I've been advocating property name restrictions for quite a while now... I'm
totally happy with at least saying they must conform to those rules.
Cheers,
-g
On Thu, Mar 01, 2001 at 11:28:42AM -0500, Jim Blandy wrote:
>
> Yoshiki's function works on binary data. At the moment, we place no
> constraints on property names.
>
> I'm willing to restrict property names somewhat, but not too much.
> Excluding the null byte would be okay. Using UTF-8 in Normalization
> form C would be okay. (And I think that would mean we could carry
> them in cleartext in XML, right?)
>
> What do people think about this?
>
> Ben Collins-Sussman <sussman@newton.ch.collab.net> writes:
>
> >
> > Yoshiki Hayashi <yoshiki@xemacs.org> writes:
> >
> > > +/* Return true iff SKEL is an atom whose data is the same as STR. */
> > > +int svn_fs__atom_matches_string (skel_t *skel, svn_string_t *str);
> >
> > Is this really necessary? Can't you just call
> >
> > svn_fs__matches_atom (skel, string->data)
> >
> > Remember than svn_string_t's are guaranteed to be null-terminated, so
> > you can use them as char *'s effortlessly.
> >
> >
--
Greg Stein, http://www.lyra.org/
Received on Sat Oct 21 14:36:24 2006