maria-developers team mailing list archive

[Sergei Golubchik <serg@xxxxxxxxxxxx>]

Hi, Sanja!

On Sep 07, sanja@xxxxxxxxxxxx wrote:
> At file:///home/bell/maria/bzr/work-maria-5.3-lb615760/
> 
> ------------------------------------------------------------
> revno: 2816
> revision-id: sanja@xxxxxxxxxxxx-20100907061732-yyuwtfg0ucyqiu8j
> parent: sanja@xxxxxxxxxxxx-20100906123424-2sco3ghittvidm6l
> committer: sanja@xxxxxxxxxxxx
> branch nick: work-maria-5.3-lb615760
> timestamp: Tue 2010-09-07 09:17:32 +0300
> message:
>   Fixed second part of the LP BUG#615760 - incorrect parameters of the
>   temporary heap table index (unique & nulls are equal).

> === modified file 'sql/sql_expression_cache.cc'
> --- a/sql/sql_expression_cache.cc	2010-09-06 12:34:24 +0000
> +++ b/sql/sql_expression_cache.cc	2010-09-07 06:17:32 +0000

I cannot review that change, I don't know sql_expression_cache code :(
Anyway, you asked me to look at the "incorrect heap table index flags"
changes only.

> === modified file 'sql/table.cc'
> --- a/sql/table.cc	2010-07-10 10:37:30 +0000
> +++ b/sql/table.cc	2010-09-07 06:17:32 +0000
> @@ -5191,7 +5191,7 @@
>    keyinfo->usable_key_parts= keyinfo->key_parts = key_parts;
>    keyinfo->key_length=0;
>    keyinfo->algorithm= HA_KEY_ALG_UNDEF;
> -  keyinfo->flags= HA_GENERATED_KEY;
> +  keyinfo->flags= HA_GENERATED_KEY | HA_NOSAME;
>    sprintf(buf, "key%i", key);
>    if (!(keyinfo->name= strdup_root(&mem_root, buf)))
>      return TRUE;
> @@ -5230,6 +5230,7 @@
>      {
>        key_part_info->store_length+= HA_KEY_NULL_LENGTH;
>        keyinfo->key_length+= HA_KEY_NULL_LENGTH;
> +      keyinfo->flags|= HA_NULL_ARE_EQUAL;	// def. that NULL == NULL
>      }
>      if ((*reg_field)->type() == MYSQL_TYPE_BLOB || 
>          (*reg_field)->real_type() == MYSQL_TYPE_VARCHAR)

Difficult to say.

Sure, temporary tables should have keyinfo->flags=HA_NOSAME and
keyinfo->flags|= HA_NULL_ARE_EQUAL. You can see that in
create_tmp_table(). HA_GENERATED_KEY has no effect, so it doesn't matter
if you specify it or not.

But I don't know what are requirements for temporary tables in your
subquery cache code.

Unrelated question - why did you duplicate creation of temporary
tables? I understand that you may've needed something slightly different
from create_tmp_table(), but then you should've changed
create_tmp_table() to use your new code, TABLE::add_tmp_key() etc.

By the way, if you had done that (reused the code, not duplicated it)
this bug would've been impossible - many tests in the test suite would
fail at once if all temporary tables would have non-unique indexes.
 
Regards,
Sergei