monty | 1 Dec 13:59 2002
Picon

bk commit into 4.1 tree

Below is the list of changes that have just been committed into a local
4.1 repository of monty. When monty does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://www.mysql.com/doc/I/n/Installing_source_tree.html

ChangeSet
  1.1392 02/12/01 14:59:06 monty <at> hundin.mysql.fi +2 -0
  Fixed that one can use ` in identifiers:  CREATE TABLE `fo``a` ...
  Added BOOLEAN as synonym for tinyint

  sql/sql_yacc.yy
    1.204 02/12/01 14:53:59 monty <at> hundin.mysql.fi +2 -0
    Added BOOLEAN as synonym for tinyint

  sql/sql_lex.cc
    1.55 02/12/01 14:53:58 monty <at> hundin.mysql.fi +43 -6
    Fixed that one can use ` in identifiers:  CREATE TABLE `fo``a` ... 

# This is a BitKeeper patch.  What follows are the unified diffs for the
# set of deltas contained in the patch.  The rest of the patch, the part
# that BitKeeper cares about, is below these diffs.
# User:	monty
# Host:	hundin.mysql.fi
# Root:	/my/mysql-4.1

--- 1.54/sql/sql_lex.cc	Wed Nov 27 11:13:02 2002
+++ 1.55/sql/sql_lex.cc	Sun Dec  1 14:53:58 2002
 <at>  <at>  -216,7 +216,7  <at>  <at> 
(Continue reading)

pem | 1 Dec 17:10 2002
Picon

bk commit into 4.1 tree

Below is the list of changes that have just been committed into a local
4.1 repository of pem. When pem does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://www.mysql.com/doc/I/n/Installing_source_tree.html

ChangeSet
  1.1411 02/12/01 17:10:13 pem <at> mysql.com +1 -0
  Got rid of the (previously) intentional SELECT/UNION reduce/reduce conflict.
  There should be no more reduce/reduce conflicts in sql_yacc.yy from now on!

  sql/sql_yacc.yy
    1.212 02/12/01 17:10:11 pem <at> mysql.com +13 -6
    Got rid of the (previously) intentional SELECT/UNION reduce/reduce conflict
    (in the optional_order_or_limit clause).

# This is a BitKeeper patch.  What follows are the unified diffs for the
# set of deltas contained in the patch.  The rest of the patch, the part
# that BitKeeper cares about, is below these diffs.
# User:	pem
# Host:	mysql.telia.com
# Root:	/home/pem/work/mysql-4.1

--- 1.211/sql/sql_yacc.yy	Sun Dec  1 16:03:40 2002
+++ 1.212/sql/sql_yacc.yy	Sun Dec  1 17:10:11 2002
 <at>  <at>  -2682,7 +2682,11  <at>  <at> 

 opt_limit_clause:
 	/* empty */ {}
(Continue reading)

Tim Bunce | 1 Dec 23:51 2002
Picon

Re: bk commit into mysqldoc tree

On Sun, Dec 01, 2002 at 12:56:18AM +0200, monty <at> mysql.com wrote:
> + <at> item
> +If once creates a too long  <at> code{CHAR}/ <at> code{VARCHAR} it's now
> +automaticly changed to  <at> code{TEXT} or  <at> code{BLOB}; One will get a
> +warning in this case.
> + <at> item
> +One can specify the different  <at> code{BLOB}/ <at> code{TEXT} types with the
> +syntax  <at> code{BLOB(length)} and  <at> code{TEXT(length)}.  MySQL will
> +automaticly change it to one of the internal  <at> code{BLOB}/ <at> code{TEXT}
> +types.
> + <at> item

Any plans to support a VARCHAR type that goes beyond 255 chars?

I ask because I've recently been bitten by BLOB types not being
supported in HASH tables. I needed a varchar(4000) field for an
app I'm porting from Oracle.

I've worked around it now but I think a two-byte-length varchar
type would be a valuable addition - avoiding the overheads and
restrictions on blobs.

Is this a change that's awaiting support for different block sizes?

Tim.

