Re: Session Handling Changes between 4.2.3 and 4.3.1
| From: | Bobby Patel | Date: | Tue, 15 Apr 2003 17:07:49 +0000 |
| Subject: | Re: Session Handling Changes between 4.2.3 and 4.3.1 | ||
| References: | 1 | Groups: | php.general |
| Request: | Send a blank email to php-general+get-143981@lists.php.net to get a copy of this message | ||
if you read the documentation about session_register() you will see that
this will not work when register_globals are set to OFF. Is this the case
for the two other servers?
http://www.php.net/manual/en/function.session-register.php
(Read the cautions)
"Sascha Braun" <saschabraun@gmx.de> wrote in message
news:1050450973.8020.13.camel@gentoo.vaio.net...
> Hi,
>
> i worked for a while with the Windows Version of PHP 4.1
> and later 4.2.
>
> After I changed my Development Environment to Linux and
> the PHP Version 4.3.1, some or all of my formerly projects
> showed some errors or are just working in a different way
> than before, due to the changes in session handling.
>
> I have read thru the session handling parts of the 4.2.3
> Docs for PHP and compared it to the way i have programmed
> the session handling in my websites. I could not really
> find any differences.
>
> I do start the sessions in an global.inc.php file with
> session_start() and after that i register them for exam-
> ple with session_register('user') for Customermanagement.
>
> After that I'm referencing the session thru
> $_SESSION['user']['username'] and so on.
>
> I have written one administration tool where everything
> works pretty fine this way, but on two other webpages
> where the usermanagement does not really work this way.
>
> In php.ini there is written some about an 4.2.3 Error
> which lets me initialize sessions in a wrong way, but
> in 4.3 this error should not occure anymore.
>
> Do i use the sessionmanagement in the right way, or do
> i do something wrong. When i use a windows PC for the
> development everything works fine, just Linux seems to
> be a little more konservative.
>
> It would be nice, to get an explaination and maybe so-
> lution for this strange problem.
>
> Thanks indeed
>
> Sascha
>
>