcd to
1397edc
Compare
| if sys.implementation.name != "cpython": | ||
| pytest.skip("jedi does not support pypy") |
There was a problem hiding this comment.
| if sys.implementation.name != "cpython": | |
| pytest.skip("jedi does not support pypy") | |
| if sys.implementation.name != "cpython": # pragma: no branch | |
| pytest.skip("jedi does not support pypy") |
There was a problem hiding this comment.
This feels weird though, like I don't think codecov should be complaining about this :(
There was a problem hiding this comment.
Ah yes, this is the issue where we have double codecov runs on some PRs - and the inline errors are taken from the "incorrect" codecov run, I've seen this a couple times.
That's also the same as #2887 (comment) not getting updated.
I was gonna try and link them .. when I noticed that one set links to https://app.codecov.io/gh/python-trio/trio/commit/c74f51bf891c940d4ec7ed5859e55ab91a2a1809 and the other links to https://github.com/python-trio/trio/pull/2887/checks?check_run_id=19006357499 in the "details" link in the summary. Which is maybe a good hint to what's causing it?
We need some more messing around with codecov 🙃, see also #2689
src/trio/socket.py
Outdated
| @@ -67,13 +67,19 @@ | |||
| ntohs as ntohs, | |||
| ) | |||
|
|
|||
| # TODO: update python docs to reflect this | |||
There was a problem hiding this comment.
mostly just this that's a blocker
There was a problem hiding this comment.
Hopefully done with python/typeshed#11073 and https://foss.heptapod.net/pypy/pypy/-/issues/4040.
There was a problem hiding this comment.
Well I made a typeshed PR thinking "oh well the stdlib docs are the spec, so PyPy is deviating from it and should not have the docs update for it" but then now I realize sys.implementation.name isn't a guard typeshed allows :(
And it looks like these functions just never exist on PyPy. Lol.
|
With ca46414 eliminating jakkdl's blocker, I am going to merge this. |
Was looking at export tests for #2885 and spotted a few changes to make!
Refs #2885