X fonts selection weirdness

X fonts selection weirdness

[Yoshiaki Kasahara]

Hello,

I just want to use a specific font for my emacs, using X font backend
(I don't like AA fonts for editing).

When I start emacs with an XLFD font like this:

src/emacs -Q -fn '-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1' -xrm 'Emacs.FontBackend: x'

'-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1' is
selected instead (the displayed characters are distorted by
auto-scaling).

I don't understand why the DPI value is changed to 100.  I explicitly
specified my X server's DPI as 75x75.  Actually I don't care about the
actual DPI setting of X server.  I just want to specify the exact font
name.

Maybe the pointsize '101' is come from X server (by auto-scaling):

% xlsfonts -fn '-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1'    
-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1
-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1
-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1
-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1
-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1
-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1
-misc-fixed-medium-r-normal--14-110-100-100-c-70-iso8859-1
-misc-fixed-medium-r-normal--14-110-100-100-c-70-iso8859-1
-misc-fixed-medium-r-normal--14-110-100-100-c-70-iso8859-1

x-list-fonts returns undesired values for me:

(x-list-fonts "-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1")
("-misc-fixed-medium-r-normal--13-*-100-100-c-70-iso8859-1" "-misc-fixed-medium-r-normal--13-*-100-100-c-80-iso8859-1" "-misc-fixed-medium-r-normal--13-*-75-75-c-70-iso8859-1" "-misc-fixed-medium-r-normal--13-*-75-75-c-80-iso8859-1" "-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1" "-misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1" "-misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1" "-misc-fixed-medium-r-normal--15-*-75-75-c-90-iso8859-1")

On the other hand, "a14" returns a better result.

(x-list-fonts "a14")
("-Misc-Fixed-medium-r-normal--14-*-75-75-c-70-ISO8859-1")

Is there a way to instruct emacs to use the exact font if available?
Also I want to avoid auto-scaling if possible (because it always
results in ugly glyphs).

By the way, even when using X font backend, Shift-Mouse-1 shows xft
font selection menu now.  Is there a way to revive the fontset
selection menu?

Regards,
-- 
Yoshiaki Kasahara
Research Institute for Information Technology, Kyushu University
kasahara@nc.kyushu-u.ac.jp

Re: X fonts selection weirdness

[Miles Bader]

Yoshiaki Kasahara <kasahara@nc.kyushu-u.ac.jp> writes:
> I just want to use a specific font for my emacs, using X font backend
> (I don't like AA fonts for editing).
...
> Is there a way to instruct emacs to use the exact font if available?
> Also I want to avoid auto-scaling if possible (because it always
> results in ugly glyphs).

Have you tried playing with the xft/fontconfig settings?  Depending on
the font, and the settings you use, the xft backend can give non-AA
fonts that are very attractive, on par with bitmap fonts (but of course
with greater scaling flexibility).

In my experience this is particularly true of CJK fonts, which seem like
they were designed for non-AA rendering at typical sizes.

-Miles

-- 
Circus, n. A place where horses, ponies and elephants are permitted to see
men, women and children acting the fool.

Re: X fonts selection weirdness

[Kenichi Handa]

In article <20080624.124226.1059534586998157705.kasahara@nc.kyushu-u.ac.jp>, Yoshiaki Kasahara <kasahara@nc.kyushu-u.ac.jp> writes:

> I just want to use a specific font for my emacs, using X font backend
> (I don't like AA fonts for editing).

> When I start emacs with an XLFD font like this:

> src/emacs -Q -fn '-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1' -xrm 'Emacs.FontBackend: x'

> '-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1' is
> selected instead (the displayed characters are distorted by
> auto-scaling).

I can't reproduce that problem.  Please start emacs as this:

% EMACS_FONT_LOG=1 src/emacs -Q -fn ...

and show me the result of M-x font-show-log RET.

---
Kenichi Handa
handa@ni.aist.go.jp

Re: X fonts selection weirdness

[Yoshiaki Kasahara]

On Tue, 24 Jun 2008 13:53:00 +0900,
	Kenichi Handa <handa@m17n.org> said:

> I can't reproduce that problem.  Please start emacs as this:
> 
> % EMACS_FONT_LOG=1 src/emacs -Q -fn ...
> 
> and show me the result of M-x font-show-log RET.

