[
https://issues.apache.org/jira/browse/SPARK-58639?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18102732#comment-18102732
]
Uroš Bojanić commented on SPARK-58639:
--------------------------------------
In progress: https://github.com/apache/spark/pull/57846.
> Replace stale listCachedTables reference in the SQL performance tuning guide
> ----------------------------------------------------------------------------
>
> Key: SPARK-58639
> URL: https://issues.apache.org/jira/browse/SPARK-58639
> Project: Spark
> Issue Type: Documentation
> Components: Documentation
> Affects Versions: 4.3.0
> Reporter: Uroš Bojanić
> Priority: Major
> Labels: pull-request-available
>
> The caching section of docs/sql-performance-tuning.md tells users to call
> spark.catalog.listCachedTables() to list cached relations. That API no longer
> exists: it was removed by the SPARK-56221 followup, which deleted
> listCachedTables and the SHOW CACHED TABLES command. The same commit updated
> the sentence to drop the SQL command but left the listCachedTables call
> behind, so the doc now points at a method that cannot be called.
> This replaces the sentence with guidance that works today:
> spark.catalog.isCached for a specific table or view, the Dataset.storageLevel
> property (which returns StorageLevel.NONE when the data is not cached), and
> the Storage tab of the web UI for an overview of everything persisted in the
> application, including data cached directly via Dataset.cache().
> This is the only remaining reference to the removed API in the repository.
> Note that the stale sentence is also present on the released 4.2 line and on
> the current maintenance branches, so a backport is worthwhile.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
