[minor] add more test for rfc46 by wzx140 · Pull Request #7003 · apache/hudi

wzx140 · 2022-10-19T14:28:36Z

Change Logs

Add HoodieSparkValidateDuplicateKeyRecordMerger behaving the same as ValidateDuplicateKeyPayload. We should use it with config "hoodie.sql.insert.mode=strict".
Fix nest field exist in HoodieCatalystExpressionUtils
Fix rewrite in HoodieInternalRowUtiles to support type promoted as avro
Fallback to avro when use "merge into" sql
Fix some schema handling issue
Support delta streamer
Convert parquet schema to spark schema and then avro schema(in org.apache.hudi.io.storage.HoodieSparkParquetReader#getSchema). Some types in avro are not compatible with parquet. For ex, decimal as int32/int64 in parquet will convert to int/long in avro. Because avro do not has decimal as int/long . We will lose the logic type info if we directly convert it to avro schema.
Support schema evolution in parquet block

Impact

Describe any public API or user-facing feature change or any performance impact.

Risk level (write none, low medium or high below)

If medium or high, explain what verification was done to mitigate the risks.

Documentation Update

Describe any necessary documentation update if there is any new feature, config, or user-facing change

The config description must be updated if new configs are added or the default value of the configs are changed
Any new feature or user-facing change requires updating the Hudi website. Please create a Jira ticket, attach the
ticket number here and follow the instruction to make
changes to the website.

Contributor's checklist

Read through contributor's guide
Change Logs and Impact were stated clearly
Adequate tests were added if applicable
CI passed

alexeykudinkin · 2022-10-19T19:32:14Z

LGTM!

Let's add cover for DeltaStreamer tests as well (w/in TestHoodieDeltaStreamer, we don't need to parameterize all of them but let's pick representative set)

wzx140 · 2022-11-16T15:04:45Z

@hudi-bot run azure

alexeykudinkin · 2022-11-17T06:05:06Z

+    }
+  }
+
+  protected def withRecordType(f: => Unit, recordConfig: Map[HoodieRecordType, Map[String, String]]=Map.empty) {


Why not structuring it the same way as witSqlConf? Like below:

def withRecordType(recordConfig)(f: => Unit)

In most test case, we do not need to pass different sql configs for recordType independently. So I used the default parameters in function .

In "Test Insert Into None Partitioned Table", SparkMerger should use "HoodieSparkValidateDuplicateKeyRecordMerger" with "hoodie.sql.insert.mode=strict".

We should not replace SparkRecordMerger with HoodieSparkValidateDuplicateKeyRecordMerger when user set "hoodie.sql.insert.mode=strict". Because sql cofig has the higher priority.

alexeykudinkin · 2022-11-17T06:06:25Z

+  }
+
+  protected def withRecordType(f: => Unit, recordConfig: Map[HoodieRecordType, Map[String, String]]=Map.empty) {
+    Seq(HoodieRecordType.SPARK).foreach { recordType =>


Why are we testing just the Spark?

alexeykudinkin · 2022-11-17T06:09:19Z

+    HoodieStorageConfig.LOGFILE_DATA_BLOCK_FORMAT.key -> "parquet"
+  )
+
+  def getOpts(recordType: HoodieRecordType, opt: Map[String, String]): (Map[String, String], Map[String, String]) = {


nit: getWriterReaderOpts

alexeykudinkin · 2022-11-17T06:10:06Z

-      commonOpts ++ sparkOpts
-    } else {
-      commonOpts
+  def getOpts(recordType: HoodieRecordType, opt: Map[String, String] = commonOpts): (Map[String, String], Map[String, String]) = {


Same as below, please ditto everywhere

alexeykudinkin · 2022-11-17T06:12:53Z

    // Create the write parameters
    val parameters = buildMergeIntoConfig(hoodieCatalogTable)
+    // TODO Remove it when we implement ExpressionPayload for SparkRecord
+    val parametersWithAvroRecordMerger = parameters ++ Map(HoodieWriteConfig.MERGER_IMPLS.key -> classOf[HoodieAvroRecordMerger].getName)


Let's just append to parameters (no need for new var)

parameters is immutable map

alexeykudinkin · 2022-11-17T06:48:23Z

-    newRecord.update(HoodieMetadataField.FILENAME_METADATA_FIELD.ordinal, CatalystTypeConverters.convertToCatalyst(fileName))
+  def compareSchema(left: StructType, right: StructType): Boolean = {
+    val schemaPair = (left, right)
+    if (!schemaCompareMap.contains(schemaPair)) {


Left this comment before, but this still seems to be bubbling up: this doesn't make sense -- keys in the map will always be checked for equality, which defeats the purpose of this mapping

Ok. rewriteRecord and rewriteRecordWithNewSchema is fixed

alexeykudinkin · 2022-11-17T06:51:49Z

+              case _: FloatType => newRow.update(pos, oldValue.asInstanceOf[Float].toDouble)
+              case _ => throw new IllegalArgumentException(s"$oldSchema and $newSchema are incompatible")
+            }
+          case _: BinaryType if oldType.isInstanceOf[StringType] => newRow.update(pos, oldValue.asInstanceOf[Array[Byte]].map(_.toChar).mkString)


These 2 seems to be inverted

alexeykudinkin · 2022-11-17T06:57:18Z

+  @tailrec
  def existField(structType: StructType, name: String): Boolean = {
-    structType.getFieldIndex(name).isDefined
+    if (name.contains(".")) {


Let's move this method to HoodieInternalRowUtils, it has nothing to do w/ expressions

Would be better to rewrite it as a single loop, instead of tail-rec (to avoid calling split N times)

Ok. I just want to call structType.getFieldIndex. So I put this func in HoodieCatalystExpressionUtils. I will move HoodieInternalRowUtils to spark.sql package.

alexeykudinkin · 2022-11-17T07:01:18Z

  private final boolean enablePointLookups;

-  protected final Schema readerSchema;
+  protected Schema readerSchema;


We should not be relaxing mutability constraints unless absolutely have to

@wzx140 we'd need to revert these changes after we rebase on the latest master (these issues should have already been addressed on master)

alexeykudinkin · 2022-11-19T01:48:27Z

-    InternalRow finalRow = new HoodieInternalRow(metaFields, data, containMetaFields);
+    // Rewrite if schema is not same
+    InternalRow finalRow = this.data;
+    StructType rawTargetSchema = HoodieInternalRowUtils.removeMetaField(targetStructType);


