linyanghao commented on code in PR #7218:
URL: https://github.com/apache/iceberg/pull/7218#discussion_r1173355010
##########
flink/v1.16/flink/src/test/java/org/apache/iceberg/flink/actions/TestRewriteDataFilesAction.java:
##########
@@ -386,4 +405,84 @@ public void testRewriteAvoidRepeateCompress() throws
IOException {
expected.add(SimpleDataUtil.createRecord(2, "b"));
SimpleDataUtil.assertTableRecords(icebergTableUnPartitioned, expected);
}
+
+ @Test
+ public void testRewriteNoConflictWithEqualityDeletes() throws IOException {
+ // Add 2 data files
+ sql("INSERT INTO %s SELECT 1, 'hello'", TABLE_NAME_WITH_PK);
+ sql("INSERT INTO %s SELECT 2, 'world'", TABLE_NAME_WITH_PK);
+
+ // Load 2 stale tables to pass to rewrite actions
+ // Since the first rewrite will refresh stale1, we need another stale2 for
the second rewrite
+ Table stale1 =
+ validationCatalog.loadTable(TableIdentifier.of(icebergNamespace,
TABLE_NAME_WITH_PK));
+ Table stale2 =
+ validationCatalog.loadTable(TableIdentifier.of(icebergNamespace,
TABLE_NAME_WITH_PK));
+
+ // Add 1 data file and 1 equality-delete file
+ sql("INSERT INTO %s /*+ OPTIONS('upsert-enabled'='true')*/ SELECT 1,
'hi'", TABLE_NAME_WITH_PK);
+
+ icebergTableWithPk.refresh();
+ Assert.assertEquals(
+ "The latest sequence number should be greater than that of the stale
snapshot",
+ stale1.currentSnapshot().sequenceNumber() + 1,
+ icebergTableWithPk.currentSnapshot().sequenceNumber());
+
+ CloseableIterable<FileScanTask> tasks =
icebergTableWithPk.newScan().planFiles();
+ List<DataFile> dataFiles =
+ Lists.newArrayList(CloseableIterable.transform(tasks,
FileScanTask::file));
+ Set<DeleteFile> deleteFiles =
+ Lists.newArrayList(CloseableIterable.transform(tasks,
FileScanTask::deletes)).stream()
+ .flatMap(Collection::stream)
+ .collect(Collectors.toSet());
+ Assert.assertEquals("Should have 3 data files before rewrite", 3,
dataFiles.size());
+ Assert.assertEquals("Should have 1 delete file before rewrite", 1,
deleteFiles.size());
+ Assert.assertSame(
+ "The 1 delete file should be an equality-delete file",
+ Iterables.getOnlyElement(deleteFiles).content(),
+ FileContent.EQUALITY_DELETES);
+ shouldHaveDataAndFileSequenceNumbers(
+ TABLE_NAME_WITH_PK,
+ ImmutableList.of(Pair.of(1L, 1L), Pair.of(2L, 2L), Pair.of(3L, 3L),
Pair.of(3L, 3L)));
+
+ Assertions.assertThatThrownBy(
+ () ->
+ Actions.forTable(stale1)
+ .rewriteDataFiles()
+ .useStartingSequenceNumber(false)
+ .execute(),
+ "Rewrite using new sequence number should fail")
+ .isInstanceOf(ValidationException.class);
+
+ // Rewrite using the starting sequence number should succeed
+ RewriteDataFilesActionResult result =
+
Actions.forTable(stale2).rewriteDataFiles().useStartingSequenceNumber(true).execute();
+
+ // Should not rewrite files from the new commit
+ Assert.assertEquals("Action should rewrite 2 data files", 2,
result.deletedDataFiles().size());
+ Assert.assertEquals("Action should add 1 data file", 1,
result.addedDataFiles().size());
+ shouldHaveDataAndFileSequenceNumbers(
+ TABLE_NAME_WITH_PK, ImmutableList.of(Pair.of(3L, 3L), Pair.of(3L, 3L),
Pair.of(2L, 4L)));
Review Comment:
Added
##########
flink/v1.16/flink/src/test/java/org/apache/iceberg/flink/actions/TestRewriteDataFilesAction.java:
##########
@@ -386,4 +405,84 @@ public void testRewriteAvoidRepeateCompress() throws
