shangxinli commented on code in PR #20353:
URL: https://github.com/apache/hudi/pull/20353#discussion_r4289842509
##########
hudi-common/src/main/java/org/apache/hudi/common/table/log/HoodieLogBlockMetadataScanner.java:
##########
@@ -36,12 +37,20 @@ public class HoodieLogBlockMetadataScanner extends
BaseHoodieLogRecordReader<Ind
public HoodieLogBlockMetadataScanner(HoodieTableMetaClient metaClient,
List<HoodieLogFile> logFiles,
int bufferSize, String maxInstantTime,
Option<InstantRange> instantRange) {
- super(getReaderContext(metaClient, maxInstantTime),
TableState.fromMetaClient(metaClient), metaClient.getStorage(), logFiles,
+ super(getReaderContext(metaClient, maxInstantTime),
TableState.fromMetaClient(metaClient), metaClient.getStorage(),
withoutCdcLogFiles(logFiles),
Review Comment:
**major:** `HoodieLogCompactionPlanGenerator:97` counts
`fileSlice.getLogFiles()` and returns true at L100 before this scanner is ever
constructed, so a slice with 5 CDC log files and 1 data log file still
schedules a log compaction with nothing to compact. That count needs the same
filter, otherwise only the block-count path is fixed and the file-count path
runs first.
--
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]
