You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hudi.apache.org by GitBox <gi...@apache.org> on 2022/03/01 01:54:36 UTC
[GitHub] [hudi] yihua commented on a change in pull request #4878: [HUDI-3465] Add validation of column stats and bloom filters in HoodieMetadataTableValidator
yihua commented on a change in pull request #4878:
URL: https://github.com/apache/hudi/pull/4878#discussion_r816389548
##########
File path: hudi-utilities/src/main/java/org/apache/hudi/utilities/HoodieMetadataTableValidator.java
##########
@@ -438,27 +481,62 @@ private void validateLatestBaseFiles(HoodieTableFileSystemView metaFsView, Hoodi
/**
* Compare getLatestFileSlices between metadata table and fileSystem.
*/
- private void validateLatestFileSlices(HoodieTableFileSystemView metaFsView, HoodieTableFileSystemView fsView, String partitionPath) {
+ private void validateLatestFileSlices(
+ HoodieMetadataValidationContext metadataTableBasedContext,
+ HoodieMetadataValidationContext fsBasedContext, String partitionPath) {
- List<FileSlice> latestFileSlicesFromMetadataTable = metaFsView.getLatestFileSlices(partitionPath).sorted(new FileSliceCompactor()).collect(Collectors.toList());
- List<FileSlice> latestFileSlicesFromFS = fsView.getLatestFileSlices(partitionPath).sorted(new FileSliceCompactor()).collect(Collectors.toList());
+ List<FileSlice> latestFileSlicesFromMetadataTable = metadataTableBasedContext.getSortedLatestFileSliceList(partitionPath);
+ List<FileSlice> latestFileSlicesFromFS = fsBasedContext.getSortedLatestFileSliceList(partitionPath);
- LOG.info("Latest file list from metadata: " + latestFileSlicesFromMetadataTable + ". For partition " + partitionPath);
- LOG.info("Latest file list from direct listing: " + latestFileSlicesFromFS + ". For partition " + partitionPath);
+ LOG.debug("Latest file list from metadata: " + latestFileSlicesFromMetadataTable + ". For partition " + partitionPath);
+ LOG.debug("Latest file list from direct listing: " + latestFileSlicesFromFS + ". For partition " + partitionPath);
- validateFileSlice(latestFileSlicesFromMetadataTable, latestFileSlicesFromFS, partitionPath);
+ validate(latestFileSlicesFromMetadataTable, latestFileSlicesFromFS, partitionPath, "file slices");
LOG.info("Validation of getLatestFileSlices succeeded for partition " + partitionPath);
}
- private HoodieTableFileSystemView createHoodieTableFileSystemView(HoodieSparkEngineContext engineContext, boolean enableMetadataTable) {
+ private void validateAllColumnStats(
+ HoodieMetadataValidationContext metadataTableBasedContext,
+ HoodieMetadataValidationContext fsBasedContext, String partitionPath) {
+ List<String> latestBaseFilenameList = fsBasedContext.getSortedLatestBaseFileList(partitionPath)
+ .stream().map(BaseFile::getFileName).collect(Collectors.toList());
+ List<HoodieColumnRangeMetadata<String>> metadataBasedColStats = metadataTableBasedContext
+ .getSortedColumnStatsList(partitionPath, latestBaseFilenameList);
+ List<HoodieColumnRangeMetadata<String>> fsBasedColStats = fsBasedContext
+ .getSortedColumnStatsList(partitionPath, latestBaseFilenameList);
Review comment:
Thanks for raising the concern. The idea is to separate the issue due to wrong column stats from the mismatched latest base file list for clarity of validation.
--
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-unsubscribe@hudi.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org