Re: Re: Please review: new output control implementation

From: Date: Fri, 17 Mar 2006 23:33:07 +0000
Subject: Re: Re: Please review: new output control implementation
References: 1 2  Groups: php.internals 
Request: Send a blank email to internals+get-22514@lists.php.net to get a copy of this message
Hello Michael, Friday, March 17, 2006, 4:09:12 PM, you wrote: > I wrote: >> References: >> [0] http://dev.iworks.at/PATCHES/output/ >> [1] http://dev.iworks.at/PATCHES/output/ob.patch.txt.gz >> [2] http://dev.iworks.at/PATCHES/output/php_output.h.txt >> [3] http://dev.iworks.at/PATCHES/output/output.c.txt >> [4] >> http://dev.iworks.at/PATCHES/output/README.NEW-OUTPUT-API > Should I interpret the sustained silence as no interest in it? I'd say the only thing that doesn't look ten times better than what we have right now is that you have exported function prefixed with underscore "PHPAPI int _php_output" and static ones without. Imo it should be the other way round. The open question i have however is how do you deal with unicode. I saw some estr*() calls where i would have expected a few unicode checks. So there is no unicode support right now. So i suggest you add that too. A think i am a bit curious yout is why some function have both *() and *_all() versions. Why didn't you choose to pass an int in? Last but not least i wonder why i am seeing use of the strange TSRMLS_FETCH_FROM_CTX() instead of passing TSRM keys using TSRMLS_* macros. Best regards, Marcus

« previous php.internals (#22514) next »