Another increment for static analysis #659
Conversation
Use names from stdlib socket as attributes of the imported module, not via runtime reexport magic
Codecov Report
@@ Coverage Diff @@
## master #659 +/- ##
==========================================
- Coverage 99.31% 99.31% -0.01%
==========================================
Files 94 94
Lines 11014 11002 -12
Branches 790 789 -1
==========================================
- Hits 10939 10927 -12
Misses 56 56
Partials 19 19
Continue to review full report at Codecov.
|
| not self._nested_child_running and not self._children | ||
| and not self._pending_starts | ||
| if not any( | ||
| [self._nested_child_running, self._children, self._pending_starts] |
There was a problem hiding this comment.
Do you think this is an improvement in readability? Consistency-wise I don't think Trio ever uses any() over a literal list of bools and I personally avoid that style because I feel it's less expressive.
There was a problem hiding this comment.
Personally I do find it reads more naturally this way, as starting with "not any" makes reading the three attributes easier. It's also closer to English grammar!
However, if there's any doubt or disagreement I'd rather take it out and get the rest merged.
There was a problem hiding this comment.
Eh, I guess I'd do it the way I originally did it? But they're both fine and let's focus on more important things :-)
|
regarding unused imports, does trio not having a lint check for this set up on the continuous build? |
I'm going in small stages to make review (and debugging) manageable, so:
sslandsocketmodules, rather than by magic