Posted by James Benson on 11/05/05 18:49

Max Belushkin wrote:
>>Try using require_once or include_once
>
> Well, here's all the fun starting. This is relatively long, but please bear
> with me.
> First of all, yes, I did include_once, require_once, even resorted to "if (!
> $inc_done) include_once("func.php")", and setting $inc_done in func.php.
> No-go, was still getting the "Cannot redefine mquery()" error message.
> Now for the fun.
> I've had a different set of scripts employing a different "func.php". Let's
> say, for simplicity, /home/test1/index.php included /home/test1/func.php,
> and /home/test2/index.php included /home/test2/func.php. However, in both of
> those, mquery() is defined. But they do *not* include anything else, and the


So you defined mquery() in both files and one include's or require's the
other file?


If so thats your problem, you cannot re-define functions



> paths where set as written above. And, what do you know? If I worked inside
> test1 for a while, and went to test2, I'd get:
> Cannot redefine mquery() in /home/test2/func.php line 9 (first defined
> in /home/test1/func.php on line 9). Voila.
>
> Going through PHP 4.4.1 release notes, I noticed the change stating "fixed a
> memory corruption bug in include()". Now, I don't know how FreeBSD ported the
> PHP code into it's ports tree, but for me, on FreeBSD 5.1, PHP 4.4.1
> *introduced* a memory corruption bug in include.
>
> I downgraded to 4.4.0, and everything works perfectly fine again.
>
> I've written to the freebsd-questions mailing list, but got no sensible
> reply so far (only one was "upgrade the ports that use PHP to newer versions,
> but my own simple set of scripts exhibits this very strange behaviour, so
> it's not a solution).
>
> I'll wait a few days to see if anyone else in FreeBSD experiences the same
> behaviour (the port was only introduced 2-3 days ago) before posting a bug
> report, else if no one else sees the same thing I did, I'll have to dig
> through PHP code to see what was modified in the include()-related code, I
> guess, and see what influences the behaviour I'm seeing...
>
> Comments?

[Back to original message]