geyanggang opened a new pull request, #14664:
URL: https://github.com/apache/gravitino/pull/14664

   ### What changes were proposed in this pull request?
   
   `SparkJdbcTypeConverter.toSparkType` now maps `Types.FixedType` to 
`DataTypes.BinaryType`.
   
   ### Why are the changes needed?
   
   Without a mapping for `Types.FixedType`, a table with a fixed-length binary 
column (for example a JDBC `binary(n)`) could not be described or read through 
Spark; the conversion fell through to `super` and threw. `BinaryType` is the 
natural target, the same type Spark reads a JDBC `BINARY` column as. The 
mapping is one-way: the fixed length is not representable in Spark.
   
   Fixes #14661
   
   ### Does this PR introduce _any_ user-facing change?
   
   Yes. A JDBC table with a `binary(n)` column can now be described and read 
through Spark, where it previously failed.
   
   ### How was this patch tested?
   
   - New unit test 
`TestSparkJdbcTypeConverter#testConvertFixedTypeToSparkBinary`.
   - Built `spark-3.5` and `spark-4.0` with `-PskipITs`; spotless clean.
   


-- 
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