[Python][Doc] Improve docs to specify that source argument on parquet.read_table can also be a list of strings

[raulcd]

Describe the enhancement requested

The docstring for parquet.read_table doesn't specify that the source can be a list of strings:

arrow/python/pyarrow/parquet/core.py

Lines 1629 to 1632 in e2a5b4e

	source : str, pyarrow.NativeFile, or file-like object
	If a string passed, can be a single file name or directory name. For
	file-like objects, only read a single file. Use pyarrow.BufferReader to
	read a file contained in a bytes or buffer-like object.

There is also a #TODO for a check on the source argument to validate is of the valid type in case dataset is not enabled.

arrow/python/pyarrow/parquet/core.py

Lines 1826 to 1829 in e2a5b4e

	if filesystem is not None:
	source = filesystem.open_input_file(path)
	# TODO test that source is not a directory or a list
	dataset = ParquetFile(

This should only tackle those. There is another issue to test the fallback case successfully:

• [Python] Exercise fallback case on tests for parquet.read_table in case dataset is not available #46373

Component(s)

Documentation, Python

[rok]

Issue resolved by pull request 47142
#47142

[AlenkaF]

Let's reopen the issue as the second PR for mentioned todo still needs to get merged: #47144. After this is merged, the issue can be closed.

[AlenkaF]

I have moved the TODO part for checking on the source argument into a follow-up issue (#47728) so this can be closed for 22.0.0.