[
https://issues.apache.org/jira/browse/SPARK-4182?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Cheng Lian updated SPARK-4182:
------------------------------
Description:
If a table contains a column whose type is binary, array, struct, map, and for
some reason, boolean, in-memory columnar caching doesn't work because a
{{NoopColumnStats}} is used to collect column statistics. {{NoopColumnStats}}
returns an empty statistics row, and thus breaks {{InMemoryRelation}}
statistics calculation.
{code}
{code}
was:If a table contains a column whose type is binary, array, struct, map,
and for some reason, boolean, in-memory columnar caching doesn't work because a
{{NoopColumnStats}} is used to collect column statistics. {{NoopColumnStats}}
returns an empty statistics row, and thus breaks {{InMemoryRelation}}
statistics calculation.
> Caching tables containing boolean, binary, array, struct and/or map columns
> doesn't work
> ----------------------------------------------------------------------------------------
>
> Key: SPARK-4182
> URL: https://issues.apache.org/jira/browse/SPARK-4182
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 1.1.1
> Reporter: Cheng Lian
> Assignee: Cheng Lian
> Priority: Blocker
>
> If a table contains a column whose type is binary, array, struct, map, and
> for some reason, boolean, in-memory columnar caching doesn't work because a
> {{NoopColumnStats}} is used to collect column statistics. {{NoopColumnStats}}
> returns an empty statistics row, and thus breaks {{InMemoryRelation}}
> statistics calculation.
> {code}
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
