danny0405 commented on code in PR #11276:
URL: https://github.com/apache/hudi/pull/11276#discussion_r1612519725
##########
hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/ColumnStatsIndexSupport.scala:
##########
@@ -116,8 +116,8 @@ class ColumnStatsIndexSupport(spark: SparkSession,
* w/in the Metadata Table
*/
def isIndexAvailable: Boolean = {
- checkState(metadataConfig.isEnabled, "Metadata Table support has to be
enabled")
-
metaClient.getTableConfig.getMetadataPartitions.contains(HoodieTableMetadataUtil.PARTITION_NAME_COLUMN_STATS)
+ metadataConfig.isEnabled &&
Review Comment:
makes sense, and another improvement that I can think of is the semantics of
empty `prunedFileNames`, there is case the files are actually empty in rare
cases if all the files are filtered out, and we should return early instead of
moving to another index.
The best way we can do is making it orthoganal, i.e. to extends the pruned
files from preceding index pruning.
--
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]
