into narttu.mysql.fi:/my/mysql-4.0

  NATIONAL CHARACTER() has latin1 charset now.
  Reasons are (by Monty):
  I think we should keepy having latin1 (default configure option)
  as the national character set.  The reason for this is simple that:

- This is ok for a large part of the world.
- It's faster and takes less space than utf8.

  LIKE was used against wrong name

into narttu.mysql.fi:/my/mysql-4.1

Fixed bug in allocation memory in key cache. (Memory was not properly aligned which cased core dumps on sparc CPU's)
Changed keycache variables to start with my_ instead of _my_

  Fixed that concat() didn't set character set in val_str
  concat_ws() now honors coercibility

  The result didn't return proper collation

  A bug has been fixed. ELT() didn't honor the second argument.
  Coercibility support has been added.

   (Thanks to Marc Liyanage for spotting this)

into mysql.com:/suse80/my/mysql-4.0

cause compatibility problems. Fix will be enabled in 4.1.

e.g. default value is not accepted with auto_increment anymore.

at the same time.

    Now it does not SEGV, but USER() is still badly replicated
    (it is replicated to ""), which is a lower priority bug.

into mysql.com:/home/mysql_src/mysql-4.1

into narttu.mysql.fi:/my/mysql-4.1

  add foreign key value update test

  IFNULL() now honors collations

  Reduce/reduce conflict
  fix. 

Some speed improvements

into bar.mysql.r18.ru:/usr/home/bar/mysql-4.1

into narttu.mysql.fi:/my/mysql-4.1

into bar.mysql.r18.ru:/usr/home/bar/mysql-4.1

"a LIKE b COLLATE c"  worked as "(a LIKE b) COLLATE c" which is not right