state of BIDI within stable GNU emacs (UTF8) nikud

[Uwe Brauer]

Hello


After a long time I have to use GNU emacs to write some hebrew text
(most likely including nikud).

    - I   am currently trying out  the  stable emacs 21.4 version
      (shipped with Debian  unstable). That  version has no   R2L
      support as far as I can see. Which version has, a snapshot
      version.

    - Could the hebrew  text be saved as UTF8  (I would need that
      urgently)

    - I remember having seen   some emacs branch which  had  even
      nikud support, but   I forgot where, could  somebody please
      give me pointer.

Regards

Uwe Brauer



_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[hossein-3]

Hi.
I want to know more about using Pango for text rendering that can
solve these kind of problems. I read in mailing list achieves that
Emacs developers have tried Pango for text rendering, but they found
it not suitable for Emcas. I want to know why? Is it too hard to make
it suitable for this job? I think it would be a nice job to do this,
integration with GNOME desktop, antialieased fonts, good
internationalization support, etc.
Thanks.


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Eli Zaretskii]

> From: Uwe Brauer <oub@...>
> Date: Fri, 18 Nov 2005 18:28:52 +0100
>
> After a long time I have to use GNU emacs to write some hebrew text
> (most likely including nikud).
>
>     - I   am currently trying out  the  stable emacs 21.4 version
>       (shipped with Debian  unstable). That  version has no   R2L
>       support as far as I can see. Which version has, a snapshot
>       version.

Sorry, there's still no support for bidi editing in any version of
Emacs, not even in the CVS.

>     - Could the hebrew  text be saved as UTF8  (I would need that
>       urgently)

Yes.

>     - I remember having seen   some emacs branch which  had  even
>       nikud support, but   I forgot where, could  somebody please
>       give me pointer.

I don't think anything like this exists.


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Eli Zaretskii]

> Date: Mon, 21 Nov 2005 02:56:37 +0330
> From: Hossein Noorikhah <hossein.ir@...>
>
> I want to know more about using Pango for text rendering that can
> solve these kind of problems. I read in mailing list achieves that
> Emacs developers have tried Pango for text rendering, but they found
> it not suitable for Emcas. I want to know why?

I once wrote here the reason: Pango, like fribidi, is a batch-mode
reordering library: you hand it a buffer with text and it returns it
reordered for display.

By contrast, Emacs needs a reordering function that could be called
repeatedly, and will on every call return the next character in the
visual order.  This is because the Emacs display engine walks the
buffer one character at a time and decides how to display it based on
data structures built by the application.

I wrote such a sequential version of bidi reordering, and it was
integrated into redisplay on a special branch of Emacs CVS.  But no
one had time to debug it and expand it, unfortunately.


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[TAKAHASHI Naoto]

Uwe Brauer writes:

>     - I remember having seen   some emacs branch which  had  even
>       nikud support, but   I forgot where, could  somebody please
>       give me pointer.

This one?

  http://www.m17n.org/emacs-bidi/heb.png

--
TAKAHASHI Naoto
ntakahas@...
http://www.m17n.org/ntakahas/


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Uwe Brauer]

>>>>> " " == TAKAHASHI Naoto <ntakahas@...> writes:

    > Uwe Brauer writes:
   >> - I remember having seen   some emacs branch which  had  even
   >> nikud support, but   I forgot where, could  somebody please
   >> give me pointer.

    > This one?

    >   http://www.m17n.org/emacs-bidi/heb.png

Most likely, is it possible to obtain a copy of the source code?



