[PHP-DEV] Current CVS segfaults on Win32

_efree(void * 0xcdcdcdcd, char * 0x1026e258 `string', unsigned int 1459, char * 0x, unsigned int 0) line 222 + 3 bytes php_rshutdown_session_globals(void * * * 0x00b32870) line 1459 + 44 bytes zm_deactivate_session(int 1, int 10, void * * * 0x00b32870) line 1519 + 9 bytes module_registry_c

Re: [PHP-DEV] Re: [PHP-CVS] cvs: php4 / php.ini-distphp.ini-recommended

I have to agree with Rasmus. Don't fix what isn't broken. While short_tags may annoy some people, it certainly doesn't break anything, while disabling it by default would. What are we trying to do? Roll out release after release making problems for people? IMHO we've filled our quota for a wh

[PHP-DEV] CVS Account Request

Would like to contribute and help with existing PEAR modules ~ Andrew Heebner, EvilWalrus.com -- PHP Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php

Re: [PHP-DEV] Who vote for this? (was Re: [PHP-DEV] Discourage use

Rasmus Lerdorf wrote: > By suddenly changing a basic characteristic of PHP that has been around > since the beginning of time in order to solve a problem almost nobody > has, we violate both principles, and on top of that we add a security > risk in that now people are likely to expose code th

Re: [PHP-DEV] Who vote for this? (was Re: [PHP-DEV] Discourage useof short tags)

The standard configuration is not such that PHP is set up to process XML or XHTML documents. It is very deliberate that we have a separate mime-type and our own .php extension as the default config. Yes, you can configure PHP to parse all .html files, or .xhtml or even .xml if you wanted to

Re: [PHP-DEV] Re: Discourage use of short tags

Looks good to me On Saturday 27 April 2002 06:24 pm, Yasuo Ohgaki wrote: > It's Zeev's version. > Everyone happy with this? > > > > Using short tags should be avoided when developing applications or > libraries that are meant for redistribution, or deployment on PHP > servers which are

[PHP-DEV] Re: Discourage use of short tags

It's Zeev's version. Everyone happy with this? Using short tags should be avoided when developing applications or libraries that are meant for redistribution, or deployment on PHP servers which are not under your control, because short tags may not be supported on the target server.

Re: [PHP-DEV] Shared PHP library

Stig S. Bakken wrote: > On Sat, 2002-04-20 at 17:53, Andrei Zmievski wrote: > >>I would like to know if we can modify the build process to build and >>install a PHP library, without any SAPI modules, so that libphp.a and >>libphp.so get installed in the appropriate locations. I need this in >>ord

[PHP-DEV] Re: cvs: php4 /ext/zlib zlib.c

Yasuo Ohgaki wrote: > Thies C. Arntzen wrote: > >> thiesSat Apr 27 13:03:53 2002 EDT >> >> Modified files: /php4/ext/zlibzlib.c Log: >> @ - If possible set Content-Length header in zlib.output_compression >> mode. (thies) > > > A little suggestion. It seems i

Re: [PHP-DEV] Who vote for this? (was Re: [PHP-DEV] Discourage use of short tags)

Hi, Rasmus gave quiet a few examples as others did. I suggest reading through the mails posted the lasty days about this topic. - Markus On Sun, Apr 28, 2002 at 07:50:38AM +0900, Yasuo Ohgaki wrote : > I really don't understand why we should keep short_tag=on. This > implie

[PHP-DEV] Who vote for this? (was Re: [PHP-DEV] Discourage use of short tags)

I really don't understand why we should keep short_tag=on. This implies we have following _standard_ and _portable_ syntax for all XML documents preprocessed by PHP. ' ?> Who vote for this _standard_ and _portable_ syntax? Don't forget to write why this syntax is good, look great, etc. -- Yasu

Re: [PHP-DEV] Shared PHP library

Hi, that would be great expandation of PHP. For example, TCL has quite a nice interface to start an interpreter, (virtually) compile a script into byte code, bind simple C types to TCL variables, eval/call function in TCL and have the output in the bound C variables ...

