rdblue commented on code in PR #16936:
URL: https://github.com/apache/iceberg/pull/16936#discussion_r4223437484
##########
core/src/test/java/org/apache/iceberg/TestTrackedFileAdapters.java:
##########
@@ -671,6 +733,312 @@ void unknownSpecIdThrows() {
.hasMessageContaining("Cannot find partition spec for spec ID");
}
+ @Test
+ void dataTrackedFileAdapterFromDataFile() {
+ TrackedFile result =
TrackedFileAdapters.forDataFile(TABLE_SCHEMA).wrap(DATA_FILE);
+
+ assertThat(result.tracking()).isNull();
+ assertWrappedDataFileMatchesFileFields(result, DATA_FILE);
+ assertThatThrownBy(result::formatVersion)
+ .isInstanceOf(IllegalStateException.class)
+ .hasMessage("Format version is assigned at write time");
+ }
+
+ @Test
+ void dataTrackedFileAdapterFromExistingManifestEntry() {
+ TrackedFile result =
+ TrackedFileAdapters.forDataFile(TABLE_SCHEMA)
+ .wrap(
+ newEntry()
+ .wrapExisting(
+ SNAPSHOT_ID, DATA_SEQUENCE_NUMBER,
FILE_SEQUENCE_NUMBER, DATA_FILE));
+
+ assertThat(result.tracking().status()).isEqualTo(EntryStatus.EXISTING);
+ assertThat(result.tracking().snapshotId()).isEqualTo(SNAPSHOT_ID);
+
assertThat(result.tracking().dataSequenceNumber()).isEqualTo(DATA_SEQUENCE_NUMBER);
+
assertThat(result.tracking().fileSequenceNumber()).isEqualTo(FILE_SEQUENCE_NUMBER);
+ assertThat(result.tracking().firstRowId()).isEqualTo(FIRST_ROW_ID);
+ assertManifestPosition(result.tracking(), DATA_FILE);
+ assertWrappedDataFileMatchesFileFields(result, DATA_FILE);
+ }
+
+ @Test
+ void dataTrackedFileAdapterReuse() {
+ TrackedFileAdapters.DataTrackedFile adapter =
TrackedFileAdapters.forDataFile(TABLE_SCHEMA);
+
+ adapter.wrap(DATA_FILE);
+ assertWrappedDataFileMatchesFileFields(adapter, DATA_FILE);
+ assertThat(adapter.tracking()).isNull();
+
+ DataFile file2 =
+ new GenericDataFile(
+ UNPARTITIONED_SPEC.specId(),
+ "s3://bucket/data/file2.parquet",
+ FileFormat.PARQUET,
+ PartitionData.EMPTY,
+ 2048L,
+ new Metrics(200L, null, null, null, null),
+ null,
+ ImmutableList.of(0L),
+ null,
+ null);
+ assignManifestPosition(file2, "s3://bucket/table/manifest-2.parquet", 8L);
+ adapter.wrap(
+ newEntry().wrapExisting(SNAPSHOT_ID, DATA_SEQUENCE_NUMBER,
FILE_SEQUENCE_NUMBER, file2));
+ assertWrappedDataFileMatchesFileFields(adapter, file2);
+ assertThat(adapter.tracking().status()).isEqualTo(EntryStatus.EXISTING);
+ assertManifestPosition(adapter.tracking(), file2);
+ }
+
+ @Test
+ void dataTrackedFileAdapterRejectsNullFile() {
+ TrackedFileAdapters.DataTrackedFile adapter =
TrackedFileAdapters.forDataFile(TABLE_SCHEMA);
+ assertThatThrownBy(() -> adapter.wrap((DataFile) null))
+ .isInstanceOf(IllegalArgumentException.class)
+ .hasMessageContaining("Invalid file: null");
+ }
+
+ @Test
+ void dataTrackedFileAdapterContentStats() {
+ TrackedFile result =
TrackedFileAdapters.forDataFile(TABLE_SCHEMA).wrap(DATA_FILE_WITH_METRICS);
+
+ ContentStats stats = result.contentStats();
+ assertThat(stats).isNotNull();
+
