Xander-run commented on issue #6697:
URL: https://github.com/apache/gravitino/issues/6697#issuecomment-2747815982

   > Thanks for posting the error stack, the main problem is Spark JDBC catalog 
passes all table properties to Gravitino JDBC catalog in 
https://github.com/apache/gravitino/blob/main/spark-connector/spark-common/src/main/java/org/apache/gravitino/spark/connector/jdbc/JdbcPropertiesConverter.java#L65-L67
 which containes `owner` property added by Spark. We could create a separate PG 
properties converter to filter out `owner` properties to fix it.
   
   I'm actually leaning towards addressing this on the Gravitino server side. 
We can create a PostgresTablePropertiesMetadata similar to what we did in 
[MysqlTablePropertiesMetadata](https://github.com/apache/gravitino/blob/6b7cb02de8f4bc84528570a5e515a6790388e23c/catalogs/catalog-jdbc-mysql/src/main/java/org/apache/gravitino/catalog/mysql/MysqlTablePropertiesMetadata.java).
   
   My 2cents:
   
   1. It aligns with the existing MySQL implementation.
   2. It removes the TODO item in 
`PostgreSqlTableOperations#generateCreateTableSql`.
   3. Handling property filtering and conversion universally on the server side 
ensures consistency across different table creation sources(spark, trino, some 
other query engines that might be supported).
   
   wdyt?


-- 
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]

Reply via email to