_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Micha Feigin]

On Mon, 21 Nov 2005 12:10:17 +0100
Uwe Brauer <oub@...> wrote:

> >>>>> " " == TAKAHASHI Naoto <ntakahas@...> writes:
>
>     > Uwe Brauer writes:
>    >> - I remember having seen   some emacs branch which  had  even
>    >> nikud support, but   I forgot where, could  somebody please
>    >> give me pointer.
>
>     > This one?
>
>     >   http://www.m17n.org/emacs-bidi/heb.png
>
> Most likely, is it possible to obtain a copy of the source code?
>

Sure, at
http://www.m17n.org/emacs-bidi/

>
>
> _______________________________________________
> emacs-bidi mailing list
> emacs-bidi@...
> http://lists.gnu.org/mailman/listinfo/emacs-bidi
>  
>  +++++++++++++++++++++++++++++++++++++++++++
>  This Mail Was Scanned By Mail-seCure System
>  at the Tel-Aviv University CC.
>

 
 +++++++++++++++++++++++++++++++++++++++++++
 This Mail Was Scanned By Mail-seCure System
 at the Tel-Aviv University CC.


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Gregg Reynolds]

Eli Zaretskii wrote:

>>From: Uwe Brauer <oub@...>
>>Date: Fri, 18 Nov 2005 18:28:52 +0100
>>
>>After a long time I have to use GNU emacs to write some hebrew text
>>(most likely including nikud).
>>
>>    - I   am currently trying out  the  stable emacs 21.4 version
>>      (shipped with Debian  unstable). That  version has no   R2L
>>      support as far as I can see. Which version has, a snapshot
>>      version.
>
>
> Sorry, there's still no support for bidi editing in any version of
> Emacs, not even in the CVS.
>
>

Sure there is!  Well, sort of.  Emacs correctly shapes Arabic *words*
(including diacritics) RTL, but sentences run LTR.  Sometimes you have
to C-l to get it to redraw the shapes.  It works surprisingly well for
many browse/edit tasks; it's not that hard to get used to reading Arabic
that way.  The advantage is of course that you can use all the standard
Emacs functionality on Arabic text.  I dunno if it would work this way
for Hebrew.

I use emacs (v 22.0.50.2, also one of the 21.3 versions) all the time
for editing (not composing, except for short bits) Arabic plaintext.  I
had to create an Arabic Quail package to get an Arabic keyboard layout, e.g.

(require 'quail)

(quail-define-package
  "arabic-ar-AR" "arabic-ar-AR" "aar<"
  t
  "arabic-ar character input method with Arabic keyboard layout

Doubling the postfix separates the letter and postfix: e.g. a'' -> a'
" nil t nil nil nil nil nil nil nil nil t)

(quail-define-rules

        ("f" ?ب) ;; baa
        ("j" ?ت)  ;; taa
        ("e" ?ث) ;; thaa
        etc.
)

BTW, you know that Vim supports RTL?  But without bidi reordering.

Hope that helps,

gregg


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Eli Zaretskii]

> Date: Mon, 21 Nov 2005 15:21:55 -0600
> From: Gregg Reynolds <gar@...>
> CC: Uwe Brauer <oub@...>, emacs-bidi@...
>
> > Sorry, there's still no support for bidi editing in any version of
> > Emacs, not even in the CVS.
>
> Sure there is!  Well, sort of.  Emacs correctly shapes Arabic *words*
> (including diacritics) RTL, but sentences run LTR.  Sometimes you have
> to C-l to get it to redraw the shapes.

Is this on MS-Windows, by any chance?  Because if it is, then Emacs
has nothing to do with this: it's the Windows function used to display
a word, it reorders the logical into visual order automatically.  But
as soon as you move the cursor with C-f or the arrow keys, the
characters are displayed in their logical order, LTR.  That's not bidi
support, it's not even close.

> BTW, you know that Vim supports RTL?  But without bidi reordering.

Yes, I know.


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

["Martin J. Dürst"]

At 13:53 05/11/21, Eli Zaretskii wrote:

 >I once wrote here the reason: Pango, like fribidi, is a batch-mode
 >reordering library: you hand it a buffer with text and it returns it
 >reordered for display.
 >
 >By contrast, Emacs needs a reordering function that could be called
 >repeatedly, and will on every call return the next character in the
 >visual order.  This is because the Emacs display engine walks the
 >buffer one character at a time and decides how to display it based on
 >data structures built by the application.

