yihua opened a new pull request, #19165: URL: https://github.com/apache/hudi/pull/19165
### Change Logs Adds `TestFileGroupReaderReadPath`, a functional suite for the default (file-group reader) Spark read path, covering gaps not exercised by the legacy reader-disabled suite or the existing COW/MOR/CDC suites: - Read-time schema evolution (add-column and int->long type promotion) with filter pushdown over evolved schemas, on COW and MOR (MOR exercises base+log merge). - Incremental read after evolution, asserting the exact changed record set. - Partition pruning via `HoodieFileIndex.listFiles` (single/multi/no filter) and typed partition-value reconstruction (String + Int) with partition columns dropped from data files. Reads keep `hoodie.file.group.reader.enabled` at its default. Targets `ParquetSchemaEvolutionUtils`, `HoodieFileIndex`/`SparkHoodieTableFileIndex`, `HoodieMergeOnReadRDDV2`, `FileFormatUtilsForFileGroupReader`, and the per-version `HoodiePartitionValues` classes. ### Impact Test-only. No production code changes. ### Risk level none ### Documentation Update None. ### Contributor's checklist - [x] Read through the contributor's guide - [x] Change is test-only and behavior-preserving - [x] Tests assert meaningful behavior -- 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]
