1. 01 Nov, 2006 1 commit
    • unknown's avatar
      Merge mysql.com:/usr/home/bar/mysql-5.0.b18908 · bfdc9c12
      unknown authored
      into  mysql.com:/usr/home/bar/mysql-5.1-rpl
      
      
      BitKeeper/etc/collapsed:
        auto-union
      mysql-test/r/ctype_recoding.result:
        Auto merged
      mysql-test/r/ctype_utf8.result:
        Auto merged
      mysql-test/r/fulltext.result:
        Auto merged
      mysql-test/r/strict_autoinc_1myisam.result:
        Auto merged
      mysql-test/r/strict_autoinc_2innodb.result:
        Auto merged
      mysql-test/r/strict_autoinc_3heap.result:
        Auto merged
      mysql-test/r/strict_autoinc_4bdb.result:
        Auto merged
      mysql-test/r/strict_autoinc_5ndb.result:
        Auto merged
      sql/sql_string.cc:
        Auto merged
      sql/sql_string.h:
        Auto merged
      sql/field.cc:
        SCCS merged
      bfdc9c12
  2. 31 Oct, 2006 2 commits
    • unknown's avatar
      Added new file · a78bdf82
      unknown authored
      a78bdf82
    • unknown's avatar
      Refactoring: Moved rli code to new file rpl_rli.cc. The idea being that rli · 69d92fb1
      unknown authored
      should be a separate module (i.e. a class) to make it easier to maintain the 
      code, e.g. by having checks within the rli checking sanity of data and making 
      member variables private.  This will also ease implementation of multi-source 
      and, at least in my fantasies :), make it possible in some future to have 
      separate replication servers.
      
      
      sql/Makefile.am:
        Added file rpl_rli.cc
      sql/repl_failsafe.cc:
        Move function
      sql/rpl_rli.h:
        moved functions
      sql/slave.cc:
        Moved rli code into rpl_rli.cc
      sql/sql_repl.cc:
        Changed function to be member function
      sql/rpl_rli.cc:
        Code for rli
      69d92fb1
  3. 30 Oct, 2006 3 commits
    • unknown's avatar
      BUG#20697 slave fails to rollback replicated transaction hang over innodb_lock_wait_timeou · b7d43470
      unknown authored
      post-merge fixes in test/result
      
      
      mysql-test/extra/rpl_tests/rpl_deadlock.test:
        recalculating BEGIN's event position.
      mysql-test/r/rpl_deadlock_innodb.result:
        results changed
      b7d43470
    • unknown's avatar
      Merge dsl-hkigw8-feaef900-46.dhcp.inet.fi:/home/elkin/MySQL/TEAM/BARE/5.0-merge · 9fdc9acb
      unknown authored
      into  dsl-hkigw8-feaef900-46.dhcp.inet.fi:/home/elkin/MySQL/TEAM/FIXES/merge_50
      
      
      mysql-test/r/rpl_deadlock_innodb.result:
        results changed
      mysql-test/extra/rpl_tests/rpl_deadlock.test:
        taking a fix for testing timed-out behaviour.
      sql/slave.cc:
        just comments are fixed due to changes in behavior of timed-out msta.
      9fdc9acb
    • unknown's avatar
      Bug#18908: ERROR 1406 (22001): Data too long for column :: using utf8 · 00c1c19e
      unknown authored
      Problem: Too confusing error message when cannot convert
      between string and column character sets on INSERT and UPDATE.
      Fix: producing a better error message, instead of "Data too long"
      in such cases
      Additional changes: Adding "DROP TABLE IF EXISTS" into several
      tests to be safe against failures in previous tests. 
      
      
      mysql-test/include/strict_autoinc.inc:
        Adding DROP TABLE to be safe against previous tests failure.
      mysql-test/r/ctype_recoding.result:
        Fixing test results
      mysql-test/r/ctype_utf8.result:
        Fixing test results
      mysql-test/r/fulltext.result:
        Fixing test results
      mysql-test/r/strict_autoinc_1myisam.result:
        Adding DROP TABLE to be safe against previous tests failure.
      mysql-test/r/strict_autoinc_2innodb.result:
        Adding DROP TABLE to be safe against previous tests failure.
      mysql-test/r/strict_autoinc_3heap.result:
        Adding DROP TABLE to be safe against previous tests failure.
      mysql-test/r/strict_autoinc_4bdb.result:
        Adding DROP TABLE to be safe against previous tests failure.
      mysql-test/r/strict_autoinc_5ndb.result:
        Adding DROP TABLE to be safe against previous tests failure.
      sql/field.cc:
        - producing better error messages than
          "DATA TRUNCATED" or "DATA TOO LONG" (in strict mode)
          in case of "not well formed source" and
          "cannot convert to field character set"
        - Performance improvements: copying directly to
          the target, instead of using an intermediate
          String.
        - Moving duplicate code into report_data_too_long() function.
      sql/sql_string.cc:
        Adding a new function to convert strings between character sets,
        but not more than "nchar" characters - a helper function for
        Field_string::store(), Field_varstring::store() and Field_blob::store().
      sql/sql_string.h:
        Adding new function prototype.
      00c1c19e
  4. 25 Oct, 2006 1 commit
  5. 23 Oct, 2006 3 commits
  6. 20 Oct, 2006 6 commits
  7. 19 Oct, 2006 19 commits
  8. 18 Oct, 2006 3 commits
    • unknown's avatar
      Merge mysql.com:/data0/bk/mysql-5.0 · ea39ca34
      unknown authored
      into  mysql.com:/data0/bk/mysql-5.0-kt
      
      
      ea39ca34
    • unknown's avatar
      BUG#23175 - MYISAM crash/repair failed during repair · 48cf65c0
      unknown authored
      Repair table could crash a server if there is not sufficient
      memory (myisam_sort_buffer_size) to operate. Affects not only
      repair, but also all statements that use create index by sort:
      repair by sort, parallel repair, bulk insert.
      
      Return an error if there is not sufficient memory to store at
      least one key per BUFFPEK.
      
      Also fixed memory leak if thr_find_all_keys returns an error.
      
      
      myisam/sort.c:
        maxbuffer is number of BUFFPEK-s for repair. It is calculated
        as records / keys. keys is number of keys that can be stored
        in memory (myisam_sort_buffer_size). There must be sufficient
        memory to store both BUFFPEK-s and keys. It was checked
        correctly before this patch. However there is another
        requirement that wasn't checked: there must be sufficient
        memory for at least one key per BUFFPEK, otherwise repair
        by sort/parallel repair cannot operate.
        
        Return an error if there is not sufficient memory to store at
        least one key per BUFFPEK.
        
        Also fixed memory leak if thr_find_all_keys returns an error.
      mysql-test/r/repair.result:
        A test case for BUG#23175.
      mysql-test/t/repair.test:
        A test case for BUG#23175.
      48cf65c0
    • unknown's avatar
      Merge abotchkov@bk-internal.mysql.com:/home/bk/mysql-5.1-kt · ddee02ec
      unknown authored
      into  mysql.com:/home/hf/mysql-5.1.mrg
      
      
      ddee02ec
  9. 17 Oct, 2006 2 commits