| # |
Jul 23rd 2009, 07:08 |
alkemann |
jperras: can u pull the bakery 2.0 branch and run the test urself? |
| # |
Jul 23rd 2009, 07:07 |
alkemann |
jperras: http://bin.cakephp.org/view/1334665431 - this one doesnt do it |
| # |
Jul 23rd 2009, 07:07 |
jperras |
alkemann: oh? |
| # |
Jul 23rd 2009, 07:06 |
jperras |
we might be able to rewrite some portions of Model and DboSource to get the query a bit more optimized |
| # |
Jul 23rd 2009, 07:06 |
alkemann |
jperras: im not able to recreate it in a lesser test |
| # |
Jul 23rd 2009, 07:06 |
jperras |
alkemann: post an enhancement ticket |
| # |
Jul 23rd 2009, 07:05 |
Phally |
alkemann: if it is query #85 of article.test.php, then i have |
| # |
Jul 23rd 2009, 07:04 |
alkemann |
update is on the parent. scope is on the counted model |
| # |
Jul 23rd 2009, 07:04 |
jperras |
that's a side-effect of updateAll, iirc |
| # |
Jul 23rd 2009, 07:03 |
alkemann |
jperras: but the scope would be on the find('count' it still does not make sense on the update |
| # |
Jul 23rd 2009, 07:03 |
jperras |
if your counterscope is not set, then there shouldn't be any joins produced |
| # |
Jul 23rd 2009, 07:03 |
alkemann |
i am using scope yes |
| # |
Jul 23rd 2009, 07:03 |
jperras |
since you can scope on the associated model |
| # |
Jul 23rd 2009, 07:03 |
jperras |
if I remember correctly, it has something to do with counterScope |
| # |
Jul 23rd 2009, 07:02 |
jperras |
hmm |
| # |
Jul 23rd 2009, 07:02 |
alkemann |
Phally: u found the same query running 2.0 branch tests? |
| # |
Jul 23rd 2009, 07:01 |
alkemann |
jperras: http://bin.cakephp.org/view/1638485376 - dont see why it makes all the joins for SET `Article`.`article_page_count` =Â 1Â WHERE `Article`.`id` |
| # |
Jul 23rd 2009, 07:01 |
jperras |
alkemann: yes |
| # |
Jul 23rd 2009, 07:01 |
Phally |
jperras: yo |
| # |
Jul 23rd 2009, 07:01 |
alkemann |
jperras: u have any familiarity with countercache? |
| # |
Jul 23rd 2009, 07:01 |
Phally |
alkemann: haha |
| # |
Jul 23rd 2009, 07:00 |
Phally |
alkemann: http://api.cakephp.org/view_source/model/#line-1439 |
| # |
Jul 23rd 2009, 07:00 |
jperras |
ahoy hoy |
| # |
Jul 23rd 2009, 07:00 |
alkemann |
morn jperras |
| # |
Jul 23rd 2009, 07:00 |
alkemann |
had a find before a write and couldnt understand why i didnt find the new stuff |
| # |
Jul 23rd 2009, 07:00 |
alkemann |
different issue. |
| # |
Jul 23rd 2009, 07:00 |
alkemann |
damn im an idiot |
| # |
Jul 23rd 2009, 06:59 |
Phally |
agreed |
| # |
Jul 23rd 2009, 06:59 |
alkemann |
but my point would actually be that all counter cache updates should be hardcoded to do no joins |
| # |
Jul 23rd 2009, 06:59 |
Phally |
alkemann: no, i think when it gets there it isn't -1 anymore |
| # |
Jul 23rd 2009, 06:58 |
alkemann |
Phally: u mean the problem of high recursiveness is that i set it to -1? |
| # |
Jul 23rd 2009, 06:57 |
Phally |
alkemann: hmm was checking an old ticket out of mine which was about a different issue, but there were no joins either, so it has to be that recursive thing |
| # |
Jul 23rd 2009, 06:54 |
alkemann |
just typo |
| # |
Jul 23rd 2009, 06:54 |
Phally |
norwegian? :p |
| # |
Jul 23rd 2009, 06:54 |
Phally |
public $alies = 'Article'; |
| # |
Jul 23rd 2009, 06:54 |
Phally |
alkemann: now, that is strange, still using 1.2? |
| # |
Jul 23rd 2009, 06:53 |
alkemann |
and it does not do the JOINs |
| # |
Jul 23rd 2009, 06:53 |
alkemann |
Phally: then i made this standalone to check it http://bin.cakephp.org/view/1334665431 |
| # |
Jul 23rd 2009, 06:52 |
alkemann |
put on debug 2 and see for urself |
| # |
Jul 23rd 2009, 06:50 |
Phally |
i see |
| # |
Jul 23rd 2009, 06:46 |
alkemann |
http://thechaw.com/bakery/source/branches/2.0/tests/cases/models/article.test.php#11 |