[SPARK-18871][SQL][TESTS] New test cases for IN/NOT IN subquery 3rd b…

…atch ## What changes were proposed in this pull request? This is 3ird batch of test case for IN/NOT IN subquery. In this PR, it has these test files: `in-having.sql` `in-joins.sql` `in-multiple-columns.sql` These are the queries and results from running on DB2. [in-having DB2 version](https://github.com/apache/spark/files/772668/in-having.sql.db2.txt) [output of in-having](https://github.com/apache/spark/files/772670/in-having.sql.db2.out.txt) [in-joins DB2 version](https://github.com/apache/spark/files/772672/in-joins.sql.db2.txt) [output of in-joins](https://github.com/apache/spark/files/772673/in-joins.sql.db2.out.txt) [in-multiple-columns DB2 version](https://github.com/apache/spark/files/772678/in-multiple-columns.sql.db2.txt) [output of in-multiple-columns](https://github.com/apache/spark/files/772680/in-multiple-columns.sql.db2.out.txt) ## How was this patch tested? This pr is adding new test cases. We compare the result from spark with the result from another RDBMS(We used DB2 LUW). If the results are the same, we assume the result is correct. Author: Kevin Yu <[email protected]> Closes apache#16841 from kevinyu98/spark-18871-33.
tusharm · Feb 16, 2017 · 3871d94 · 3871d94
1 parent f041e55
commit 3871d94
Show file tree

