Add file with NaN in statistics

[wjones127]

File was generated with:

import pyarrow as pa # version 0.8.0
import pyarrow.parquet as pq
from numpy import NaN

tab = pa.Table.from_arrays(
    [pa.array([1.0, NaN])],
    names="x"
)

pq.write_table(tab, "nan_in_stats.parquet")

metadata = pq.read_metadata("nan_in_stats.parquet")
metadata.row_group(0).column(0)
# <pyarrow._parquet.ColumnChunkMetaData object at 0x7f28539e58f0>
#   file_offset: 88
#   file_path: 
#   type: DOUBLE
#   num_values: 2
#   path_in_schema: x
#   is_stats_set: True
#   statistics:
#     <pyarrow._parquet.RowGroupStatistics object at 0x7f28539e5738>
#       has_min_max: True
#       min: 1
#       max: nan
#       null_count: 0
#       distinct_count: 0
#       num_values: 2
#       physical_type: DOUBLE
#   compression: 1
#   encodings: <map object at 0x7f28539eb4e0>
#   has_dictionary_page: True
#   dictionary_page_offset: 4
#   data_page_offset: 36
#   index_page_offset: 0
#   total_compressed_size: 84
#   total_uncompressed_size: 80

[pitrou]

Hmm, can you remove the merge markers here?

[wjones127]

🤦

[pitrou]

"Previous versions" is quite unspecific, can you be more explciit?

[wjones127]

I noted the Parquet version that changed in.

[pitrou]

I think it would be useful to add the "File was generated with: [...]" snippet that is part of the PR description.

[wjones127]

I was thinking it would always be in the PR history, but your are right that in the text would be more convenient.

[pitrou]

Thanks @wjones127 !