fix doc and logging mark_channels

[sappelhoff]

PR Description

• closes docstring for mark_channels is misleading #1292

Merge checklist

Maintainer, please confirm the following before merging.
If applicable:

• All comments are resolved
• This is not your own PR
• All CIs are happy
• PR title starts with [MRG]
• whats_new.rst is updated
• New contributors have been added to CITATION.cff
• PR description includes phrase "closes <#issue-number>"

[sappelhoff]

no idea why this was just bad before 🤔 especially because the column is initialized with all good if it wasn't present before.

    # Read sidecar and create required columns if they do not exist.
    if "status" not in tsv_data:
        logger.info('No "status" column found in input file. Creating.')
        tsv_data["status"] = ["good"] * len(tsv_data["name"])

I think this is also why in the docstring it said something about "default is 'bad'". If we start writing the status for a few channels only, the status for all other channels gets automatically set. However it gets set to "good", not "bad", ... which I find reasonable, but which was misleading in the function docstr.

[sappelhoff]

ok, I followed up on this in 5c04279

[sappelhoff]

Thanks for making me aware of this @drammock, I think the present PR should fix a thing or two.

[drammock]

I'm about to open a separate issue about this, but FWIW the behavior of "empty list means mark all channels" was really surprising and caused me to have to alter and re-run my dataset preparation script (no big deal, this is why we dry-run those scripts with only 2 subjs/2 sessions).

[sappelhoff]

Thanks for the review, yes let's discuss your concerns in a separate issue.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 97.45%. Comparing base (87eea28) to head (8f77101).
Report is 55 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1293      +/-   ##
==========================================
- Coverage   97.61%   97.45%   -0.17%     
==========================================
  Files          40       40              
  Lines        8685     8717      +32     
==========================================
+ Hits         8478     8495      +17     
- Misses        207      222      +15     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[sappelhoff]

test failures are unrelated, merging.