It looks to me like you could just use fribidi or so, with an
additional layer that caches the batch-mode results and returns
them one-by-one. You'd want to cache quite a bit of stuff even
in a character walking implementation, I guess, although there
are probably some shortcuts you can make (haven't thought about
that yet in detail).

 >I wrote such a sequential version of bidi reordering, and it was
 >integrated into redisplay on a special branch of Emacs CVS.  But no
 >one had time to debug it and expand it, unfortunately.

Can you give a definitive pointer (ideally URI) to that branch?
Thanks!

Regards,    Martin.



_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Eli Zaretskii]

> Date: Tue, 22 Nov 2005 10:30:06 +0900
> From: Martin Duerst <duerst@...>
> Cc: emacs-bidi@...
>
> It looks to me like you could just use fribidi or so, with an
> additional layer that caches the batch-mode results and returns
> them one-by-one.

This was considered.  Such a caching would be a performance hit, and
it also requires significant changes in the logic and structure of the
Emacs display code.

>  >I wrote such a sequential version of bidi reordering, and it was
>  >integrated into redisplay on a special branch of Emacs CVS.  But no
>  >one had time to debug it and expand it, unfortunately.
>
> Can you give a definitive pointer (ideally URI) to that branch?

The branch name is emacs-bidi.  The instructions to use the Emacs CVS
can be found here:

   http://savannah.gnu.org/cvs/?group=emacs


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Gregg Reynolds]

Eli Zaretskii wrote:

>>Date: Mon, 21 Nov 2005 15:21:55 -0600
>>From: Gregg Reynolds <gar@...>
>>CC: Uwe Brauer <oub@...>, emacs-bidi@...
>>
>>
>>>Sorry, there's still no support for bidi editing in any version of
>>>Emacs, not even in the CVS.
>>
>>Sure there is!  Well, sort of.  Emacs correctly shapes Arabic *words*
>>(including diacritics) RTL, but sentences run LTR.  Sometimes you have
>>to C-l to get it to redraw the shapes.
>
>
> Is this on MS-Windows, by any chance?  Because if it is, then Emacs
> has nothing to do with this: it's the Windows function used to display
> a word, it reorders the logical into visual order automatically.  But
> as soon as you move the cursor with C-f or the arrow keys, the
> characters are displayed in their logical order, LTR.  That's not bidi
> support, it's not even close.
>

Aha.  I always wondered why Emacs had this strange capability.  Yes,
this is just nt-emacs; I've never been able to figure out how to control
fonts on my other machine (OS X), so I've not tested it elsewhere.
Well, in any case it's quite useful.  VERY eagerly awaiting rtl arabic
support in Emacs -

thanks,

gregg


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Uwe Brauer]

>>>>> "Micha" == Micha Feigin <michf@...> writes:

   Micha> On Mon, 21 Nov 2005 12:10:17 +0100
   Micha> Uwe Brauer <oub@...> wrote:

   >>
   >> > This one?
   >>
   >> >   http://www.m17n.org/emacs-bidi/heb.png
   >>
   >> Most likely, is it possible to obtain a copy of the source code?
   >>

   Micha> Sure, at
   Micha> http://www.m17n.org/emacs-bidi/

Ok thanks, one question remains:
why didn't enter  this into main emacs, lack of support,
incompatibility of Mule internals?


Regards

Uwe Brauer




_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Eli Zaretskii]

> From: Uwe Brauer <oub@...>
> Date: Wed, 23 Nov 2005 13:30:04 +0100
>
>    Micha> Sure, at
>    Micha> http://www.m17n.org/emacs-bidi/
>
> Ok thanks, one question remains:
> why didn't enter  this into main emacs, lack of support,
> incompatibility of Mule internals?