Here it is:

xfont-list: -misc-fixed-*-*-*--*-*-*-*-c-*-iso8859-1
  -misc-fixed-medium-r-semicondensed--13-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-semicondensed--13-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-semicondensed--12-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-semicondensed--12-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--9-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--9-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-normal--8-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--8-*-100-100-c-50-iso8859-1
  -misc-fixed-medium-r-normal--7-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--7-*-100-100-c-50-iso8859-1
  -misc-fixed-medium-r-normal--6-*-75-75-c-40-iso8859-1
  -misc-fixed-medium-r-normal--20-*-75-75-c-100-iso8859-1
  -misc-fixed-medium-r-normal--20-*-100-100-c-100-iso8859-1
  -misc-fixed-medium-r-normal--18-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--18-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--15-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  ...
list: -misc-fixed-medium-r-normal--*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--15-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--18-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--18-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--20-*-100-100-c-100-iso8859-1
  -misc-fixed-medium-r-normal--20-*-75-75-c-100-iso8859-1
  -misc-fixed-medium-r-normal--6-*-75-75-c-40-iso8859-1
  -misc-fixed-medium-r-normal--7-*-100-100-c-50-iso8859-1
  -misc-fixed-medium-r-normal--7-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--8-*-100-100-c-50-iso8859-1
  ...
sort-by: -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  x:-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
open: -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  x:-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1
xfont-list: -misc-fixed-*-*-*--*-*-*-*-c-*-iso8859-1
  -misc-fixed-medium-r-semicondensed--13-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-semicondensed--13-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-semicondensed--12-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-semicondensed--12-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--9-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--9-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-normal--8-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--8-*-100-100-c-50-iso8859-1
  -misc-fixed-medium-r-normal--7-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--7-*-100-100-c-50-iso8859-1
  -misc-fixed-medium-r-normal--6-*-75-75-c-40-iso8859-1
  -misc-fixed-medium-r-normal--20-*-75-75-c-100-iso8859-1
  -misc-fixed-medium-r-normal--20-*-100-100-c-100-iso8859-1
  -misc-fixed-medium-r-normal--18-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--18-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--15-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  ...
list: -misc-fixed-medium-*-normal--*-100-100-c-70-iso8859-1
  -misc-fixed-medium-i-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-i-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-o-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-o-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--15-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--18-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--18-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--20-*-100-100-c-100-iso8859-1
  -misc-fixed-medium-r-normal--20-*-75-75-c-100-iso8859-1
  ...
sort-by: -misc-fixed-medium-italic-normal--14-*-100-100-c-70-iso8859-1
  x:-misc-fixed-medium-i-normal--14-*-75-75-c-70-iso8859-1
open: -misc-fixed-medium-i-normal--14-*-75-75-c-70-iso8859-1
  x:-misc-fixed-medium-i-normal--14-130-75-75-c-70-iso8859-1
list: -misc-fixed-medium-r-normal--*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--15-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--18-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--18-*-75-75-c-90-iso8859-1
  -misc-fixed-medium-r-normal--20-*-100-100-c-100-iso8859-1
  -misc-fixed-medium-r-normal--20-*-75-75-c-100-iso8859-1
  -misc-fixed-medium-r-normal--6-*-75-75-c-40-iso8859-1
  -misc-fixed-medium-r-normal--7-*-100-100-c-50-iso8859-1
  -misc-fixed-medium-r-normal--7-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--8-*-100-100-c-50-iso8859-1
  ...
sort-by: -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  x:-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
list: -misc-fixed-*-r-normal--*-100-100-c-70-iso8859-1
  -misc-fixed-bold-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-bold-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-bold-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-bold-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-bold-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-bold-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-bold-r-normal--15-*-100-100-c-90-iso8859-1
  -misc-fixed-bold-r-normal--15-*-75-75-c-90-iso8859-1
  -misc-fixed-bold-r-normal--18-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1
  ...
sort-by: -misc-fixed-light-r-normal--14-*-100-100-c-70-iso8859-1
  x:-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
open: -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  x:-misc-fixed-medium-r-normal--14-101-100-100-c-70-iso8859-1
list: -misc-fixed-*-r-normal--*-100-100-c-70-iso8859-1
  -misc-fixed-bold-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-bold-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-bold-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-bold-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-bold-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-bold-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-bold-r-normal--15-*-100-100-c-90-iso8859-1
  -misc-fixed-bold-r-normal--15-*-75-75-c-90-iso8859-1
  -misc-fixed-bold-r-normal--18-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1
  ...
sort-by: -misc-fixed-light-r-normal--14-*-100-100-c-70-iso8859-1
  x:-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
list: -misc-fixed-*-r-normal--*-100-100-c-70-iso8859-1
  -misc-fixed-bold-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-bold-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-bold-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-bold-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-bold-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-bold-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-bold-r-normal--15-*-100-100-c-90-iso8859-1
  -misc-fixed-bold-r-normal--15-*-75-75-c-90-iso8859-1
  -misc-fixed-bold-r-normal--18-*-100-100-c-90-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-100-100-c-60-iso8859-1
  -misc-fixed-medium-r-normal--10-*-75-75-c-50-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-100-100-c-80-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--13-*-75-75-c-80-iso8859-1
  -misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1
  -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  -misc-fixed-medium-r-normal--15-*-100-100-c-90-iso8859-1
  ...
sort-by: -misc-fixed-bold-r-normal--14-*-100-100-c-70-iso8859-1
  x:-misc-fixed-bold-r-normal--14-*-75-75-c-70-iso8859-1
open: -misc-fixed-bold-r-normal--14-*-75-75-c-70-iso8859-1
  x:-misc-fixed-bold-r-normal--14-130-75-75-c-70-iso8859-1

