Bug#15098: CAST(column double TO signed int), wrong result
which was fixed originally in 4.1.

into  mysql.com:/usr/home/bar/mysql-5.0

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0

  After merge fix.

  Ndbapi can reorder recattr when doing sorted scan (for faster comparision)
    make sure handler maps correct recattr to correct mysql column

into  mysql.com:/usr/home/bar/mysql-5.0
Bug#15098

into  rurik.mysql.com:/home/igor/mysql-5.0

An invalid assertion in Item_direct_view_ref::eq caused 
an assertion abort in the debug version.

into  zim.(none):/home/ian/workspace/mysql-5.0

into mysql.com:/Users/kent/mysql/bk/mysql-5.0-new

into mysql.com:/Users/kent/mysql/bk/mysql-5.0-new

  Remove Solaris -xc99=none C option as C++ compiler can't handle it

in older version.

The problem is that TRN-files created in "old" versions contain
junk in trigger_table field, which is not acceptable in "new" versions.

into mysql.com:/Users/kent/mysql/bk/mysql-5.0-new

  We are not to control what malloc package others are to use, remove -lmtmalloc from --libs(_r), bug#18322

into  mysql.com:/home/my/mysql-5.0

into  mysql.com:/space/pekka/ndb/version/my50

into  mysql.com:/home/my/mysql-5.0

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0

  Reproduced in 5.0.16-nt. Tested to work in 5.0.20
  

into  mysql.com:/home/my/mysql-5.0

This fixes the problem if someone is using struct MYSQL as part of another structure together with a shared
library

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0

into  perch.ndb.mysql.com:/home/jonas/src/mysql-4.1

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0

into  perch.ndb.mysql.com:/home/jonas/src/50-work

  Change semantic on DumpStateOrd::CmvmiSetRestartOnErrorInsert()
    Called wo/ args it resets to value in configuration (previously it set to 1 if called wo/ args)

into  rurik.mysql.com:/home/igor/dev/mysql-5.0-0

If the WHERE condition of a query contained an ORed FALSE term
then the set of tables whose rows cannot serve for null complements
in outer joins was determined incorrectly. It resulted in blocking 
possible conversions of outer joins into joins for such queries.

tables corrupt triggers".

It turned out that we also have relied at certain places that
(new_table != table_name) were always true on Windows and for transactional
tables. Since our fix for the bug brakes this assumption we have to add new
flag to pass this information around.
This code needs to be refactored but I dare not to do this in 5.0.

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0

into  mysql.com:/home/dlenev/mysql-5.0-bg18153