Bug #30391 Result file for "nist" suite must be updated after bug fix

• Submitted: 13 Aug 2007 15:34
• Modified: 13 Aug 2007 20:32
• Reporter: Joerg Bruehe
• Status: Closed
• Category: MySQL Server: Tests
• Severity: S3 (Non-critical)
• Version: 5.1.21
• OS: Any
• Assigned to: Matthias Leich
• CPU Architecture: Any
• Tags: sr5_1

[13 Aug 2007 15:34] Joerg Bruehe

Description:
The NIST suite needs its result files updated:

bug#30310 is fixed, so this test now gives the correct result:

-------------------------------------------------------
*** /PATH/mysql-test/suite/nist/r/nist_all.result
--- /PATH/mysql-test/suite/nist/r/nist_all.reject
***************
*** 2269,2281
  SELECT COUNT(*) FROM INFORMATION_SCHEMA.SCHEMATA
  WHERE SCHEMA_NAME = 'CTS1B' ;
  COUNT(*)
! 1

! TEST-INFO: NIST expectations violated (was to be expected), result got:1 expected: 0
  Please check nist-MySQL.txt  48

! TEST-INFO: passed (MYSQL expectations)

  CREATE SCHEMA CTS1B;
  CREATE TABLE CTS1B.STAFF7 (EMPNUM    CHAR(3) NOT NULL,
--- 2269,2284
  SELECT COUNT(*) FROM INFORMATION_SCHEMA.SCHEMATA
  WHERE SCHEMA_NAME = 'CTS1B' ;
  COUNT(*)
! 0

! TEST-INFO: passed (NIST expectations)
  Please check nist-MySQL.txt  48

! FIXME: The last statement failed in history, but is now successful.
! That means MySQL improved his behaviour and we should alter the test case.
! Please drop the implemented circumviation and generate
! new reference files

  CREATE SCHEMA CTS1B;
  CREATE TABLE CTS1B.STAFF7 (EMPNUM    CHAR(3) NOT NULL,
-------------------------------------------------------

How to repeat:
Run the NIST suite using 5.1.21 (or higher).

[13 Aug 2007 20:32] Matthias Leich

A fix for this bug is pushed to mysql-test-extra-5.1.
  ChangeSet@1.317.1.1, 2007-08-13
There is no need to document anything, because this
was a combination of a most probably somewhere documented 
server improvement and bugs within a testsuite.