kbendick opened a new pull request #3090:
URL: https://github.com/apache/iceberg/pull/3090


   …rted
   
   The (non-Iceberg) tables generated during testing in `TestAddFiles` 
procedure are partitioned on date columns that are not casts to strings.
   
   Most metastores will handle this just fine, but Derby and some others will 
throw an exception. MySQL and Postgres backed metastores will handle this fine 
and won't need to fall back or generate an exception.
   
   Without setting this, many of the tests in `TestAddFiles` fail with 
`MetaException(message:Filtering is supported only on partition keys of type 
string)`
   
   This is related to https://issues.apache.org/jira/browse/SPARK-36128. The 
consensus in the community was that `false` is the best value for this in 
production environments as this can in theory have impact on performance, to 
let users know and adjust their data accordingly.
   
   For tests though, it should probably be set everywhere.
   
   Setting it here now as this is the only place that I've encountered that 
will need it once https://issues.apache.org/jira/browse/SPARK-36128 is part of 
a supported version (should be Spark 3.2 which has release candidates though is 
not GA).


-- 
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.

To unsubscribe, e-mail: [email protected]

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