[
https://issues.apache.org/jira/browse/SPARK-7743?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14551579#comment-14551579
]
Cheng Lian commented on SPARK-7743:
-----------------------------------
We probably want to upgrade to the newly release 1.7.0, which is identical to
1.6.0 except for some renaming:
{quote}
This release is identical to the 1.6.0-incubating release, except that the
packages and artifacts have been renamed to org.apache.parquet. There are no
behavioral changes to make transitioning to the new names easier.
{quote}
> Upgrade parquet dependency
> --------------------------
>
> Key: SPARK-7743
> URL: https://issues.apache.org/jira/browse/SPARK-7743
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Reporter: Thomas Omans
>
> There are many outstanding issues with the parquet format that have been
> resolved between the version depended on by spark (1.6.0rc3 as of spark
> 1.3.1) and the most recent parquet release (1.6.0).
> Some of these are things include not supporting schema migration when using
> parquet with avro, not supporting summary metadata in the parquet footers
> causing null pointer exceptions reading, and many others.
> See https://github.com/apache/parquet-mr/blob/master/CHANGES.md#version-160
> for the full list of fixes.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
