On Sat, Mar 07, 2009 at 05:52:55AM -0500, Robert Cummings wrote: > On Sat, 2009-03-07 at 02:12 -0500, Paul M Foster wrote: > > On Sat, Mar 07, 2009 at 11:29:41AM +1100, Clancy wrote: > > > > > On Fri, 6 Mar 2009 08:53:44 -0500, danbr...@php.net (Daniel Brown) wrote: > > > > > > >On Fri, Mar 6, 2009 at 00:12, Clancy <clanc...@cybec.com.au> wrote: > > > > <snip> > > > > > > > > > > Then you'll be happy with the advent of PHP6: > > > > > > > > http://php.net/goto > > > > > > > Someone would add gotos to a language *on purpose*?! > > The demonization of goto was due to the kind of goto found in older > languages that would either a) goto an arbitrary line number where line > numbers might change as new line numbers were needed (who invented that > concept anyways???). The other was where you might jump willie nillie to > a globally defined goto label that might appear anywhere in a program > versus jumping to a goto label within a well defined context. The goto > added to PHP allows jumping only within the current context... you can't > jump out of a function or into a function, nor as the help says can you > jump into a loop, or into a switch statement. What this does is greatly > simplify some types of coding where you might otherwise have to use > multi level breaks, or define a temporary variable whose value is > checked at various breakpoints. Goto resolves these issues by allowing > jumping directly to a defined label within the context. Additionally, > implementation of things likes parsers are extremely succinct using goto > semantics. Goto has a place in languages, just not ad-hoc poorly thought > out goto.
My deities (Kernighan and Ritchie) told me not to use it, and I don't want to offend them. They also told me how to indent my code. ;-} Once we have goto, it's a short slide down a slippery slope to setjmp and longjmp. And thence, the apocalypse. ;-} Paul -- Paul M. Foster -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php