Uploaded image for project: 'MariaDB ColumnStore'
  1. MariaDB ColumnStore
  2. MCOL-4613

Garbage result of a union between huge narrow DECIMAL and BIGINT

Details

    • 2021-7

    Description

      This ticket is for 5.6.1 only. Testing in 6.1.1 will need to wait until MCOL-4612 is MERGED into develop branch

      DROP TABLE IF EXISTS t1;
      		 
      CREATE TABLE t1 (a DECIMAL(17,1), b BIGINT) ENGINE=ColumnStore;
      		 
      INSERT INTO t1 VALUES (9999999999999999.9, 999999999999999999);
      		 
      SELECT * FROM (SELECT a FROM t1 UNION SELECT b FROM t1) tu;


      +-----------------------+
      		 
      | a                     |
      		 
      +-----------------------+
      		 
      |    9999999999999999.9 |
      		 
      | -844674407370955162.6 |
      		 
      +-----------------------+

      Looks wrong.

      Note, the problem cannot be repeated in 6.x because of MCOL-4612
      But perhaps it still exists. MCOL-4612 needs to be fixed first to verify it further in 6.x.

      Attachments

        Issue Links

          is blocked by

          Bug - A problem which impairs or prevents the functions of the product. MCOL-4612 A subquery with a union for DECIMAL and BIGINT returns zeros

          • Blocker - Blocks development and/or testing work, production could not run.
          • Closed
          relates to

          Task - A task that needs to be done. MCOL-4361 Replace pow(10.0, (double)scale) expressions with a static dictionary lookup.

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Closed

          New Feature - A new feature of the product, which has yet to be developed. MCOL-641 Full DECIMAL support in ColumnStore

          • Major - Major loss of function.
          • Closed

          Bug - A problem which impairs or prevents the functions of the product. MCOL-4700 Wrong result of a UNION for INT and INT UNSIGNED

          • Major - Major loss of function.
          • Closed
          split to

          Task - A task that needs to be done. MCOL-4705 Retest MCOL-4613 in 6.1.1 given that the blocking fix for MCOL-4612 is merged into develop branch

          • Blocker - Blocks development and/or testing work, production could not run.
          • Closed

          Activity

            Ascending order - Click to sort in descending order
            tntnatbry Gagan Goel (Inactive) added a comment -

            For QA: Queries to confirm the issue and verify the fix works are in the issue description.

            tntnatbry Gagan Goel (Inactive) added a comment - For QA: Queries to confirm the issue and verify the fix works are in the issue description.
            tntnatbry Gagan Goel (Inactive) added a comment -

            The fix has some loss of precision for large values. This is because we set the union type for a BIGINT and a DECIMAL column as a DOUBLE. This limitation will be addressed in MCOL-4612 when we set the union type to a wide decimal.

            tntnatbry Gagan Goel (Inactive) added a comment - The fix has some loss of precision for large values. This is because we set the union type for a BIGINT and a DECIMAL column as a DOUBLE. This limitation will be addressed in MCOL-4612 when we set the union type to a wide decimal.
            dleeyh Daniel Lee (Inactive) added a comment -

            Build verified: 5.6.1 ( Drone #2245)

            Precision issue will be tracked in mentioned tickets.

            MariaDB [mytest]> INSERT INTO t1 VALUES (9999999999999999.9, 999999999999999999);
            		 
            Query OK, 1 row affected (0.354 sec)
            		 
             
            		 
            MariaDB [mytest]> SELECT * FROM (SELECT a FROM t1 UNION SELECT b FROM t1) tu;
            		 
            +----------------------+
            		 
            | a                    |
            		 
            +----------------------+
            		 
            | 999999999999999999.0 |
            		 
            |   9999999999999999.0 |
            		 
            +----------------------+
            		 
            2 rows in set (0.070 sec)

            dleeyh Daniel Lee (Inactive) added a comment - Build verified: 5.6.1 ( Drone #2245) Precision issue will be tracked in mentioned tickets. MariaDB [mytest]> INSERT INTO t1 VALUES (9999999999999999.9, 999999999999999999); Query OK, 1 row affected (0.354 sec)   MariaDB [mytest]> SELECT * FROM (SELECT a FROM t1 UNION SELECT b FROM t1) tu; +----------------------+ | a | +----------------------+ | 999999999999999999.0 | | 9999999999999999.0 | +----------------------+ 2 rows in set (0.070 sec)

            People

              dleeyh Daniel Lee (Inactive)
              bar Alexander Barkov
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved:

                Git Integration

                  Error rendering 'com.xiplink.jira.git.jira_git_plugin:git-issue-webpanel'. Please contact your Jira administrators.