-- 
Yoshiaki Kasahara
Research Institute for Information Technology, Kyushu University
kasahara@nc.kyushu-u.ac.jp
  

Re: X fonts selection weirdness

[Yoshiaki Kasahara]

On Tue, 24 Jun 2008 13:21:45 +0900,
	Miles Bader <miles.bader@necel.com> said:

> Have you tried playing with the xft/fontconfig settings?

Sort of...  I read a message in this ML how to enable bitmap fonts in
xft/fontconfig (70-yes-bitmaps.conf trick), but I don't find how to
change directory preferences like 'xset +fp/fp+'.  I don't want to
touch the default fonts in X11, so I override the same font names by
my customized font files and put the directory in front of the font
path.

Actually I just want to use (somewhat customized) 7x14.pcf and k14.pcf
(with bold/italic variants) for my Emacs, and not very interested in
other fonts such as IPAGothic etc....

Anyway, thank you for your suggestion.  I think I have to fight with
new font scheme anyway.

Regards,
-- 
Yoshiaki Kasahara
Research Institute for Information Technology, Kyushu University
kasahara@nc.kyushu-u.ac.jp

Re: X fonts selection weirdness

[Kenichi Handa]

In article <20080624.145703.566779761171555788.kasahara@nc.kyushu-u.ac.jp>, Yoshiaki Kasahara <kasahara@nc.kyushu-u.ac.jp> writes:

> On Tue, 24 Jun 2008 13:53:00 +0900,
> 	Kenichi Handa <handa@m17n.org> said:

> > I can't reproduce that problem.  Please start emacs as this:
> > 
> > % EMACS_FONT_LOG=1 src/emacs -Q -fn ...
> > 
> > and show me the result of M-x font-show-log RET.

> Here it is:

Thank you.  I've just installed a fix to pay attention to
DPI on font sorting.  Please try with the latest code again.

---
Kenichi Handa
handa@ni.aist.go.jp

Re: X fonts selection weirdness

[Yoshiaki Kasahara]

On Tue, 24 Jun 2008 16:02:05 +0900,
	Kenichi Handa <handa@m17n.org> said:

> Thank you.  I've just installed a fix to pay attention to
> DPI on font sorting.  Please try with the latest code again.

Thank you very much.  It worked for '-fn' option.  It seems that
create-fontset-from-ascii-font still ignores specified DPI, though...

(create-fontset-from-ascii-font
        "-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1" nil "misc_14")
"-misc-fixed-medium-r-normal--13-*-100-100-c-70-fontset-misc_14"

---

By the way, I have one more unsolved question... Is this a correct way
to specify Japanese kanji characters to use Japanese kanji fonts?
(after emacs -fn '-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1')

(set-fontset-font "fontset-startup"
		  'japanese-jisx0208
		  '("misc-fixed" . "jisx0208.1983"))

Without explicit set-fontset-font, some symbol characters are
displayed using iso10646 font instead of jisx0208 font, such as
-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso10646-1.

