Re: session bug or feature

register_global is currently on. The problem is, I am hacking into someone else's (awful awful) code, and if I turned off register_globals, the whole application would go kaput. I am not even using the old session_x functions, just the new associative session array $_SESSION. "Curt Zirzow" <curt@zirzow.dyndns.org> wrote in message news:20030807171325.GZ21938@bagend.shire... > * Thus wrote Christian Calloway (callowaylc@yahoo.com): > > Hey Everyone, > > > > I am running Apache 2.047 with PHP (as module) 4.3.2. I ran into something > > interesting and I wanted to know if it was a bug, or actually supposed to be > > that way. Given the following lines of code: > > > > <?php > > > > session_start(); > > > > // lets say this equals "bar" and it was set on a previous page > > $_SESSION["foo"]; > > > > $foo = "rab"; > > > > > > echo $_SESSION["foo"]; > > > > ?> > > > > > > The problem is, when I set the global variable $foo="rab", when I echo the > > session variable $_SESSION["foo"], it outputs "rab" instead of > > "bar"?! This > > doesn't seem right? If that is the intended behavior, is there anyway to > > avoid this (perhaps a php.ini directive)? It was a nightmare finding out why > > my session variables values were being changed by like-named global > > variables ;-) Hey PHP still rocks though, later > > ---- > php4-3.3RC1 > Apache 1.3 > register_globals: off > ---- > Hm.. I can't reproduce that, what are you're php settings that > affect session stuff, and your register_globals setting? > > My guess is that you have a session_register('foo') somewhere. > > Curt > -- > "I used to think I was indecisive, but now I'm not so sure."