Hi!
<cut>
>3.23.10 Running on PPRO w RH Linux 6.1
>
>mysql> select * from ct00l02site_groups where group_id=177;
>+---------+----------+--------+-----------+-----------+------------------
>---+------+
>| SITE_ID | GROUP_ID | SHADOW | MODERATOR | MODERATED | MD_TS
> | PRIM |
>+---------+----------+--------+-----------+-----------+------------------
>---+------+
>| 393 | 177 | 0 | 0 | 0 | 0000-00-00
>00:00:00 | 0 |
>| 1197 | 177 | 0 | 0 | 0 | 0000-00-00
>00:00:00 | 0 |
>| 415 | 177 | 0 | 0 | 0 | 0000-00-00
>00:00:00 | 0 |
>+---------+----------+--------+-----------+-----------+------------------
>---+------+
>3 rows in set (0.01 sec)
>
>mysql>
>
>mysql> select id from sites where id=393;
>+-----+
>| id |
>+-----+
>| 393 |
>+-----+
>1 row in set (0.00 sec)
>
>
>So If I'm not mistaken there should be at least one row returned by the
>following query:
>
>
>select * from ct00l02site_groups,sites where group_id=177 and ID=SITE_ID;
> But What we have here is an empty set therefore the explain says the
> following:
This is the problem with living on the bleeding edge :) The problem
was a bug in the code that I used to solve one of your previous
problems.
Here is a patch for this:
*** /my/monty/master/mysql-3.23.10-alpha/sql/sql_select.cc Wed Jan 19 03:11:03 2000
--- ./sql_select.cc Wed Feb 16 12:36:34 2000
***************
*** 1574,1580 ****
tmp=record_count*min(tmp,s->worst_seeks);
}
else
! tmp=best; // Do nothing
}
if (tmp < best_time - records/(double) TIME_FOR_COMPARE)
{
--- 1574,1580 ----
tmp=record_count*min(tmp,s->worst_seeks);
}
else
! tmp=best_time; // Do nothing
}
if (tmp < best_time - records/(double) TIME_FOR_COMPARE)
{
Regards,
Monty