assertThat(stats.fieldStats()).extracting(FieldStats::fieldId).containsExactlyInAnyOrder(1,
2);
+
+ FieldStats<?> idStats = stats.statsFor(1);
+ assertThat(idStats.valueCount()).isEqualTo(100L);
+ assertThat(idStats.lowerBound()).isEqualTo(1);
+ assertThat(idStats.upperBound()).isEqualTo(1000);
+ }
+
+ @Test
+ void dataTrackedFileAdapterWithoutMetricsHasNoContentStats() {
+ TrackedFile result =
TrackedFileAdapters.forDataFile(TABLE_SCHEMA).wrap(DATA_FILE);
+
+ assertThat(result.contentStats()).isNull();
+ }
+
+ @Test
+ void dataTrackedFileAdapterKeepsPartitionTuple() {
+ DataFile partitioned =
+ new GenericDataFile(
+ PARTITIONED_SPEC.specId(),
+ DATA_FILE_LOCATION,
+ FileFormat.PARQUET,
+ PARTITION,
+ 1024L,
+ new Metrics(100L, null, null, null, null),
+ null,
+ ImmutableList.of(0L),
+ null,
+ null);
+ TrackedFile result =
TrackedFileAdapters.forDataFile(TABLE_SCHEMA).wrap(partitioned);
+
+ assertThat(result.partition())
+ .usingComparator(Comparators.forType(PARTITIONED_SPEC.partitionType()))
+ .isEqualTo(PARTITION);
+ assertThat(TrackedFileAdapters.asDataFile(result,
specsById(PARTITIONED_SPEC)).partition())
+ .usingComparator(Comparators.forType(PARTITIONED_SPEC.partitionType()))
+ .isEqualTo(PARTITION);
+ }
+
+ @Test
+ void dataTrackedFileAdapterUnwrapsToOriginalTrackedFile() {
+ TrackedFile source = trackedFile(FileContent.DATA);
+ DataFile dataFile = TrackedFileAdapters.asDataFile(source, UNPARTITIONED);
+ TrackedFile roundTripped =
TrackedFileAdapters.forDataFile(TABLE_SCHEMA).wrap(dataFile);
+ assertThat(roundTripped).isSameAs(source);
+ }
+
+ @Test
+ void manifestTrackedFileAdapterUnwrapsToOriginalTrackedFile() {
+ TrackedFile original = trackedFile(FileContent.DATA_MANIFEST, 0);
+ ManifestFile adapted = TrackedFileAdapters.asManifestFile(original);
+ TrackedFile result = TrackedFileAdapters.forManifestFile().wrap(adapted);
+ assertThat(result).isSameAs(original);
+ assertThat(result.formatVersion()).isZero();
+ }
+
+ @Test
+ void dataManifestTrackedFileAdapter() {
+ ManifestFile manifest = writeManifestFile(ManifestContent.DATA);
+ TrackedFile result = TrackedFileAdapters.forManifestFile().wrap(manifest);
+
+ assertThat(result.contentType()).isEqualTo(FileContent.DATA_MANIFEST);
+ assertThat(result.formatVersion()).isZero();
+ assertThat(result.location()).isEqualTo(MANIFEST_LOCATION);
+ assertThat(result.fileFormat()).isEqualTo(FileFormat.AVRO);
+ assertThat(result.tracking().status()).isEqualTo(EntryStatus.EXISTING);
+ assertThat(result.tracking().snapshotId()).isEqualTo(SNAPSHOT_ID);
+ assertThat(result.tracking().firstRowId()).isNull();
+ assertThat(result.recordCount())
+ .isEqualTo(ADDED_FILES_COUNT + EXISTING_FILES_COUNT +
DELETED_FILES_COUNT);
+ assertThat(result.manifestInfo()).isNotNull();
+
assertThat(result.manifestInfo().addedFilesCount()).isEqualTo(ADDED_FILES_COUNT);
+
assertThat(result.manifestInfo().existingFilesCount()).isEqualTo(EXISTING_FILES_COUNT);
+