The above page spells this out (under "Future Plans"):

    The Emacs core developers concluded that the bi-directional drawing
    method used here is not suitable to be integrated into the original
    GNU Emacs.  Another attempt to implement a bidi algorithm in Emacs'
    display engine is ongoing and it is likely to be a part of future GNU
    Emacs.  An overview is described in the DisplayEngineForBidi section of
    the Emacs Wiki web site.

The ``other ongoing attempt'' it refers to is the code I wrote, but
unfortunately no one had time to work on it since it was written and
integrated into the Emacs display engine.  It needs a lot of
debugging.


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Thamer Mahmoud]

[I had some trouble accessing this list through gnu.org, so my
apologies if this message appeared twice, or if someone is
inappropriately CCed]

Gregg Reynolds <gar@...> writes:

> I had to create an Arabic Quail package to get an Arabic keyboard
> layout, e.g.
>

I have previously submitted a patch to emacs-dev with such additions
and other minor improvements for Arabic. It did not receive any
responses from emacs-dev, however.

Eli, any thoughts on the attached patch? Is there an interest in
seeing it checked in to Emacs proper? It's a trivial patch, but I'd be
willing to sign the GNU papers if necessary.

This makes distributing a minor mode for Arabic editing, or adapting
other Hebrew-specific packages for use with Arabic, alot easier.

The attached files include the following:

arabic.patch:
      - Enable support for iso-8859-6.
      - Add properly encoded Arabic characters to the HELLO file.
      - Remove a block of code from code-pages.el that conflicts
        with the new iso-8859-6 support.  

arabic.el: Language file for Arabic.

arabic.el: quail input modes for Arabic and Persian.

Regards,
Thamer Mahmoud


[arabic22.patch]

---

? arabic22.patch
? leim/quail/arabic.el
? lisp/language/arabic.el
Index: etc/HELLO
===================================================================
RCS file: /cvsroot/emacs/emacs/etc/HELLO,v
retrieving revision 1.40
diff -u -u -r1.40 HELLO
--- etc/HELLO 3 Apr 2004 04:41:21 -0000 1.40
+++ etc/HELLO 22 Nov 2005 05:59:14 -0000
@@ -2,7 +2,7 @@
 Its purpose is to illustrate a number of scripts.
 ---------------------------------------------------------
 Amharic ($(3"c!<!N"^(B) $(3!A!,!>(B
-Arabic (38R(47d(3T!JSa(4W(3W(B
+Arabic (,GGdYQHjI(B) (,GGdSdGe(B ,GYdjce(B)
 Braille $,2(3(1('('(5(B
 C printf ("Hello, world!\n");
 Czech (,Bh(Be,B9(Btina) Dobr,B}(B den
@@ -57,6 +57,7 @@
 A short test for characters represented by the character set
 mule-unicode-0100-24ff:
 
+Arabic ($,1-g.$-y-q-h.*-i(B) ($,1-g.$-s.$-g.%(B $,1-y.$.*.#.%(B)
 Czech ($,1 ,(Besky) Dobr,A}(B den
 Esperanto Saluton (E$,1 E(Bo$,1 }(Ban$,1 =(Bo $,1 )(Biu$,1 U(Ba$,1!-(Bde)
 Greek ($,1'5';';'7'='9':',(B) $,1&s'5'9',(B $,1'C'1'B(B
Index: leim/Makefile.in
===================================================================
RCS file: /cvsroot/emacs/emacs/leim/Makefile.in,v
retrieving revision 1.72
diff -u -u -r1.72 Makefile.in
--- leim/Makefile.in 3 Nov 2005 17:00:51 -0000 1.72
+++ leim/Makefile.in 22 Nov 2005 05:59:14 -0000
@@ -125,6 +125,7 @@
 RUSSIAN=${srcdir}/quail/cyrillic.elc ${srcdir}/quail/cyril-jis.elc
 
