JingsongLi opened a new pull request, #8379: URL: https://github.com/apache/paimon/pull/8379
## Summary Refactor Variant shredding writes onto a generic per-file shredding write plan framework, and prepare the same framework for MAP shared-shredding writes. ## Changes - Introduce generic shredding write plan abstractions and writer factory wrappers for logical-to-physical row conversion. - Move Variant shredding write inference onto the generic framework while keeping Parquet Variant read/write behavior intact. - Add MAP shared-shredding write plan/factory support, including physical row conversion and per-file field metadata generation. - Add Parquet and ORC support for creating writers from shredding physical row types and committing Arrow schema metadata. - Add a helper to wrap writer factories from table options for MAP shared-shredding preparation. ## Testing - mvn -pl paimon-common,paimon-format -am -Pfast-build -DfailIfNoTests=false -Dtest=MapSharedShreddingWritePlanTest,MapSharedShreddingRowConverterTest,MapSharedShreddingUtilsTest,MapSharedShreddingContextTest,ShreddingWritePlanFormatTest,InferVariantShreddingWriteTest,VariantShreddingReadTest test - mvn -pl paimon-common,paimon-format -am -DskipTests validate - mvn -pl paimon-common,paimon-format,paimon-arrow -am -Pfast-build -DskipTests compile - git diff --check ## Notes This PR only prepares the MAP write framework and metadata path. It does not enable MAP shared-shredding automatically in table writes before the corresponding read-side reconstruction is implemented. -- 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]
