Github user gatorsmile commented on a diff in the pull request: https://github.com/apache/spark/pull/11846#discussion_r56933487 --- Diff: sql/core/src/test/scala/org/apache/spark/sql/SQLQuerySuite.scala --- @@ -459,25 +460,96 @@ class SQLQuerySuite extends QueryTest with SharedSQLContext { Seq(Row(1, 3), Row(2, 3), Row(3, 3))) } - test("literal in agg grouping expressions") { + test("Group By Ordinal - basic") { checkAnswer( - sql("SELECT a, count(1) FROM testData2 GROUP BY a, 1"), - Seq(Row(1, 2), Row(2, 2), Row(3, 2))) - checkAnswer( - sql("SELECT a, count(2) FROM testData2 GROUP BY a, 2"), - Seq(Row(1, 2), Row(2, 2), Row(3, 2))) + sql("SELECT a, sum(b) FROM testData2 GROUP BY 1"), + sql("SELECT a, sum(b) FROM testData2 GROUP BY a")) + // duplicate group-by columns checkAnswer( sql("SELECT a, 1, sum(b) FROM testData2 GROUP BY a, 1"), sql("SELECT a, 1, sum(b) FROM testData2 GROUP BY a")) + + checkAnswer( + sql("SELECT a, 1, sum(b) FROM testData2 GROUP BY 1, 2"), + sql("SELECT a, 1, sum(b) FROM testData2 GROUP BY a")) + } + + test("Group By Ordinal - non aggregate expressions") { + checkAnswer( + sql("SELECT a, b + 2, count(2) FROM testData2 GROUP BY a, 2"), + sql("SELECT a, b + 2, count(2) FROM testData2 GROUP BY a, b + 2")) + + checkAnswer( + sql("SELECT a, b + 2, count(2) FROM testData2 GROUP BY a, b + 2"), + Seq(Row(1, 3, 1), Row(1, 4, 1), Row(2, 3, 1), Row(2, 4, 1), Row(3, 3, 1), Row(3, 4, 1))) + } + + test("Group By Ordinal - non-foldable constant expression") { + checkAnswer( + sql("SELECT a, b, sum(b) FROM testData2 GROUP BY a, b, 1 + 0"), + sql("SELECT a, b, sum(b) FROM testData2 GROUP BY a, b")) + checkAnswer( sql("SELECT a, 1, sum(b) FROM testData2 GROUP BY a, 1 + 2"), sql("SELECT a, 1, sum(b) FROM testData2 GROUP BY a")) + } + + test("Group By Ordinal - alias") { + checkAnswer( + sql("SELECT a, (b + 2) as c, count(2) FROM testData2 GROUP BY a, 2"), + sql("SELECT a, b + 2, count(2) FROM testData2 GROUP BY a, b + 2")) + + checkAnswer( + sql("SELECT a as b, b as a, sum(b) FROM testData2 GROUP BY 1, 2"), + sql("SELECT a, b, sum(b) FROM testData2 GROUP BY a, b")) + } + + test("Group By Ordinal - constants") { checkAnswer( sql("SELECT 1, 2, sum(b) FROM testData2 GROUP BY 1, 2"), sql("SELECT 1, 2, sum(b) FROM testData2")) } + test("Group By Ordinal - negative cases") { + intercept[UnresolvedException[Aggregate]] { + sql("SELECT a, b FROM testData2 GROUP BY -1") + } + + intercept[UnresolvedException[Aggregate]] { + sql("SELECT a, b FROM testData2 GROUP BY 3") + } + + val e = intercept[UnresolvedException[Aggregate]]( + sql("SELECT SUM(a) FROM testData2 GROUP BY 1")) + assert(e.getMessage contains + "Invalid call to Group by position: the '1'th column in the select is an aggregate function") + + var ae = intercept[AnalysisException]( + sql("SELECT a, rand(0), sum(b) FROM testData2 GROUP BY a, 2")) + assert(ae.getMessage contains + "nondeterministic expression rand(0) should not appear in grouping expression") + + ae = intercept[AnalysisException]( + sql("SELECT * FROM testData2 GROUP BY a, b, 1")) --- End diff -- The latest merge removed it. : ) Let me add it back. Thanks! BTW, the latest test build will fail.
--- 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