Github user gatorsmile commented on a diff in the pull request:
https://github.com/apache/spark/pull/15158#discussion_r79705605
--- Diff:
sql/hive/src/main/scala/org/apache/spark/sql/hive/MetastoreRelation.scala ---
@@ -109,39 +109,59 @@ private[hive] case class MetastoreRelation(
}
@transient override lazy val statistics: Statistics = {
- catalogTable.stats.getOrElse(Statistics(
- sizeInBytes = {
- val totalSize =
hiveQlTable.getParameters.get(StatsSetupConst.TOTAL_SIZE)
- val rawDataSize =
hiveQlTable.getParameters.get(StatsSetupConst.RAW_DATA_SIZE)
- // TODO: check if this estimate is valid for tables after
partition pruning.
- // NOTE: getting `totalSize` directly from params is kind of
hacky, but this should be
- // relatively cheap if parameters for the table are populated into
the metastore.
- // Besides `totalSize`, there are also `numFiles`, `numRows`,
`rawDataSize` keys
- // (see StatsSetupConst in Hive) that we can look at in the future.
- BigInt(
- // When table is external,`totalSize` is always zero, which will
influence join strategy
- // so when `totalSize` is zero, use `rawDataSize` instead
- // when `rawDataSize` is also zero, use
`HiveExternalCatalog.STATISTICS_TOTAL_SIZE`,
- // which is generated by analyze command.
- if (totalSize != null && totalSize.toLong > 0L) {
- totalSize.toLong
- } else if (rawDataSize != null && rawDataSize.toLong > 0) {
- rawDataSize.toLong
- } else if
(sparkSession.sessionState.conf.fallBackToHdfsForStatsEnabled) {
- try {
- val hadoopConf = sparkSession.sessionState.newHadoopConf()
- val fs: FileSystem =
hiveQlTable.getPath.getFileSystem(hadoopConf)
- fs.getContentSummary(hiveQlTable.getPath).getLength
- } catch {
- case e: IOException =>
- logWarning("Failed to get table size from hdfs.", e)
- sparkSession.sessionState.conf.defaultSizeInBytes
- }
- } else {
- sparkSession.sessionState.conf.defaultSizeInBytes
- })
+ catalogTable.stats.getOrElse {
+ // For non-partitioned tables, Hive-generated statistics are stored
in table properties
+ // For partitioned tables, Hive-generated statistics are stored in
partition properties
+ val (totalSize, rawDataSize) = if
(catalogTable.partitionColumnNames.isEmpty) {
+ val properties =
Option(hiveQlTable.getParameters).map(_.asScala.toMap).orNull
+ (properties.get(StatsSetupConst.TOTAL_SIZE).map(_.toLong),
+ properties.get(StatsSetupConst.RAW_DATA_SIZE).map(_.toLong))
+ } else {
+ (getTotalTableSize(StatsSetupConst.TOTAL_SIZE),
+ getTotalTableSize(StatsSetupConst.RAW_DATA_SIZE))
}
- ))
+ Statistics(
+ sizeInBytes = {
+ BigInt(
+ // When table is external,`totalSize` is always zero, which
will influence join strategy
+ // so when `totalSize` is zero, use `rawDataSize` instead.
When `rawDataSize` is also
+ // zero, fall back to filesystem to get file sizes, if enabled.
+ if (totalSize.isDefined && totalSize.get > 0L) {
+ totalSize.get
+ } else if (rawDataSize.isDefined && rawDataSize.get > 0L) {
+ rawDataSize.get
+ } else if
(sparkSession.sessionState.conf.fallBackToHdfsForStatsEnabled) {
+ try {
+ val hadoopConf = sparkSession.sessionState.newHadoopConf()
+ val fs: FileSystem =
hiveQlTable.getPath.getFileSystem(hadoopConf)
+ fs.getContentSummary(hiveQlTable.getPath).getLength
+ } catch {
+ case e: IOException =>
+ logWarning("Failed to get table size from hdfs.", e)
+ sparkSession.sessionState.conf.defaultSizeInBytes
+ }
+ } else {
+ sparkSession.sessionState.conf.defaultSizeInBytes
+ })
+ }
+ )
+ }
+ }
+
+ // For partitioned tables, get the size of all the partitions.
+ // Note: the statistics might not be gathered for all the partitions.
+ // For partial collection, we will not utilize the Hive-generated
statistics.
+ private def getTotalTableSize(statType: String): Option[Long] = {
--- End diff --
: ) We are facing the same issue in both data source tables and hive
tables. That requires another PR to change how we use the statistics of leaf
nodes. Partition filtering on statistics should be considered by the `Filter`
nodes. Is my understanding right?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
