Re: [xslt] rand() .vs. random()

[Vincent Lefevre <vincent+gnome vinc17 org>]

On 2010-11-05 21:30:02 +0100, Daniel Veillard wrote:
> On Fri, Oct 08, 2010 at 11:07:34AM -0400, Phil Shafer wrote:
> > The math:random() function ends up calling rand(), not random():
[...]
>  we had a thread about this earlier:
> 
>   http://mail.gnome.org/archives/xml/2008-February/msg00092.html
> 
> what is crucially needed is an srand initialization there, I think
> that's the most important point, but we could detect random() from
> configure and plug it in too.

For some applications, reproducibility is important, so that the
default seed at the beginning must be the same. So, I'd say that
math:random is under-specified in the EXSLT spec (it should also
say whether 0 and 1 are inclusive or not). The description of
random:random-sequence is more clear. If you intend to implement
this one too, you may think about it, as random() would be
incorrect as not necessarily being thread-safe.

-- 
Vincent Lefèvre <vincent vinc17 net> - Web: <http://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / Arénaire project (LIP, ENS-Lyon)