character: ๑ (3665, #o7121, #xe51)
preferred charset: gb18030 (GB18030)
code point: 0x8132D639
syntax: w which means: word
category: 6:digit t:Thai
buffer code: #xE0 #xB9 #x91
file code: #xE0 #xB9 #x91 (encoded by coding system utf-8-unix)
display: no font available

How to configure emacs to show this symbol? I'm quite sure I have the
font that contains it, because urxvt and opera (yes, I use opera to test
it ;p) could. urxvt use "wenquanyi zen hei", I don't know how to get the
font info opera uses.

My .emacs font setting:
(set-default-font "Bitstream Vera Sans Mono-12")
(set-fontset-font (frame-parameter nil 'font)
'latin '("Bitstream Vera Sans Mono" . "unicode-bmp"))
(set-fontset-font...

Sorry, I forgot to give my OS and emacs info:
OS: freebsd_stable7
emacs: 23.0.60.2, yesterdaya cvsed from trunk. font-backend has been enabled.

On Tue, 26 Feb 2008 12:05:03 +0800 anhnmncb gmail.com> wrote:

-Misc-Fixed-Medium-R-SemiCondense

I don't seem it's font's problem. I think that urxvt's font
selection mechanism is great, if a font doesn't contain a symbol in that
encoding, it will find another font to display it, don't know whether
emacs could implement this technique?

Am 26.02.2008 um 02:11 schrieb anhnmncb:

By adding a clause for Thai? GNU Emacs tells you it's Thai:

character: ๑ (3665, #o7121, #xe51)
[...]
category: 6:digit t:Thai

which is not mentioned in the excerpt from your ~/.emacs file.
Code2000, Bitstream Cyberbase, TITUS Cyberbit, and more provide it.

Be aware also that Opera will use a proportional font while in GNU
Emacs monospaced fonts are in regular use.

Peter Dyballa writes:

How to add a clause for Thai? Could you give me example?
I have the font contain it, so I dont' want to add another font to my
system ;p

Am 26.02.2008 um 12:00 schrieb anhnmncb:

It would not be that easy.

A terminal emulation has an easy job: display things in one encoding.
GNU Emacs supports some dozens. Less than a handful are sensible,
IMO, i.e. Unicode based. Some encodings are 8 bit, left from the
child days of some operating systems, other are 16 bit or more, most
of them for some restricted national use.

Thank you for very detail explanation!

th or thai? In my system, using thai produces nothing, but th.
The output is:
/usr/home/anhnmncb $ fc-list :lang=th
Fixed:style=Bold
Fixed:style=SemiCondensed
Fixed:style=Oblique
Fixed:style=Regular

ok, I have done it, perfectly!
Now I find another symbol emac couldn't recognize:

character: ✺ (10042, #o23472, #x273a)
preferred charset: gb18030 (GB18030)
code point: 0x8137C136
syntax: w which means: word
buffer code: #xE2 #x9C #xBA
file code: #xE2 #x9C #xBA (encoded by coding system utf-8-unix)
display: no font available

Even does not show which language it is belonged to.

Am 28.02.2008 um 01:10 schrieb anhnmncb:

It is according to ISO 639 and ISO 3166 th. You can try:

fc-list : file family lang

It will be a long listing in *shell* buffer ...

The file fonts.conf configures the libfontconfig based system. I am
confident that there is a way to "correct" the information a font
gives by making it responsible for something it can do.