STYLE/CI: implement incremental linting for asv benchmarks #18647
Conversation
|
👍 Sounds good to me @jbrockmendel |
asv_bench/vbench_to_asv.py
Outdated
| @@ -15,7 +15,8 @@ def vbench_to_asv_source(bench, kinds=None): | |||
| output += tab + 'goal_time = 0.2\n\n' | |||
|
|
|||
| if bench.setup: | |||
| indented_setup = [tab * 2 + '{}\n'.format(x) for x in bench.setup.splitlines()] | |||
| indented_setup = [tab * 2 + '{}\n'.format(x) | |||
There was a problem hiding this comment.
Slight preference for keyword arguments for clarity vs. bare {}.
@jreback @jorisvandenbossche : were we just trying to transition to using .format OR .format with keyword?
Codecov Report
@@ Coverage Diff @@
## master #18647 +/- ##
==========================================
- Coverage 91.58% 91.57% -0.01%
==========================================
Files 153 153
Lines 51250 51210 -40
==========================================
- Hits 46935 46895 -40
Misses 4315 4315
Continue to review full report at Codecov.
|
|
@jbrockmendel No need to clean up the vbench_to_asv.py file (it is not used anymore, it is rather there for historical reasons. We could remove it, but let's do that in another PR). So I would revert the changes of that file here. |
jorisvandenbossche
changed the title
|
Thanks! |
Another try at #18620
@mroeschke since it looks like you are going through the asv files one-by-one, the idea here is that when file X gets cleaned up, it can be un-excluded in the lint.sh command.
git diff upstream/master -u -- "*.py" | flake8 --diff