HeartSaVioR opened a new pull request #1432:
URL: https://github.com/apache/iceberg/pull/1432


   This patch invalidates metadata tables in cache when the base table is 
modified, via leveraging listener. 
   
   Iceberg sends `CreateSnapshotEvent` event when the table adds a new 
snapshot, which means the table is modified. CachingCatalog will listen the 
event and find the relevant metadata tables in the cache, and invalidate them.
   
   With Spark 3 the tricky part appears: listener provides table name as full 
identifier which may add catalog name if you don't use default Spark catalog, 
whereas CachingCatalog caches the table identifiers which don't contain the 
catalog name part. Since we just invalidate the cache entities which won't hurt 
much, the patch applies a bit broader criteria to match with.
   
   * compare with case-insensitive, regardless of the option of case-sensitive 
in CachingCatalog
   * compare via endsWith(), as we assume caching catalog has table identifiers 
which don't have the catalog part
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to