[ https://issues.apache.org/jira/browse/HUDI-5453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
sivabalan narayanan updated HUDI-5453: -------------------------------------- Sprint: 0.13.0 Final Sprint > Ensure new fileId format is good across all code paths and backwards > compatible > ------------------------------------------------------------------------------- > > Key: HUDI-5453 > URL: https://issues.apache.org/jira/browse/HUDI-5453 > Project: Apache Hudi > Issue Type: Improvement > Components: metadata > Reporter: sivabalan narayanan > Assignee: sivabalan narayanan > Priority: Blocker > Fix For: 0.13.0 > > > as per master, our file groups in MDT are of format > String.format("%s%04d", metadataPartition.getFileIdPrefix(), i) > > But w/ record level index, we are switching it to > String.format("%s%04d-0", metadataPartition.getFileIdPrefix(), i) > > # Ensure this works across all code paths i.e. we don't rely on exact char > length to determine the file id. > # Ensure this is backwards compatible. for an already populated table, file > group name will be in older format. So, we need to ensure our code handles > both format. -- This message was sent by Atlassian Jira (v8.20.10#820010)