Why do we need to remove meta fields?

alexeykudinkin · 2022-11-19T01:51:08Z

+    StructType rawTargetSchema = HoodieInternalRowUtils.removeMetaField(targetStructType);
+    if (!HoodieInternalRowUtils.compareSchema(structType, rawTargetSchema)) {
+      InternalRow rewriteRecord = HoodieInternalRowUtils.rewriteRecord(this.data, structType, rawTargetSchema);
+      this.data = null;


We should not modify existing record

alexeykudinkin · 2022-11-19T01:53:38Z

-    HoodieInternalRow finalRow = new HoodieInternalRow(metaFields, rewrittenRow, containMetaFields);
+    // Rewrite if schema is not same
+    InternalRow finalRow = this.data;
+    StructType rawNewSchema = HoodieInternalRowUtils.removeMetaField(newStructType);


Same comments as above

alexeykudinkin · 2022-11-19T01:54:11Z

@@ -56,6 +63,8 @@ public class HoodieSparkParquetReader implements HoodieSparkFileReader {
  public HoodieSparkParquetReader(Configuration conf, Path path) {
    this.path = path;
    this.conf = conf;


We need to make sure we make the copy of the config first, before modifying it

alexeykudinkin · 2022-11-19T02:24:43Z


-            new HoodieSparkRecord(key, processedRow, structType, false)
+            val processedRow = if (reconcileSchema) {
+              HoodieInternalRowUtils.getCachedUnsafeProjection(processedStructType, processedStructType)


We'd need to reconcile this with #6358

alexeykudinkin · 2022-11-19T02:26:15Z

+ * Validate the duplicate key for insert statement without enable the INSERT_DROP_DUPS_OPT
+ * config.
+ */
+class HoodieSparkValidateDuplicateKeyRecordMerger extends HoodieSparkRecordMerger {


I see now. Please update the java-doc to reflect that

alexeykudinkin · 2022-11-19T02:33:30Z

+        return new HoodieAvroRecord<>(keyGenerator.getKey(record), payload);
+      });
+    } else if (recordType == HoodieRecordType.SPARK) {
+      // TODO we should remove it if we can read InternalRow from source.


We actually already have Jira for it, let me find it

alexeykudinkin · 2022-11-19T02:34:07Z

+    SerializableSchema avroSchema = new SerializableSchema(schemaProvider.getTargetSchema());
+    SerializableSchema processedAvroSchema = new SerializableSchema(isDropPartitionColumns() ? HoodieAvroUtils.removeMetadataFields(avroSchema.get()) : avroSchema.get());
+    if (recordType == HoodieRecordType.AVRO) {
+      records = avroRDD.map(record -> {


Code seems unchanged, but want to double-check with you: this section didn't change, right?

Yes. We do not change the avro section.

alexeykudinkin

@wzx140 LGTM! Looking for some minor clarifications

alexeykudinkin · 2022-11-21T21:11:32Z

  }

  test("Test Insert Exception") {
-    val tableName = generateTableName


Why do these tests change?

We just add withRecordType() { here

@wzx140 we didn't modify the test itself, did we?

GH UI shows it as if the whole test had been rewritten which is very hard to decipher

@alexeykudinkin We just add withRecordType() and put sql config in withSQLConf. Maybe We have also changed the indentation. This make GH UI shows that the whole test had been rewritten.

alexeykudinkin · 2022-11-21T21:24:56Z

    UTF8String[] metaFields = tryExtractMetaFields(data, structType);
+    InternalRow rewriteRecord = HoodieInternalRowUtils.rewriteRecord(this.data, structType, targetStructType);
+    UnsafeRow unsafeRow = HoodieInternalRowUtils.getCachedUnsafeProjection(targetStructType, targetStructType).apply(rewriteRecord);
+    InternalRow finalRow = copy ? unsafeRow.copy() : unsafeRow;


I appreciate the intent, but we should keep all copying decisions to be committed by the user of HoodieSparkRecord -- we should only do the copying, where's absolutely necessary. Let's move this copying to a place where it's actually required.

I think operation on record should not change the copy state. When the record is copied, it means the record need to be copied. So we should do copy on the rewrited record too. Does this make sence?

alexeykudinkin · 2022-11-21T21:25:08Z

    UTF8String[] metaFields = tryExtractMetaFields(data, structType);
+    InternalRow rewriteRecord = HoodieInternalRowUtils.rewriteRecordWithNewSchema(this.data, structType, newStructType, renameCols);
+    UnsafeRow unsafeRow = HoodieInternalRowUtils.getCachedUnsafeProjection(newStructType, newStructType).apply(rewriteRecord);
+    InternalRow finalRow = copy ? unsafeRow.copy() : unsafeRow;


Same comment as above

alexeykudinkin · 2022-11-21T21:42:44Z

  private final boolean enablePointLookups;

-  protected final Schema readerSchema;
+  protected Schema readerSchema;


alexeykudinkin · 2022-11-28T21:39:30Z

@wzx140 LGTM. Let's resolve copying discussion and we should be good to go.

alexeykudinkin · 2022-11-28T23:44:38Z

@wzx140 let's disable TestCleaner in this PR and re-run the CI to make sure no other tests are failing

wzx140 · 2022-11-29T09:41:26Z

@hudi-bot run azure

wzx140 · 2022-11-29T16:24:16Z

@alexeykudinkin "UT FT other modules" was cancel because of timeout. We add some UTs in delta streamer. https://dev.azure.com/apache-hudi-ci-org/apache-hudi-ci/_build/results?buildId=13314&view=results

This test was run before "fix copy". It seems that there is no failure in "UT FT other modules". https://dev.azure.com/apache-hudi-ci-org/apache-hudi-ci/_build/results?buildId=13283&view=results

wzx140 · 2022-11-29T16:24:25Z

@hudi-bot run azure

alexeykudinkin · 2022-11-30T01:25:38Z

-        HoodieRecord record = schemaOption.isPresent() ? currentRecord.rewriteRecordWithNewSchema(dataBlock.getSchema(), new Properties(), schemaOption.get()) : currentRecord;
+        HoodieRecord record = schemaOption.isPresent()
+            ? currentRecord
+            .rewriteRecordWithNewSchema(dataBlock.getSchema(), new Properties(), schemaOption.get())