 OTHERS= \
+ ${srcdir}/quail/arabic.elc \
  ${srcdir}/quail/ethiopic.elc \
  ${srcdir}/quail/ipa.elc \
  ${srcdir}/quail/hebrew.elc \
Index: leim/makefile.w32-in
===================================================================
RCS file: /cvsroot/emacs/emacs/leim/makefile.w32-in,v
retrieving revision 1.29
diff -u -u -r1.29 makefile.w32-in
--- leim/makefile.w32-in 4 Jul 2005 15:27:32 -0000 1.29
+++ leim/makefile.w32-in 22 Nov 2005 05:59:15 -0000
@@ -114,6 +114,7 @@
 RUSSIAN=$(srcdir)/quail/cyrillic.elc $(srcdir)/quail/cyril-jis.elc
 
 MISC= \
+ $(srcdir)/quail/arabic.elc \
  $(srcdir)/quail/ethiopic.elc \
  $(srcdir)/quail/ipa.elc \
  $(srcdir)/quail/hebrew.elc \
Index: lib-src/makefile.w32-in
===================================================================
RCS file: /cvsroot/emacs/emacs/lib-src/makefile.w32-in,v
retrieving revision 2.38
diff -u -u -r2.38 makefile.w32-in
--- lib-src/makefile.w32-in 11 Sep 2005 22:28:28 -0000 2.38
+++ lib-src/makefile.w32-in 22 Nov 2005 05:59:15 -0000
@@ -203,6 +203,7 @@
  $(lispsource)case-table.elc
 
 lisp2 = \
+ $(lispsource)language/arabic.el \
  $(lispsource)language/chinese.elc \
  $(lispsource)language/cyrillic.elc \
  $(lispsource)language/indian.elc \
Index: lisp/ldefs-boot.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/ldefs-boot.el,v
retrieving revision 1.11
diff -u -u -r1.11 ldefs-boot.el
--- lisp/ldefs-boot.el 28 Oct 2005 17:06:02 -0000 1.11
+++ lisp/ldefs-boot.el 22 Nov 2005 05:59:38 -0000
@@ -29877,7 +29877,7 @@
 ;;;;;;  "international/subst-gb2312.el" "international/subst-jis.el"
 ;;;;;;  "international/subst-ksc.el" "international/ucs-tables.el"
 ;;;;;;  "international/utf-16.el" "international/utf-8.el" "isearch.el"
-;;;;;;  "jka-cmpr-hook.el" "kermit.el" "language/chinese.el" "language/cyrillic.el"
+;;;;;;  "jka-cmpr-hook.el" "kermit.el" "language/arabic.el" "language/chinese.el" "language/cyrillic.el"
 ;;;;;;  "language/czech.el" "language/devanagari.el" "language/english.el"
 ;;;;;;  "language/ethiopic.el" "language/european.el" "language/georgian.el"
 ;;;;;;  "language/greek.el" "language/hebrew.el" "language/indian.el"
Index: lisp/loadup.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/loadup.el,v
retrieving revision 1.144
diff -u -u -r1.144 loadup.el
--- lisp/loadup.el 16 Nov 2005 02:26:37 -0000 1.144
+++ lisp/loadup.el 22 Nov 2005 05:59:39 -0000
@@ -98,6 +98,7 @@
   (load "international/latin-8")
   (load "international/latin-9"))
 ;; Load language-specific files.
+(load "language/arabic")
 (load "language/chinese")
 (load "language/cyrillic")
 (load "language/indian")
Index: lisp/international/characters.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/international/characters.el,v
retrieving revision 1.54
diff -u -u -r1.54 characters.el
--- lisp/international/characters.el 4 Jul 2005 17:22:27 -0000 1.54
+++ lisp/international/characters.el 22 Nov 2005 05:59:40 -0000
@@ -1235,7 +1235,7 @@
    (latin-iso8859-4 . iso-latin-4)
    (thai-tis620 . thai-tis620)
    (greek-iso8859-7 . greek-iso-8bit)
-   (arabic-iso8859-6 . iso-2022-7bit)
+   (arabic-iso8859-6 . arabic-iso-8bit)
    (hebrew-iso8859-8 . hebrew-iso-8bit)
    (katakana-jisx0201 . japanese-shift-jis)
    (latin-jisx0201 . japanese-shift-jis)
Index: lisp/international/code-pages.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/international/code-pages.el,v
retrieving revision 1.31
diff -u -u -r1.31 code-pages.el
--- lisp/international/code-pages.el 24 Sep 2005 13:43:59 -0000 1.31
+++ lisp/international/code-pages.el 22 Nov 2005 05:59:41 -0000
@@ -3791,72 +3791,6 @@
 (define-coding-system-alias 'iso-8859-16 'iso-latin-10)
 (define-coding-system-alias 'latin-10 'iso-latin-10)
 