Hide file tree

Showing 6 changed files with 1,297 additions and 0 deletions.
diff --git a/sql/core/src/test/resources/sql-tests/inputs/subquery/in-subquery/in-having.sql b/sql/core/src/test/resources/sql-tests/inputs/subquery/in-subquery/in-having.sql
@@ -0,0 +1,152 @@
+-- A test suite for IN HAVING in parent side, subquery, and both predicate subquery
+-- It includes correlated cases.
+
+create temporary view t1 as select * from values
+  ("val1a", 6S, 8, 10L, float(15.0), 20D, 20E2, timestamp '2014-04-04 01:00:00.000', date '2014-04-04'),
+  ("val1b", 8S, 16, 19L, float(17.0), 25D, 26E2, timestamp '2014-05-04 01:01:00.000', date '2014-05-04'),
+  ("val1a", 16S, 12, 21L, float(15.0), 20D, 20E2, timestamp '2014-06-04 01:02:00.001', date '2014-06-04'),
+  ("val1a", 16S, 12, 10L, float(15.0), 20D, 20E2, timestamp '2014-07-04 01:01:00.000', date '2014-07-04'),
+  ("val1c", 8S, 16, 19L, float(17.0), 25D, 26E2, timestamp '2014-05-04 01:02:00.001', date '2014-05-05'),
+  ("val1d", null, 16, 22L, float(17.0), 25D, 26E2, timestamp '2014-06-04 01:01:00.000', null),
+  ("val1d", null, 16, 19L, float(17.0), 25D, 26E2, timestamp '2014-07-04 01:02:00.001', null),
+  ("val1e", 10S, null, 25L, float(17.0), 25D, 26E2, timestamp '2014-08-04 01:01:00.000', date '2014-08-04'),
+  ("val1e", 10S, null, 19L, float(17.0), 25D, 26E2, timestamp '2014-09-04 01:02:00.001', date '2014-09-04'),
+  ("val1d", 10S, null, 12L, float(17.0), 25D, 26E2, timestamp '2015-05-04 01:01:00.000', date '2015-05-04'),
+  ("val1a", 6S, 8, 10L, float(15.0), 20D, 20E2, timestamp '2014-04-04 01:02:00.001', date '2014-04-04'),
+  ("val1e", 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);
+
+create temporary view t2 as select * from values
+  ("val2a", 6S, 12, 14L, float(15), 20D, 20E2, timestamp '2014-04-04 01:01:00.000', date '2014-04-04'),
+  ("val1b", 10S, 12, 19L, float(17), 25D, 26E2, timestamp '2014-05-04 01:01:00.000', date '2014-05-04'),
+  ("val1b", 8S, 16, 119L, float(17), 25D, 26E2, timestamp '2015-05-04 01:01:00.000', date '2015-05-04'),
+  ("val1c", 12S, 16, 219L, float(17), 25D, 26E2, timestamp '2016-05-04 01:01:00.000', date '2016-05-04'),
+  ("val1b", null, 16, 319L, float(17), 25D, 26E2, timestamp '2017-05-04 01:01:00.000', null),
+  ("val2e", 8S, null, 419L, float(17), 25D, 26E2, timestamp '2014-06-04 01:01:00.000', date '2014-06-04'),
+  ("val1f", 19S, null, 519L, float(17), 25D, 26E2, timestamp '2014-05-04 01:01:00.000', date '2014-05-04'),
+  ("val1b", 10S, 12, 19L, float(17), 25D, 26E2, timestamp '2014-06-04 01:01:00.000', date '2014-06-04'),
+  ("val1b", 8S, 16, 19L, float(17), 25D, 26E2, timestamp '2014-07-04 01:01:00.000', date '2014-07-04'),
+  ("val1c", 12S, 16, 19L, float(17), 25D, 26E2, timestamp '2014-08-04 01:01:00.000', date '2014-08-05'),
+  ("val1e", 8S, null, 19L, float(17), 25D, 26E2, timestamp '2014-09-04 01:01:00.000', date '2014-09-04'),
+  ("val1f", 19S, null, 19L, float(17), 25D, 26E2, timestamp '2014-10-04 01:01:00.000', date '2014-10-04'),
+  ("val1b", 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);
+
+create temporary view t3 as select * from values
+  ("val3a", 6S, 12, 110L, float(15), 20D, 20E2, timestamp '2014-04-04 01:02:00.000', date '2014-04-04'),
+  ("val3a", 6S, 12, 10L, float(15), 20D, 20E2, timestamp '2014-05-04 01:02:00.000', date '2014-05-04'),
+  ("val1b", 10S, 12, 219L, float(17), 25D, 26E2, timestamp '2014-05-04 01:02:00.000', date '2014-05-04'),
+  ("val1b", 10S, 12, 19L, float(17), 25D, 26E2, timestamp '2014-05-04 01:02:00.000', date '2014-05-04'),
+  ("val1b", 8S, 16, 319L, float(17), 25D, 26E2, timestamp '2014-06-04 01:02:00.000', date '2014-06-04'),
+  ("val1b", 8S, 16, 19L, float(17), 25D, 26E2, timestamp '2014-07-04 01:02:00.000', date '2014-07-04'),
+  ("val3c", 17S, 16, 519L, float(17), 25D, 26E2, timestamp '2014-08-04 01:02:00.000', date '2014-08-04'),
+  ("val3c", 17S, 16, 19L, float(17), 25D, 26E2, timestamp '2014-09-04 01:02:00.000', date '2014-09-05'),
+  ("val1b", null, 16, 419L, float(17), 25D, 26E2, timestamp '2014-10-04 01:02:00.000', null),
+  ("val1b", null, 16, 19L, float(17), 25D, 26E2, timestamp '2014-11-04 01:02:00.000', null),
+  ("val3b", 8S, null, 719L, float(17), 25D, 26E2, timestamp '2014-05-04 01:02:00.000', date '2014-05-04'),
+  ("val3b", 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);
+
+-- correlated IN subquery
+-- HAVING in the subquery
+-- TC 01.01
+SELECT t1a,
+       t1b,
+       t1h
+FROM   t1
+WHERE  t1b IN (SELECT t2b
+               FROM   t2
+               GROUP BY t2b
+               HAVING t2b < 10);
+
+-- TC 01.02
+SELECT t1a,
+       t1b,
+       t1c
+FROM   t1
+WHERE  t1b IN (SELECT Min(t2b)
+               FROM   t2
+               WHERE  t1a = t2a
+               GROUP  BY t2b
+               HAVING t2b > 1);
+
+-- HAVING in the parent
+-- TC 01.03
+SELECT t1a, t1b, t1c
+FROM   t1
+WHERE  t1b IN (SELECT t2b
+               FROM   t2
+               WHERE t1c < t2c)
+GROUP BY t1a, t1b, t1c
+HAVING t1b < 10;
+
+-- TC 01.04
+SELECT t1a, t1b, t1c
+FROM   t1
+WHERE  t1b IN (SELECT t2b
+               FROM   t2
+               WHERE t1c = t2c)
+GROUP BY t1a, t1b, t1c
+HAVING COUNT (DISTINCT t1b) < 10;
+
+-- BOTH
+-- TC 01.05
+SELECT Count(DISTINCT( t1a )),
+       t1b
+FROM   t1
+WHERE  t1c IN (SELECT t2c
+               FROM   t2
+               WHERE  t1a = t2a
+               GROUP BY t2c
+               HAVING t2c > 10)
+GROUP  BY t1b
+HAVING t1b >= 8;
+
+-- TC 01.06
+SELECT t1a,
+       Max(t1b)
+FROM   t1
+WHERE  t1b > 0
+GROUP  BY t1a
+HAVING t1a IN (SELECT t2a
+               FROM   t2
+               WHERE  t2b IN (SELECT t3b
+                              FROM   t3
+                              WHERE  t2c = t3c)
+               );
+
+-- HAVING clause with NOT IN
+-- TC 01.07
+SELECT t1a,
+       t1c,
+       Min(t1d)
+FROM   t1
+WHERE  t1a NOT IN (SELECT t2a
+                   FROM   t2
+                   GROUP BY t2a
+                   HAVING t2a > 'val2a')
+GROUP BY t1a, t1c
+HAVING Min(t1d) > t1c;
+
+-- TC 01.08
+SELECT t1a,
+       t1b
+FROM   t1
+WHERE  t1d NOT IN (SELECT t2d
+                   FROM   t2
+                   WHERE  t1a = t2a
+                   GROUP BY t2c, t2d
+                   HAVING t2c > 8)
+GROUP  BY t1a, t1b
+HAVING t1b < 10;
+
+-- TC 01.09
+SELECT t1a,
+       Max(t1b)
+FROM   t1
+WHERE  t1b > 0
+GROUP  BY t1a
+HAVING t1a NOT IN (SELECT t2a
+                   FROM   t2
+                   WHERE  t2b > 3);
+