Please add a comment elaborating why copy is necessary (please ditto everywhere, where we make a copy)

alexeykudinkin · 2022-11-30T01:26:34Z

  private final boolean enablePointLookups;

-  protected final Schema readerSchema;
+  protected Schema readerSchema;


@wzx140 we'd need to revert these changes after we rebase on the latest master (these issues should have already been addressed on master)

alexeykudinkin · 2022-11-30T01:26:47Z

        blockContentLoc.getBlockSize());

+    Schema writerSchema = new Schema.Parser().parse(this.getLogBlockHeader().get(HeaderMetadataType.SCHEMA));
+    if (!internalSchema.isEmptySchema()) {


Same as https://github.com/apache/hudi/pull/7003/files#r1024817901

alexeykudinkin · 2022-11-30T01:32:09Z

-        HoodieRecord recordCopy = record.copy();
        if (!externalSchemaTransformation) {
-          return recordCopy;
+          return record.copy();


Let's add a comment elaborating why we're making a copy here

alexeykudinkin · 2022-11-30T01:32:48Z

+          String recKey = record.getRecordKey(reader.getSchema(), Option.of(keyGenerator));
+          HoodieRecord hoodieRecord = record
+              .rewriteRecord(reader.getSchema(), config.getProps(), HoodieAvroUtils.RECORD_KEY_SCHEMA)
+              .copy();


Same comment as above

hudi-bot · 2022-11-30T03:08:18Z

CI report:

4a74177 Azure: FAILURE Azure: FAILURE Azure: CANCELED
6835303 Azure: PENDING

Bot commands

@hudi-bot supports the following commands:

@hudi-bot run azure re-run the last Azure build

wzx140 · 2022-11-30T06:08:42Z

ci pass https://dev.azure.com/apache-hudi-ci-org/apache-hudi-ci/_build/results?buildId=13332&view=results

[minor] add more test for rfc46 (apache#7003) ## Change Logs - Add HoodieSparkValidateDuplicateKeyRecordMerger behaving the same as ValidateDuplicateKeyPayload. We should use it with config "hoodie.sql.insert.mode=strict". - Fix nest field exist in HoodieCatalystExpressionUtils - Fix rewrite in HoodieInternalRowUtiles to support type promoted as avro - Fallback to avro when use "merge into" sql - Fix some schema handling issue - Support delta streamer - Convert parquet schema to spark schema and then avro schema(in org.apache.hudi.io.storage.HoodieSparkParquetReader#getSchema). Some types in avro are not compatible with parquet. For ex, decimal as int32/int64 in parquet will convert to int/long in avro. Because avro do not has decimal as int/long . We will lose the logic type info if we directly convert it to avro schema. - Support schema evolution in parquet block [Minor] fix multi deser avro payload (apache#7021) In HoodieAvroRecord, we will call isDelete, shouldIgnore before we write it to the file. Each method will deserialize HoodiePayload. So we add deserialization method in HoodieRecord and call this method once before calling isDelete or shouldIgnore. Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: Alexey Kudinkin <alexey.kudinkin@gmail.com> [MINOR] Properly registering target classes w/ Kryo (apache#7026) * Added `HoodieKryoRegistrar` registering necessary Hudi's classes w/ Kryo to make their serialization more efficient (by serializing just the class id, in-liue the fully qualified class-name) * Redirected Kryo registration to `HoodieKryoRegistrar` * Registered additional classes likely to be serialized by Kryo * Updated tests * Fixed serialization of Avro's `Utf8` to serialize just the bytes * Added tests * Added custom `AvroUtf8Serializer`; Tidying up * Extracted `HoodieCommonKryoRegistrar` to leverage in `SerializationUtils` * `HoodieKryoRegistrar` > `HoodieSparkKryoRegistrar`; Rebased `HoodieSparkKryoRegistrar` onto `HoodieCommonKryoRegistrar` * `lint` * Fixing compilation for Spark 2.x * Disabling flaky test [MINOR] Make sure all `HoodieRecord`s are appropriately serializable by Kryo (apache#6977) * Make sure `HoodieRecord`, `HoodieKey`, `HoodieRecordLocation` are all `KryoSerializable` * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; Implemented serialization hooks for `HoodieAvroRecord`; * Revisited `HoodieSparkRecord` to transiently hold on to the schema so that it could project row * Implemented serialization hooks for `HoodieSparkRecord` * Added `TestHoodieSparkRecord` * Added tests for Avro-based records * Added test for `HoodieEmptyRecord` * Fixed sealing/unsealing for `HoodieRecord` in `HoodieBackedTableMetadataWriter` * Properly handle deflated records * Fixing `Row`s encoding * Fixed `HoodieRecord` to be properly sealed/unsealed * Fixed serialization of the `HoodieRecordGlobalLocation` [MINOR] Additional fixes for apache#6745 (apache#6947) * Tidying up * Tidying up more * Cleaning up duplication * Tidying up * Revisited legacy operating mode configuration * Tidying up * Cleaned up `projectUnsafe` API * Fixing compilation * Cleaning up `HoodieSparkRecord` ctors; Revisited mandatory unsafe-projection * Fixing compilation * Cleaned up `ParquetReader` initialization * Revisited `HoodieSparkRecord` to accept either `UnsafeRow` or `HoodieInternalRow`, and avoid unnecessary copying after unsafe-projection * Cleaning up redundant exception spec * Make sure `updateMetadataFields` properly wraps `InternalRow` into `HoodieInternalRow` if necessary; Cleaned up `MetadataValues` * Fixed meta-fields extraction and `HoodieInternalRow` composition w/in `HoodieSparkRecord` * De-duplicate `HoodieSparkRecord` ctors; Make sure either only `UnsafeRow` or `HoodieInternalRow` are permitted inside `HoodieSparkRecord` * Removed unnecessary copying * Cleaned up projection for `HoodieSparkRecord` (dropping partition columns); Removed unnecessary copying * Fixing compilation * Fixing compilation (for Flink) * Cleaned up File Raders' interfaces: - Extracted `HoodieSeekingFileReader` interface (for key-ranged reads) - Pushed down concrete implementation methods into `HoodieAvroFileReaderBase` from the interfaces * Cleaned up File Readers impls (inline with then new interfaces) * Rebsaed `HoodieBackedTableMetadata` onto new `HoodieSeekingFileReader` * Tidying up * Missing licenses * Re-instate custom override for `HoodieAvroParquetReader`; Tidying up * Fixed missing cloning w/in `HoodieLazyInsertIterable` * Fixed missing cloning in deduplication flow * Allow `HoodieSparkRecord` to hold `ColumnarBatchRow` * Missing licenses * Fixing compilation * Missing changes * Fixed Spark 2.x validation whether the row was read as a batch Fix comment in RFC46 (apache#6745) * rename * add MetadataValues in updateMetadataValues * remove singleton in fileFactory * add truncateRecordKey * remove hoodieRecord#setData * rename HoodieAvroRecord * fix code style * fix HoodieSparkRecordSerializer * fix benchmark * fix SparkRecordUtils * instantiate HoodieWriteConfig on the fly * add test * fix HoodieSparkRecordSerializer. Replace Java's object serialization with kryo * add broadcast * fix comment * remove unnecessary broadcast * add unsafe check in spark record * fix getRecordColumnValues * remove spark.sql.parquet.writeLegacyFormat * fix unsafe projection * fix * pass external schema * update doc * rename back to HoodieAvroRecord * fix * remove comparable wrapper * fix comment * fix comment * fix comment * fix comment * simplify row copy * fix ParquetReaderIterator Co-authored-by: Shawy Geng <gengxiaoyu1996@gmail.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [RFC-46][HUDI-4414] Update the RFC-46 doc to fix comments feedback (apache#6132) * Update the RFC-46 doc to fix comments feedback * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer.(apache#5629) * [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer. * add schema finger print * add benchmark * a new way to config the merger * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: gengxiaoyu <gengxiaoyu@bytedance.com> [HUDI-3350][HUDI-3351] Support HoodieMerge API and Spark engine-specific HoodieRecord (apache#5627) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4344] fix usage of HoodieDataBlock#getRecordIterator (apache#6005) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4292][RFC-46] Update doc to align with the Record Merge API changes (apache#5927) [MINOR] Fix type casting in TestHoodieHFileReaderWriter [HUDI-3378][HUDI-3379][HUDI-3381] Migrate usage of HoodieRecordPayload and raw Avro payload to HoodieRecord (apache#5522) Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com>

