yihua commented on code in PR #13254:
URL: https://github.com/apache/hudi/pull/13254#discussion_r2076409128
##########
hudi-client/hudi-client-common/src/main/java/org/apache/hudi/metadata/HoodieBackedTableMetadataWriter.java:
##########
@@ -665,38 +670,37 @@ private Pair<Integer, HoodieData<HoodieRecord>>
initializeSecondaryIndexPartitio
return Pair.of(fileGroupCount, records);
}
- private List<Pair<String, FileSlice>> getPartitionFileSlicePairs() throws
IOException {
- String latestInstant =
dataMetaClient.getActiveTimeline().filterCompletedAndCompactionInstants().lastInstant()
- .map(HoodieInstant::requestedTime).orElse(SOLO_COMMIT_TIMESTAMP);
- try (HoodieTableFileSystemView fsView = getMetadataView()) {
- // Collect the list of latest file slices present in each partition
- List<String> partitions = metadata.getAllPartitionPaths();
- fsView.loadAllPartitions();
- List<Pair<String, FileSlice>> partitionFileSlicePairs = new
ArrayList<>();
- partitions.forEach(partition ->
fsView.getLatestMergedFileSlicesBeforeOrOn(partition, latestInstant)
- .forEach(fs -> partitionFileSlicePairs.add(Pair.of(partition, fs))));
- return partitionFileSlicePairs;
- }
+ private Lazy<List<Pair<String, FileSlice>>>
getLazyLatestMergedPartitionFileSliceList() {
+ return Lazy.lazily(() -> {
+ String latestInstant =
dataMetaClient.getActiveTimeline().filterCompletedAndCompactionInstants().lastInstant()
+ .map(HoodieInstant::requestedTime).orElse(SOLO_COMMIT_TIMESTAMP);
+ try (HoodieTableFileSystemView fsView = getMetadataView()) {
+ // Collect the list of latest file slices present in each partition
+ List<String> partitions = metadata.getAllPartitionPaths();
+ fsView.loadAllPartitions();
+ List<Pair<String, FileSlice>> partitionFileSlicePairs = new
ArrayList<>();
+ partitions.forEach(partition ->
fsView.getLatestMergedFileSlicesBeforeOrOn(partition, latestInstant)
+ .forEach(fs -> partitionFileSlicePairs.add(Pair.of(partition,
fs))));
+ return partitionFileSlicePairs;
+ } catch (IOException e) {
+ throw new UncheckedIOException(e);
+ }
+ });
}
- private Pair<Integer, HoodieData<HoodieRecord>>
initializeRecordIndexPartition() throws IOException {
- final HoodieTableFileSystemView fsView = getMetadataView();
+ private Pair<Integer, HoodieData<HoodieRecord>>
initializeRecordIndexPartition(
+ Lazy<List<Pair<String, FileSlice>>>
lazyLatestMergedPartitionFileSliceList) {
Review Comment:
The idea is to avoid all these checks and have a single list of files or
file slices passed down for generating index records so there is no confusion
on the input. That also makes the indexer API much simpler and unit tests
easier.
--
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
