Ticket Hash: | ced41c7c7d6b4d362bea7b4283515d77962fdb2a | ||
Title: | Query with DISTINCT does not fetch all distinct rows | ||
Status: | Fixed | Type: | Code_Defect |
Severity: | Severe | Priority: | Immediate |
Subsystem: | Unknown | Resolution: | Fixed |
Last Modified: | 2019-07-30 01:17:23 | ||
Version Found In: | 3.29.0 |
User Comments: | ||||
mrigger added on 2019-07-29 22:42:21:
Consider the test case below: CREATE TABLE t1 (c1 , c2, c3, c4 , PRIMARY KEY (c4, c3)); INSERT INTO t1(c3) VALUES (0), (0), (0), (0), (0), (0), (0), (0), (0), (0), (NULL), (1), (0); UPDATE t1 SET c2 = 0; INSERT INTO t1(c1) VALUES (0), (0), (NULL), (0), (0); ANALYZE t1; UPDATE t1 SET c3 = 1; SELECT DISTINCT * FROM t1 WHERE t1.c3 = 1; -- expected: |0|1|, 0||1|, ||1|, actual: |0|1| I would expect the query to fetch the three distinct rows, however, only one is fetched. If the ANALYZE is omitted, then all three distinct rows are fetched. |