unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
* Changes in mapconv
       [not found] <E1MF1EC-0000Qj-2r@monty-python.gnu.org>
@ 2009-06-13  2:58 ` Eli Zaretskii
  2009-06-15  6:19   ` Kenichi Handa
  0 siblings, 1 reply; 2+ messages in thread
From: Eli Zaretskii @ 2009-06-13  2:58 UTC (permalink / raw)
  To: handa; +Cc: emacs-devel

> Date: Fri, 12 Jun 2009 03:31:00 -0400
> From: emacs-diffs-request@gnu.org
> Reply-To: emacs-diffs@gnu.org
> 
> RCS file: /cvsroot/emacs/emacs/admin/charsets/mapconv,v
> retrieving revision 1.5
> retrieving revision 1.6
> diff -u -b -r1.5 -r1.6
> --- mapconv	8 Jan 2009 04:24:45 -0000	1.5
> +++ mapconv	12 Jun 2009 07:22:48 -0000	1.6
> @@ -30,34 +30,42 @@
>  #   $1: source map file
>  #   $2: address pattern for sed (optionally with substitution command)
>  #   $3: format of source map file
> -#	GLIBC-1 GLIBC-2 GLIBC-2-7 CZYBORRA IANA UNICODE YASUOKA MICROSOFT
> +#	GLIBC-1 GLIBC-2 GLIBC-2-7 CZYBORRA IANA UNICODE YASUOKA
> [...]
>      YASUOKA)
> -	SOURCE="http://kanji.zinbun.kyoto-u.ac.jp/~yasuoka/.../${BASE}";;
> -    MICROSOFT)
> -	SOURCE="http://www.microsoft.com/globaldev/reference/oem/${BASE}";;
> +	BASE="$BASE.Z";
> +	SOURCE="http://kanji.zinbun.kyoto-u.ac.jp/~yasuoka/ftp/CJKtable/${BASE}";;
>      KANJI-DATABASE)

Why were the MICROSOFT parts removed?  The log message does not reveal
any reason, nor even tells about the removal.

Btw, is there any documentation on how to add support for character
maps not in the current CVS?  The only information I see is this
comment in the Makefile:

    #  If your system doesn't have the directory /usr/share/i18n/charmaps,
    #  get the source of the latest glibc, gzip all the charmap files in
    #  the directory "localedate/charmaps", and set the variable
    #  GLIBC_CHARMAPS to that directory.

This has two disadvantages:

  . You have no way of adding a charmap that is not in the glibc
    sources.

  . Downloading the humongous glibc distro just to get a small part of
    it is not my idea of fun.

Suppose I wanted to add codepage 853 support to Emacs: how would I go
about doing that?




^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: Changes in mapconv
  2009-06-13  2:58 ` Changes in mapconv Eli Zaretskii
@ 2009-06-15  6:19   ` Kenichi Handa
  0 siblings, 0 replies; 2+ messages in thread
From: Kenichi Handa @ 2009-06-15  6:19 UTC (permalink / raw)
  To: Eli Zaretskii; +Cc: emacs-devel

In article <E1MFJRT-00076Q-Ti@fencepost.gnu.org>, Eli Zaretskii <eliz@gnu.org> writes:

> > Date: Fri, 12 Jun 2009 03:31:00 -0400
> > From: emacs-diffs-request@gnu.org
> > Reply-To: emacs-diffs@gnu.org
> > 
> > RCS file: /cvsroot/emacs/emacs/admin/charsets/mapconv,v
> > retrieving revision 1.5
> > retrieving revision 1.6
> > diff -u -b -r1.5 -r1.6
> > --- mapconv	8 Jan 2009 04:24:45 -0000	1.5
> > +++ mapconv	12 Jun 2009 07:22:48 -0000	1.6
> > @@ -30,34 +30,42 @@
> >  #   $1: source map file
> >  #   $2: address pattern for sed (optionally with substitution command)
> >  #   $3: format of source map file
> > -#	GLIBC-1 GLIBC-2 GLIBC-2-7 CZYBORRA IANA UNICODE YASUOKA MICROSOFT
> > +#	GLIBC-1 GLIBC-2 GLIBC-2-7 CZYBORRA IANA UNICODE YASUOKA
> > [...]
> >      YASUOKA)
> > -	SOURCE="http://kanji.zinbun.kyoto-u.ac.jp/~yasuoka/.../${BASE}";;
> > -    MICROSOFT)
> > -	SOURCE="http://www.microsoft.com/globaldev/reference/oem/${BASE}";;
> > +	BASE="$BASE.Z";
> > +	SOURCE="http://kanji.zinbun.kyoto-u.ac.jp/~yasuoka/ftp/CJKtable/${BASE}";;
> >      KANJI-DATABASE)

> Why were the MICROSOFT parts removed?  The log message does not reveal
> any reason, nor even tells about the removal.

The map files for CP720 and CP858 are now not available in
Microsoft's page.  So, I made them manually based on the
information in Wikipedia as written in
admin/charsets/mapfiles/README.  As a result,
admin/charset/Makefile doesn't call mapconv with MICROSOFT
as an arg, so I deleted that path from mapconv.

> Btw, is there any documentation on how to add support for character
> maps not in the current CVS?  The only information I see is this
> comment in the Makefile:

>     #  If your system doesn't have the directory /usr/share/i18n/charmaps,
>     #  get the source of the latest glibc, gzip all the charmap files in
>     #  the directory "localedate/charmaps", and set the variable
>     #  GLIBC_CHARMAPS to that directory.

> This has two disadvantages:

>   . You have no way of adding a charmap that is not in the glibc
>     sources.

You can do that by modifying Makefile, and the recent
changes don't change that aspect.

>   . Downloading the humongous glibc distro just to get a small part of
>     it is not my idea of fun.

> Suppose I wanted to add codepage 853 support to Emacs: how would I go
> about doing that?

(1) The most direct way is to add a define-charset call with
    a vector as the value of :map property.

(2) The 2nd direct way is to add a map file in etc/charsets.

(3) Create a file admin/charsets/mapfiles/CP853.map.gz manyally,
    and modify Makefile as these:
	Add CP853.map in CODEPAGE.
	Add these lines:

CP853.map: mapfiles/CP853.map.gz
	# Generating $@...
	@zcat $< > $@

(4) Store a freely available source file in
    admin/charset/mapfiles, and modify Makefile somehow to
    generate CP853.map from that file.

And, after any of (2)..(4), add define-charset call in
lisp/international/mule-conf to define a charset that use
that map file.

---
Kenichi Handa
handa@m17n.org




^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2009-06-15  6:19 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <E1MF1EC-0000Qj-2r@monty-python.gnu.org>
2009-06-13  2:58 ` Changes in mapconv Eli Zaretskii
2009-06-15  6:19   ` Kenichi Handa

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).