[minor] add more test for rfc46 (apache#7003) - Add HoodieSparkValidateDuplicateKeyRecordMerger behaving the same as ValidateDuplicateKeyPayload. We should use it with config "hoodie.sql.insert.mode=strict". - Fix nest field exist in HoodieCatalystExpressionUtils - Fix rewrite in HoodieInternalRowUtiles to support type promoted as avro - Fallback to avro when use "merge into" sql - Fix some schema handling issue - Support delta streamer - Convert parquet schema to spark schema and then avro schema(in org.apache.hudi.io.storage.HoodieSparkParquetReader#getSchema). Some types in avro are not compatible with parquet. For ex, decimal as int32/int64 in parquet will convert to int/long in avro. Because avro do not has decimal as int/long . We will lose the logic type info if we directly convert it to avro schema. - Support schema evolution in parquet block [Minor] fix multi deser avro payload (apache#7021) In HoodieAvroRecord, we will call isDelete, shouldIgnore before we write it to the file. Each method will deserialize HoodiePayload. So we add deserialization method in HoodieRecord and call this method once before calling isDelete or shouldIgnore. Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: Alexey Kudinkin <alexey.kudinkin@gmail.com> [MINOR] Properly registering target classes w/ Kryo (apache#7026) * Added `HoodieKryoRegistrar` registering necessary Hudi's classes w/ Kryo to make their serialization more efficient (by serializing just the class id, in-liue the fully qualified class-name) * Redirected Kryo registration to `HoodieKryoRegistrar` * Registered additional classes likely to be serialized by Kryo * Updated tests * Fixed serialization of Avro's `Utf8` to serialize just the bytes * Added tests * Added custom `AvroUtf8Serializer`; Tidying up * Extracted `HoodieCommonKryoRegistrar` to leverage in `SerializationUtils` * `HoodieKryoRegistrar` > `HoodieSparkKryoRegistrar`; Rebased `HoodieSparkKryoRegistrar` onto `HoodieCommonKryoRegistrar` * `lint` * Fixing compilation for Spark 2.x * Disabling flaky test [MINOR] Make sure all `HoodieRecord`s are appropriately serializable by Kryo (apache#6977) * Make sure `HoodieRecord`, `HoodieKey`, `HoodieRecordLocation` are all `KryoSerializable` * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; Implemented serialization hooks for `HoodieAvroRecord`; * Revisited `HoodieSparkRecord` to transiently hold on to the schema so that it could project row * Implemented serialization hooks for `HoodieSparkRecord` * Added `TestHoodieSparkRecord` * Added tests for Avro-based records * Added test for `HoodieEmptyRecord` * Fixed sealing/unsealing for `HoodieRecord` in `HoodieBackedTableMetadataWriter` * Properly handle deflated records * Fixing `Row`s encoding * Fixed `HoodieRecord` to be properly sealed/unsealed * Fixed serialization of the `HoodieRecordGlobalLocation` [MINOR] Additional fixes for apache#6745 (apache#6947) * Tidying up * Tidying up more * Cleaning up duplication * Tidying up * Revisited legacy operating mode configuration * Tidying up * Cleaned up `projectUnsafe` API * Fixing compilation * Cleaning up `HoodieSparkRecord` ctors; Revisited mandatory unsafe-projection * Fixing compilation * Cleaned up `ParquetReader` initialization * Revisited `HoodieSparkRecord` to accept either `UnsafeRow` or `HoodieInternalRow`, and avoid unnecessary copying after unsafe-projection * Cleaning up redundant exception spec * Make sure `updateMetadataFields` properly wraps `InternalRow` into `HoodieInternalRow` if necessary; Cleaned up `MetadataValues` * Fixed meta-fields extraction and `HoodieInternalRow` composition w/in `HoodieSparkRecord` * De-duplicate `HoodieSparkRecord` ctors; Make sure either only `UnsafeRow` or `HoodieInternalRow` are permitted inside `HoodieSparkRecord` * Removed unnecessary copying * Cleaned up projection for `HoodieSparkRecord` (dropping partition columns); Removed unnecessary copying * Fixing compilation * Fixing compilation (for Flink) * Cleaned up File Raders' interfaces: - Extracted `HoodieSeekingFileReader` interface (for key-ranged reads) - Pushed down concrete implementation methods into `HoodieAvroFileReaderBase` from the interfaces * Cleaned up File Readers impls (inline with then new interfaces) * Rebsaed `HoodieBackedTableMetadata` onto new `HoodieSeekingFileReader` * Tidying up * Missing licenses * Re-instate custom override for `HoodieAvroParquetReader`; Tidying up * Fixed missing cloning w/in `HoodieLazyInsertIterable` * Fixed missing cloning in deduplication flow * Allow `HoodieSparkRecord` to hold `ColumnarBatchRow` * Missing licenses * Fixing compilation * Missing changes * Fixed Spark 2.x validation whether the row was read as a batch Fix comment in RFC46 (apache#6745) * rename * add MetadataValues in updateMetadataValues * remove singleton in fileFactory * add truncateRecordKey * remove hoodieRecord#setData * rename HoodieAvroRecord * fix code style * fix HoodieSparkRecordSerializer * fix benchmark * fix SparkRecordUtils * instantiate HoodieWriteConfig on the fly * add test * fix HoodieSparkRecordSerializer. Replace Java's object serialization with kryo * add broadcast * fix comment * remove unnecessary broadcast * add unsafe check in spark record * fix getRecordColumnValues * remove spark.sql.parquet.writeLegacyFormat * fix unsafe projection * fix * pass external schema * update doc * rename back to HoodieAvroRecord * fix * remove comparable wrapper * fix comment * fix comment * fix comment * fix comment * simplify row copy * fix ParquetReaderIterator Co-authored-by: Shawy Geng <gengxiaoyu1996@gmail.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [RFC-46][HUDI-4414] Update the RFC-46 doc to fix comments feedback (apache#6132) * Update the RFC-46 doc to fix comments feedback * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer.(apache#5629) * [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer. * add schema finger print * add benchmark * a new way to config the merger * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: gengxiaoyu <gengxiaoyu@bytedance.com> [HUDI-3350][HUDI-3351] Support HoodieMerge API and Spark engine-specific HoodieRecord (apache#5627) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4344] fix usage of HoodieDataBlock#getRecordIterator (apache#6005) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4292][RFC-46] Update doc to align with the Record Merge API changes (apache#5927) [MINOR] Fix type casting in TestHoodieHFileReaderWriter [HUDI-3378][HUDI-3379][HUDI-3381] Migrate usage of HoodieRecordPayload and raw Avro payload to HoodieRecord (apache#5522) Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com>

