bug#4129: Support for dual spacing fonts

bug#4129: Support for dual spacing fonts

[Takanori MATSUURA]

When I use IPAGothic font with emacs-23.1, ASCII characters are shown
as character itself and extra space (ie. "A ", "t ").  The font have
the spacing value of "90" which means dual spacing. If I explicitly
define the spacing to "0", this font-width issue is solved.

Other software which use fontconfig can treat IPAGothic fonts
properly.  So I'm afraid that emacs doesn't handle it.


Fonts are shown with wrong spacing:
Add ~/.emacs to
(set-frame-font "IPAGothic")

Fonts are shown properly:
Add ~/.emacs to
(set-frame-font "IPAGothic:spacing=0")

And add the following fontconfig configuration
<match target="font">
  <test name="family">
     <string>IPAGothic</string>
  </test>
  <edit name="spacing">
     <int>0</int>
  </edit>
  <edit name="globaladvance">
     <bool>false</bool>
  </edit>
</match>


IPAGothic font is available from:
http://ossipedia.ipa.go.jp/ipafont/
(Both Japanese and English contents included in one page. same meaning)


Regards,
Takanori

bug#4129: Support for dual spacing fonts

[Takanori MATSUURA]

[-- Attachment #1: Type: text/plain, Size: 69 bytes --]

Please find the file attached.

The patch fixes bug#4129.


Takanori

[-- Attachment #2: emacs-23.1-spacing.patch --]
[-- Type: text/x-patch, Size: 949 bytes --]

diff --git a/src/ftfont.c b/src/ftfont.c
index 4ebc4be..bac9a4f 100644
--- a/src/ftfont.c
+++ b/src/ftfont.c
@@ -1262,7 +1262,7 @@ ftfont_open (f, entity, pixel_size)
     spacing = XINT (AREF (entity, FONT_SPACING_INDEX));
   else
     spacing = FC_PROPORTIONAL;
-  if (spacing != FC_PROPORTIONAL)
+  if (spacing != FC_PROPORTIONAL && spacing != FC_DUAL)
     font->min_width = font->average_width = font->space_width
       = (scalable ? ft_face->max_advance_width * size / upEM
 	 : ft_face->size->metrics.max_advance >> 6);
diff --git a/src/xftfont.c b/src/xftfont.c
index 82701ce..e74ce31 100644
--- a/src/xftfont.c
+++ b/src/xftfont.c
@@ -418,7 +418,7 @@ xftfont_open (f, entity, pixel_size)
 	ascii_printable[i] = ' ' + i;
     }
   BLOCK_INPUT;
-  if (spacing != FC_PROPORTIONAL)
+  if (spacing != FC_PROPORTIONAL && spacing != FC_DUAL)
     {
       font->min_width = font->average_width = font->space_width
 	= xftfont->max_advance_width;

bug#4129: Emacs still have font spacing problem

[Naohiro Aota]

Hi, list

As I post last year [1], Emacs still doesn't deal with fonts whose
spacing is set to FC_DUAL. Please see this screen-shot I've sent last
year [2]. You see disgusting Elisp indentation, don't you?

This is all because Emacs assume "space width" is as same as "font max
width" only just if the font is not "PROPORTIONAL". Some Japanese are
using "DUAL" fonts (M+ font, IPA font and so on). These fonts have
"glyphs in precisely two widths, one twice as wide as the other" [3],
Emacs shouldn't use "max width" for "min width" and "space width" when
there is two kinds of width!

Please please think applying the patch.

Regards,

[1] http://article.gmane.org/gmane.emacs.devel/111557/
[2] http://article.gmane.org/gmane.emacs.devel/111572/
[3] http://fontconfig.org/fontconfig-devel/fcfreetypecharsetandspacing.html

bug#4129: Emacs still have font spacing problem

[Kenichi Handa]

In article <20100626.021033.1233076457334905269.naota@elisp.net>, Naohiro Aota <naota@elisp.net> writes:

> As I post last year [1], Emacs still doesn't deal with fonts whose
> spacing is set to FC_DUAL. Please see this screen-shot I've sent last
> year [2]. You see disgusting Elisp indentation, don't you?

> This is all because Emacs assume "space width" is as same as "font max
> width" only just if the font is not "PROPORTIONAL". Some Japanese are
> using "DUAL" fonts (M+ font, IPA font and so on). These fonts have
> "glyphs in precisely two widths, one twice as wide as the other" [3],
> Emacs shouldn't use "max width" for "min width" and "space width" when
> there is two kinds of width!

> Please please think applying the patch.

Thank you for the patch, and I'm very sorry for the late
response on this matter.  It seems that I somehow missed
your first report.

I've just applied your patch in this mail:

> [1] http://article.gmane.org/gmane.emacs.devel/111557/

---
Kenichi Handa
handa@m17n.org