---------------------------------------------------------------------
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/           (the list archive)
(Continue reading)

jimw | 2 Dec 02:11 2002
Picon

bk commit into mysqldoc tree

Below is the list of changes that have just been committed into a local
mysqldoc repository of jimw. When jimw does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://www.mysql.com/doc/I/n/Installing_source_tree.html

ChangeSet
  1.111 02/12/01 17:11:53 jimw <at> mysql.com +2 -0
  chunk-manual-for-web:
    generate simple text version of table of contents
  index-manual-for-web:
    another index to optimize away a slow query

  Docs/Support/chunk-manual-for-web
    1.4 02/12/01 17:11:21 jimw <at> mysql.com +6 -0
    generate simple text version of table of contents

  Docs/Support/index-manual-for-web
    1.3 02/11/21 22:44:08 jimw <at> mysql.com +1 -0
    another index to optimize away a slow query

# This is a BitKeeper patch.  What follows are the unified diffs for the
# set of deltas contained in the patch.  The rest of the patch, the part
# that BitKeeper cares about, is below these diffs.
# User:	jimw
# Host:	sugriva.trainedmonkey.com
# Root:	/Users/jimw/my/mysqldoc

--- 1.3/Docs/Support/chunk-manual-for-web	Wed Nov 20 14:52:29 2002
(Continue reading)

jocelyn fournier | 2 Dec 09:44 2002

Re: bk commit into 4.1 tree

Hi,

About this problem, I think the following :

As inline functions are not included in object file if -O3 optimisation is
used, the problem we encounter is inside sql_yacc.h and item_uniq.cc we
included item_sum.h which explicitely calls mark_as_sum_func() inside
constructors.
However mark_as_sum_func() is only defined inside item_sum.cc, and is not
included in the object file with -O3 optimisation. Thus, the linker is
unable to find this function, and report the undefined reference problem.
In conclusion, I think to solve the problem, we need to redefined the
mark_as_sum_func function inside sql_yacc.cc and item_uniq.cc (and it works
fine for me with these modifications) :

===== item_uniq.cc 1.5 vs edited =====
--- 1.5/sql/item_uniq.cc        Thu Dec  6 13:10:45 2001
+++ edited/item_uniq.cc Mon Dec  2 09:33:23 2002
 <at>  <at>  -20,3 +20,8  <at>  <at> 
 #endif

 #include "mysql_priv.h"
+
+inline void Item_sum::mark_as_sum_func()
+{
+  current_thd->lex.current_select->with_sum_func= with_sum_func= 1;
+}
===== sql_yacc.yy 1.210 vs edited =====
--- 1.210/sql/sql_yacc.yy       Sat Nov 30 18:54:47 2002
+++ edited/sql_yacc.yy  Mon Dec  2 09:27:06 2002
(Continue reading)

monty | 2 Dec 09:47 2002
Picon

bk commit into 3.23 tree

Below is the list of changes that have just been committed into a local
3.23 repository of monty. When monty does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://www.mysql.com/doc/I/n/Installing_source_tree.html

ChangeSet
  1.1271 02/12/02 10:47:15 monty <at> hundin.mysql.fi +1 -0
  Fixed bug in <=> NULL handling

  sql/opt_range.cc
    1.47 02/12/02 10:47:13 monty <at> hundin.mysql.fi +1 -1
    Fixed bug in <=> NULL handling

# This is a BitKeeper patch.  What follows are the unified diffs for the
# set of deltas contained in the patch.  The rest of the patch, the part
# that BitKeeper cares about, is below these diffs.
# User:	monty
# Host:	hundin.mysql.fi
# Root:	/my/mysql-3.23

--- 1.46/sql/opt_range.cc	Fri Nov 22 13:35:41 2002
+++ 1.47/sql/opt_range.cc	Mon Dec  2 10:47:13 2002
 <at>  <at>  -1030,7 +1030,7  <at>  <at> 
     {
       /* convert column_name <=> NULL -> column_name IS NULL */
       char *str= (char*) sql_alloc(1);		// Get local copy of key
-      if (!*str)
+      if (!str)
(Continue reading)

monty | 2 Dec 09:58 2002
Picon

bk commit into mysqldoc tree

Below is the list of changes that have just been committed into a local
mysqldoc repository of monty. When monty does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://www.mysql.com/doc/I/n/Installing_source_tree.html

ChangeSet
  1.111 02/12/02 10:58:28 monty <at> hundin.mysql.fi +2 -0
  Changelog and IFNULL() update

  BitKeeper/etc/logging_ok
    1.14 02/12/02 10:58:28 monty <at> hundin.mysql.fi +2 -1
    Logging to logging <at> openlogging.org accepted

  Docs/manual.texi
    1.104 02/12/02 10:58:13 monty <at> hundin.mysql.fi +24 -2
    Changelog and IFNULL() update

# This is a BitKeeper patch.  What follows are the unified diffs for the
# set of deltas contained in the patch.  The rest of the patch, the part
# that BitKeeper cares about, is below these diffs.
# User:	monty
# Host:	hundin.mysql.fi
# Root:	/my/mysqldoc

