the-other-tim-brown commented on code in PR #13508:
URL: https://github.com/apache/hudi/pull/13508#discussion_r2205538374
##########
hudi-cli/src/main/java/org/apache/hudi/cli/commands/HoodieLogFileCommand.java:
##########
@@ -222,36 +226,38 @@ storage, new StoragePath(logFilePathPattern)).stream()
}
Objects.requireNonNull(readerSchema);
List<IndexedRecord> allRecords = new ArrayList<>();
-
if (shouldMerge) {
System.out.println("===========================> MERGING RECORDS
<===================");
- HoodieMergedLogRecordScanner scanner =
- HoodieMergedLogRecordScanner.newBuilder()
- .withStorage(storage)
- .withBasePath(client.getBasePath())
- .withLogFilePaths(logFilePaths)
- .withReaderSchema(readerSchema)
- .withLatestInstantTime(
- client.getActiveTimeline()
-
.getCommitAndReplaceTimeline().lastInstant().get().requestedTime())
- .withReverseReader(
- Boolean.parseBoolean(
-
HoodieReaderConfig.COMPACTION_REVERSE_LOG_READ_ENABLE.defaultValue()))
-
.withBufferSize(HoodieMemoryConfig.MAX_DFS_STREAM_BUFFER_SIZE.defaultValue())
- .withMaxMemorySizeInBytes(
-
HoodieMemoryConfig.DEFAULT_MAX_MEMORY_FOR_SPILLABLE_MAP_IN_BYTES)
-
.withSpillableMapBasePath(FileIOUtils.getDefaultSpillableMapBasePath())
-
.withDiskMapType(HoodieCommonConfig.SPILLABLE_DISK_MAP_TYPE.defaultValue())
-
.withBitCaskDiskMapCompressionEnabled(HoodieCommonConfig.DISK_MAP_BITCASK_COMPRESSION_ENABLED.defaultValue())
-
.withOptimizedLogBlocksScan(Boolean.parseBoolean(HoodieReaderConfig.ENABLE_OPTIMIZED_LOG_BLOCKS_SCAN.defaultValue()))
- .build();
- for (HoodieRecord hoodieRecord : scanner) {
- Option<HoodieAvroIndexedRecord> record =
hoodieRecord.toIndexedRecord(readerSchema, new Properties());
- if (allRecords.size() < limit) {
- allRecords.add(record.get().getData());
- }
+ HoodieReaderContext<IndexedRecord> readerContext = new
HoodieAvroReaderContext(
+ HoodieCLI.getTableMetaClient().getStorage().getConf(),
+ HoodieCLI.getTableMetaClient().getTableConfig(),
+ Option.empty(),
+ Option.empty());
+ StoragePath firstLogFile = new StoragePath(logFilePaths.get(0));
+ HoodieFileGroupId fileGroupId = new
HoodieFileGroupId(FSUtils.getRelativePartitionPath(HoodieCLI.getTableMetaClient().getBasePath(),
firstLogFile), FSUtils.getFileIdFromLogPath(firstLogFile));
+ FileSlice fileSlice = new FileSlice(fileGroupId, "000", null,
logFilePaths.stream()
+ .map(l -> new HoodieLogFile(new
StoragePath(l))).collect(Collectors.toList()));
+ try (HoodieFileGroupReader<IndexedRecord> fileGroupReader =
HoodieFileGroupReader.<IndexedRecord>newBuilder()
+ .withReaderContext(readerContext)
+ .withHoodieTableMetaClient(HoodieCLI.getTableMetaClient())
+ .withFileSlice(fileSlice)
+ .withDataSchema(readerSchema)
+ .withRequestedSchema(readerSchema)
+ .withLatestCommitTime("99999999999999")
Review Comment:
That is because the timeline has no commits in the test. You can use:
`client.getActiveTimeline().getCommitAndReplaceTimeline().lastInstant().map(HoodieInstant::requestedTime).orElse("99999999999999")`
or have some constant for all those `9`s.