[minor] add more test for rfc46 (apache#7003) ## Change Logs - Add HoodieSparkValidateDuplicateKeyRecordMerger behaving the same as ValidateDuplicateKeyPayload. We should use it with config "hoodie.sql.insert.mode=strict". - Fix nest field exist in HoodieCatalystExpressionUtils - Fix rewrite in HoodieInternalRowUtiles to support type promoted as avro - Fallback to avro when use "merge into" sql - Fix some schema handling issue - Support delta streamer - Convert parquet schema to spark schema and then avro schema(in org.apache.hudi.io.storage.HoodieSparkParquetReader#getSchema). Some types in avro are not compatible with parquet. For ex, decimal as int32/int64 in parquet will convert to int/long in avro. Because avro do not has decimal as int/long . We will lose the logic type info if we directly convert it to avro schema. - Support schema evolution in parquet block [Minor] fix multi deser avro payload (apache#7021) In HoodieAvroRecord, we will call isDelete, shouldIgnore before we write it to the file. Each method will deserialize HoodiePayload. So we add deserialization method in HoodieRecord and call this method once before calling isDelete or shouldIgnore. Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: Alexey Kudinkin <alexey.kudinkin@gmail.com> [MINOR] Properly registering target classes w/ Kryo (apache#7026) * Added `HoodieKryoRegistrar` registering necessary Hudi's classes w/ Kryo to make their serialization more efficient (by serializing just the class id, in-liue the fully qualified class-name) * Redirected Kryo registration to `HoodieKryoRegistrar` * Registered additional classes likely to be serialized by Kryo * Updated tests * Fixed serialization of Avro's `Utf8` to serialize just the bytes * Added tests * Added custom `AvroUtf8Serializer`; Tidying up * Extracted `HoodieCommonKryoRegistrar` to leverage in `SerializationUtils` * `HoodieKryoRegistrar` > `HoodieSparkKryoRegistrar`; Rebased `HoodieSparkKryoRegistrar` onto `HoodieCommonKryoRegistrar` * `lint` * Fixing compilation for Spark 2.x * Disabling flaky test [MINOR] Make sure all `HoodieRecord`s are appropriately serializable by Kryo (apache#6977) * Make sure `HoodieRecord`, `HoodieKey`, `HoodieRecordLocation` are all `KryoSerializable` * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; Implemented serialization hooks for `HoodieAvroRecord`; * Revisited `HoodieSparkRecord` to transiently hold on to the schema so that it could project row * Implemented serialization hooks for `HoodieSparkRecord` * Added `TestHoodieSparkRecord` * Added tests for Avro-based records * Added test for `HoodieEmptyRecord` * Fixed sealing/unsealing for `HoodieRecord` in `HoodieBackedTableMetadataWriter` * Properly handle deflated records * Fixing `Row`s encoding * Fixed `HoodieRecord` to be properly sealed/unsealed * Fixed serialization of the `HoodieRecordGlobalLocation` [MINOR] Additional fixes for apache#6745 (apache#6947) * Tidying up * Tidying up more * Cleaning up duplication * Tidying up * Revisited legacy operating mode configuration * Tidying up * Cleaned up `projectUnsafe` API * Fixing compilation * Cleaning up `HoodieSparkRecord` ctors; Revisited mandatory unsafe-projection * Fixing compilation * Cleaned up `ParquetReader` initialization * Revisited `HoodieSparkRecord` to accept either `UnsafeRow` or `HoodieInternalRow`, and avoid unnecessary copying after unsafe-projection * Cleaning up redundant exception spec * Make sure `updateMetadataFields` properly wraps `InternalRow` into `HoodieInternalRow` if necessary; Cleaned up `MetadataValues` * Fixed meta-fields extraction and `HoodieInternalRow` composition w/in `HoodieSparkRecord` * De-duplicate `HoodieSparkRecord` ctors; Make sure either only `UnsafeRow` or `HoodieInternalRow` are permitted inside `HoodieSparkRecord` * Removed unnecessary copying * Cleaned up projection for `HoodieSparkRecord` (dropping partition columns); Removed unnecessary copying * Fixing compilation * Fixing compilation (for Flink) * Cleaned up File Raders' interfaces: - Extracted `HoodieSeekingFileReader` interface (for key-ranged reads) - Pushed down concrete implementation methods into `HoodieAvroFileReaderBase` from the interfaces * Cleaned up File Readers impls (inline with then new interfaces) * Rebsaed `HoodieBackedTableMetadata` onto new `HoodieSeekingFileReader` * Tidying up * Missing licenses * Re-instate custom override for `HoodieAvroParquetReader`; Tidying up * Fixed missing cloning w/in `HoodieLazyInsertIterable` * Fixed missing cloning in deduplication flow * Allow `HoodieSparkRecord` to hold `ColumnarBatchRow` * Missing licenses * Fixing compilation * Missing changes * Fixed Spark 2.x validation whether the row was read as a batch Fix comment in RFC46 (apache#6745) * rename * add MetadataValues in updateMetadataValues * remove singleton in fileFactory * add truncateRecordKey * remove hoodieRecord#setData * rename HoodieAvroRecord * fix code style * fix HoodieSparkRecordSerializer * fix benchmark * fix SparkRecordUtils * instantiate HoodieWriteConfig on the fly * add test * fix HoodieSparkRecordSerializer. Replace Java's object serialization with kryo * add broadcast * fix comment * remove unnecessary broadcast * add unsafe check in spark record * fix getRecordColumnValues * remove spark.sql.parquet.writeLegacyFormat * fix unsafe projection * fix * pass external schema * update doc * rename back to HoodieAvroRecord * fix * remove comparable wrapper * fix comment * fix comment * fix comment * fix comment * simplify row copy * fix ParquetReaderIterator Co-authored-by: Shawy Geng <gengxiaoyu1996@gmail.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [RFC-46][HUDI-4414] Update the RFC-46 doc to fix comments feedback (apache#6132) * Update the RFC-46 doc to fix comments feedback * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer.(apache#5629) * [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer. * add schema finger print * add benchmark * a new way to config the merger * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: gengxiaoyu <gengxiaoyu@bytedance.com> [HUDI-3350][HUDI-3351] Support HoodieMerge API and Spark engine-specific HoodieRecord (apache#5627) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4344] fix usage of HoodieDataBlock#getRecordIterator (apache#6005) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4292][RFC-46] Update doc to align with the Record Merge API changes (apache#5927) [MINOR] Fix type casting in TestHoodieHFileReaderWriter [HUDI-3378][HUDI-3379][HUDI-3381] Migrate usage of HoodieRecordPayload and raw Avro payload to HoodieRecord (apache#5522) Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com>