-;; Unicode-based alternative which has the possible advantage of
-;; having its relative sparseness specified.
-;;;###autoload(autoload-coding-system 'iso-8859-6 '(require 'code-pages))
-(cp-make-coding-system
- ;; The base system uses arabic-iso-8bit, but that's not a MIME charset.
- iso-8859-6
- [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
-  nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
-  ?\ 
-  ?\¤
-  ?\،
-  ?\­
-  ?\؛
-  ?\؟
-  ?\Ø¡
-  ?\Ø¢
-  ?\Ø£
-  ?\ؤ
-  ?\Ø¥
-  ?\ئ
-  ?\ا
-  ?\ب
-  ?\Ø©
-  ?\ت
-  ?\Ø«
-  ?\ج
-  ?\Ø­
-  ?\Ø®
-  ?\د
-  ?\Ø°
-  ?\ر
-  ?\ز
-  ?\س
-  ?\Ø´
-  ?\ص
-  ?\ض
-  ?\Ø·
-  ?\ظ
-  ?\ع
-  ?\غ
-  ?\ـ
-  ?\ف
-  ?\ق
-  ?\ك
-  ?\ل
-  ?\م
-  ?\ن
-  ?\ه
-  ?\و
-  ?\ى
-  ?\ي
-  ?\ً
-  ?\ٌ
-  ?\ٍ
-  ?\َ
-  ?\ُ
-  ?\ِ
-  ?\ّ
-  ?\ْ
-  nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
-  nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil
-  nil nil nil nil nil nil nil nil nil nil nil]
- "Unicode-based Arabic ISO/IEC 8859-6 (MIME: ISO-8859-6)"
- ?6)
-(define-coding-system-alias 'arabic-iso-8bit 'iso-8859-6)
-
 ;;;###autoload(autoload-coding-system 'iso-8859-10 '(require 'code-pages))
 (cp-make-coding-system
  iso-latin-6
Index: lisp/international/mule-cmds.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/international/mule-cmds.el,v
retrieving revision 1.287
diff -u -u -r1.287 mule-cmds.el
--- lisp/international/mule-cmds.el 20 Nov 2005 07:45:19 -0000 1.287
+++ lisp/international/mule-cmds.el 22 Nov 2005 05:59:44 -0000
@@ -2103,7 +2103,7 @@
     ("af" . "Latin-1") ; Afrikaans
     ("am" "Ethiopic" utf-8) ; Amharic
     ("an" . "Latin-9") ; Aragonese
-    ; ar Arabic glibc uses 8859-6
+    ("ar" "Arabic" iso-8859-6)
     ; as Assamese
     ; ay Aymara
     ("az" . "UTF-8") ; Azerbaijani
Index: lisp/international/ucs-tables.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/international/ucs-tables.el,v
retrieving revision 1.43
diff -u -u -r1.43 ucs-tables.el
--- lisp/international/ucs-tables.el 16 Nov 2005 22:21:16 -0000 1.43
+++ lisp/international/ucs-tables.el 22 Nov 2005 05:59:46 -0000
@@ -129,6 +129,10 @@
   "Used as `translation-table-for-encode' for iso-8859-5.
 Translates from the iso8859 charsets and `mule-unicode-0100-24ff'.")
 
