Verify compression type in Parquet reader

[vuule]

Closes #10602

This PR adds a compression type check for each chunk in the input file.
Reader throws in an unsupported compression is used.

[codecov]

Codecov Report

Merging #10610 (f1dcfd8) into branch-22.06 (956c7b5) will increase coverage by 0.03%.
The diff coverage is 88.97%.

❗ Current head f1dcfd8 differs from pull request most recent head 82b437d. Consider uploading reports for the commit 82b437d to get more accurate results

@@               Coverage Diff                @@
##           branch-22.06   #10610      +/-   ##
================================================
+ Coverage         86.30%   86.34%   +0.03%     
================================================
  Files               140      140              
  Lines             22255    22280      +25     
================================================
+ Hits              19207    19237      +30     
+ Misses             3048     3043       -5     

Impacted Files	Coverage Δ
python/cudf/cudf/core/frame.py	94.75% <ø> (+1.02%)	⬆️
python/dask_cudf/dask_cudf/tests/test_accessor.py	98.41% <ø> (ø)
python/cudf/cudf/core/indexed_frame.py	91.77% <87.93%> (-0.87%)	⬇️
python/cudf/cudf/core/column/lists.py	90.62% <100.00%> (+0.57%)	⬆️
python/cudf/cudf/core/dataframe.py	93.59% <100.00%> (ø)
python/cudf/cudf/core/series.py	95.28% <100.00%> (-0.01%)	⬇️
python/cudf/cudf/core/column/column.py	89.45% <0.00%> (+0.10%)	⬆️
python/cudf/cudf/core/column/string.py	89.10% <0.00%> (+0.12%)	⬆️
python/cudf/cudf/core/groupby/groupby.py	91.72% <0.00%> (+0.22%)	⬆️
python/cudf/cudf/core/tools/datetimes.py	84.49% <0.00%> (+0.30%)	⬆️
... and 1 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 956c7b5...82b437d. Read the comment docs.

[brandon-b-miller]

Fairly minor comment, usually we don't propagate libcudf errors directly to the user from python. Is there a way the reader code can query the codec before it hits libcudf and error there?

[bdice]

Fairly minor comment, usually we don't propagate libcudf errors directly to the user from python. Is there a way the reader code can query the codec before it hits libcudf and error there?

The current design looks fine to me. We pass quite a few low-level errors from libcudf, to my knowledge (perhaps especially in I/O code?). I would avoid re-implementing this codec check in Python if we can let it fail in C++ and bubble through Cython's exception handling. The exception would be cases that need to be pre-emptively stopped at a higher layer (Python), but that doesn't seem to apply here.

[vuule]

Fairly minor comment, usually we don't propagate libcudf errors directly to the user from python. Is there a way the reader code can query the codec before it hits libcudf and error there?

The current design looks fine to me. We pass quite a few low-level errors from libcudf, to my knowledge (perhaps especially in I/O code?). I would avoid re-implementing this codec check in Python if we can let it fail in C++ and bubble through Cython's exception handling. The exception would be cases that need to be pre-emptively stopped at a higher layer (Python), but that doesn't seem to apply here.

Thank you for the comments!

Unfortunately, there's no cheap way to catch this error in Python. I don't think we can avoid propagating the C++ exception here.

[vuule]

@gpucibot merge