to auto_increment in 4.1".
Now we are enforcing NO_AUTO_VALUE_ON_ZERO mode during ALTER TABLE only
if we are converting one auto_increment column to another auto_increment
column (this also includes most common case when we don't do anything
with such column).

Also now when we convert some column to TIMESTAMP NOT NULL column with
ALTER TABLE we convert NULL values to current timestamp, (as we do this
in INSERT). One can still get old behavior by setting system TIMESTAMP
variable to 0.


mysql-test/r/auto_increment.result:
  Added tests for ALTER TABLE converting columns containing NULL and 0
  values to AUTO_INCREMENT columns.
mysql-test/r/type_timestamp.result:
  Removed test for creation of TIMESTAMP(19) columns (it is 4.0 specific).
  Added test for ALTER TABLE converting columns containing NULL values
  to TIMESTAMP columns.
mysql-test/t/auto_increment.test:
  Added tests for ALTER TABLE converting columns containing NULL and 0
  values to AUTO_INCREMENT columns.
mysql-test/t/type_timestamp.test:
  Removed test for creation of TIMESTAMP(19) columns (it is 4.0 specific).
  Added test for ALTER TABLE converting columns containing NULL values
  to TIMESTAMP columns.
sql/field_conv.cc:
  Fix bug #5915 "ALTER TABLE behaves differently when converting column
  to auto_increment in 4.1". Also now when we are converting some column
  to TIMESTAMP column, we are converting NULL values to CURRENT_TIMESTAMP
  (as it was initially planned).
  
  do_copy_timestamp(): Fixed comment.
  do_copy_next_number(): We should also set auto_increment_field_not_null
    to FALSE if we have NULL in source field.
  Copy_field::set(): Moved setting of copy functions for TIMESTAMP and
    AUTO_INCREMENT fields to proper place (this was dead code before).
sql/sql_table.cc:
  Fix for bug #5915 "ALTER TABLE behaves differently when converting column
  to auto_increment in 4.1".
  
  Instead of always forcing NO_AUTO_VALUE_ON_ZERO in ALTER TABLE it is
  better to do this only if we are converting one auto_increment column
  to another auto_increment column (this also includes most common case
  when we don't do anything with such column).

into mysql.com:/home/dlenev/src/mysql-4.1-bf


sql/field.cc:
  Auto merged

introduced during implementation of TIMESTAMP columns, which are able 
to store NULLs (Unfortunately it is impossible to write test case for
this. Kudos to Holyfoot for noticing it!)


sql/field.cc:
  Field_timestamp::get_auto_set_type(): To preserve compatibility 
  with pre-4.1 we have to handle TIMESTAMP_OLD_FIELD columns properly. 
  (It is ok to handle them as TIMESTAMP_DNUN_FIELD since value of first
   such column should be set to current timestamp on both INSERT and UPDATE
   operations, and we don't call this method for rest of them).

into deer.(none):/home/hf/work/mysql-4.1.5730

  Fix typos in previous push


innobase/include/dict0mem.h:
  Fix typos in previous push

into hundin.mysql.fi:/home/heikki/mysql-4.1

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

  Raise maximum column prefix len to 767 bytes, so that MySQL can create a column prefix index of 255 UTF-8 characters (each takes 3 bytes at the maximum); add comments about why innobase_get_at_most_n_mbchars() works ok
dict0mem.h:
  Raise maximum column prefix len to 767 bytes, so that MySQL can create a column prefix index of 255 UTF-8 characters (each takes 3 bytes at the maximum)
row0mysql.c:
  If MySQL tries to create a column prefix index longer that 255 UTF-8 characters, give an error, and drop the table from the InnoDB internal data dictionary. MySQL did not drop the table there in its own error handling.