IOException {
expected.add(SimpleDataUtil.createRecord(2, "b"));
SimpleDataUtil.assertTableRecords(icebergTableUnPartitioned, expected);
}
+
+ @Test
+ public void testRewriteNoConflictWithEqualityDeletes() throws IOException {
+ // Add 2 data files
+ sql("INSERT INTO %s SELECT 1, 'hello'", TABLE_NAME_WITH_PK);
+ sql("INSERT INTO %s SELECT 2, 'world'", TABLE_NAME_WITH_PK);
+
+ // Load 2 stale tables to pass to rewrite actions
+ // Since the first rewrite will refresh stale1, we need another stale2 for
the second rewrite
+ Table stale1 =
+ validationCatalog.loadTable(TableIdentifier.of(icebergNamespace,
TABLE_NAME_WITH_PK));
+ Table stale2 =
+ validationCatalog.loadTable(TableIdentifier.of(icebergNamespace,
TABLE_NAME_WITH_PK));
+
+ // Add 1 data file and 1 equality-delete file
+ sql("INSERT INTO %s /*+ OPTIONS('upsert-enabled'='true')*/ SELECT 1,
'hi'", TABLE_NAME_WITH_PK);
+
+ icebergTableWithPk.refresh();
+ Assert.assertEquals(
+ "The latest sequence number should be greater than that of the stale
snapshot",
+ stale1.currentSnapshot().sequenceNumber() + 1,
+ icebergTableWithPk.currentSnapshot().sequenceNumber());
+
+ CloseableIterable<FileScanTask> tasks =
icebergTableWithPk.newScan().planFiles();
+ List<DataFile> dataFiles =
+ Lists.newArrayList(CloseableIterable.transform(tasks,
FileScanTask::file));
+ Set<DeleteFile> deleteFiles =
+ Lists.newArrayList(CloseableIterable.transform(tasks,
FileScanTask::deletes)).stream()
+ .flatMap(Collection::stream)
+ .collect(Collectors.toSet());
+ Assert.assertEquals("Should have 3 data files before rewrite", 3,
dataFiles.size());
+ Assert.assertEquals("Should have 1 delete file before rewrite", 1,
deleteFiles.size());
+ Assert.assertSame(
+ "The 1 delete file should be an equality-delete file",
+ Iterables.getOnlyElement(deleteFiles).content(),
+ FileContent.EQUALITY_DELETES);
+ shouldHaveDataAndFileSequenceNumbers(
+ TABLE_NAME_WITH_PK,
+ ImmutableList.of(Pair.of(1L, 1L), Pair.of(2L, 2L), Pair.of(3L, 3L),
Pair.of(3L, 3L)));
+
+ Assertions.assertThatThrownBy(
+ () ->
+ Actions.forTable(stale1)
+ .rewriteDataFiles()
+ .useStartingSequenceNumber(false)
+ .execute(),
+ "Rewrite using new sequence number should fail")
+ .isInstanceOf(ValidationException.class);
+
+ // Rewrite using the starting sequence number should succeed
+ RewriteDataFilesActionResult result =
+
Actions.forTable(stale2).rewriteDataFiles().useStartingSequenceNumber(true).execute();
+
+ // Should not rewrite files from the new commit
+ Assert.assertEquals("Action should rewrite 2 data files", 2,
result.deletedDataFiles().size());
+ Assert.assertEquals("Action should add 1 data file", 1,
result.addedDataFiles().size());
+ shouldHaveDataAndFileSequenceNumbers(
+ TABLE_NAME_WITH_PK, ImmutableList.of(Pair.of(3L, 3L), Pair.of(3L, 3L),
Pair.of(2L, 4L)));
+
+ // Assert the table records as expected.
+ SimpleDataUtil.assertTableRecords(
+ icebergTableWithPk,
+ Lists.newArrayList(
+ SimpleDataUtil.createRecord(1, "hi"),
SimpleDataUtil.createRecord(2, "world")));
+ }
+
+ private void shouldHaveDataAndFileSequenceNumbers(
Review Comment:
Added
--
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: issues-unsubscr...@iceberg.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For additional commands, e-mail: issues-h...@iceberg.apache.org
