alexeykudinkin commented on a change in pull request #5077:
URL: https://github.com/apache/hudi/pull/5077#discussion_r831514560
##########
File path:
hudi-spark-datasource/hudi-spark/src/test/scala/org/apache/hudi/functional/TestColumnStatsIndex.scala
##########
@@ -63,6 +69,91 @@ class TestColumnStatsIndex extends HoodieClientTestBase {
cleanupSparkContexts()
}
+ @Test
+ def testMetadataColumnStatsIndex(): Unit = {
Review comment:
In general, i think we should test against all operations that can
modify Col Stats index (commit, delta_commit, clean, compact, cluster) to make
sure they leave it in consistent state.
##########
File path:
hudi-spark-datasource/hudi-spark/src/test/scala/org/apache/hudi/functional/TestColumnStatsIndex.scala
##########
@@ -63,6 +69,91 @@ class TestColumnStatsIndex extends HoodieClientTestBase {
cleanupSparkContexts()
}
+ @Test
+ def testMetadataColumnStatsIndex(): Unit = {
+ setTableName("hoodie_test")
+ initMetaClient()
+ val sourceJSONTablePath =
getClass.getClassLoader.getResource("index/zorder/input-table-json").toString
+ val inputDF =
+ // NOTE: Schema here is provided for validation that the input date is in
the appropriate format
+ spark.read
+ .schema(sourceTableSchema)
+ .json(sourceJSONTablePath)
+
+ val opts = Map(
+ "hoodie.insert.shuffle.parallelism" -> "4",
+ "hoodie.upsert.shuffle.parallelism" -> "4",
+ HoodieWriteConfig.TBL_NAME.key -> "hoodie_test",
+ RECORDKEY_FIELD.key -> "c1",
+ PRECOMBINE_FIELD.key -> "c1",
+ HoodieMetadataConfig.ENABLE.key -> "true",
+ HoodieMetadataConfig.ENABLE_METADATA_INDEX_COLUMN_STATS.key -> "true",
+
HoodieMetadataConfig.ENABLE_METADATA_INDEX_COLUMN_STATS_FOR_ALL_COLUMNS.key ->
"true",
+ HoodieTableConfig.POPULATE_META_FIELDS.key -> "true"
+ )
+
+ inputDF.repartition(4)
+ .write
+ .format("hudi")
+ .options(opts)
+ .option(DataSourceWriteOptions.OPERATION.key,
DataSourceWriteOptions.INSERT_OPERATION_OPT_VAL)
+ .option(HoodieStorageConfig.PARQUET_MAX_FILE_SIZE.key, 100 * 1024)
+ .mode(SaveMode.Overwrite)
+ .save(basePath)
+
+ metaClient = HoodieTableMetaClient.reload(metaClient)
+
+ val metadataTablePath =
HoodieTableMetadata.getMetadataTableBasePath(basePath)
+
+ val targetColStatsIndexColumns = Seq(
+ HoodieMetadataPayload.COLUMN_STATS_FIELD_FILE_NAME,
+ HoodieMetadataPayload.COLUMN_STATS_FIELD_MIN_VALUE,
+ HoodieMetadataPayload.COLUMN_STATS_FIELD_MAX_VALUE,
+ HoodieMetadataPayload.COLUMN_STATS_FIELD_NULL_COUNT)
+
+ val requiredMetadataIndexColumns =
+ (targetColStatsIndexColumns :+
HoodieMetadataPayload.COLUMN_STATS_FIELD_COLUMN_NAME).map(colName =>
+ s"${HoodieMetadataPayload.SCHEMA_FIELD_ID_COLUMN_STATS}.${colName}")
+
+ // Read Metadata Table's Column Stats Index into Spark's [[DataFrame]]
+ val metadataTableDF = spark.read.format("org.apache.hudi")
+
.load(s"$metadataTablePath/${MetadataPartitionType.COLUMN_STATS.getPartitionPath}")
+
+ val colStatsDF =
metadataTableDF.where(col(HoodieMetadataPayload.SCHEMA_FIELD_ID_COLUMN_STATS).isNotNull)
+ .select(requiredMetadataIndexColumns.map(col): _*)
+
+ // assert min/max for some columns
+ val minC1 =
colStatsDF.select(HoodieMetadataPayload.COLUMN_STATS_FIELD_MIN_VALUE)
Review comment:
Instead of selectively asserting min/max for columns, let's use the
final table fixture and match it as a whole (like is being done in
`testZIndexTableComposition`)
--
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]
