diff options
Diffstat (limited to 'sql/core/src/test')
-rw-r--r-- | sql/core/src/test/scala/org/apache/spark/sql/StatisticsSuite.scala | 26 |
1 files changed, 26 insertions, 0 deletions
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/StatisticsSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/StatisticsSuite.scala index 2c81cbf15f..264a2ffbeb 100644 --- a/sql/core/src/test/scala/org/apache/spark/sql/StatisticsSuite.scala +++ b/sql/core/src/test/scala/org/apache/spark/sql/StatisticsSuite.scala @@ -18,6 +18,7 @@ package org.apache.spark.sql import org.apache.spark.sql.catalyst.plans.logical.{GlobalLimit, Join, LocalLimit} +import org.apache.spark.sql.execution.datasources.LogicalRelation import org.apache.spark.sql.test.SharedSQLContext import org.apache.spark.sql.types._ @@ -75,4 +76,29 @@ class StatisticsSuite extends QueryTest with SharedSQLContext { } } + test("test table-level statistics for data source table created in InMemoryCatalog") { + def checkTableStats(tableName: String, expectedRowCount: Option[BigInt]): Unit = { + val df = sql(s"SELECT * FROM $tableName") + val relations = df.queryExecution.analyzed.collect { case rel: LogicalRelation => + assert(rel.catalogTable.isDefined) + assert(rel.catalogTable.get.stats.flatMap(_.rowCount) === expectedRowCount) + rel + } + assert(relations.size === 1) + } + + val tableName = "tbl" + withTable(tableName) { + sql(s"CREATE TABLE $tableName(i INT, j STRING) USING parquet") + Seq(1 -> "a", 2 -> "b").toDF("i", "j").write.mode("overwrite").insertInto("tbl") + + // noscan won't count the number of rows + sql(s"ANALYZE TABLE $tableName COMPUTE STATISTICS noscan") + checkTableStats(tableName, expectedRowCount = None) + + // without noscan, we count the number of rows + sql(s"ANALYZE TABLE $tableName COMPUTE STATISTICS") + checkTableStats(tableName, expectedRowCount = Some(2)) + } + } } |