COMPAT: handle pyarrow deprecation of timestamps_to_ms in .from_pandas with pyarrow < 0.6.0

[jreback]

closes #17438

[codecov]

Codecov Report

Merging #17447 into master will decrease coverage by 0.02%.
The diff coverage is 75%.

@@            Coverage Diff             @@
##           master   #17447      +/-   ##
==========================================
- Coverage   91.16%   91.14%   -0.03%     
==========================================
  Files         163      163              
  Lines       49581    49586       +5     
==========================================
- Hits        45199    45193       -6     
- Misses       4382     4393      +11

Flag	Coverage Δ
#multiple	88.92% <75%> (-0.01%)	⬇️
#single	40.25% <12.5%> (-0.07%)	⬇️

Impacted Files	Coverage Δ
pandas/io/parquet.py	65.38% <75%> (-0.37%)	⬇️
pandas/io/gbq.py	25% <0%> (-58.34%)	⬇️
pandas/core/frame.py	97.72% <0%> (-0.1%)	⬇️

---

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 5bca6ce...ea2ff0c. Read the comment docs.

[codecov]

Codecov Report

Merging #17447 into master will decrease coverage by 0.02%.
The diff coverage is 75%.

@@            Coverage Diff             @@
##           master   #17447      +/-   ##
==========================================
- Coverage   91.16%   91.14%   -0.03%     
==========================================
  Files         163      163              
  Lines       49581    49586       +5     
==========================================
- Hits        45199    45193       -6     
- Misses       4382     4393      +11

Flag	Coverage Δ
#multiple	88.92% <75%> (-0.01%)	⬇️
#single	40.25% <12.5%> (-0.07%)	⬇️

Impacted Files	Coverage Δ
pandas/io/parquet.py	65.38% <75%> (-0.37%)	⬇️
pandas/io/gbq.py	25% <0%> (-58.34%)	⬇️
pandas/core/frame.py	97.72% <0%> (-0.1%)	⬇️

---

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 5bca6ce...ea2ff0c. Read the comment docs.

[wesm]

👍, thanks @jreback!

[jorisvandenbossche]

Possibly related to this (didn't look into detail, but just saw this error in the doc builds):

>>>-------------------------------------------------------------------------
Exception in /tmp/doc/source/io.rst at block ending on line 4530
Specify :okexcept: as an option in the ipython:: block to suppress this message
---------------------------------------------------------------------------
ArrowInvalid                              Traceback (most recent call last)
<ipython-input-501-033b23d0b657> in <module>()
----> 1 df.to_parquet('example_pa.parquet', engine='pyarrow')
~/build/pandas-dev/pandas/pandas/core/frame.py in to_parquet(self, fname, engine, compression, **kwargs)
   1620         from pandas.io.parquet import to_parquet
   1621         to_parquet(self, fname, engine,
-> 1622                    compression=compression, **kwargs)
   1623 
   1624     @Substitution(header='Write out the column names. If a list of strings '
~/build/pandas-dev/pandas/pandas/io/parquet.py in to_parquet(df, path, engine, compression, **kwargs)
    176         raise ValueError("parquet must have string column names")
    177 
--> 178     return impl.write(df, path, compression=compression)
    179 
    180 
~/build/pandas-dev/pandas/pandas/io/parquet.py in write(self, df, path, compression, coerce_timestamps, **kwargs)
     75             self.api.parquet.write_table(
     76                 table, path, compression=compression,
---> 77                 coerce_timestamps=coerce_timestamps, **kwargs)
     78 
     79     def read(self, path):
~/miniconda3/envs/pandas/lib/python3.6/site-packages/pyarrow/parquet.py in write_table(table, where, row_group_size, version, use_dictionary, compression, use_deprecated_int96_timestamps, coerce_timestamps, **kwargs)
    794     try:
    795         writer = ParquetWriter(where, table.schema, **options)
--> 796         writer.write_table(table, row_group_size=row_group_size)
    797     except:
    798         if writer is not None:
_parquet.pyx in pyarrow._parquet.ParquetWriter.write_table()
error.pxi in pyarrow.lib.check_status()
ArrowInvalid: Casting from timestamp[ns] to timestamp[ms] would lose data: 1356998400000000001

<<<-------------------------------------------------------------------------

Failing example in the docs: https://pandas-docs.github.io/pandas-docs-travis/io.html#parquet

[jreback]

I pushed a fix for #17447 (comment) @jorisvandenbossche