You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@kafka.apache.org by GitBox <gi...@apache.org> on 2022/04/24 03:04:39 UTC

[GitHub] [kafka] ruanliang-hualun commented on a diff in pull request #12092: KAFKA-13834: add normal test case

ruanliang-hualun commented on code in PR #12092:
URL: https://github.com/apache/kafka/pull/12092#discussion_r857058649


##########
clients/src/test/java/org/apache/kafka/clients/producer/internals/RecordAccumulatorTest.java:
##########
@@ -142,15 +144,26 @@ public void testDrainBatches() throws Exception {
         // drain batches from 2 nodes: node1 => tp2, node2 => tp3 (because tp4 is muted)
         Map<Integer, List<ProducerBatch>> batches4 = accum.drain(cluster, new HashSet<Node>(Arrays.asList(node1, node2)), (int) batchSize, 0);
         verifyTopicPartitionInBatches(batches4, tp2, tp3);
+
+
+        // add record for tp1, tp2, tp3, and unmute tp4
+        accum.append(tp1, 0L, key, value, Record.EMPTY_HEADERS, null, maxBlockTimeMs, false, time.milliseconds());
+        accum.append(tp2, 0L, key, value, Record.EMPTY_HEADERS, null, maxBlockTimeMs, false, time.milliseconds());
+        accum.append(tp3, 0L, key, value, Record.EMPTY_HEADERS, null, maxBlockTimeMs, false, time.milliseconds());
+        accum.unmutePartition(tp4);
+        // set maxSize as a max value, so that the all partitions in 2 nodes should be drained: node1 => [tp1, tp2], node2 => [tp3, tp4]
+        Map<Integer, List<ProducerBatch>> batches5 = accum.drain(cluster, new HashSet<Node>(Arrays.asList(node1, node2)), Integer.MAX_VALUE, 0);
+        verifyTopicPartitionInBatches(batches5, tp1, tp2, tp3, tp4);
     }
 
-    private void verifyTopicPartitionInBatches(Map<Integer, List<ProducerBatch>> batches, TopicPartition... tp) {
-        assertEquals(tp.length, batches.size());
+    private void verifyTopicPartitionInBatches(Map<Integer, List<ProducerBatch>> nodeBatches, TopicPartition... tp) {
+        int allTpBatchCount = nodeBatches.values().stream().flatMap(Collection::stream).collect(Collectors.toList()).size();

Review Comment:
   allTpBatchCount  represent the count of the producerbatch from the nodeBatches map



-- 
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: jira-unsubscribe@kafka.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org