innobase/row/row0mysql.c:
  If MySQL tries to create a column prefix index longer that 255 UTF-8 characters, give an error, and drop the table from the InnoDB internal data dictionary. MySQL did not drop the table there in its own error handling.
innobase/include/dict0mem.h:
  Raise maximum column prefix len to 767 bytes, so that MySQL can create a column prefix index of 255 UTF-8 characters (each takes 3 bytes at the maximum)
sql/ha_innodb.cc:
  Raise maximum column prefix len to 767 bytes, so that MySQL can create a column prefix index of 255 UTF-8 characters (each takes 3 bytes at the maximum); add comments about why innobase_get_at_most_n_mbchars() works ok

into hundin.mysql.fi:/home/marko/j/mysql-4.1

into hundin.mysql.fi:/home/marko/j/mysql-4.1

removed staus flag

innobase/include/ut0mem.h:
  Improve documentation of ut_strlenq(), ut_strcpyq() and ut_memcpyq()
innobase/include/ut0mem.ic:
  Improve documentation of ut_strlenq()
innobase/row/row0mysql.c:
  row_discard_tablespace_for_mysql(): Add quotes around table_name
  in DISCARD_TABLESPACE_PROC (Bug #5851)
innobase/ut/ut0mem.c:
  Improve documentation of ut_strcpyq() and ut_memcpyq()

  Fix compiler warning


innobase/lock/lock0lock.c:
  Fix compiler warning

into mysql.com:/home/dlenev/src/mysql-4.1-ryan

into mysql.com:/M41/mysql-4.1

into mysql.com:/home/dlenev/src/mysql-4.1-ryan

into mysql.com:/home/dlenev/src/mysql-4.1-ryan

treated as DEFAULT NULL columns (independently from their position in
table).

(still to be discussed with Monty, Brian, Trudy et al. before push)


mysql-test/r/type_timestamp.result:
  Update test after making TIMESTAMP NULL columns without explicit 
  DEFAULT value to be always treated as DEFAULT NULL columns 
  (independently from their position in table).
mysql-test/t/type_timestamp.test:
  Update test after making TIMESTAMP NULL columns without explicit 
  DEFAULT value to be always treated as DEFAULT NULL columns 
  (independently from their position in table).
sql/sql_parse.cc:
  add_field_to_list(): made TIMESTAMP NULL columns without explicit 
  DEFAULT value to be always treated as DEFAULT NULL columns (independently
  from their position in table). This also simplifies code a bit.

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


sql/ha_ndbcluster.cc:
  Auto merged

sql/sql_show.cc:
  Auto merged
sql/ha_ndbcluster.cc:
  SCCS merged

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

bug #5872, transactions should only be restarted with transaction.on flag off if execute_commit has been performed
added testcase for this
use force send for all executes


mysql-test/r/ndb_blob.result:
  added testcase for alter table of blob from ndb to myisam
mysql-test/t/ndb_blob.test:
  added testcase for alter table of blob from ndb to myisam
sql/ha_ndbcluster.cc:
  bug #5872, transactions should only be restarted with transaction.on flag off if execute_commit has been performed
  use force send for all executes

sql/item.cc:
  A typo fixed in a comment, which is probably not valid any more
sql/sql_class.h:
  no such class Ttem_subselect.

with the system-reserved (and system-defined) '_REENTRANT' 
CPP identifier.
Especially, it must not be "undefine"d, because the Tru64 compiler
needs it if threads are enabled.


include/my_pthread.h:
  On Tru64, '_REENTRANT' is set by the compiler when the "-pthread" 
  option is given, and it must be set when the system's 'pthread.h' is 
  included (or else a '#error' is activated).
  So it must not be undef'ed, and it should not be def'ed either, as it
  is a system-reserved CPP identifier with which we should not mess
  anyway - it is risky to use it!
  Build on all platforms was checked privately, change caused no errors!

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


ndb/src/mgmclient/CommandInterpreter.cpp:
  Auto merged

switch to using my_ for heap allocations


ndb/include/util/NdbAutoPtr.hpp:
  addes auto pointer class for using with my_ functions
ndb/src/mgmclient/CommandInterpreter.cpp:
  switch to using my_ for heap allocations

  ctype_recode does not hang anymore. Small fix after Marko's change.


sql/sql_show.cc:
  ctype_recode does not hang anymore. Small fix after Marko's change.

mysql-test/mysql-test-run.sh:
  SCCS merged

mysql-test/mysql-test-run.sh:
  SCCS merged