On Mon, Apr 01, 2002 at 05:32:50PM +0100, Nick Ing-Simmons wrote:
> Dan Kogai <[EMAIL PROTECTED]> writes:
> >On Tuesday, April 2, 2002, at 12:26 , Jarkko Hietaniemi wrote:
> >> Just checking: is enc2xs really a widely enough useful tool to install
> >> for everybody? It's pretty specialized, after all.
> >
> > It is special indeed for those wanting to add ucm-based new
> >encodings. However, External encoding modules stand no chance without
> >it. So far Encode::HanExtra works fine but once for example, NI-S
> >decides to rehack encengine.c, the Encode::HanExtra need to be updated.
> >with enc2xs, all you need is recompilation should that occur.
> > Don't expect those who want to add encodings are programmers.
> >Autrijus happens to be a programmer (and darn good one!) so HanExtra
> >would be ok with or without enc2xs. But with more modules to come (such
> >as proposed Encode::JISX0213), enc2xs is needed and it is rather
> >ludicrous to get the whole Encode distro just to get enc2xs.....
>
> NI-S's Tk extension may decide to add encodings for fonts it comes across.
> (FWIW I note that in Tcl/Tk the Tk part adds some encodings that are
> apparently only of interest for fonts - in their case ucs-2be.
> So _maybe_ perl/Tk can take back any "raw" encodings which are only
> of use for fonts.)
>
> That said we don't need to install enc2xs is a public bin dir, so long
> as it is installed somewhere. Compare xsubpp itself - that is not
Ahhh, that makes more sense. $Config{privlibexp}/Extutils/enc2xs?
> put in a 'bin' but MakeMaker knows where to find it for .xs -> .c
> conversion. We could do the same for .ucm -> .xs ??
>
> >
> >Dan the Encode Maintainer
> --
> Nick Ing-Simmons
> http://www.ni-s.u-net.com/
--
$jhi++; # http://www.iki.fi/jhi/
# There is this special biologist word we use for 'stable'.
# It is 'dead'. -- Jack Cohen
