Support more units in cudf.DateOffset

[brandon-b-miller]

Enhancements for cudf.DateOffset

• Adds more units for DateOffset
• Allows specifying multiple units when constructing a DateOffset
• Fix for reflected binaryops with DateOffsets

[codecov]

Codecov Report

Merging #7078 (b40930e) into branch-0.20 (51336df) will decrease coverage by 0.44%.
The diff coverage is 82.22%.

❗ Current head b40930e differs from pull request most recent head e0f1b5c. Consider uploading reports for the commit e0f1b5c to get more accurate results

@@               Coverage Diff               @@
##           branch-0.20    #7078      +/-   ##
===============================================
- Coverage        82.88%   82.44%   -0.45%     
===============================================
  Files              103      103              
  Lines            17668    17432     -236     
===============================================
- Hits             14645    14371     -274     
- Misses            3023     3061      +38     

Impacted Files	Coverage Δ
python/cudf/cudf/core/column/__init__.py	100.00% <ø> (ø)
python/cudf/cudf/io/orc.py	86.80% <ø> (-0.10%)	⬇️
python/cudf/cudf/utils/cudautils.py	55.04% <25.00%> (-2.72%)	⬇️
python/cudf/cudf/utils/dtypes.py	81.87% <41.66%> (-1.57%)	⬇️
python/cudf/cudf/core/column/numerical.py	94.41% <70.00%> (-0.02%)	⬇️
python/cudf/cudf/core/tools/datetimes.py	80.00% <74.07%> (-4.54%)	⬇️
python/cudf/cudf/core/column/column.py	88.48% <75.00%> (-0.17%)	⬇️
python/cudf/cudf/core/groupby/groupby.py	91.39% <76.92%> (-0.06%)	⬇️
python/cudf/cudf/core/column/decimal.py	90.08% <78.26%> (-2.85%)	⬇️
python/dask_cudf/dask_cudf/backends.py	89.20% <80.00%> (-0.38%)	⬇️
... and 61 more

---

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 b8baed5...e0f1b5c. Read the comment docs.

[shwina]

rerun tests

[shwina]

rerun tests

[kkraus14]

rerun tests

[vyasr]

I believe all CI is stalled right now due to the connectivity problems we're experiencing across the board preventing environments from even installing correctly.

[shwina]

rerun tests

[vyasr]

Minor note, if #7845 gets merged before this PR then we should address this comment before this PR gets merged. I'll take care of it in #7845 if this PR gets merged first.

@brandon-b-miller since #7845 has been merged, when you next merge branch-0.20 into this PR you'll want to address this change. Specifically, since in this PR cudf.DateOffset no longer inherits from pandas.DateOffset and cudf._DateOffsetScalars has been deleted, you can replace the object.__setattr__ call in this line with just a normal self._scalars = scalars while resolving conflicts. This is already what the code looks like here AFAICT, just letting you know in case the merge looks unexpected.

[brandon-b-miller]

this should be ready to merge :)

[brandon-b-miller]

@kkraus14 @galipremsagar

[galipremsagar]

@brandon-b-miller should we just rerun tests before we hit merge because the run seems to be 3 days old? Just being cautious here.

[brandon-b-miller]

rerun tests

[galipremsagar]

@brandon-b-miller should we just rerun tests before we hit merge because the run seems to be 3 days old? Just being cautious here.

🤦 It got triggered because of the phrase I used

[galipremsagar]

@gpucibot merge