Re: Announcement: Homage to SWI-Prolog

On 07/20/2013 07:45 PM, Michael Hendricks wrote:
> On Sat, Jul 20, 2013 at 9:22 AM, Jan Wielemaker <[email protected]
> <mailto:[email protected]>> wrote:
> 
>     The sandbox code is since a few versions available in the core Prolog
>     library.
> 
> 
> Fantastic!
> 
>     Please note that this code is not extensively reviewed.  Use at your own
>     risk.  It might be wise to run such systems also in an OS provided
>     sandbox ...
> 
> 
> Is this why library(sandbox) is not listed in the manual in section A or
> section F?

It is still experimental.  There isn't really a good rule which libraries
end up as part of the main PDF and which not.  There used to be a lot of
simply undocumented code in the libs.  That has improved a lot.  There is
still a lot left to wish for though.

Eventually, if library(sandbox) proves to be a safe and workable solution,
it probably ends up in the main docs.  If may also be that we must conclude
that this is not workable, and that we need something else.  The main
drawback is that the library needs to be able to decide on the set of
reachable predicates from a given goal.  Thus programs using complicate
meta-calling will not pass.

	--- Jan

> 
> -- 
> Michael