Details

    • Type: Task
    • Status: Open
    • Priority: Major
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None

      Description

      I got the following failure in RexProgramFuzzyTest. Is it a duplicate issue?

      Failure #1

      2018-09-28 03:52:59,737 [pool-1-thread-1] INFO  - Using seed 44 for rex fuzzing
      2018-09-28 03:52:59,748 [pool-1-thread-1] INFO  - Rex fuzzing results: number of cases tested=18, failed cases=0, duplicate failures=0, fuzz rate=1636 per second
      2018-09-28 03:52:59,751 [pool-1-thread-1] INFO  - Rex fuzzing results: number of cases tested=2, failed cases=1, duplicate failures=0, fuzz rate=666 per second
      2018-09-28 03:52:59,754 [pool-1-thread-1] INFO  - Randomized test identified a potential defect. Feel free to fix that issue
      java.lang.AssertionError: COALESCE(CASE(IS NOT FALSE(true), -(COALESCE(CAST(-1):INTEGER, null, ?0.notNullInt1), *(CAST(100500):INTEGER, ?0.notNullInt0)), -(+(-1990568273))), +(-(COALESCE(null, ?0.notNullInt0, ?0.notNullInt0))))
      coalesce(case_(isNotFalse(trueLiteral), sub(coalesce(rexBuilder.makeCall(SqlStdOperatorTable.CAST, literal(-1)), nullInt, vIntNotNull(1)), rexBuilder.makeCall(SqlStdOperatorTable.*, rexBuilder.makeCall(SqlStdOperatorTable.CAST, literal(100500)), vIntNotNull(0))), unaryMinus(unaryPlus(literal(-1990568273)))), unaryPlus(unaryMinus(coalesce(nullInt, vIntNotNull(0), vIntNotNull(0))))) had non-nullable type INTEGER, and it was optimized to CASE(IS NOT FALSE(true), -(COALESCE(CAST(-1):INTEGER, null, ?0.notNullInt1), *(CAST(100500):INTEGER, ?0.notNullInt0)), -(+(-1990568273)))
      case_(isNotFalse(trueLiteral), sub(coalesce(rexBuilder.makeCall(SqlStdOperatorTable.CAST, literal(-1)), nullInt, vIntNotNull(1)), rexBuilder.makeCall(SqlStdOperatorTable.*, rexBuilder.makeCall(SqlStdOperatorTable.CAST, literal(100500)), vIntNotNull(0))), unaryMinus(unaryPlus(literal(-1990568273)))) that has nullable type INTEGER, unknownAsFalse 

      I suggest that we append new failures to this case. If they are new issues, log new cases, but there is also a good chance that they are variants of, or related to, other open cases.

      In this case, it could be one of the two COALESCE issues already open.

        Attachments

          Activity

            People

            • Assignee:
              Unassigned
              Reporter:
              julianhyde Julian Hyde
            • Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated: