11:10 p.m.
Hi,
Gnu libidn, http://www.gnu.org/software/libidn/, includes a simple
Elisp interface consisting of two files, idna.el and punycode.el. They
seem to work out of the box for XEmacs, (not tested with XEmacs 21.4),
so it would be easy to add them to XEmacs Packages.
The author Simon Josefsson thinks that it is OK.
I have one problem though, where to put it? [1]. Since these files
requires utf8 support should they be provided by a mule-package or is
it OK to add them to the net-lib package where they seem to belong
logically.
Yours
Footnotes:
[1] I have been using a mule enabled XEmacs for so long now so I have
forgotten all about what limits non mule XEmacs has.
--
%% Mats
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
8:31 a.m.
Mats Lidell writes:
I have one problem though, where to put it? [1]. Since these files
requires utf8 support should they be provided by a mule-package or is
it OK to add them to the net-lib package where they seem to belong
logically.
If they *contain* UTF-8, they must go in mule-packages. Non-mule
XEmacs 21.4 is not prepared to handle that at all in Lisp libraries
(although modern XEmacs 21.5 can handle UTF-8 and UTF-16, I believe,
at least with an appropriate compile switch but without Mule).
If they don't actually contain UTF-8, but compute it on the fly, they
can go in net-lib, but probably should error on load if Unicode
support is not available.
Steve
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
11:21 p.m.
>>>> Stephen J Turnbull <stephen(a)xemacs.org> writes:
If they *contain* UTF-8, they must go in mule-packages. [...]
Aha. The files them selves are Latin-1 because they contains some
internationalized examples using swedish chars. So that ought to be
OK.
If they don't actually contain UTF-8, but compute it on the fly,
they can go in net-lib, but probably should error on load if Unicode
support is not available.
What is the proper way to check for that?
Yours
--
%% Mats
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
12:56 a.m.
Mats Lidell writes:
>>>>> Stephen J Turnbull <stephen(a)xemacs.org>
writes:
> If they *contain* UTF-8, they must go in mule-packages. [...]
Aha. The files them selves are Latin-1 because they contains some
internationalized examples using swedish chars. So that ought to be
OK.
Yes, that's basically OK because XEmacs --without-mule really only
demands a unibyte encoding. Make sure it has a proper -*- coding:
iso-8859-1 -*- local variable, so Mule will handle it correctly too.
> If they don't actually contain UTF-8, but compute it on the
fly,
> they can go in net-lib, but probably should error on load if Unicode
> support is not available.
What is the proper way to check for that?
(get-coding-system 'utf-8)
is all you need to do. Font support is the user's problem (well, of
course we should document it it, but it's not really a problem on most
platforms these days).
We could do something more complicated to avoid loading mule-ucs on
21.4, but I don't think it's worth it. If somebody loads one of those
libraries, they presumably intend to use them and mule-ucs will be
necessary. If somebody complains, I'll fix it.
Steve
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
11:29 p.m.
>>>> Stephen J Turnbull <stephen(a)xemacs.org> writes:
(get-coding-system 'utf-8)
is all you need to do.
[...]
Are you saying I should just do a top-level call to get-coding-system?
(And let a possible error stop further processing?)
That didn't feel right so I tried to steel the code from somewhere
else but I fail to find any other place that checks for this. So this
would be the first files to check for utf-8 coding system in order to
refuse to load!?
This is what came up with. OK?
(eval-when (load eval)
(unless (find-coding-system 'utf-8)
(error 'invalid-operation "`idna.el' requires support for the utf-8
coding system")))
Yours
--
%% Mats
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
2:50 a.m.
Mats Lidell writes:
>>>>> Stephen J Turnbull <stephen(a)xemacs.org>
writes:
> (get-coding-system 'utf-8)
> is all you need to do.
[...]
Are you saying I should just do a top-level call to get-coding-system?
(And let a possible error stop further processing?)
Exactly.
That didn't feel right so I tried to steel the code from
somewhere
else but I fail to find any other place that checks for this. So this
would be the first files to check for utf-8 coding system in order to
refuse to load!?
Probably. Most Mule functions require coding system support in
general except for ASCII and Latin-1. These are the only libraries I
know that requires a specific encoding other than those two basic
ones.
This is what came up with. OK?
(eval-when (load eval)
(unless (find-coding-system 'utf-8)
(error 'invalid-operation "`idna.el' requires support for the utf-8
coding system")))
OK.
Steve
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta
3761
days inactive
3765
days old
5 comments
2 participants
participants (2)
-
Stephen J. Turnbull -
The XEmacs Package Smoketest