This is to enable table handlers to implement online create/drop index.
It consists of some parts:
- New default handler methods in handler.h
- Split of mysql_alter_table. It decides if only one kind of
  alteration is to be done (e.g. only create indexes or only drop
  indexes etc.) It then calls the specialized new handler method if
  the handler implements it. Otherwise it calls real_alter_table.
- The parser sets flags for each alter operation detected in a
  command. These are used by mysql_alter_table for the decision.
- mysql_prepare_table is pulled out of mysql_create_table. This is
  also used by mysql_create_index to prepare the key structure array
  for the handler. It is also used by mysql_create_index and
  mysql_drop_index to prepare a call to mysql_create_frm.
- mysql_create_frm is pulled out of rea_create_table for use by
  mysql_create_index and mysql_drop_index after the index is
  created/dropped.
Thanks to Antony who supplied most of the changes.

into mysql.com:/home/mydev/mysql-4.1-wl1648

  new file

into gluh.mysql.r18.ru:/home/gluh/mysql-4.1.clear

UCS2 now has its own my_like_range function.

into mysql.com:/space/my/mysql-4.1-build

   not work as expected) - need to investigate (after 4.1.2 is released)

I am fixing it here in 4.1, not in 4.0, as it goes with another change
which I did in 4.1 (fix for BUG#3204
""mysqlbinlog --read-from-remote-server this_binlog.001" prints all binlogs").

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

we now detect that the server is sending us a log which we did not request
by testing the info in the fake Rotate event.
I also changed code to not print the fake Rotate which describes the
log we asked for (it's always the first received event but old masters
may not send it).

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

  Update an obsolete comment about trx commit: we can no longer call trx_undo_update_cleanup_by_discard(), and actually, the idea to call it was erroneous, it cannot work

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

Changed _XXX to _MY_XXX to solve conflict problem on Mac OS X

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

   instead of calling them directly (WL#1521) - this also fixed a compile
   error with libtool-1.5.2 on FreeBSD (error: tag name "CXX" already 
   exists)

into gluh.mysql.r18.ru:/home/gluh/mysql-4.1.clear

into serg.mylan:/usr/home/serg/Abk/mysql-4.1

  Remove includes of sys/types.h and sys/stat.h because these are already included through univ.i

Portability fixes

  Add a comment that InnoDB always retrieves the row id, if the clustered index was internally generated by InnoDB. Then we know the 'ref' to the row.

into teton.kitebird.com:/home/paul/mysql-4.1

  Fix Bug #1942: do not retrieve all columns in a table if we only need the 'ref' of the row (usually, the PRIMARY KEY) to calculate an ORDER BY

into teton.kitebird.com:/home/paul/mysql-4.1

into teton.kitebird.com:/home/paul/mysql-4.1

Added test case for the bug #3155: strange results with index (x, y) ... WHERE x=val_1 AND y>=val_2 ORDER BY pk;
(thankee Monty)

- added unsigned_flag= 1; to Item_uint constructors
- removed Item_uint::fix_fields()
- added a comment about fix_fields() call

into gluh.mysql.r18.ru:/home/gluh/mysql-4.1.clear

Fix for the bug #3155: strange results with index (x, y) ... WHERE x=val_1 AND y>=val_2 ORDER BY pk;
(test case is too big to be included, take a look at the bug database)