smjn commented on code in PR #17149:
URL: https://github.com/apache/kafka/pull/17149#discussion_r1759247979
##########
share-coordinator/src/main/java/org/apache/kafka/coordinator/share/ShareCoordinatorShard.java:
##########
@@ -537,38 +539,308 @@ private static ShareGroupOffset merge(ShareGroupOffset
soFar, ShareUpdateValue n
.setStartOffset(newStartOffset)
.setLeaderEpoch(newLeaderEpoch)
.setStateBatches(combineStateBatches(currentBatches,
newData.stateBatches().stream()
- .map(PersisterOffsetsStateBatch::from)
- .collect(Collectors.toCollection(LinkedHashSet::new)),
newStartOffset))
- .build();
+ .map(ShareCoordinatorShard::toPersisterStateBatch)
+ .collect(Collectors.toList()), newStartOffset))
+ .build();
}
/**
- * Util method which takes in 2 collections containing {@link
PersisterOffsetsStateBatch}
+ * Util method which takes in 2 collections containing {@link
PersisterStateBatch}
* and the startOffset.
- * It removes all batches from the 1st collection which have the same
first and last offset
- * as the batches in 2nd collection. It then creates a final list of
batches which contains the
- * former result and all the batches in the 2nd collection. In set
notation (A - B) U B (we prefer batches in B
- * which have same first and last offset in A).
- * Finally, it removes any batches where the lastOffset < startOffset, if
the startOffset > -1.
- * @param currentBatch - collection containing current soft state of
batches
- * @param newBatch - collection containing batches in incoming request
+ * This method checks any overlap between current state batches and new
state batches.
+ * Based on various conditions it creates new non-overlapping records
preferring new batches.
+ * Finally, it removes any batches where the lastOffset < startOffset, if
the startOffset > -1 and
+ * merges any contiguous intervals with same state.
+ * @param batchesSoFar - collection containing current soft state of
batches
+ * @param newBatches - collection containing batches in incoming request
* @param startOffset - startOffset to consider when removing old batches.
* @return List containing combined batches
*/
- private static List<PersisterOffsetsStateBatch> combineStateBatches(
- Collection<PersisterOffsetsStateBatch> currentBatch,
- Collection<PersisterOffsetsStateBatch> newBatch,
+ // visibility for testing
+ static List<PersisterStateBatch> combineStateBatches(
+ List<PersisterStateBatch> batchesSoFar,
+ List<PersisterStateBatch> newBatches,
long startOffset
) {
- currentBatch.removeAll(newBatch);
- List<PersisterOffsetsStateBatch> batchesToAdd = new
LinkedList<>(currentBatch);
- batchesToAdd.addAll(newBatch);
- // Any batches where the last offset is < the current start offset
- // are now expired. We should remove them from the persister.
+ List<PersisterStateBatch> combinedList = new
ArrayList<>(batchesSoFar.size() + newBatches.size());
+ combinedList.addAll(batchesSoFar);
+ combinedList.addAll(newBatches);
+
+ // sort keeping delivery state in mind
+ combinedList.sort(PersisterStateBatch::compareTo);
+
+ return mergeBatches(
+ pruneBatches(
+ combinedList,
+ startOffset
+ )
+ );
+ }
+
+ private static class BatchOverlapState {
+ private final PersisterStateBatch last;
+ private final PersisterStateBatch candidate;
+ private final List<PersisterStateBatch> nonOverlapping;
+ public static final BatchOverlapState SENTINEL = new
BatchOverlapState(null, null, Collections.emptyList());
+
+ public BatchOverlapState(
+ PersisterStateBatch last,
+ PersisterStateBatch candidate,
+ List<PersisterStateBatch> nonOverlapping
+ ) {
+ this.last = last;
+ this.candidate = candidate;
+ this.nonOverlapping = nonOverlapping;
+ }
+
+ public PersisterStateBatch last() {
+ return last;
+ }
+
+ public PersisterStateBatch candidate() {
+ return candidate;
+ }
+
+ public List<PersisterStateBatch> nonOverlapping() {
+ return nonOverlapping;
+ }
+
+ @Override
+ public boolean equals(Object o) {
+ if (this == o) return true;
+ if (!(o instanceof BatchOverlapState)) return false;
+ BatchOverlapState that = (BatchOverlapState) o;
+ return Objects.equals(last, that.last) &&
Objects.equals(candidate, that.candidate) && Objects.equals(nonOverlapping,
that.nonOverlapping);
+ }
+
+ @Override
+ public int hashCode() {
+ return Objects.hash(last, candidate, nonOverlapping);
+ }
+ }
+
+ static List<PersisterStateBatch> mergeBatches(List<PersisterStateBatch>
batches) {
+ if (batches.size() < 2) {
+ return batches;
+ }
+ TreeSet<PersisterStateBatch> sortedBatches = new TreeSet<>(batches);
+ List<PersisterStateBatch> finalBatches = new
ArrayList<>(batches.size() * 2); // heuristic size
+
+ BatchOverlapState overlapState = getOverlappingState(sortedBatches);
+
+ while (overlapState != BatchOverlapState.SENTINEL) {
Review Comment:
then we'll make it
--
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]
