On Tue, May 31, 2011 at 9:58 AM, Tom Samplonius <t...@samplonius.org> wrote:
> > > > The presentation implied that there was vast goals for the project, > > > including a lot of localization features. It seems like some of the > > > smaller features can be worked into a Son-of-Unicode project, and > > > maybe rolled into 5.5? > > > > it would be a good thing, but nobody stepped up for that, and it > > seems that somehow it isn't really important for the internals. > > This happens in every open source project. You can't expect a group of > open source developers to develop what you ask for. So, it isn't the fault > of "internals" that no one is stepping up. Go onto any open source > developers mailing list, and send a "You guys should do X", and you'll see > the exact same thing happen. > > you was the one who proposed that we should have this rolled. I just said that nobody took the responsibility to push this forward. so actually you are the one who said that "You guys should do X" > On top of that, where is no concept for what Unicode-for-PHP should be. > There was for the previous (now deemed to have failed) project, including > UTF16. But should it be UTF16 or UTF8 or should both be supported? Unicode > identifiers...yes or no? I'm quite interested what everyone's Unicode > wishlist would look like. > > without somobody who is commited to the cause and has the neccessary technical knowledge both in the php core and in unicode stuff, this kind of discussions is deemed to fail. personally I would go for UTF-8, as Andrei suggested, and also expose the ICU facility to the php userland(extend/polish the intl ext.) the main problem was with the previous attempt that we had to rewrite a whole lot of code to make them unicode aware and that didn't work out. we should take this into account, and probably come up with some middle-ground implementation, which is maybe a second class citizen, or slower that, both it would be better that we have with mbstring, and we won't screw up the people who don't need it. > > > But right now, there is this impression of failure, and implication > > > that PHP has no Unicode, because a project to add UTF16 to > > > everything in PHP wasn't completed. > > > > yes there is the impression, Unicode support was the flagship feature > > of the PHP6 release. > > Actually, for me, the deprecated features were far more important than > Unicode, as I can create Unicode web apps right now. thats weird, I mean I agree that killing some bad decisions after 10 years is a good thing, but I wouldn't upgrade to PHP6 for that alone. > The PHP Wikipedia page is wrong about this too, and states that > addslashes() can be used instead of magic quotes. addslashes() should > probably be deprecated too (and it isn't Unicode aware either, so removing > it solves two problems). > magic_quotes did the same thing than manually addslashing every argument AFAIK, so I can't see why you say they couldn't be used interchangeably. the problem with magic was the magic part, addslashes is a valid feature on its own. Tyrael
