yihua commented on code in PR #11761:
URL: https://github.com/apache/hudi/pull/11761#discussion_r1720350785
##########
hudi-hadoop-common/src/main/java/org/apache/hudi/hadoop/fs/HadoopFSUtils.java:
##########
@@ -171,6 +171,17 @@ public static StoragePathInfo
convertToStoragePathInfo(FileStatus fileStatus) {
fileStatus.getModificationTime());
}
+ public static StoragePathInfo convertToStoragePathInfo(FileStatus
fileStatus, String[] locations) {
Review Comment:
`LocatedFileStatus` (which extends `FileStatus`) stores `BlockLocation[]
locations`. As a follow-up, see if we want to leverage that.
##########
hudi-common/src/main/java/org/apache/hudi/common/engine/HoodieReaderContext.java:
##########
@@ -165,10 +167,50 @@ public void setShouldMergeUseRecordPosition(boolean
shouldMergeUseRecordPosition
* @param storage {@link HoodieStorage} for reading records.
* @return {@link ClosableIterator<T>} that can return all records through
iteration.
*/
- public abstract ClosableIterator<T> getFileRecordIterator(
+ protected abstract ClosableIterator<T> getFileRecordIterator(
StoragePath filePath, long start, long length, Schema dataSchema, Schema
requiredSchema,
HoodieStorage storage) throws IOException;
+ /**
+ * Gets the record iterator based on the type of engine-specific record
representation from the
+ * file.
+ *
+ * @param file {@link StorageFile} instance of a file.
+ * @param start Starting byte to start reading.
+ * @param length Bytes to read.
+ * @param dataSchema Schema of records in the file in {@link Schema}.
+ * @param requiredSchema Schema containing required fields to read in {@link
Schema} for projection.
+ * @param storage {@link HoodieStorage} for reading records.
+ * @return {@link ClosableIterator<T>} that can return all records through
iteration.
+ */
+ public final ClosableIterator<T> getFileRecordIterator(
+ StorageFile file, long start, long length, Schema dataSchema, Schema
requiredSchema,
+ HoodieStorage storage) throws IOException {
+ if (file.getPathInfo() != null) {
+ return getFileRecordIterator(file.getPathInfo(), start, length,
dataSchema, requiredSchema, storage);
+ } else {
+ return getFileRecordIterator(file.getStoragePath(), start, length,
dataSchema, requiredSchema, storage);
+ }
+ }
+
+ /**
+ * Gets the record iterator based on the type of engine-specific record
representation from the
+ * file.
+ *
+ * @param storagePathInfo {@link StoragePathInfo} instance of a file.
+ * @param start Starting byte to start reading.
+ * @param length Bytes to read.
+ * @param dataSchema Schema of records in the file in {@link Schema}.
+ * @param requiredSchema Schema containing required fields to read in
{@link Schema} for projection.
+ * @param storage {@link HoodieStorage} for reading records.
+ * @return {@link ClosableIterator<T>} that can return all records through
iteration.
+ */
+ protected ClosableIterator<T> getFileRecordIterator(
+ StoragePathInfo storagePathInfo, long start, long length, Schema
dataSchema, Schema requiredSchema,
+ HoodieStorage storage) throws IOException {
+ return getFileRecordIterator(storagePathInfo.getPath(), start, length,
dataSchema, requiredSchema, storage);
Review Comment:
Since this is overridden by the `HiveHoodieReaderContext` and has a default
implementation, I think it's OK to keep it now.
##########
hudi-common/src/main/java/org/apache/hudi/common/table/read/HoodieFileGroupReader.java:
##########
@@ -145,10 +146,19 @@ private ClosableIterator<T> makeBaseFileIterator() throws
IOException {
return makeBootstrapBaseFileIterator(baseFile);
}
- return readerContext.getFileRecordIterator(
- baseFile.getStoragePath(), start, length,
- readerContext.getSchemaHandler().getDataSchema(),
- readerContext.getSchemaHandler().getRequiredSchema(), storage);
+ StoragePathInfo baseFileStoragePathInfo = baseFile.getPathInfo();
+ if (baseFileStoragePathInfo != null) {
+ return readerContext.getFileRecordIterator(
+ baseFileStoragePathInfo, start, length,
+ readerContext.getSchemaHandler().getDataSchema(),
+ readerContext.getSchemaHandler().getRequiredSchema(), storage);
+ } else {
+ return readerContext.getFileRecordIterator(
+ baseFile.getStoragePath(), start, length,
+ readerContext.getSchemaHandler().getDataSchema(),
+ readerContext.getSchemaHandler().getRequiredSchema(), storage);
+ }
+
Review Comment:
```suggestion
```
--
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]