[minor] add more test for rfc46 (apache#7003) - Add HoodieSparkValidateDuplicateKeyRecordMerger behaving the same as ValidateDuplicateKeyPayload. We should use it with config "hoodie.sql.insert.mode=strict". - Fix nest field exist in HoodieCatalystExpressionUtils - Fix rewrite in HoodieInternalRowUtiles to support type promoted as avro - Fallback to avro when use "merge into" sql - Fix some schema handling issue - Support delta streamer - Convert parquet schema to spark schema and then avro schema(in org.apache.hudi.io.storage.HoodieSparkParquetReader#getSchema). Some types in avro are not compatible with parquet. For ex, decimal as int32/int64 in parquet will convert to int/long in avro. Because avro do not has decimal as int/long . We will lose the logic type info if we directly convert it to avro schema. - Support schema evolution in parquet block [Minor] fix multi deser avro payload (apache#7021) In HoodieAvroRecord, we will call isDelete, shouldIgnore before we write it to the file. Each method will deserialize HoodiePayload. So we add deserialization method in HoodieRecord and call this method once before calling isDelete or shouldIgnore. Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: Alexey Kudinkin <alexey.kudinkin@gmail.com> [MINOR] Properly registering target classes w/ Kryo (apache#7026) * Added `HoodieKryoRegistrar` registering necessary Hudi's classes w/ Kryo to make their serialization more efficient (by serializing just the class id, in-liue the fully qualified class-name) * Redirected Kryo registration to `HoodieKryoRegistrar` * Registered additional classes likely to be serialized by Kryo * Updated tests * Fixed serialization of Avro's `Utf8` to serialize just the bytes * Added tests * Added custom `AvroUtf8Serializer`; Tidying up * Extracted `HoodieCommonKryoRegistrar` to leverage in `SerializationUtils` * `HoodieKryoRegistrar` > `HoodieSparkKryoRegistrar`; Rebased `HoodieSparkKryoRegistrar` onto `HoodieCommonKryoRegistrar` * `lint` * Fixing compilation for Spark 2.x * Disabling flaky test [MINOR] Make sure all `HoodieRecord`s are appropriately serializable by Kryo (apache#6977) * Make sure `HoodieRecord`, `HoodieKey`, `HoodieRecordLocation` are all `KryoSerializable` * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; Implemented serialization hooks for `HoodieAvroRecord`; * Revisited `HoodieSparkRecord` to transiently hold on to the schema so that it could project row * Implemented serialization hooks for `HoodieSparkRecord` * Added `TestHoodieSparkRecord` * Added tests for Avro-based records * Added test for `HoodieEmptyRecord` * Fixed sealing/unsealing for `HoodieRecord` in `HoodieBackedTableMetadataWriter` * Properly handle deflated records * Fixing `Row`s encoding * Fixed `HoodieRecord` to be properly sealed/unsealed * Fixed serialization of the `HoodieRecordGlobalLocation` [MINOR] Additional fixes for apache#6745 (apache#6947) * Tidying up * Tidying up more * Cleaning up duplication * Tidying up * Revisited legacy operating mode configuration * Tidying up * Cleaned up `projectUnsafe` API * Fixing compilation * Cleaning up `HoodieSparkRecord` ctors; Revisited mandatory unsafe-projection * Fixing compilation * Cleaned up `ParquetReader` initialization * Revisited `HoodieSparkRecord` to accept either `UnsafeRow` or `HoodieInternalRow`, and avoid unnecessary copying after unsafe-projection * Cleaning up redundant exception spec * Make sure `updateMetadataFields` properly wraps `InternalRow` into `HoodieInternalRow` if necessary; Cleaned up `MetadataValues` * Fixed meta-fields extraction and `HoodieInternalRow` composition w/in `HoodieSparkRecord` * De-duplicate `HoodieSparkRecord` ctors; Make sure either only `UnsafeRow` or `HoodieInternalRow` are permitted inside `HoodieSparkRecord` * Removed unnecessary copying * Cleaned up projection for `HoodieSparkRecord` (dropping partition columns); Removed unnecessary copying * Fixing compilation * Fixing compilation (for Flink) * Cleaned up File Raders' interfaces: - Extracted `HoodieSeekingFileReader` interface (for key-ranged reads) - Pushed down concrete implementation methods into `HoodieAvroFileReaderBase` from the interfaces * Cleaned up File Readers impls (inline with then new interfaces) * Rebsaed `HoodieBackedTableMetadata` onto new `HoodieSeekingFileReader` * Tidying up * Missing licenses * Re-instate custom override for `HoodieAvroParquetReader`; Tidying up * Fixed missing cloning w/in `HoodieLazyInsertIterable` * Fixed missing cloning in deduplication flow * Allow `HoodieSparkRecord` to hold `ColumnarBatchRow` * Missing licenses * Fixing compilation * Missing changes * Fixed Spark 2.x validation whether the row was read as a batch Fix comment in RFC46 (apache#6745) * rename * add MetadataValues in updateMetadataValues * remove singleton in fileFactory * add truncateRecordKey * remove hoodieRecord#setData * rename HoodieAvroRecord * fix code style * fix HoodieSparkRecordSerializer * fix benchmark * fix SparkRecordUtils * instantiate HoodieWriteConfig on the fly * add test * fix HoodieSparkRecordSerializer. Replace Java's object serialization with kryo * add broadcast * fix comment * remove unnecessary broadcast * add unsafe check in spark record * fix getRecordColumnValues * remove spark.sql.parquet.writeLegacyFormat * fix unsafe projection * fix * pass external schema * update doc * rename back to HoodieAvroRecord * fix * remove comparable wrapper * fix comment * fix comment * fix comment * fix comment * simplify row copy * fix ParquetReaderIterator Co-authored-by: Shawy Geng <gengxiaoyu1996@gmail.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [RFC-46][HUDI-4414] Update the RFC-46 doc to fix comments feedback (apache#6132) * Update the RFC-46 doc to fix comments feedback * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer.(apache#5629) * [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer. * add schema finger print * add benchmark * a new way to config the merger * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: gengxiaoyu <gengxiaoyu@bytedance.com> [HUDI-3350][HUDI-3351] Support HoodieMerge API and Spark engine-specific HoodieRecord (apache#5627) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4344] fix usage of HoodieDataBlock#getRecordIterator (apache#6005) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4292][RFC-46] Update doc to align with the Record Merge API changes (apache#5927) [MINOR] Fix type casting in TestHoodieHFileReaderWriter [HUDI-3378][HUDI-3379][HUDI-3381] Migrate usage of HoodieRecordPayload and raw Avro payload to HoodieRecord (apache#5522) Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com>