After set-fontset-font, almost all characters are displayed by desired
fonts, but still there are a few exceptions.  For example, '×' is
shown as a half-width character.  C-u C-x = shows:

        character: × (215, #o327, #xd7)
preferred charset: japanese-jisx0208
		   (JISX0208.1983/1990 Japanese Kanji: ISO-IR-87)
       code point: 0x215F
           syntax: _ 	which means: symbol
         category: c:Chinese h:Korean j:Japanese l:Latin
      buffer code: #xC3 #x97
        file code: ESC #x24 #x42 #x21 #x5F
		   (encoded by coding system iso-2022-7bit)
          display: by this font (glyph code)
     -misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1 (#xD7)

I wonder why iso8859 font is selected when the preferred charset is
japanese-jisx0208.  Maybe I overlooked something.... can I specify a
charset preference for font selection?

Regards,
--
Yoshiaki Kasahara
Research Institute for Information Technology, Kyushu University
kasahara@nc.kyushu-u.ac.jp

Re: X fonts selection weirdness

[Stefan Monnier]

> Thank you.  I've just installed a fix to pay attention to
> DPI on font sorting.  Please try with the latest code again.

I couldn't find the corresponding patch, so just to make sure: by "pay
attention" you don't mean that you compare various fonts DPI, right?
DPIs are irrelevant, what we want instead is to sort based on the pixel
size or on pointsize/dpi.


        Stefan

Re: X fonts selection weirdness

[Kenichi Handa]

In article <jwvwskfm2ij.fsf-monnier+emacs@gnu.org>, Stefan Monnier <monnier@iro.umontreal.ca> writes:

> > Thank you.  I've just installed a fix to pay attention to
> > DPI on font sorting.  Please try with the latest code again.

> I couldn't find the corresponding patch, so just to make sure: by "pay
> attention" you don't mean that you compare various fonts DPI, right?
> DPIs are irrelevant, what we want instead is to sort based on the pixel
> size or on pointsize/dpi.

What I modified is the function font_score.  I made it to
prefer the same DPI if the pixelsize is the same.
Previously, as this log shows:

sort-by: -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
  x:-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1

the difference of DPI was ignored.

---
Kenichi Handa
handa@ni.aist.go.jp
 

Re: X fonts selection weirdness

[Stefan Monnier]

> What I modified is the function font_score.  I made it to
> prefer the same DPI if the pixelsize is the same.
> Previously, as this log shows:

> sort-by: -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
>   x:-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1

> the difference of DPI was ignored.

Oops, sorry, I see I was just confused.  His case is a really messed up
one, and giving preference to the same DPI is probably the best/only
thing we can do.


        Stefan

Re: X fonts selection weirdness

[Kenichi Handa]

In article <jwv7icen04k.fsf-monnier+emacs@gnu.org>, Stefan Monnier <monnier@IRO.UMontreal.CA> writes:

> > What I modified is the function font_score.  I made it to
> > prefer the same DPI if the pixelsize is the same.
> > Previously, as this log shows:

> > sort-by: -misc-fixed-medium-r-normal--14-*-75-75-c-70-iso8859-1
> >   x:-misc-fixed-medium-r-normal--14-*-100-100-c-70-iso8859-1

> > the difference of DPI was ignored.

> Oops, sorry, I see I was just confused.  His case is a really messed up
> one, and giving preference to the same DPI is probably the best/only
> thing we can do.

I found that I've forgot to commit src/ChangeLog for that.
Just committed.

---
Kenichi Handa
handa@ni.aist.go.jp

Re: X fonts selection weirdness

[Kenichi Handa]

In article <20080624.180802.154682915676368965.kasahara@nc.kyushu-u.ac.jp>, Yoshiaki Kasahara <kasahara@nc.kyushu-u.ac.jp> writes:

> Thank you very much.  It worked for '-fn' option.  It seems that
> create-fontset-from-ascii-font still ignores specified DPI, though...

> (create-fontset-from-ascii-font
>         "-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1" nil "misc_14")
> "-misc-fixed-medium-r-normal--13-*-100-100-c-70-fontset-misc_14"

I installed a fix.  Please try again.

