Github user gatorsmile commented on a diff in the pull request: https://github.com/apache/spark/pull/16337#discussion_r94718082 --- Diff: sql/core/src/test/resources/sql-tests/results/subquery/in-subquery/in-group-by.sql.out --- @@ -0,0 +1,357 @@ +-- Automatically generated by SQLQueryTestSuite +-- Number of queries: 19 + + +-- !query 0 +create temporary view t1 as select * from values + ("t1a", 6S, 8, 10L, float(15.0), 20D, 20E2, timestamp '2014-04-04 01:00:00.000', date '2014-04-04'), + ("t1b", 8S, 16, 19L, float(17.0), 25D, 26E2, timestamp '2014-05-04 01:01:00.000', date '2014-05-04'), + ("t1a", 16S, 12, 21L, float(15.0), 20D, 20E2, timestamp '2014-06-04 01:02:00.001', date '2014-06-04'), + ("t1a", 16S, 12, 10L, float(15.0), 20D, 20E2, timestamp '2014-07-04 01:01:00.000', date '2014-07-04'), + ("t1c", 8S, 16, 19L, float(17.0), 25D, 26E2, timestamp '2014-05-04 01:02:00.001', date '2014-05-05'), + ("t1d", null, 16, 22L, float(17.0), 25D, 26E2, timestamp '2014-06-04 01:01:00.000', null), + ("t1d", null, 16, 19L, float(17.0), 25D, 26E2, timestamp '2014-07-04 01:02:00.001', null), + ("t1e", 10S, null, 25L, float(17.0), 25D, 26E2, timestamp '2014-08-04 01:01:00.000', date '2014-08-04'), + ("t1e", 10S, null, 19L, float(17.0), 25D, 26E2, timestamp '2014-09-04 01:02:00.001', date '2014-09-04'), + ("t1d", 10S, null, 12L, float(17.0), 25D, 26E2, timestamp '2015-05-04 01:01:00.000', date '2015-05-04'), + ("t1a", 6S, 8, 10L, float(15.0), 20D, 20E2, timestamp '2014-04-04 01:02:00.001', date '2014-04-04'), + ("t1e", 10S, null, 19L, float(17.0), 25D, 26E2, timestamp '2014-05-04 01:01:00.000', date '2014-05-04') + as t1(t1a, t1b, t1c, t1d, t1e, t1f, t1g, t1h, t1i) +-- !query 0 schema +struct<> +-- !query 0 output + + + +-- !query 1 +create temporary view t2 as select * from values + ("t2a", 6S, 12, 14L, float(15), 20D, 20E2, timestamp '2014-04-04 01:01:00.000', date '2014-04-04'), + ("t1b", 10S, 12, 19L, float(17), 25D, 26E2, timestamp '2014-05-04 01:01:00.000', date '2014-05-04'), + ("t1b", 8S, 16, 119L, float(17), 25D, 26E2, timestamp '2015-05-04 01:01:00.000', date '2015-05-04'), + ("t1c", 12S, 16, 219L, float(17), 25D, 26E2, timestamp '2016-05-04 01:01:00.000', date '2016-05-04'), + ("t1b", null, 16, 319L, float(17), 25D, 26E2, timestamp '2017-05-04 01:01:00.000', null), + ("t2e", 8S, null, 419L, float(17), 25D, 26E2, timestamp '2014-06-04 01:01:00.000', date '2014-06-04'), + ("t1f", 19S, null, 519L, float(17), 25D, 26E2, timestamp '2014-05-04 01:01:00.000', date '2014-05-04'), + ("t1b", 10S, 12, 19L, float(17), 25D, 26E2, timestamp '2014-06-04 01:01:00.000', date '2014-06-04'), + ("t1b", 8S, 16, 19L, float(17), 25D, 26E2, timestamp '2014-07-04 01:01:00.000', date '2014-07-04'), + ("t1c", 12S, 16, 19L, float(17), 25D, 26E2, timestamp '2014-08-04 01:01:00.000', date '2014-08-05'), + ("t1e", 8S, null, 19L, float(17), 25D, 26E2, timestamp '2014-09-04 01:01:00.000', date '2014-09-04'), + ("t1f", 19S, null, 19L, float(17), 25D, 26E2, timestamp '2014-10-04 01:01:00.000', date '2014-10-04'), + ("t1b", null, 16, 19L, float(17), 25D, 26E2, timestamp '2014-05-04 01:01:00.000', null) + as t2(t2a, t2b, t2c, t2d, t2e, t2f, t2g, t2h, t2i) +-- !query 1 schema +struct<> +-- !query 1 output + + + +-- !query 2 +create temporary view t3 as select * from values + ("t3a", 6S, 12, 110L, float(15), 20D, 20E2, timestamp '2014-04-04 01:02:00.000', date '2014-04-04'), + ("t3a", 6S, 12, 10L, float(15), 20D, 20E2, timestamp '2014-05-04 01:02:00.000', date '2014-05-04'), + ("t1b", 10S, 12, 219L, float(17), 25D, 26E2, timestamp '2014-05-04 01:02:00.000', date '2014-05-04'), + ("t1b", 10S, 12, 19L, float(17), 25D, 26E2, timestamp '2014-05-04 01:02:00.000', date '2014-05-04'), + ("t1b", 8S, 16, 319L, float(17), 25D, 26E2, timestamp '2014-06-04 01:02:00.000', date '2014-06-04'), + ("t1b", 8S, 16, 19L, float(17), 25D, 26E2, timestamp '2014-07-04 01:02:00.000', date '2014-07-04'), + ("t3c", 17S, 16, 519L, float(17), 25D, 26E2, timestamp '2014-08-04 01:02:00.000', date '2014-08-04'), + ("t3c", 17S, 16, 19L, float(17), 25D, 26E2, timestamp '2014-09-04 01:02:00.000', date '2014-09-05'), + ("t1b", null, 16, 419L, float(17), 25D, 26E2, timestamp '2014-10-04 01:02:00.000', null), + ("t1b", null, 16, 19L, float(17), 25D, 26E2, timestamp '2014-11-04 01:02:00.000', null), + ("t3b", 8S, null, 719L, float(17), 25D, 26E2, timestamp '2014-05-04 01:02:00.000', date '2014-05-04'), + ("t3b", 8S, null, 19L, float(17), 25D, 26E2, timestamp '2015-05-04 01:02:00.000', date '2015-05-04') + as t3(t3a, t3b, t3c, t3d, t3e, t3f, t3g, t3h, t3i) +-- !query 2 schema +struct<> +-- !query 2 output + + + +-- !query 3 +SELECT t1a, + Avg(t1b) +FROM t1 +WHERE t1a IN (SELECT t2a + FROM t2) +GROUP BY t1a +-- !query 3 schema +struct<t1a:string,avg(t1b):double> +-- !query 3 output +t1b 8.0 +t1c 8.0 +t1e 10.0 --- End diff -- When comparing with the DB2 results, the data type of `Avg` is different. Just for the other reviewers, [the document of DB2 ](https://www.ibm.com/support/knowledgecenter/SSEPGG_9.7.0/com.ibm.db2.luw.sql.ref.doc/doc/r0000758.html) explains the behavior: ``` The data type of the result is the same as the data type of the argument values, except that: - The result is a large integer if the argument values are small integers. - The result is double-precision floating point if the argument values are single-precision floating point. - The result is DECFLOAT(34) if the argument is DECFLOAT(n). ```
--- 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