RE: [PHP-DEV] feature request

I agree Phil, that is the mature way to examine this. Except, I would be surprised if we _cannot_ apply the same reasoning to the other is_* functions, it's the exact same concept. I'm just not familiar enough with the internal workings to conclude whether is_*($x, $y, $z) would be efficient or even used for that matter. Although I might consider using it like if (isSet($a, $b) && is_int($a, $b)). -Chris > -----Original Message----- > From: Phil Driscoll [mailto:phil@dialsolutions.co.uk] > Sent: Monday, 19 March, 2001 2-11 pM > To: Chris Newbill; Zeev Suraski; Andi Gutmans; usrgre-php@tds.net > Cc: PHP DEV > Subject: Re: [PHP-DEV] feature request > > > Andi said: > > >isset() is not an internal function but a language construct. I > disagree on > >the "using additional arguments for other stuff part". It's often very > >weird that functions behave differently according to the way they are > >called > > I've just done a quick count of the php functions which take an optional > extra argument to change their behaviour, and I stopped when I > reached 100. > > Fair enough, isset is a language construct, but most of the other > isblahs/ext_isblahs are functions, however your average php > programmer won't > even know what this means, let alone be able to discriminate. > > My point is that if you are going to do it to isset, AND the > language is to > become more orthogonal then it MUST also be done to isanythingelse. I can > see that there are no nasty repercussions with using the technique for > isset, however there are many many isanythingelses and I'd be surprised if > we can apply the > reasoning across the board. > > I don't want to stifle innovation here, I just want us to think beyond the > immediate problem when the language gets extended. > > Cheers > -- > Phil Driscoll > Dial Solutions > +44 (0)113 294 5112 > http://www.dialsolutions.com > http://www.dtonline.org > >