assertThat(result.manifestInfo().deletedFilesCount()).isEqualTo(DELETED_FILES_COUNT);
+
assertThat(result.manifestInfo().addedRowsCount()).isEqualTo(ADDED_ROWS_COUNT);
+ assertThat(result.manifestInfo().replacedFilesCount()).isEqualTo(0);
+ assertThat(result.manifestInfo().replacedRowsCount()).isEqualTo(0L);
+ assertThat(result.manifestInfo().modifiedFilesCount()).isEqualTo(0);
+ assertThat(result.manifestInfo().modifiedRowsCount()).isEqualTo(0L);
+ }
+
+ @Test
+ void deleteManifestTrackedFileAdapter() {
+ ManifestFile manifest = writeManifestFile(ManifestContent.DELETES);
+ TrackedFile result = TrackedFileAdapters.forManifestFile().wrap(manifest);
+
+ assertThat(result.contentType()).isEqualTo(FileContent.DELETE_MANIFEST);
+ assertThat(result.formatVersion()).isZero();
+ assertThat(result.recordCount())
+ .isEqualTo(ADDED_FILES_COUNT + EXISTING_FILES_COUNT +
DELETED_FILES_COUNT);
+ assertThat(result.tracking().status()).isEqualTo(EntryStatus.EXISTING);
+ assertThat(result.tracking().firstRowId()).isNull();
+ }
+
+ @Test
+ void manifestTrackedFileAdapterFailsWhenAddedFilesCountMissing() {
+ ManifestFile manifest = mock(ManifestFile.class);
+ when(manifest.path()).thenReturn(MANIFEST_LOCATION);
+ when(manifest.content()).thenReturn(ManifestContent.DATA);
+ when(manifest.addedFilesCount()).thenReturn(null);
+ when(manifest.existingFilesCount()).thenReturn(1);
+ when(manifest.deletedFilesCount()).thenReturn(0);
+ when(manifest.replacedFilesCount()).thenReturn(0);
+
+ assertThatThrownBy(() ->
TrackedFileAdapters.forManifestFile().wrap(manifest))
+ .isInstanceOf(NullPointerException.class)
+ .hasMessageContaining("missing added files count");
+ }
+
+ @Test
+ void manifestTrackedFileAdapterRejectsNullReplacedFilesCount() {
+ ManifestFile manifest = manifestWithCounts(1, 1, 0, null, 0);
+
+ assertThatThrownBy(() ->
TrackedFileAdapters.forManifestFile().wrap(manifest))
+ .isInstanceOf(IllegalArgumentException.class)
+ .hasMessage(
+ "Cannot convert manifest %s: Invalid replaced file count: null",
MANIFEST_LOCATION);
+ }
+
+ @Test
+ void manifestTrackedFileAdapterRejectsNonZeroReplacedFilesCount() {
+ ManifestFile manifest = manifestWithCounts(1, 1, 0, 1, 0);
+
+ assertThatThrownBy(() ->
TrackedFileAdapters.forManifestFile().wrap(manifest))
+ .isInstanceOf(IllegalArgumentException.class)
+ .hasMessage(
+ "Cannot convert manifest %s: Invalid replaced file count: 1",
MANIFEST_LOCATION);
+ }
+
+ @Test
+ void manifestTrackedFileAdapterRejectsNullModifiedFilesCount() {
+ ManifestFile manifest = manifestWithCounts(1, 1, 0, 0, null);
+
+ assertThatThrownBy(() ->
TrackedFileAdapters.forManifestFile().wrap(manifest))
+ .isInstanceOf(IllegalArgumentException.class)
+ .hasMessage(
+ "Cannot convert manifest %s: Invalid modified file count: null",
MANIFEST_LOCATION);
+ }
+
+ @Test
+ void manifestTrackedFileAdapterRejectsNonZeroModifiedFilesCount() {
+ ManifestFile manifest = manifestWithCounts(1, 1, 0, 0, 1);
+
+ assertThatThrownBy(() ->
TrackedFileAdapters.forManifestFile().wrap(manifest))