> By the way, I have one more unsolved question... Is this a correct way
> to specify Japanese kanji characters to use Japanese kanji fonts?
> (after emacs -fn '-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1')

> (set-fontset-font "fontset-startup"
> 		  'japanese-jisx0208
> 		  '("misc-fixed" . "jisx0208.1983"))

> Without explicit set-fontset-font, some symbol characters are
> displayed using iso10646 font instead of jisx0208 font, such as
> -misc-fixed-medium-r-normal--14-130-75-75-c-70-iso10646-1.

The above should work, but I'm now thinking about
introducing non-default-font-script-list:

non-default-font-script-list is a variable defined in `C source code'.
Its value is 
(arabic syriac thaana devanagari before gurmukhi gurmukhi oriya tamil telugu kannada malayalam sinhala thai lao tibetan myanmar hangul khmer mongolian kana han ideographic-description cjk-misc hangul cham tai-viet)

Documentation:
List of scripts not to display by the default font.
Usually, the font specified for ASCII characters is also used
to display a non-ASCII character if the font supports that character.
But, if a character belongs to one of these scripts, a font for it is
determined by looking up the current fontset and the default fontset.

But, I have not yet decided the detail of the
implementation.   Could you please wait for a while?

---
Kenichi Handa
handa@ni.aist.go.jp

Re: X fonts selection weirdness

[Miles Bader]

Kenichi Handa <handa@m17n.org> writes:
> The above should work, but I'm now thinking about
> introducing non-default-font-script-list:
..
> Documentation:
> List of scripts not to display by the default font.
> Usually, the font specified for ASCII characters is also used
> to display a non-ASCII character if the font supports that character.
> But, if a character belongs to one of these scripts, a font for it is
> determined by looking up the current fontset and the default fontset.

Why would you do that though?

If the user choose a default font capable of displaying one of those
charsets, it seems like he would _want_ the default font to be used!

-Miles

-- 
Friendship, n. A ship big enough to carry two in fair weather, but only one
in foul.

Re: X fonts selection weirdness

[Jason Rumney]

Miles Bader wrote:
> Kenichi Handa <handa@m17n.org> writes:
>> The above should work, but I'm now thinking about
>> introducing non-default-font-script-list:
> ..
>> Documentation:
>> List of scripts not to display by the default font.
>> Usually, the font specified for ASCII characters is also used
>> to display a non-ASCII character if the font supports that character.
>> But, if a character belongs to one of these scripts, a font for it is
>> determined by looking up the current fontset and the default fontset.
> 
> Why would you do that though?
> 
> If the user choose a default font capable of displaying one of those
> charsets, it seems like he would _want_ the default font to be used!

Conversely, if the user has explicitly specified fonts in a fontset,
then they expect those fonts to be used, not just for specific scripts.

Re: X fonts selection weirdness

[David Kastrup]

Kenichi Handa <handa@m17n.org> writes:

> I installed a fix.  Please try again.

[...]

> The above should work, but I'm now thinking about
> introducing non-default-font-script-list:

[...]

I can't help feeling that we are currently stumbling without much
orientation through a maze of ad-hoc local patches and fixes.

This does not feel like it should be the intended mode of operation:
other applications have to use fontsets, too.  So I have the suspicion
that at least for some platforms, we are approaching the problem from an
angle that has not been intended by the creators of the rendering
infrastructures we are using.

It may be that the current Emacs APIs don't allow us to match our font
stuff to what is Xft's intended operation.  In that case, it might be
better to interface at a different level.  If that means that the
behavior and customization of Xft- and non-Xft rendering will differ
more than it does currently, that would appear a reasonable price to pay
in my opinion.

In any way: our current rate of creating yet another ad-hoc variable
intended to patch around the font system in a manner that no other
application does is not useful for platform integration.  And it does
not help people understand what happens.

I have no expertise at all in that area.  I just recognize a pattern
here that feels like a bad idea.  It is a pattern occuring when
something needs to get shipped, and sound judgment is suspended while
"just make it work, we won't touch this again" is the only left
priority.  But "we won't touch this again" is not what we are aiming for
here.

If doing this right needs more time and plan and thought, then we should
invest that.  Or it is going to bite us later.