[minor] add more test for rfc46 (apache#7003) ## Change Logs - Add HoodieSparkValidateDuplicateKeyRecordMerger behaving the same as ValidateDuplicateKeyPayload. We should use it with config "hoodie.sql.insert.mode=strict". - Fix nest field exist in HoodieCatalystExpressionUtils - Fix rewrite in HoodieInternalRowUtiles to support type promoted as avro - Fallback to avro when use "merge into" sql - Fix some schema handling issue - Support delta streamer - Convert parquet schema to spark schema and then avro schema(in org.apache.hudi.io.storage.HoodieSparkParquetReader#getSchema). Some types in avro are not compatible with parquet. For ex, decimal as int32/int64 in parquet will convert to int/long in avro. Because avro do not has decimal as int/long . We will lose the logic type info if we directly convert it to avro schema. - Support schema evolution in parquet block [Minor] fix multi deser avro payload (apache#7021) In HoodieAvroRecord, we will call isDelete, shouldIgnore before we write it to the file. Each method will deserialize HoodiePayload. So we add deserialization method in HoodieRecord and call this method once before calling isDelete or shouldIgnore. Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: Alexey Kudinkin <alexey.kudinkin@gmail.com> [MINOR] Properly registering target classes w/ Kryo (apache#7026) * Added `HoodieKryoRegistrar` registering necessary Hudi's classes w/ Kryo to make their serialization more efficient (by serializing just the class id, in-liue the fully qualified class-name) * Redirected Kryo registration to `HoodieKryoRegistrar` * Registered additional classes likely to be serialized by Kryo * Updated tests * Fixed serialization of Avro's `Utf8` to serialize just the bytes * Added tests * Added custom `AvroUtf8Serializer`; Tidying up * Extracted `HoodieCommonKryoRegistrar` to leverage in `SerializationUtils` * `HoodieKryoRegistrar` > `HoodieSparkKryoRegistrar`; Rebased `HoodieSparkKryoRegistrar` onto `HoodieCommonKryoRegistrar` * `lint` * Fixing compilation for Spark 2.x * Disabling flaky test [MINOR] Make sure all `HoodieRecord`s are appropriately serializable by Kryo (apache#6977) * Make sure `HoodieRecord`, `HoodieKey`, `HoodieRecordLocation` are all `KryoSerializable` * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; * Revisited `HoodieRecord` serialization hooks to make sure they a) could not be overridden, b) provide for hooks to properly serialize record's payload; Implemented serialization hooks for `HoodieAvroIndexedRecord`; Implemented serialization hooks for `HoodieEmptyRecord`; Implemented serialization hooks for `HoodieAvroRecord`; * Revisited `HoodieSparkRecord` to transiently hold on to the schema so that it could project row * Implemented serialization hooks for `HoodieSparkRecord` * Added `TestHoodieSparkRecord` * Added tests for Avro-based records * Added test for `HoodieEmptyRecord` * Fixed sealing/unsealing for `HoodieRecord` in `HoodieBackedTableMetadataWriter` * Properly handle deflated records * Fixing `Row`s encoding * Fixed `HoodieRecord` to be properly sealed/unsealed * Fixed serialization of the `HoodieRecordGlobalLocation` [MINOR] Additional fixes for apache#6745 (apache#6947) * Tidying up * Tidying up more * Cleaning up duplication * Tidying up * Revisited legacy operating mode configuration * Tidying up * Cleaned up `projectUnsafe` API * Fixing compilation * Cleaning up `HoodieSparkRecord` ctors; Revisited mandatory unsafe-projection * Fixing compilation * Cleaned up `ParquetReader` initialization * Revisited `HoodieSparkRecord` to accept either `UnsafeRow` or `HoodieInternalRow`, and avoid unnecessary copying after unsafe-projection * Cleaning up redundant exception spec * Make sure `updateMetadataFields` properly wraps `InternalRow` into `HoodieInternalRow` if necessary; Cleaned up `MetadataValues` * Fixed meta-fields extraction and `HoodieInternalRow` composition w/in `HoodieSparkRecord` * De-duplicate `HoodieSparkRecord` ctors; Make sure either only `UnsafeRow` or `HoodieInternalRow` are permitted inside `HoodieSparkRecord` * Removed unnecessary copying * Cleaned up projection for `HoodieSparkRecord` (dropping partition columns); Removed unnecessary copying * Fixing compilation * Fixing compilation (for Flink) * Cleaned up File Raders' interfaces: - Extracted `HoodieSeekingFileReader` interface (for key-ranged reads) - Pushed down concrete implementation methods into `HoodieAvroFileReaderBase` from the interfaces * Cleaned up File Readers impls (inline with then new interfaces) * Rebsaed `HoodieBackedTableMetadata` onto new `HoodieSeekingFileReader` * Tidying up * Missing licenses * Re-instate custom override for `HoodieAvroParquetReader`; Tidying up * Fixed missing cloning w/in `HoodieLazyInsertIterable` * Fixed missing cloning in deduplication flow * Allow `HoodieSparkRecord` to hold `ColumnarBatchRow` * Missing licenses * Fixing compilation * Missing changes * Fixed Spark 2.x validation whether the row was read as a batch Fix comment in RFC46 (apache#6745) * rename * add MetadataValues in updateMetadataValues * remove singleton in fileFactory * add truncateRecordKey * remove hoodieRecord#setData * rename HoodieAvroRecord * fix code style * fix HoodieSparkRecordSerializer * fix benchmark * fix SparkRecordUtils * instantiate HoodieWriteConfig on the fly * add test * fix HoodieSparkRecordSerializer. Replace Java's object serialization with kryo * add broadcast * fix comment * remove unnecessary broadcast * add unsafe check in spark record * fix getRecordColumnValues * remove spark.sql.parquet.writeLegacyFormat * fix unsafe projection * fix * pass external schema * update doc * rename back to HoodieAvroRecord * fix * remove comparable wrapper * fix comment * fix comment * fix comment * fix comment * simplify row copy * fix ParquetReaderIterator Co-authored-by: Shawy Geng <gengxiaoyu1996@gmail.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [RFC-46][HUDI-4414] Update the RFC-46 doc to fix comments feedback (apache#6132) * Update the RFC-46 doc to fix comments feedback * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer.(apache#5629) * [HUDI-4301] [HUDI-3384][HUDI-3385] Spark specific file reader/writer. * add schema finger print * add benchmark * a new way to config the merger * fix Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> Co-authored-by: gengxiaoyu <gengxiaoyu@bytedance.com> [HUDI-3350][HUDI-3351] Support HoodieMerge API and Spark engine-specific HoodieRecord (apache#5627) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4344] fix usage of HoodieDataBlock#getRecordIterator (apache#6005) Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com> [HUDI-4292][RFC-46] Update doc to align with the Record Merge API changes (apache#5927) [MINOR] Fix type casting in TestHoodieHFileReaderWriter [HUDI-3378][HUDI-3379][HUDI-3381] Migrate usage of HoodieRecordPayload and raw Avro payload to HoodieRecord (apache#5522) Co-authored-by: Alexey Kudinkin <alexey@infinilake.com> Co-authored-by: wangzixuan.wzxuan <wangzixuan.wzxuan@bytedance.com>

