[
https://issues.apache.org/jira/browse/PARQUET-502?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15128516#comment-15128516
]
Wes McKinney commented on PARQUET-502:
--------------------------------------
Can you check if this also fails with the patch for PARQUET-485? I fixed some
bugs on the scan path. Either way we need to get that patch merged ASAP for
unit testing purposes.
> Scanner segfaults when reading a FIXED_LEN_BYTE_ARRAY column
> ------------------------------------------------------------
>
> Key: PARQUET-502
> URL: https://issues.apache.org/jira/browse/PARQUET-502
> Project: Parquet
> Issue Type: Bug
> Components: parquet-cpp
> Affects Versions: cpp-0.1
> Reporter: Aliaksei Sandryhaila
> Assignee: Deepak Majeti
> Fix For: cpp-0.1
>
> Attachments: all_types.parquet
>
>
> TypedScanner<>::HasNext() does not work properly for FIXED_LEN_BYTE_ARRAY.
> When it reaches the end of the column, HasNext() still returns true and a
> subsequent attempt to get the value (e.g. in PrintNext()) causes a segfault.
> To reproduce, use the attached file and run
> parquet-reader all_types.parquet
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