+(defvar ucs-8859-6-encode-table nil
+  "Used as `translation-table-for-encode' for iso-8859-6.
+Translates from the iso8859 charsets and `mule-unicode-0100-24ff'.")
+
 (defvar ucs-8859-7-encode-table nil
   "Used as `translation-table-for-encode' for iso-8859-7.
 Translates from the iso8859 charsets and `mule-unicode-0100-24ff'.")
@@ -1114,8 +1118,7 @@
 
   ;; Derive tables that can be used as per-coding-system
   ;; `translation-table-for-encode's.
-  ;; N.B., there's no 8859-6 coding system.
-  (dolist (n (list 15 14 9 8 7 5 4 3 2 1))
+  (dolist (n (list 15 14 9 8 7 6 5 4 3 2 1))
     (let* ((alist (symbol-value (intern (format "ucs-8859-%d-alist" n))))
    (encode-translator (set (intern (format "ucs-8859-%d-encode-table"
    n))
@@ -1194,7 +1197,7 @@
 
     ;; Adjust the 8859 coding systems to fragment the unified characters
     ;; on encoding.
-    (dolist (n '(1 2 3 4 5 7 8 9 14 15))
+    (dolist (n '(1 2 3 4 5 6 7 8 9 14 15))
       (let* ((coding-system
       (coding-system-base (intern (format "iso-8859-%d" n))))
      (table (symbol-value
@@ -1237,7 +1240,7 @@
 
     ;; For each charset, remove the parent of `safe-chars' property of
     ;; the corresponding coding system.
-    (dolist (n '(1 2 3 4 5 7 8 9 14 15))
+    (dolist (n '(1 2 3 4 5 6 7 8 9 14 15))
       (let* ((coding-system
       (coding-system-base (intern (format "iso-8859-%d" n))))
      (safe (coding-system-get coding-system 'safe-chars)))
Index: src/Makefile.in
===================================================================
RCS file: /cvsroot/emacs/emacs/src/Makefile.in,v
retrieving revision 1.318
diff -u -u -r1.318 Makefile.in
--- src/Makefile.in 16 Nov 2005 02:38:24 -0000 1.318
+++ src/Makefile.in 22 Nov 2005 05:59:48 -0000
@@ -745,6 +745,7 @@
  ${lispsource}international/latin-8.el \
  ${lispsource}international/latin-9.el \
  ${lispsource}case-table.elc \
+ ${lispsource}language/arabic.el \
  ${lispsource}language/chinese.elc \
  ${lispsource}language/cyrillic.elc \
  ${lispsource}language/indian.elc \
@@ -841,6 +842,7 @@
  ../lisp/international/latin-8.el \
  ../lisp/international/latin-9.el \
  ../lisp/case-table.elc \
+ ../lisp/language/arabic.el \
  ../lisp/language/chinese.elc \
  ../lisp/language/cyrillic.elc \
  ../lisp/language/indian.elc \

---

_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

["Martin J. Dürst"]

At 04:48 05/11/24, Eli Zaretskii wrote:
 >>  http://www.m17n.org/emacs-bidi/

 >The ``other ongoing attempt'' it refers to is the code I wrote, but
 >unfortunately no one had time to work on it since it was written and
 >integrated into the Emacs display engine.  It needs a lot of
 >debugging.

What kind of debugging exactly? Do you have a known list of bugs?
Are there particular features of the implementation that you know
need testing more than others? What kind and amount of testing or
debugging did you do so far? What's the best way to debug the code?
Has anybody done anything like a code review or so? Any other info
that might help anybody to make progress?

Regards,   Martin.



_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Eli Zaretskii]

> Date: Thu, 24 Nov 2005 00:10:17 +0300
> From: Thamer Mahmoud <thamer@...>
> Cc: Gregg Reynolds <gar@...>, emacs-bidi@...
>
> Eli, any thoughts on the attached patch? Is there an interest in
> seeing it checked in to Emacs proper? It's a trivial patch, but I'd be
> willing to sign the GNU papers if necessary.