wzx140 changed the title ~~[minor] add more test~~ [minor] add more test for rfc46 Oct 19, 2022

wzx140 force-pushed the add-test branch 3 times, most recently from 4a4d0e7 to 8a5e66b Compare October 19, 2022 17:43

wzx140 force-pushed the add-test branch 3 times, most recently from 01c4965 to 6efb6b4 Compare October 25, 2022 17:24

xushiyan added the priority:blocker Production down; release blocker label Oct 31, 2022

nsivabalan assigned alexeykudinkin and xushiyan Nov 2, 2022

wzx140 force-pushed the add-test branch 5 times, most recently from 277ad5d to f91a0b4 Compare November 13, 2022 17:28

minihippo force-pushed the add-test branch 9 times, most recently from 1c1d6e2 to f09dab9 Compare November 15, 2022 17:29

alexeykudinkin reviewed Nov 17, 2022

View reviewed changes

alexeykudinkin reviewed Nov 19, 2022

View reviewed changes

wzx140 force-pushed the add-test branch from 3be6021 to 790c32c Compare November 20, 2022 16:27

alexeykudinkin reviewed Nov 21, 2022

View reviewed changes

apache deleted a comment from hudi-bot Nov 28, 2022

alexeykudinkin approved these changes Nov 28, 2022

View reviewed changes

wangzixuan.wzxuan added 3 commits November 29, 2022 10:09

add more test

8dbd156

disable TestCleaner

ab7f9df

fix copy

45cfd6e

wzx140 force-pushed the add-test branch from 48e7b5b to 45cfd6e Compare November 29, 2022 02:13

add copy

4a74177

alexeykudinkin approved these changes Nov 30, 2022

View reviewed changes

wangzixuan.wzxuan added 2 commits November 30, 2022 10:27

add comments

803594c

change ci timeout

6835303

alexeykudinkin merged commit 9c691f0 into apache:release-feature-rfc46 Nov 30, 2022

wzx140 mentioned this pull request Jan 29, 2023

[HUDI-5633] Fixing performance regression in HoodieSparkRecord #7769

Merged

4 tasks

Uh oh!

Conversation

wzx140 commented Oct 19, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Change Logs

Impact

Risk level (write none, low medium or high below)

Documentation Update

Contributor's checklist

Uh oh!

alexeykudinkin commented Oct 19, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

wzx140 commented Nov 16, 2022

Uh oh!

Choose a reason for hiding this comment

Uh oh!

wzx140 Nov 17, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

wzx140 commented Oct 19, 2022 •

edited

Loading

alexeykudinkin commented Oct 19, 2022 •

edited

Loading

wzx140 Nov 17, 2022 •

edited

Loading