nsivabalan commented on code in PR #9308: URL: https://github.com/apache/hudi/pull/9308#discussion_r1283907863
########## hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/HoodieFileIndex.scala: ########## @@ -223,10 +225,13 @@ case class HoodieFileIndex(spark: SparkSession, // nothing CSI in particular could be applied for) lazy val queryReferencedColumns = collectReferencedColumns(spark, queryFilters, schema) - if (!isMetadataTableEnabled || !isDataSkippingEnabled || !columnStatsIndex.isIndexAvailable) { + if (!isMetadataTableEnabled || !isDataSkippingEnabled) { validateConfig() Option.empty - } else if (queryFilters.isEmpty || queryReferencedColumns.isEmpty) { + } else if (recordLevelIndex.isIndexApplicable(queryFilters)) { + Option.apply(recordLevelIndex.getCandidateFiles(allFiles, queryFilters)) + } else if (!columnStatsIndex.isIndexAvailable || queryFilters.isEmpty || queryReferencedColumns.isEmpty) { + validateConfig() Review Comment: lets atleast fix the filtering such that just for record key we look up in RLI, for, but if there are other columns, we do route to col stats. and combine both together. currently, this seems very limited. -- 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: commits-unsubscr...@hudi.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org