[HUDI-4053] Flaky ITTestHoodieDataSource.testStreamWriteBatchReadOpti… (

apache#5526) * [HUDI-4053] Flaky ITTestHoodieDataSource.testStreamWriteBatchReadOptimized Co-authored-by: xicm <xicm@asiainfo.com>
yihua · Jun 3, 2022 · 38c1390 · 38c1390
1 parent 2f0f3b5
commit 38c1390
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 1 deletion.
diff --git a/...ink-datasource/hudi-flink/src/test/java/org/apache/hudi/table/ITTestHoodieDataSource.java b/...ink-datasource/hudi-flink/src/test/java/org/apache/hudi/table/ITTestHoodieDataSource.java
@@ -240,7 +240,15 @@ void testStreamWriteBatchReadOptimized() {
 
     List<Row> rows = CollectionUtil.iterableToList(
         () -> streamTableEnv.sqlQuery("select * from t1").execute().collect());
-    assertRowsEquals(rows, TestData.DATA_SET_SOURCE_INSERT);
+
+    // the test is flaky based on whether the first compaction is pending when
+    // scheduling the 2nd compaction.
+    // see details in CompactionPlanOperator#scheduleCompaction.
+    if (rows.size() < TestData.DATA_SET_SOURCE_INSERT.size()) {
+      assertRowsEquals(rows, TestData.DATA_SET_SOURCE_INSERT_FIRST_COMMIT);
+    } else {
+      assertRowsEquals(rows, TestData.DATA_SET_SOURCE_INSERT);
+    }
   }
 
   @Test

diff --git a/hudi-flink-datasource/hudi-flink/src/test/java/org/apache/hudi/utils/TestData.java b/hudi-flink-datasource/hudi-flink/src/test/java/org/apache/hudi/utils/TestData.java
@@ -164,6 +164,18 @@ public class TestData {
           TimestampData.fromEpochMillis(8000), StringData.fromString("par4"))
   );
 
+  // data set of test_source.data first commit.
+  public static List<RowData> DATA_SET_SOURCE_INSERT_FIRST_COMMIT = Arrays.asList(
+      insertRow(StringData.fromString("id1"), StringData.fromString("Danny"), 23,
+          TimestampData.fromEpochMillis(1000), StringData.fromString("par1")),
+      insertRow(StringData.fromString("id2"), StringData.fromString("Stephen"), 33,
+          TimestampData.fromEpochMillis(2000), StringData.fromString("par1")),
+      insertRow(StringData.fromString("id3"), StringData.fromString("Julian"), 53,
+          TimestampData.fromEpochMillis(3000), StringData.fromString("par2")),
+      insertRow(StringData.fromString("id4"), StringData.fromString("Fabian"), 31,
+          TimestampData.fromEpochMillis(4000), StringData.fromString("par2"))
+  );
+
   // data set of test_source.data latest commit.
   public static List<RowData> DATA_SET_SOURCE_INSERT_LATEST_COMMIT = Arrays.asList(
       insertRow(StringData.fromString("id5"), StringData.fromString("Sophia"), 18,