Re: [PHP4BETA] cvs: php4 /ext/standard/ levenshtein.c

On Tue, 23 May 2000, Hartmut Holzgraefe wrote:

> Zeev Suraski wrote:
> > The problem with the 'dirty trick' is that sizeof(void *) isn't necessarily
> > equal to sizeof(int).  On many platforms it isn't.
> 
> sure, i know that, but chances that a long is smaler than a pointer are
> not 
> that bad,i even thought about protecting it with an assert() instead of 
> doing it right in the first place, stupid me ...

There is no guarantee that casting a pointer to long and casting the
result to a pointer will yield the original pointer.

That means the code must not rely on this assumption, or it is considered
broken.

> but i never thought of this as a _compile_ time problem ... :(

GCC is not the right tool for testing source code on conformance to
existing standards. Most other compilers are stricter in one or another
way.

> PS: seems that cvs.php.net doesn't know about my new address
>     ([email protected] instead of [email protected])
>     could somebody change this or tell me where to ... ?

I've changed it.

- Sascha