On Sat, Aug 23, 2014 at 9:29 PM, Andrea Faulds <a...@ajf.me> wrote: > > On 23 Aug 2014, at 20:19, Kris Craig <kris.cr...@gmail.com> wrote: > >> I have a quick question for the people debating this issue: Aside from the >> dispute over whether or not it's necessary / unimportant, are there any >> pressing reasons *not* to implement the changes that Pierre is advocating? >> I.e. would it break anything, waste a large amount of time, make the code >> harder to maintain, etc? Or is it just that you don't think it's worth >> doing? > > I can think of two: > > 1. It breaks most extensions out there, perhaps unnecessarily.
Please try to port one. That will solve this never ending ping pong game. Extensions are broken per se with ng, almost every zval macros usage must change (some disappeared, like the _PP ones), all hash APIs call must be change (a must, not detectable at compile time), etc. IS_LONG to IS_INT is a joke in comparison. But as nobody agrees on that, I won't discuss it to death. > 2. If bigints are implemented, we’d have to rename everything again. Ah, and that will be acceptable then, right? ;-) Also hurry up with that, even if not totally completed. Many extensions may have to deal with it and it will just double the porting work if it is not done soon. Cheers, -- Pierre @pierrejoye | http://www.libgd.org -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
