Do not try to resolve the table name of a corrupt page if the code is in ibbackup, i.e., the dict system is not initialized

Fixed bug in LOCK TABLE + DROP TABLE when other thread was waiting for a table that was locked bug not droped

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

  Fix a hang on the adaptive hash S-latch if an application program uses mysql_use_result() and performs queries on two connections at the same time

   packaging
 - enclosed some variables in quotes for safeguarding
 - add a license file (GPL or MySQLEULA) to the package so it will be
   displayed during installation (needs to be tested)

  Print a more precise error message if log files or data files are of a size different from what is specified in my.cnf

+ fix for #70 
+ test result fix for LOAD DATA LOCAL bug #82

  Print trx rollback progress info in crash recovery

into serg.mysql.com:/usr/home/serg/Abk/mysql-4.0

remove variable rli->log_pos_current.

("Event too big" etc), stop this thread instead of going on
with the next event, which would certainly lead to slave's data
corruption.

LOAD DATA LOCAL INFILE was not replicated correctly

   Monty)

increased compatibility
postmerge changing

Fixed bug in RAND() usage in mysqlbinlog

regarded as unsigned longlongss, not as floats. This make these values work
similar to values between 10000000000000000000 and 18446744073709551615.

  Remove DEBUG_ASSERT if MySQL internally names a UNIQUE index as the PRIMARY KEY, but inside InnoDB there is no primary key for the table

  Add docs to store_lock()

into sinisa.nasamreza.org:/mnt/work/mysql-4.0

command:

load  data infile 'filename' into table db.table;

contain NULL values (which is the case for a crashed .frm file).