Re: cvs: php-src /main main.ct
| From: | Rasmus Lerdorf | Date: | Sun, 07 Aug 2005 00:00:00 +0000 |
| Subject: | Re: cvs: php-src /main main.ct | ||
| References: | 1 2 3 4 5 | Groups: | php.cvs |
| Request: | Send a blank email to php-cvs+get-33203@lists.php.net to get a copy of this message | ||
Jani Taskinen wrote:
> On Sat, 6 Aug 2005, Rasmus Lerdorf wrote:
>
>> Jani Taskinen wrote:
>>
>>> On Sat, 6 Aug 2005, Rasmus Lerdorf wrote:
>>>
>>>> Jani Taskinen wrote:
>>>>
>>>>> sniper Fri Aug 5 17:44:27 2005 EDT
>>>>>
>>>>> Modified files:
>>>>> /php-src/main main.c
>>>>> Log:
>>>>> - This ini option is only effective when set in php.ini or
>>>>> httpd.conf
>>>>> or .htacces, etc. as GPC variables are set before any ini_set()
>>>>> is called
>>>>
>>>>
>>>>
>>>> Are you sure?
>>>
>>>
>>>
>>> Sure about what? That this option has no effect when set with
>>> ini_set()?
>>> Yes, I'm sure. PG(variables_order) is set before any ini_set()'s are
>>> called.
>>
>>
>> Sure, but calling ini_set() on it will reset PG(variables_order) to the
>> new value. The question is whether we want the jit stuff to respect
>
>
> And PG(variables_order) is reset to whatever you have in php.ini
> or to the default value in the next request start.
> You can't control it with ini_set(). And whatever it is set to during
> script compile/execute makes absolutely no difference. I only has an
> effect during request startup.
Of course it doesn't bleed across requests. But we could easily check
jit and set the 'armed' jit flag in the variables_order onupdate handler
so people could control the jittable autoglobals.
-Rasmus