[
https://issues.apache.org/jira/browse/PARQUET-62?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14083332#comment-14083332
]
Ryan Blue commented on PARQUET-62:
----------------------------------
Added a fix in [PR #29|https://github.com/apache/incubator-parquet-mr/pull/29].
> DictionaryValuesWriter dictionaries are corrupted by user changes.
> ------------------------------------------------------------------
>
> Key: PARQUET-62
> URL: https://issues.apache.org/jira/browse/PARQUET-62
> Project: Parquet
> Issue Type: Bug
> Components: parquet-mr
> Reporter: Ryan Blue
> Assignee: Ryan Blue
> Priority: Blocker
>
> DictionaryValuesWriter passes incoming Binary objects directly to
> Object2IntMap to accumulate dictionary values. If the arrays backing the
> Binary objects passed in are reused by the caller, then the values are
> corrupted but still written without an error.
> Because Hadoop reuses objects passed to mappers and reducers, this can happen
> easily. For example, Avro reuses the byte arrays backing Utf8 objects, which
> parquet-avro passes wrapped in a Binary object to writeBytes.
> The fix is to make defensive copies of the values passed to the Dictionary
> writer code. I think this only affects the Binary dictionary classes because
> Strings, floats, longs, etc. are immutable.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
