lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-2176, 2178, 2179] Adding virtual key support to COW table (#3306)

Browse Source
This commit is contained in:
Sivabalan Narayanan
2021-07-26 17:21:04 -04:00
committed by GitHub
parent 5353243449
commit 61148c1c43
57 changed files with 969 additions and 413 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
hudi-cli/src/main/scala/org/apache/hudi/cli/SparkHelpers.scala
View File

@@ -45,13 +45,14 @@ object SparkHelpers {
val schema: Schema = sourceRecords.get(0).getSchema
val filter: BloomFilter = BloomFilterFactory.createBloomFilter(HoodieIndexConfig.BLOOM_FILTER_NUM_ENTRIES.defaultValue.toInt, HoodieIndexConfig.BLOOM_FILTER_FPP.defaultValue.toDouble,
HoodieIndexConfig.HOODIE_BLOOM_INDEX_FILTER_DYNAMIC_MAX_ENTRIES.defaultValue.toInt, HoodieIndexConfig.BLOOM_INDEX_FILTER_TYPE.defaultValue);
val writeSupport: HoodieAvroWriteSupport = new HoodieAvroWriteSupport(new AvroSchemaConverter(fs.getConf).convert(schema), schema, filter)
val writeSupport: HoodieAvroWriteSupport = new HoodieAvroWriteSupport(new AvroSchemaConverter(fs.getConf).convert(schema), schema, org.apache.hudi.common.util.Option.of(filter))
val parquetConfig: HoodieAvroParquetConfig = new HoodieAvroParquetConfig(writeSupport, CompressionCodecName.GZIP, HoodieStorageConfig.PARQUET_BLOCK_SIZE_BYTES.defaultValue.toInt, HoodieStorageConfig.PARQUET_PAGE_SIZE_BYTES.defaultValue.toInt, HoodieStorageConfig.PARQUET_FILE_MAX_BYTES.defaultValue.toInt, fs.getConf, HoodieStorageConfig.PARQUET_COMPRESSION_RATIO.defaultValue.toDouble)
// Add current classLoad for config, if not will throw classNotFound of 'HoodieWrapperFileSystem'.
parquetConfig.getHadoopConf().setClassLoader(Thread.currentThread.getContextClassLoader)
val writer = new HoodieParquetWriter[HoodieJsonPayload, IndexedRecord](instantTime, destinationFile, parquetConfig, schema, new SparkTaskContextSupplier())
val writer = new HoodieParquetWriter[HoodieJsonPayload, IndexedRecord](instantTime, destinationFile, parquetConfig, schema, new SparkTaskContextSupplier(),
true)
for (rec <- sourceRecords) {
val key: String = rec.get(HoodieRecord.RECORD_KEY_METADATA_FIELD).toString
if (!keysToSkip.contains(key)) {