Again: I lack the technical skills to judge the subsystems here.  I
might be all wrong about this.  But it feels increasingly that we are
getting complexity that will be beyond all users, even if perhaps not
all developers.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: X fonts selection weirdness

[Kenichi Handa]

In article <buoskv254pv.fsf@dhapc248.dev.necel.com>, Miles Bader <miles.bader@necel.com> writes:

> Kenichi Handa <handa@m17n.org> writes:
> > The above should work, but I'm now thinking about
> > introducing non-default-font-script-list:
..
> > Documentation:
> > List of scripts not to display by the default font.
> > Usually, the font specified for ASCII characters is also used
> > to display a non-ASCII character if the font supports that character.
> > But, if a character belongs to one of these scripts, a font for it is
> > determined by looking up the current fontset and the default fontset.

> Why would you do that though?

> If the user choose a default font capable of displaying one of those
> charsets, it seems like he would _want_ the default font to be used!

Not for non-Latin scripts, especially CJK and those require
CTL (Complex Text Layout).  Usually the default font, even
if it claims that it has the specific character, the design
is not good or doesn't contain proper OTF tables.

Jason Rumney <jasonr@gnu.org> writes:

> Conversely, if the user has explicitly specified fonts in a fontset,
> then they expect those fonts to be used, not just for specific scripts.

Yes.  As written in the docstring,
non-default-font-script-list doesn't prevent such a case.

Currently, if a user specifies a font (not fontset), a new
fontset is created, and in it, all Latin characters are
mapped to the specified font, and the iso10646-1 version of
that font is set in the fallback font of the fontset.

So, if a user specifies XXX-iso8859-1, XXX-iso10646-1 is
used for all characters supported by that font.  This works
well for most characters except for those in the above
scripts.  Current idea of mine is to, instead of setting
XXX-iso10646-1 to the fallback font, map all characters
(except for latin-1 and those belongs to the above scripts)
to XXX-iso10646-1.

---
Kenichi Handa
handa@ni.aist.go.jp

Re: X fonts selection weirdness

[Miles Bader]

Kenichi Handa <handa@m17n.org> writes:
>> If the user choose a default font capable of displaying one of those
>> charsets, it seems like he would _want_ the default font to be used!
>
> Not for non-Latin scripts, especially CJK and those require
> CTL (Complex Text Layout).  Usually the default font, even
> if it claims that it has the specific character, the design
> is not good or doesn't contain proper OTF tables.

I'm confused.

If I specify, for instance, a default font of "Sazanami 12", that works
great for both ascii and CJK characters, and has the rather nice
property that the 2-1 width ration of CJK and ascii characters is
preserved exactly (so, e.g., character graphics look correct).

Morever, since such a font's main repertoire is CJK characters, if
somebody specifies it as the _default_ font, it would seem fairly
obvious that they want to use it to display both ascii and CJK
characters.

When a user specifies such a default font, which has a very wide
repertoire, that's a very simple way for them to get consistent
rendering.

I gather you're instead talking about fonts that mainly cover ascii, and
"accidentally" happen to include some other characters?

-Miles

-- 
Run away!  Run away!

Re: X fonts selection weirdness

[Miles Bader]

Miles Bader <miles@gnu.org> writes:
> If I specify, for instance, a default font of "Sazanami 12", that works
> great for both ascii and CJK characters

Er, that should be "Sazanami Gothic 12".

-Miles

-- 
Friendship, n. A ship big enough to carry two in fair weather, but only one
in foul.

Re: X fonts selection weirdness

[Yoshiaki Kasahara]

On Wed, 25 Jun 2008 16:47:22 +0900,
	Kenichi Handa <handa@m17n.org> said:

> > (create-fontset-from-ascii-font
> >         "-misc-fixed-medium-r-normal--14-130-75-75-c-70-iso8859-1" nil "misc_14")
> > "-misc-fixed-medium-r-normal--13-*-100-100-c-70-fontset-misc_14"
> 
> I installed a fix.  Please try again.

I confirmed it worked.  Thank you very much.  I didn't realize my case
is a kind of edge case...  Maybe my X configuration is too dusty (so
many tweaks piled up in a decade).

> But, I have not yet decided the detail of the
> implementation.   Could you please wait for a while?

Of course I'll wait with expectations.  I hope you can find a good
solution....