Re: [PHP-DEV]

Don't worry, you're not. :-) - Stig On Sat, 2002-04-27 at 03:44, Zeev Suraski wrote: > Yes, but I thought it was SGML compliant (as in, some sort of a subset of > SGML with lots of predefined rules, but still, falls into the SGML language > category). > > But then, I could very well be wrong

Re: [PHP-DEV]

At 18:44 27/04/2002, Stig S. Bakken wrote: >On Sat, 2002-04-27 at 03:30, Zeev Suraski wrote: > > At 03:18 27/04/2002, Rasmus Lerdorf wrote: > > >It looks like we can. I was assuming the SGML characteristics for XML and > > >it looks like I was wrong. A '>' is ok inside the tags. > > > > Ok, so

Re: [PHP-DEV]

On Fri, 2002-04-26 at 20:07, Sterling Hughes wrote: > > Ok. #1 is the first logical, technical reason I've seen against the > > shorthand being fully implemented (though it begs the question why it was > > partially implemented in the first place). > > > > I'm not to knowledgeable about SGML spe

Re: [PHP-DEV]

On Sat, 2002-04-27 at 03:30, Zeev Suraski wrote: > At 03:18 27/04/2002, Rasmus Lerdorf wrote: > >It looks like we can. I was assuming the SGML characteristics for XML and > >it looks like I was wrong. A '>' is ok inside the tags. > > Ok, so that's actually useful. But it sounds odd - XML is n

Re: [PHP-DEV] Removal of "php_fopen_wrapper" in HEAD will break binary modules between 4.2 and HEAD (4.3)

> > hi, > is there any way we could avoid this? > > i think we should aim for keeping binary compatiblity as long > as it makes sense. so, question is, would it make sense to > have a function called php_fopen_wrapper in HEAD that behaves > _exactly_ like it did in 4.2.

Re: [PHP-DEV] Re: [ZEND-ENGINE-CVS] cvs: Zend / zend_execute_API.c

On Sat, Apr 27, 2002 at 06:55:46PM +0200, Sebastian Bergmann wrote: > Rasmus Lerdorf wrote: > > rasmus Sat Apr 27 12:21:00 2002 EDT > > > > Modified files: > > /Zend zend_execute_API.c > > Log: > > # How do I get at a php core_global from here? > > Isn't the Zend Engine

[PHP-DEV] Removal of "php_fopen_wrapper" in HEAD will break binary modules between 4.2 and HEAD (4.3)

hi, is there any way we could avoid this? i think we should aim for keeping binary compatiblity as long as it makes sense. so, question is, would it make sense to have a function called php_fopen_wrapper in HEAD that behaves _exactly_ like it did in 4.2. elsewise we nee

[PHP-DEV] Re: [ZEND-ENGINE-CVS] cvs: Zend / zend_execute_API.c

Rasmus Lerdorf wrote: > rasmus Sat Apr 27 12:21:00 2002 EDT > > Modified files: > /Zend zend_execute_API.c > Log: > # How do I get at a php core_global from here? Isn't the Zend Engine supposed to be independent from PHP? -- Sebastian Bergmann http://sebastian-ber

[PHP-DEV] Re: [ZEND-ENGINE-CVS] cvs: Zend / zend_execute_API.c

Rasmus Lerdorf wrote: > rasmus Sat Apr 27 10:07:35 2002 EDT > > Modified files: > /Zend zend_execute_API.c > Log: > Fix for #14542 c:\home\php\php4\Zend\zend_execute_API.c(666): warning C4013: 'PG' undefined c:\home\php\php4\Zend\zend_execute_API.c(666):

Re: [PHP-DEV] Discourage use of short tags

Hi, Removing the short tags from future php releases, or disabling them by default, is like stripping functions from glibc because they do not exist on other platforms. Stefan -- PHP Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.

Re: [PHP-DEV] socket_create() doesn't expose error if one occurs (nor does socket_select)

Hi, This is an addition to my last mail. I just saw that socket_select() basically suffers from the same problem (programmatically catching the error == not possible). This function would also benefit from the suggested behaviour to store the last error in a global error

[PHP-DEV] socket_create() doesn't expose error if one occurs [CVS HEAD]

Hi, socket_create() doesn't expose an error if one occurs when creating a new socket with socket(). The current code is: [...] php_sock->bsd_socket = socket(arg1, arg2, arg3); php_sock->type = arg1; if (IS_INVALID_SOCKET(php_sock)) { efree(php_sock); RETU

Re: [PHP-DEV] Discourage use of short tags

** Reply to note from Zeev Suraski <[EMAIL PROTECTED]> Sat, 27 Apr 2002 05:52:11 +0300 > > I don't think we should do it in PHP 5 either. I strongly agree! Rick Rick Widmer Internet Marketing Specialists http://www.developersdesk.com -- PHP Development Mailing List

RE: [PHP-DEV] Discourage use of short tags

Well the point for me is that I want "clean php code" (the relevant part of my definition of what this is for me should be apparent by the rest of the email) when I see a script starting with " -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] > Sent: Friday, April 26