+ .isInstanceOf(IllegalArgumentException.class)
+ .hasMessage(
+ "Cannot convert manifest %s: Invalid modified file count: 1",
MANIFEST_LOCATION);
+ }
+
+ @Test
+ void manifestTrackedFileAdapterFailsWhenAddedRowsCountMissing() {
+ ManifestFile manifest =
+ writeManifestFile(
+ ManifestContent.DATA, MANIFEST_SEQUENCE_NUMBER,
MANIFEST_MIN_SEQUENCE_NUMBER, null);
+ TrackedFile tracked = TrackedFileAdapters.forManifestFile().wrap(manifest);
+
+ assertThatThrownBy(() -> tracked.manifestInfo().addedRowsCount())
+ .isInstanceOf(NullPointerException.class)
+ .hasMessageContaining("null");
+ }
+
+ private static void assertWrappedDataFileMatchesFileFields(TrackedFile
result, DataFile file) {
+ assertThat(result.contentType()).isEqualTo(FileContent.DATA);
+ assertThat(result.location()).isEqualTo(file.location());
+ assertThat(result.fileFormat()).isEqualTo(file.format());
+ assertThat(result.recordCount()).isEqualTo(file.recordCount());
+ assertThat(result.fileSizeInBytes()).isEqualTo(file.fileSizeInBytes());
+ assertThat(result.specId()).isEqualTo(file.specId());
+ assertThat(result.sortOrderId()).isEqualTo(file.sortOrderId());
+ assertThat(result.keyMetadata()).isEqualTo(file.keyMetadata());
+ assertThat(result.splitOffsets()).isEqualTo(file.splitOffsets());
+ assertThat(result.manifestInfo()).isNull();
+ assertThat(result.deletionVector()).isNull();
+ assertThat(result.equalityIds()).isNull();
+ }
+
+ private static ManifestFile manifestWithCounts(
+ Integer addedFilesCount,
+ Integer existingFilesCount,
+ Integer deletedFilesCount,
+ Integer replacedFilesCount,
+ Integer modifiedFilesCount) {
+ ManifestFile manifest = mock(ManifestFile.class);
+ when(manifest.path()).thenReturn(MANIFEST_LOCATION);
+ when(manifest.content()).thenReturn(ManifestContent.DATA);
+ when(manifest.addedFilesCount()).thenReturn(addedFilesCount);
+ when(manifest.existingFilesCount()).thenReturn(existingFilesCount);
+ when(manifest.deletedFilesCount()).thenReturn(deletedFilesCount);
+ when(manifest.replacedFilesCount()).thenReturn(replacedFilesCount);
+ when(manifest.modifiedFilesCount()).thenReturn(modifiedFilesCount);
+ return manifest;
+ }
+
+ private static ManifestFile writeManifestFile(ManifestContent content) {
+ return writeManifestFile(
+ content, MANIFEST_SEQUENCE_NUMBER, MANIFEST_MIN_SEQUENCE_NUMBER,
ADDED_ROWS_COUNT);
+ }
+
+ private static ManifestFile writeManifestFile(
+ ManifestContent content, long sequenceNumber, long minSequenceNumber,
Long addedRowsCount) {
+ List<ManifestFile.PartitionFieldSummary> partitions = ImmutableList.of();
+ return new GenericManifestFile(
+ MANIFEST_LOCATION,
+ MANIFEST_FILE_SIZE,
+ UNPARTITIONED_SPEC.specId(),
+ content,
+ sequenceNumber,
+ minSequenceNumber,
+ SNAPSHOT_ID,
+ partitions,
+ null,
Review Comment:
Nit: always nice to identify what these null args represent:
```suggestion
null, // sort order id
```
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]