Re: Re: Question about superglobals

On Sat, 2007-11-17 at 01:40 -0500, Sam Barrow wrote: > I don't understand, people keep telling me this can cause problems so > it's trash. Really? Functions and variables can cause problems too. Not > to mention user input. Programming is not easy, but the more flexible it > is the better. I think it's up to the programmer to keep track of his > variables, rather than us limiting his functionality, just incase he > makes a mistake. > > Any programmer can write bad code and it won't work. If he uses > superglobals he should understand that he needs to document them, it's > that simple. If that's too difficult, don't create a superglobal then, > great. > > And if this isn't good enough, fine. Require an underscore for > superglobals to differentiate them even more. I'll write the code > myself, no problem. This isn't the same issue as functions and classes residing in the same global namespace. This is a problem with a single declaration affecting variables having the same name in EVERY scoping context. If you redefine a function --> ERROR If you redefine a class --> ERROR If you define a super global named $cfg --> NO ERROR, NO WARNING, NO NOTICE GLOBAL CLOBBERING of ALL variables names $cfg in global scope, local function scope, class method scope. That's one hell of a problem to track down to save a few lazy people some keystrokes that in their naivety they think is for the greater good. Cheers, Rob. -- ........................................................... SwarmBuy.com - http://www.swarmbuy.com Leveraging the buying power of the masses! ...........................................................