Regards,
-- 
Yoshiaki Kasahara
Research Institute for Information Technology, Kyushu University
kasahara@nc.kyushu-u.ac.jp

Re: X fonts selection weirdness

[Kenichi Handa]

In article <87wskdtyq4.fsf@catnip.gol.com>, Miles Bader <miles@gnu.org> writes:

> If I specify, for instance, a default font of "Sazanami 12", that works
> great for both ascii and CJK characters, and has the rather nice
> property that the 2-1 width ration of CJK and ascii characters is
> preserved exactly (so, e.g., character graphics look correct).

In such a case, that font is anyway selected for CJK
characters in the normal fontset looking up because the
default fontset doesn't specify a family and thus the
default family sazanami scores high in font sorting.

> I gather you're instead talking about fonts that mainly cover ascii, and
> "accidentally" happen to include some other characters?

That's one reason.  Another is, as I wrote, to use a font of
proper OpenType tables.

---
Kenichi Handa
handa@ni.aist.go.jp

Re: X fonts selection weirdness

[Richard M Stallman]

    > If the user choose a default font capable of displaying one of those
    > charsets, it seems like he would _want_ the default font to be used!

    Not for non-Latin scripts, especially CJK and those require
    CTL (Complex Text Layout).  Usually the default font, even
    if it claims that it has the specific character, the design
    is not good or doesn't contain proper OTF tables.

How about putting that explanation in the doc string?

Re: X fonts selection weirdness

[Kenichi Handa]

In article <E1KBytl-0003eN-92@fencepost.gnu.org>, Richard M Stallman <rms@gnu.org> writes:

> If the user choose a default font capable of displaying one of those
> charsets, it seems like he would _want_ the default font to be used!

>     Not for non-Latin scripts, especially CJK and those require
>     CTL (Complex Text Layout).  Usually the default font, even
>     if it claims that it has the specific character, the design
>     is not good or doesn't contain proper OTF tables.

> How about putting that explanation in the doc string?

Docstring of which function/variable?  Once the font
selection gets stable, I intend to add a new Info section
describing the detail of the font selection strategy.

---
Kenichi Handa
handa@ni.aist.go.jp

Re: X fonts selection weirdness

[Richard M Stallman]

    > If the user choose a default font capable of displaying one of those
    > charsets, it seems like he would _want_ the default font to be used!

    >     Not for non-Latin scripts, especially CJK and those require
    >     CTL (Complex Text Layout).  Usually the default font, even
    >     if it claims that it has the specific character, the design
    >     is not good or doesn't contain proper OTF tables.

    > How about putting that explanation in the doc string?

    Docstring of which function/variable?

Whichever ones you use to choose a different default font.

Re: X fonts selection weirdness

[Kenichi Handa]

Sorry for the late response on this matter.

In article <85ej6mnd5u.fsf@lola.goethe.zz>, David Kastrup <dak@gnu.org> writes:

> Kenichi Handa <handa@m17n.org> writes:
> > I installed a fix.  Please try again.

> [...]

> > The above should work, but I'm now thinking about
> > introducing non-default-font-script-list:

> [...]

> I can't help feeling that we are currently stumbling without much
> orientation through a maze of ad-hoc local patches and fixes.

I've just installed a change that doesn't use
non-default-font-script-list but respects the specification
of fontset while giving the highest priority to the default
font.

We have these four font groups for non-ASCII characters.

(1) What specified in the current fontset.
(2) What specified in the default fontset
(3) What specified as the fallbacks in the current fontset.
(3) What specified as the fallbacks in the default fontset.

The latest Emacs tries each of the above font group in the
above order.  And while selecting a font from a font group,
if the default font (i.e. what selected for ASCII characters
of the current face) supports the target character, and if
any of a font in the group matches with the default font,
select the default font.

This way, Emacs uses the default font for all characters
except for such characters that the fontset doesn't specify
a font matching with it.

---
Kenichi Handa
handa@ni.aist.go.jp

Re: X fonts selection weirdness

[Yoshiaki Kasahara]

On Wed, 09 Jul 2008 10:42:08 +0900,
	Kenichi Handa <handa@m17n.org> said:

> I've just installed a change that doesn't use
> non-default-font-script-list but respects the specification
> of fontset while giving the highest priority to the default
> font.

Thanks to this change, all of my problems about font setting had gone,
and I'm really satisfied with the appearance of Emacs now :-)

Thank you very much!
-- 
Yoshiaki Kasahara
Research Institute for Information Technology, Kyushu University
kasahara@nc.kyushu-u.ac.jp