Given the size of the changes, you will indeed need to sign papers.

In any case, please send the patches to emacs-devel@..., where
more developers will have a chance to review them and comment.

Personally, I don't have any objections to including this, although
I'm not sure how this will help, given that the display engine does
not support Arabic at this time.  Perhaps it is a good idea to
explain, when you post this to emacs-devel, why you think it is useful
to add this input method.

Finally, please include ChangeLog entries with all your changes.

> This makes distributing a minor mode for Arabic editing, or adapting
> other Hebrew-specific packages for use with Arabic, alot easier.

When you post the changes to emacs-devel, please elaborate on how they
make this easier.

>       - Add properly encoded Arabic characters to the HELLO file.

What change is that? how they were encoded before, and how do you
suggest to encode them?

Thanks.


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Eli Zaretskii]

> Date: Thu, 24 Nov 2005 10:56:15 +0900
> From: Martin Duerst <duerst@...>
> Cc: emacs-bidi@...
>
> At 04:48 05/11/24, Eli Zaretskii wrote:
>  >>  http://www.m17n.org/emacs-bidi/
>
>  >The ``other ongoing attempt'' it refers to is the code I wrote, but
>  >unfortunately no one had time to work on it since it was written and
>  >integrated into the Emacs display engine.  It needs a lot of
>  >debugging.
>
> What kind of debugging exactly?

Start using it with both LTR and RTL languages, and explore and debug
any problems (incorrect display, crashes, etc.) that you find.

> Do you have a known list of bugs?

No.  But the code was only tested and debugged outside Emacs, as a
stand-alone reordering program, so I'd guess it has gobs of (unknown)
bugs when used in the context of Emacs display engine.

> Has anybody done anything like a code review or so?

Not that I know of, no.

TIA


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

["Martin J. Dürst"]

At 04:48 05/11/24, Eli Zaretskii wrote:
 >>  http://www.m17n.org/emacs-bidi/

 >The ``other ongoing attempt'' it refers to is the code I wrote, but
 >unfortunately no one had time to work on it since it was written and
 >integrated into the Emacs display engine.  It needs a lot of
 >debugging.

Hello Eli,

Here's yet another question re. your implementation:

At the last Internationalization and Unicode Conference,
I presented a paper that looked at the problems of displaying
(X)HTML or XML with bidi text.
(http://www.sw.it.aoyama.ac.jp/2005/pub/IUC28-bidi/IUC28.html,
see also http://www.sw.it.aoyama.ac.jp/2005/pub/IUC28-bidi/
for more pointers)

Around August, we had an extensive discussion with Ken'ichi
and Naoto in Tokyo, and came to the conclusion that it should
probably be possible to implement this approach in emacs using
the bidi version at http://www.m17n.org/emacs-bidi/.

I'd like to know your oppinion on whether this might also be
possible with your version (the ``other ongoing attempt''
above). What's necessary is either that your algorithm looks
at some well-known character properties that e.g. a mode for
a particular document format could set, or that it provides
callbacks. There may of course be also other ways to do this.

Regards,    Martin.



_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi

[Gregg Reynolds]

Martin Duerst wrote:
...
> At the last Internationalization and Unicode Conference,
> I presented a paper that looked at the problems of displaying
> (X)HTML or XML with bidi text.
> (http://www.sw.it.aoyama.ac.jp/2005/pub/IUC28-bidi/IUC28.html,
> see also http://www.sw.it.aoyama.ac.jp/2005/pub/IUC28-bidi/
> for more pointers)
>

An implementation technique I'd like to see is to simply use background
color to indicate which weakly-directional characters in a bidi string
have been reordered by the algorithm.  That way if I see e.g. >" with
the "reordered-background-color" I can either do the rearrangement in my
head or add a direction override.

-g


_______________________________________________
emacs-bidi mailing list
emacs-bidi@...
http://lists.gnu.org/mailman/listinfo/emacs-bidi