[
https://issues.apache.org/jira/browse/PARQUET-1537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16774598#comment-16774598
]
Valery Meleshkin commented on PARQUET-1537:
-------------------------------------------
I can't attach the original file, but I was able to reproduce it by writing
something like the following with pyarrow:
{{df3 = pd.DataFrame({'id': [x for x in range(1000000)],
'str': [''] + [None for _ in range(1000000 - 1)]})
table3 = pa.Table.from_pandas(df3)
with pq.ParquetWriter('example.parquet', table3.schema, use_dictionary=False)
as writer:
writer.write_table(table3)
}}
> [C++] The patch for PARQUET-1508 leads to infinite loop and infinite memory
> allocation when reading very sparse ByteArray columns
> ---------------------------------------------------------------------------------------------------------------------------------
>
> Key: PARQUET-1537
> URL: https://issues.apache.org/jira/browse/PARQUET-1537
> Project: Parquet
> Issue Type: Bug
> Components: parquet-cpp
> Reporter: Valery Meleshkin
> Assignee: Hatem Helal
> Priority: Major
> Fix For: cpp-1.6.0
>
>
> In this loop
> [https://github.com/apache/arrow/commit/3d435e4f8d5fb7a54a4a9d285e1a42d60186d8dc#diff-47fe879cb9baad6c633c55f0a34a09c3R739]
> The branch of if dealing with null values does not increment variable 'i'.
> Therefore on chunks containing only NULLs once a thread enters the loop, it
> stays in that loop forever. I'm not entirely sure whether 'num_values'
> variable was meant to be the number of non-NULL values, yet the total number
> of values is passed here
> [https://github.com/apache/arrow/blob/3d435e4f8d5fb7a54a4a9d285e1a42d60186d8dc/cpp/src/parquet/arrow/record_reader.cc#L528]
>
> On my local machine adding `++i` to the NULL-handling branch seems to fix the
> problem. Unfortunately, I'm not familiar with the codebase enough to be
> certain it's a proper fix.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
