Added push_back(void *, MEM_ROOT *) to make some list-handling code easier that needs to be allocated in a different mem-root
(Before one had to change thd->mem_root ; push_back(); restore mem_root.

locks all rows (BUG #3300). When using innobase_locks_unsafe_for_binlog
option InnoDB does not take locks for those rows which do not
belong to the result set or werent changed by the query. This fix removes
unnecessary locks also from SELECT and DELETE queries.

into sanja.is.com.ua:/home/bell/mysql/bk/work-update-5.0

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

Now thd->mem_root is a pointer to thd->main_mem_root and THR_MALLOC is a pointer to thd->mem_root.
This gives us the following benefits:
- Allow us to easily detect if arena has already been swapped before (this fixes a bug in setup_conds() where arena was swaped twice in some cases)
- Faster swaps of arenas (as we don't have to copy the whole MEM_ROOT)
- We don't anymore have to call my_pthread_setspecific_ptr(THR_MALLOC,...) to change where memory is alloced. Now it's enough to set thd->mem_root

into poseidon.ndb.mysql.com:/home/tomas/mysql-4.1

  Included header fnmatch.h on Unix
  Changed C++ comments to C comments
  Corrected indentation of code written on Windows
  Split up lines to fit into 80 columns
  Initiated some variables to avoid warnings
  Added __attribute__((unused)) to unused function parameters
  Replace tab characters with space
  Put space after 'for', 'while' etc
  Added value to 'return' from non void function removef()
  On Unix strlwr() was incorrectly declared and a no op,
  replaced it with a macro that does nothing
  Split several statements on the same line
  Other minor changes to conform to coding standard

into avenger.(none):/export/brian/mysql/archive-5.0

Removed duplicate code where defaults in handler.h will work just as well. 

  Added cases for bugs #6307 and #6460.
sql_select.cc:
  Fixed the problem of bug reports #6307 and #6460.
  The reported wrong result sets were due to the fact that
  the added call of the fix_fields method for the built
  AND condition that joined WHERE and ON conditions
  broke ON expression, as it removed extra AND levels
  in the built condition. 
  It looks like that no attributes of the built condition
  are needed, so we don't have to call fix_fields here.

into poseidon.ndb.mysql.com:/home/tomas/mysql-4.1

and blank after function name".
Crop fruits of copy-paste programming: pre-caching of stored functions
wasn't performed for prepared statements just because implementation
of prepared statements is done as an add-on to the main execution flow, 
and the preload was originally implemented for main execution branch
only (mysql_execute_command).

of client equals to character set of connection, possibly required
conversion to character set of column is not performed
(prepared statements, data is supplied using placeholders).

  in 4.1: Dump only VARBINARY() in hex,
  while VARCHAR() COLLATE xxx_bin as a string.

- open and create derived tables
- detect which tables should be locked for write
- lock and fill derived tables
some unitialized variables fixed

into mysql.com:/M41/mysql-4.1

  My previous change that "set @A=NULL" doesn't change charset
  fixed 'Bug #6321' as well. Prove with a new test that
  FIELD(<uservariable content NULL>, ...) now works fine too.

_all_ versions. It tries to remove "Docs/Images/Makefile*" as part of the actions to
correct bug number 6350. If the source tree does not contain a dummy file matching
that pattern, this "rm" will fail, and so "Bootstrap" will fail.
The "Makefile.am" introduced with this changeset is identical to the one that will
arrive here when "joerg:1.2057" is propagated from 4.0,
so this separate introduction here is just to allow "Bootstraps" in the meantime.

character set with NULL, @A should be latin2
after this query sequence:

   SET @A=_latin2'string';
   SET @A=NULL;

I.e. the second query should not change the charset
to the current default value, but should keep the
original value assigned during the first query.
In order to do it, we don't copy charset
from the argument if the argument is NULL
and the variable has previously been initialized.

_all_ versions. It tries to remove "Docs/Images/Makefile*" as part of the actions to
correct bug number 6350. If the source tree does not contain a dummy file matching
that pattern, this "rm" will fail, and so "Bootstrap" will fail.
The "Makefile.am" introduced with this changeset is identical to the one that will
arrive in the 5.0 tree when "joerg:1.2057" is propagated from 4.0 via 4.1 to here,
so this separate introduction here is just to allow "Bootstraps" in the meantime.

into mysql.com:/usr/local/home/marty/MySQL/test/mysql-4.1

into rurik.mysql.com:/home/igor/mysql-4.1

  Bug #6351 make test failure "Unknown character set"
  UCS2 related tests were moved into ctype_ucs.

  - 'mysqldump --help' comment that --xeh-blob
    doesn't work with --extended-inserts was removed.
    It does work now. Thanks to Lachlan
    who noticed this wrong help message.
  - Switched to use the recently introduced
    mysql_hex_string() instead of slow sprintf().
    Thanks to Sinisa for the idea.
  - 'mysqldump --hex-blob' dumps in HEX not only
    BLOBs but also BINARY(x) columns. 
    Thanks to Paul.

  Minor clean-ups

into rurik.mysql.com:/home/igor/dev/mysql-4.1-0

  Added a case for bug #6365.
item_cmpfunc.cc:
  Fixed bug #6365 : Server crashed when list of values
  in IN predicate contains NULL while the tested field is
  of the character type and not of the default set;
  e.g. when f in 'f IN (NULL,'aa') belongs to binary
  character set, while the default character set is latin1.

  Missing initialized for "lengths" element.
  ,