STYLE: #22885, Moved all setup imports to bottom of benchmark files and added noqa: F401 to each

[battlesdev]

• closes STYLE: Fix lint error "F811 redefinition of unused 'setup'" in benchmarks  #22885
• tests added / passed
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

#22885

Following the suggestions of @datapythonista, I moved all the setup imports for the benchmark files to the end of each file and added a # noqa: F401 like so:

from .pandas_vb_common import setup # noqa: F401

Now, flake8 only returns the following two errors in asv_bench:

./benchmarks/strings.py:95:9: E731 do not assign a lambda expression, use a def
./benchmarks/io/excel.py:6:1: F401 '..pandas_vb_common.BaseIO' imported but unused

These may be expected, but I wanted to mention them here just in case they need addressing.

Also, running pytest, I got this error:

usage: pytest [options] [file_or_dir] [file_or_dir] [...]
pytest: error: unrecognized arguments: --strict-data-files
  inifile: /pandas/setup.cfg
  rootdir: /pandas

I found a discussion on it at #22700, but there didn't seem to be a definitive solution. Open to guidance on this.

Please let me know if any corrections are needed. Thanks!

[pep8speaks]

Hello @jerodestapa! Thanks for submitting the PR.

• There are no PEP8 issues in the file asv_bench/benchmarks/algorithms.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/attrs_caching.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/binary_ops.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/categoricals.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/ctors.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/eval.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/frame_ctor.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/frame_methods.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/gil.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/groupby.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/index_object.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/indexing.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/inference.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/io/csv.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/io/excel.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/io/hdf.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/io/json.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/io/msgpack.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/io/pickle.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/io/sql.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/io/stata.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/join_merge.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/multiindex_object.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/panel_ctor.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/panel_methods.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/plotting.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/reindex.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/replace.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/reshape.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/rolling.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/series_methods.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/sparse.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/stat_ops.py !

• There are no PEP8 issues in the file asv_bench/benchmarks/timeseries.py !

[datapythonista]

lgtm. The approach is not ideal of course, but I think it's the best we can do.

Once this and #22863 are merged, we can activate the linting of the benchmarks.

[datapythonista]

Thanks for the work on this @jerodestapa

[TomAugspurger]

For the pytest issue, pytest pandas should work.

I think we're doing something wrong with how we register new command line options, but I'm not familiar enough with pytest plugins to know how to fix it.

[battlesdev]

You bet, @datapythonista. Thanks for suggestions. I agree it's less than ideal, but as you say, it's the best solution given the constraints around flake8.

@TomAugspurger, it's not liking something about setup.cfg and the default option addopts = --strict-data-files --durations=10, but it may just be a local issue on my end. I'll let you know if I figure it out, though.

[jreback]

can you rebase

[codecov]

Codecov Report

Merging #22947 into master will decrease coverage by <.01%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master   #22947      +/-   ##
==========================================
- Coverage   92.19%   92.19%   -0.01%     
==========================================
  Files         169      169              
  Lines       50904    50911       +7     
==========================================
+ Hits        46933    46939       +6     
- Misses       3971     3972       +1

Flag	Coverage Δ
#multiple	90.61% <ø> (-0.01%)	⬇️
#single	42.3% <ø> (-0.01%)	⬇️

Impacted Files	Coverage Δ
pandas/core/internals/concat.py	98% <0%> (-0.38%)	⬇️
pandas/core/dtypes/dtypes.py	95.58% <0%> (+0.03%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ca7d518...a04762d. Read the comment docs.

[battlesdev]

My apologies on the wait, @jreback. Let me know if anything else needs fixing.

Thanks.

[datapythonista]

Not sure if you've seen it @jerodestapa, but seems like you've got unrelated changes in this PR.

If you don't find a better way to fix it, usually the next steps could fix it (in the PR branch):

• git fetch upstream
• git merge upstream/master
• git reset --soft upstream/master
• git commit -m "your pr comment"
• git push -f

This will delete the history of your changes, but should leave your changes and nothing else.

[battlesdev]

Thanks, @datapythonista. I must have done something weird when rebasing. I just reset and pushed again.

[jreback]

this looks fine. can you run the asv's just to be sure things still work.

[battlesdev]

@jreback: The asv's are still running, but it looks like the four files that updated when I rebased (indexing.py, join_merge.py, panel_ctor.py, and panel_methods.py) still have the from .pandas_vb_common import Panel line instead of importing Panel from pandas. Should I remove those lines and make another commit, or rebase again?

[datapythonista]

I think the PR is ok as it is. Not sure why Panel was being imported from pandas_vb_common.py instead of directly, but as your changes already fix the linting, and Panel will be removed soon anyway, I think it's worth just leaving it as it was.

I think this is ready to be merged, thanks for the work @jerodestapa

[battlesdev]

Awesome. Thanks for all the help, guys. This was my first PR for pandas, and it's just exciting to contribute. Thanks, again.

[datapythonista]

Hopefully not the last one @jerodestapa ;)