RE: [PHP-CVS] cvs: php-src /ext/standard assert.c basic_functions.c

2007-02-16 Thread Dmitry Stogov
t; To: Dmitry Stogov; 'Ilia Alshanetsky' > Cc: php-cvs@lists.php.net > Subject: Re: [PHP-CVS] cvs: php-src /ext/standard assert.c > basic_functions.c > > > So where is the test? *hint* :-) > > - Original Message - > > Really this patch fixes

Re: [PHP-CVS] cvs: php-src /ext/standard assert.c basic_functions.c

2007-02-16 Thread Nuno Lopes
ov Cc: php-cvs@lists.php.net Subject: Re: [PHP-CVS] cvs: php-src /ext/standard assert.c basic_functions.c Dmitry, Please revert this patch. The code adds RINIT that hits every request for functionality that is only needed for testing of the code and is otherwise rarely used. I do not belie

RE: [PHP-CVS] cvs: php-src /ext/standard assert.c basic_functions.c

2007-02-16 Thread Dmitry Stogov
Really this patch fixes a SIGSEGV. I'll try to look how to optimize it. Thanks. Dmitry. > -Original Message- > From: Ilia Alshanetsky [mailto:[EMAIL PROTECTED] > Sent: Friday, February 16, 2007 4:47 PM > To: Dmitry Stogov > Cc: php-cvs@lists.php.net > Subject: Re:

Re: [PHP-CVS] cvs: php-src /ext/standard assert.c basic_functions.c

2007-02-16 Thread Ilia Alshanetsky
On 16-Feb-07, at 9:55 AM, Steph Fox wrote: Ilia, hi, There's no way to query the current callback setting there. If RINIT's the only way to achieve it, and we can't have RINIT, the manual needs fixing to reflect reality. So - are you officially calling it a 'won't fix'?' yes! I already

Re: [PHP-CVS] cvs: php-src /ext/standard assert.c basic_functions.c

2007-02-16 Thread Steph Fox
x in the wings? (Read: I've no idea whether promoting OnChangeCallback to ZEND_INI_MH() would do the trick.) - Steph - Original Message - From: "Ilia Alshanetsky" <[EMAIL PROTECTED]> To: "Dmitry Stogov" <[EMAIL PROTECTED]> Cc: Sent: Friday, February

Re: [PHP-CVS] cvs: php-src /ext/standard assert.c basic_functions.c

2007-02-16 Thread Ilia Alshanetsky
Dmitry, Please revert this patch. The code adds RINIT that hits every request for functionality that is only needed for testing of the code and is otherwise rarely used. I do not believe it warrants slowing down every single request, even by a little bit. On 16-Feb-07, at 8:12 AM, Dmitry