ENH: enabled 'on' keyword for groupby(..).resample()

[verhalenn]

• closes ENH: enable 'on' keyword for groupby(..).resample() #15021
• tests added / passed
• passes git diff upstream/master | flake8 --diff
• whatsnew entry

[codecov-io]

Codecov Report

Merging #15326 into master will increase coverage by <.01%.

@@            Coverage Diff             @@
##           master   #15326      +/-   ##
==========================================
+ Coverage   86.32%   86.33%   +<.01%     
==========================================
  Files         141      141              
  Lines       51163    51164       +1     
==========================================
+ Hits        44169    44170       +1     
  Misses       6994     6994

Impacted Files	Coverage Δ
pandas/core/groupby.py	95.15% <100%> (ø)	✅

---

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 34cdfa4...9fc3b4f. Read the comment docs.

[jreback]

thanks @verhalenn

changed slightly to put in a more logical place.

we have had this dichotomy between .groupby and .resample where the former uses .key and latter .on. We had a discussion about this to normalize that, but didn't take any action.

@chrisaycock @jorisvandenbossche

worth deprecating key in favor of on in pd.Grouper? or just leave it (I happen to like on better as we use this in .merge as well).

[chrisaycock]

worth deprecating key in favor of on in pd.Grouper?

Makes sense to me. Consistent parameter names makes things easy.

[jorisvandenbossche]

I would personally just leave this.
The current 'sounds' also OK: "resample on a column", group by a key" (not "group by on")

[jreback]

ok, so split opinion! going to make an issue in any event.