Github user mbasmanova commented on a diff in the pull request:
https://github.com/apache/spark/pull/18421#discussion_r132493566
--- Diff:
sql/hive/src/test/scala/org/apache/spark/sql/hive/StatisticsSuite.scala ---
@@ -256,6 +257,201 @@ class StatisticsSuite extends
StatisticsCollectionTestBase with TestHiveSingleto
}
}
+ test("analyze single partition") {
+ val tableName = "analyzeTable_part"
+
+ def queryStats(ds: String): CatalogStatistics = {
+ val partition =
+
spark.sessionState.catalog.getPartition(TableIdentifier(tableName), Map("ds" ->
ds))
+ partition.stats.get
+ }
+
+ def createPartition(ds: String, query: String): Unit = {
+ sql(s"INSERT INTO TABLE $tableName PARTITION (ds='$ds') $query")
+ }
+
+ withTable(tableName) {
+ sql(s"CREATE TABLE $tableName (key STRING, value STRING) PARTITIONED
BY (ds STRING)")
+
+ createPartition("2010-01-01", "SELECT '1', 'A' from src")
+ createPartition("2010-01-02", "SELECT '1', 'A' from src UNION ALL
SELECT '1', 'A' from src")
+ createPartition("2010-01-03", "SELECT '1', 'A' from src")
+
+ sql(s"ANALYZE TABLE $tableName PARTITION (ds='2010-01-01') COMPUTE
STATISTICS NOSCAN")
+
+ sql(s"ANALYZE TABLE $tableName PARTITION (ds='2010-01-02') COMPUTE
STATISTICS NOSCAN")
+
+ assert(queryStats("2010-01-01").rowCount === None)
+ assert(queryStats("2010-01-01").sizeInBytes === 2000)
+
+ assert(queryStats("2010-01-02").rowCount === None)
+ assert(queryStats("2010-01-02").sizeInBytes === 2*2000)
+
+ sql(s"ANALYZE TABLE $tableName PARTITION (ds='2010-01-01') COMPUTE
STATISTICS")
+
+ sql(s"ANALYZE TABLE $tableName PARTITION (ds='2010-01-02') COMPUTE
STATISTICS")
+
+ assert(queryStats("2010-01-01").rowCount.get === 500)
+ assert(queryStats("2010-01-01").sizeInBytes === 2000)
+
+ assert(queryStats("2010-01-02").rowCount.get === 2*500)
+ assert(queryStats("2010-01-02").sizeInBytes === 2*2000)
+ }
+ }
+
+ test("analyze a set of partitions") {
+ val tableName = "analyzeTable_part"
+
+ def queryStats(ds: String, hr: String): Option[CatalogStatistics] = {
+ val tableId = TableIdentifier(tableName)
+ val partition =
+ spark.sessionState.catalog.getPartition(tableId, Map("ds" -> ds,
"hr" -> hr))
+ partition.stats
+ }
+
+ def assertPartitionStats(
+ ds: String,
+ hr: String,
+ rowCount: Option[BigInt],
+ sizeInBytes: BigInt): Unit = {
+ val stats = queryStats(ds, hr).get
+ assert(stats.rowCount === rowCount)
+ assert(stats.sizeInBytes === sizeInBytes)
+ }
+
+ def createPartition(ds: String, hr: Int, query: String): Unit = {
+ sql(s"INSERT INTO TABLE $tableName PARTITION (ds='$ds', hr=$hr)
$query")
+ }
+
+ withTable(tableName) {
+ sql(s"CREATE TABLE $tableName (key STRING, value STRING) PARTITIONED
BY (ds STRING, hr INT)")
+
+ createPartition("2010-01-01", 10, "SELECT '1', 'A' from src")
+ createPartition("2010-01-01", 11, "SELECT '1', 'A' from src")
+ createPartition("2010-01-02", 10, "SELECT '1', 'A' from src")
+ createPartition("2010-01-02", 11,
+ "SELECT '1', 'A' from src UNION ALL SELECT '1', 'A' from src")
+
+ sql(s"ANALYZE TABLE $tableName PARTITION (ds='2010-01-01') COMPUTE
STATISTICS NOSCAN")
+
+ assertPartitionStats("2010-01-01", "10", rowCount = None,
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-01", "11", rowCount = None,
sizeInBytes = 2000)
+ assert(queryStats("2010-01-02", "10") === None)
+ assert(queryStats("2010-01-02", "11") === None)
+
+ sql(s"ANALYZE TABLE $tableName PARTITION (ds='2010-01-02') COMPUTE
STATISTICS NOSCAN")
+
+ assertPartitionStats("2010-01-01", "10", rowCount = None,
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-01", "11", rowCount = None,
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-02", "10", rowCount = None,
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-02", "11", rowCount = None,
sizeInBytes = 2*2000)
+
+ sql(s"ANALYZE TABLE $tableName PARTITION (ds='2010-01-01') COMPUTE
STATISTICS")
+
+ assertPartitionStats("2010-01-01", "10", rowCount = Some(500),
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-01", "11", rowCount = Some(500),
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-02", "10", rowCount = None,
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-02", "11", rowCount = None,
sizeInBytes = 2*2000)
+
+ sql(s"ANALYZE TABLE $tableName PARTITION (ds='2010-01-02') COMPUTE
STATISTICS")
+
+ assertPartitionStats("2010-01-01", "10", rowCount = Some(500),
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-01", "11", rowCount = Some(500),
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-02", "10", rowCount = Some(500),
sizeInBytes = 2000)
+ assertPartitionStats("2010-01-02", "11", rowCount = Some(2*500),
sizeInBytes = 2*2000)
+ }
--- End diff --
@gatorsmile , the installation of Hive I have access fails the above query:
> Partition columns in partition specification are not the same as that
defined in the table schema. The names and orders have to be exactly the same.
Partition columns in the table schema are: (ds, hr), while the partitions
specified in the query are: (hr).
It also fails any partition specification which doesn't list all partition
columns. E.g., all of (ds), (hr), (ds='2017-01-01'), (hr=10) fail with the
above error message.
This installation is not using the latest open source Hive though.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
