attilapiros commented on a change in pull request #23988: [SPARK-26509][SQL]
Parquet DELTA_BYTE_ARRAY is not supported in Spark 2.x's Vectorized Reader
URL: https://github.com/apache/spark/pull/23988#discussion_r269107368
##########
File path:
sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetEncodingSuite.scala
##########
@@ -114,4 +118,36 @@ class ParquetEncodingSuite extends
ParquetCompatibilityTest with SharedSQLContex
}
}
}
+
+ test("parquet v2 pages - delta encoding") {
+ val extraOptions = Map[String, String](
+ ParquetOutputFormat.WRITER_VERSION ->
ParquetProperties.WriterVersion.PARQUET_2_0.toString,
+ ParquetOutputFormat.ENABLE_DICTIONARY -> "false"
+ )
+
+ val hadoopConf = spark.sessionState.newHadoopConfWithOptions(extraOptions)
+ withSQLConf(SQLConf.PARQUET_VECTORIZED_READER_ENABLED.key -> "true",
+ ParquetOutputFormat.JOB_SUMMARY_LEVEL -> "ALL") {
+ withTempPath { dir =>
+ val path = s"${dir.getCanonicalPath}/test.parquet"
+ val data = (1 to 3).map { i =>
+ (i, i.toLong, s"test_${i}")
+ }
+
+
spark.createDataFrame(data).write.options(extraOptions).mode("overwrite").parquet(path)
+
+ val blockMetadata = readFooter(new Path(path),
hadoopConf).getBlocks.asScala.head
+ val columnChunkMetadataList = blockMetadata.getColumns.asScala
+
+ // Verify that indeed delta encoding is used for each column
+ assert(columnChunkMetadataList.length === 3)
+
assert(columnChunkMetadataList(0).getEncodings.contains(Encoding.DELTA_BINARY_PACKED))
+
assert(columnChunkMetadataList(1).getEncodings.contains(Encoding.DELTA_BINARY_PACKED))
+
assert(columnChunkMetadataList(2).getEncodings.contains(Encoding.DELTA_BYTE_ARRAY))
Review comment:
As I have see `DELTA_BYTE_ARRAY` will used for the types `BINARY` and
`FIXED_LEN_BYTE_ARRAY` and they are also handled differently in
`VectorizedDeltaByteArrayReader#readBinary`.
Is it possible to test both cases?
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
