Davis Zhang created HUDI-9530:
---------------------------------

             Summary: AffineBloomIndexFileGroupPartitioner should consume bloom 
index version from the definition file
                 Key: HUDI-9530
                 URL: https://issues.apache.org/jira/browse/HUDI-9530
             Project: Apache Hudi
          Issue Type: Bug
            Reporter: Davis Zhang


org.apache.hudi.index.bloom.SparkHoodieBloomIndexHelper.AffineBloomIndexFileGroupPartitioner#getPartition

 

it calls mapRecordKeyToFileGroupIndex which needs to take bloom index version 
as input, we hard code it with some value but in reality we need to populate 
the field from index def json file.

 

It would be a blocker in case we have multiple bloom filer index versions



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to