--- 1.13/BitKeeper/etc/logging_ok	Thu Nov 28 03:24:20 2002
+++ 1.14/BitKeeper/etc/logging_ok	Mon Dec  2 10:58:28 2002
 <at>  <at>  -4,13 +4,14  <at>  <at> 
 heikki <at> hundin.mysql.fi
(Continue reading)

Sinisa | 2 Dec 13:16 2002
Picon

bk commit into 3.23 tree

Below is the list of changes that have just been committed into a local
3.23 repository of Sinisa. When Sinisa does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://www.mysql.com/doc/I/n/Installing_source_tree.html

ChangeSet
  1.1274 02/12/02 14:16:02 Sinisa <at> sinisa.nasamreza.org +4 -0
  Fix for sum_func IN (..)

  sql/item_cmpfunc.h
    1.13 02/12/02 14:16:01 Sinisa <at> sinisa.nasamreza.org +2 -1
    Fix for sum_func IN (..)

  sql/item_cmpfunc.cc
    1.25 02/12/02 14:16:01 Sinisa <at> sinisa.nasamreza.org +5 -1
    Fix for sum_func IN (..)

  mysql-test/t/group_by.test
    1.8 02/12/02 14:16:01 Sinisa <at> sinisa.nasamreza.org +2 -0
    Fix for sum_func IN (..)

  mysql-test/r/group_by.result
    1.8 02/12/02 14:16:01 Sinisa <at> sinisa.nasamreza.org +5 -0
    Fix for sum_func IN (..)

# This is a BitKeeper patch.  What follows are the unified diffs for the
# set of deltas contained in the patch.  The rest of the patch, the part
# that BitKeeper cares about, is below these diffs.
(Continue reading)

Sinisa Milivojevic | 2 Dec 13:31 2002
Picon

Re: bk commit into 4.1 tree

jocelyn fournier writes:
> Hi,
> 
> About this problem, I think the following :
> 
> As inline functions are not included in object file if -O3 optimisation is
> used, the problem we encounter is inside sql_yacc.h and item_uniq.cc we
> included item_sum.h which explicitely calls mark_as_sum_func() inside
> constructors.
> However mark_as_sum_func() is only defined inside item_sum.cc, and is not
> included in the object file with -O3 optimisation. Thus, the linker is
> unable to find this function, and report the undefined reference problem.
> In conclusion, I think to solve the problem, we need to redefined the
> mark_as_sum_func function inside sql_yacc.cc and item_uniq.cc (and it works
> fine for me with these modifications) :
> 
> 
> Regards,
>   Jocelyn

I do not think we should do that. That would mean lot's of
repetitions, plus lot's of maintenance problems !!

If they can't solve it, then I will move all code in the include file.

--

-- 
Regards,
   __  ___     ___ ____  __
  /  |/  /_ __/ __/ __ \/ /    Mr. Sinisa Milivojevic <sinisa <at> mysql.com>
 / /|_/ / // /\ \/ /_/ / /__   MySQL AB, Fulltime Developer
(Continue reading)

monty | 2 Dec 12:41 2002
Picon

bk commit into 3.23 tree

Below is the list of changes that have just been committed into a local
3.23 repository of monty. When monty does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://www.mysql.com/doc/I/n/Installing_source_tree.html

ChangeSet
  1.1272 02/12/02 13:41:08 monty <at> hundin.mysql.fi +5 -0
  Fixed bug in MAX() optimization when used with JOIN and ON expressions
  (Patch from 4.0.5)

  sql/opt_sum.cc
    1.10 02/12/02 13:41:06 monty <at> hundin.mysql.fi +13 -0
    Fixed bug in MAX() optimization when used with JOIN and ON expressions

  sql/item_cmpfunc.h
    1.12 02/12/02 13:41:06 monty <at> hundin.mysql.fi +2 -0
    Fixed bug in MAX() optimization when used with JOIN and ON expressions

  sql/item_cmpfunc.cc
    1.24 02/12/02 13:41:06 monty <at> hundin.mysql.fi +39 -0
    Fixed bug in MAX() optimization when used with JOIN and ON expressions

  mysql-test/t/group_by.test
    1.7 02/12/02 13:41:06 monty <at> hundin.mysql.fi +40 -1
    Test for bug

  mysql-test/r/group_by.result
    1.7 02/12/02 13:41:06 monty <at> hundin.mysql.fi +6 -0
(Continue reading)


Gmane