Skip to content

[core] Skip file index for changelog files #7697

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
JingsongLi merged 1 commit into from
Apr 26, 2026
+54 −4
Merged

[core] Skip file index for changelog files #7697

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 4 additions & 2 deletions paimon-core/src/main/java/org/apache/paimon/io/KeyValueFileWriterFactory.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,6 +149,8 @@ private KeyValueClusteringFileWriter createKvSeparatedFileWriter(

private KeyValueDataFileWriter createDataFileWriter(
Path path, WriteFormatKey key, FileSource fileSource, boolean isExternalPath) {
// Changelog is sequentially consumed, file index is unnecessary.
FileIndexOptions indexOptions = key.isChangelog ? new FileIndexOptions() : fileIndexOptions;
return formatContext.thinModeEnabled
? new KeyValueThinDataFileWriterImpl(
fileIO,
Expand All @@ -161,7 +163,7 @@ private KeyValueDataFileWriter createDataFileWriter(
key.level,
options,
fileSource,
fileIndexOptions,
indexOptions,
isExternalPath)
: new KeyValueDataFileWriterImpl(
fileIO,
Expand All @@ -174,7 +176,7 @@ private KeyValueDataFileWriter createDataFileWriter(
key.level,
options,
fileSource,
fileIndexOptions,
indexOptions,
isExternalPath);
}

Expand Down
52 changes: 50 additions & 2 deletions paimon-core/src/test/java/org/apache/paimon/io/KeyValueFileReadWriteTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -308,11 +308,16 @@ public void testFileSuffix(@TempDir java.nio.file.Path tempDir) throws Exception
}

protected KeyValueFileWriterFactory createWriterFactory(String pathStr, String format) {
Options options = new Options();
options.set(CoreOptions.METADATA_STATS_MODE, "FULL");
return createWriterFactory(pathStr, format, options);
}

protected KeyValueFileWriterFactory createWriterFactory(
String pathStr, String format, Options options) {
Path path = new Path(pathStr);
int suggestedFileSize = ThreadLocalRandom.current().nextInt(8192) + 1024;
FileIO fileIO = FileIOFinder.find(path);
Options options = new Options();
options.set(CoreOptions.METADATA_STATS_MODE, "FULL");

Function<String, FileStorePathFactory> pathFactoryMap =
format1 ->
Expand Down Expand Up @@ -455,6 +460,49 @@ private void checkRollingFiles(
}
}

@Test
void testChangelogFile() throws Exception {
Options options = new Options();
options.set(CoreOptions.METADATA_STATS_MODE, "FULL");
options.setString("file-index.bloom-filter.columns", "comment");
options.setString("file-index.in-manifest-threshold", "1B");

KeyValueFileWriterFactory writerFactory =
createWriterFactory(tempDir.toString(), "avro", options);

DataFileTestDataGenerator.Data data = gen.next();
RollingFileWriter<KeyValue, DataFileMeta> dataWriter =
writerFactory.createRollingMergeTreeFileWriter(0, FileSource.APPEND);
dataWriter.write(CloseableIterator.fromList(data.content, kv -> {}));
dataWriter.close();
List<DataFileMeta> dataFileMetas = dataWriter.result();

assertThat(dataFileMetas).isNotEmpty();
assertThat(
dataFileMetas.stream()
.anyMatch(
meta ->
meta.extraFiles().stream()
.anyMatch(
f ->
f.endsWith(
DataFilePathFactory
.INDEX_PATH_SUFFIX))))
.isTrue();

RollingFileWriter<KeyValue, DataFileMeta> changelogWriter =
writerFactory.createRollingChangelogFileWriter(0);
changelogWriter.write(CloseableIterator.fromList(data.content, kv -> {}));
changelogWriter.close();
List<DataFileMeta> changelogMetas = changelogWriter.result();

assertThat(changelogMetas).isNotEmpty();
for (DataFileMeta meta : changelogMetas) {
assertThat(meta.extraFiles())
.noneMatch(f -> f.endsWith(DataFilePathFactory.INDEX_PATH_SUFFIX));
}
}

@ParameterizedTest
@ValueSource(strings = {"parquet", "orc", "avro"})
public void testReaderUseFileSizeFromMetadata(String format) throws Exception {
Expand Down