You may also need `withAllowInflightInstants` to be true if the command is
meant to read any log file
##########
hudi-cli/src/main/java/org/apache/hudi/cli/commands/HoodieLogFileCommand.java:
##########
@@ -222,36 +227,38 @@ storage, new StoragePath(logFilePathPattern)).stream()
}
Objects.requireNonNull(readerSchema);
List<IndexedRecord> allRecords = new ArrayList<>();
-
if (shouldMerge) {
System.out.println("===========================> MERGING RECORDS
<===================");
- HoodieMergedLogRecordScanner scanner =
- HoodieMergedLogRecordScanner.newBuilder()
- .withStorage(storage)
- .withBasePath(client.getBasePath())
- .withLogFilePaths(logFilePaths)
- .withReaderSchema(readerSchema)
- .withLatestInstantTime(
- client.getActiveTimeline()
-
.getCommitAndReplaceTimeline().lastInstant().get().requestedTime())
- .withReverseReader(
- Boolean.parseBoolean(
-
HoodieReaderConfig.COMPACTION_REVERSE_LOG_READ_ENABLE.defaultValue()))
-
.withBufferSize(HoodieMemoryConfig.MAX_DFS_STREAM_BUFFER_SIZE.defaultValue())
- .withMaxMemorySizeInBytes(
-
HoodieMemoryConfig.DEFAULT_MAX_MEMORY_FOR_SPILLABLE_MAP_IN_BYTES)
-
.withSpillableMapBasePath(FileIOUtils.getDefaultSpillableMapBasePath())
-
.withDiskMapType(HoodieCommonConfig.SPILLABLE_DISK_MAP_TYPE.defaultValue())
-
.withBitCaskDiskMapCompressionEnabled(HoodieCommonConfig.DISK_MAP_BITCASK_COMPRESSION_ENABLED.defaultValue())
-
.withOptimizedLogBlocksScan(Boolean.parseBoolean(HoodieReaderConfig.ENABLE_OPTIMIZED_LOG_BLOCKS_SCAN.defaultValue()))
- .build();
- for (HoodieRecord hoodieRecord : scanner) {
- Option<HoodieAvroIndexedRecord> record =
hoodieRecord.toIndexedRecord(readerSchema, new Properties());
- if (allRecords.size() < limit) {
- allRecords.add(record.get().getData());
+ HoodieReaderContext<IndexedRecord> readerContext = new
HoodieAvroReaderContext(
+ HoodieCLI.getTableMetaClient().getStorage().getConf(),
+ HoodieCLI.getTableMetaClient().getTableConfig(),
+ Option.empty(),
+ Option.empty());
+ StoragePath firstLogFile = new StoragePath(logFilePaths.get(0));
+ HoodieFileGroupId fileGroupId = new
HoodieFileGroupId(FSUtils.getRelativePartitionPath(HoodieCLI.getTableMetaClient().getBasePath(),
firstLogFile), FSUtils.getFileIdFromLogPath(firstLogFile));
+ FileSlice fileSlice = new FileSlice(fileGroupId,
HoodieTimeline.INIT_INSTANT_TS, null, logFilePaths.stream()
+ .map(l -> new HoodieLogFile(new
StoragePath(l))).collect(Collectors.toList()));
+ try (HoodieFileGroupReader<IndexedRecord> fileGroupReader =
HoodieFileGroupReader.<IndexedRecord>newBuilder()
+ .withReaderContext(readerContext)
+ .withHoodieTableMetaClient(HoodieCLI.getTableMetaClient())
+ .withFileSlice(fileSlice)
+ .withDataSchema(readerSchema)
+ .withRequestedSchema(readerSchema)
+ .withLatestCommitTime("99999999999999")
+ .withProps(buildFileGroupReaderProperties())
+ .withShouldUseRecordPosition(false)
+ .build()) {
+
+ try (ClosableIterator<IndexedRecord> recordIterator =
fileGroupReader.getClosableIterator()) {
Review Comment:
You can do multi-line try with resources here to avoid the extra indentation
--
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]
