Re: A very strange thing about Emacs working with flyspell!

Re: A very strange thing about Emacs working with flyspell!

[zhaohs]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 4311 bytes --]

On Fri, Nov 17, 2006 at 08:11:40PM +0900, Kenichi Handa wrote:
> In article <20061116113746.GA5373@localdomain>, zhaohs <zhaohscas@yahoo.com.cn> writes:
> 
> > Though I can use the following command to run emacs and display full unicode chinese characters correctly:
> 
> > ======================
> > emacs --enable-font-backend -fn "Dejavu Sans Mono-14"
> > ====================== 
> 
> > But I find that if I use the above command, the *SCIM* input method won't be
> > activated by *Ctrl+SPACE*,
> 
> Again, I can't reproduce that bug, but I suspect that your
> emacs fails to create a fontset for XIC (XInputContext).
> As I've just installed some workaround for such a case,
> please try again with the latest code.

I have done this, but the problem is still the same.

So I Run emacs under gdb again, the following is the debuging results:

====================================================
zhaohs@ubuntu-edgy:~/cvs-svn/emacs$ cd src/
zhaohs@ubuntu-edgy:~/cvs-svn/emacs/src$ gdb emacs
GNU gdb 6.4.90-debian
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i486-linux-gnu"...Using host libthread_db
library "/lib/tls/i686/cmov/libthread_db.so.1".

DISPLAY = :0.0
TERM = xterm
Breakpoint 1 at 0x81035b6: file emacs.c, line 464.
Breakpoint 2 at 0x811c3c6: file sysdep.c, line 1385.
(gdb) run --enable-font-backend
Starting program: /media/hdb6/MyHomeFiles/cvs-svn/emacs/src/emacs
--enable-font-backend
[Thread debugging using libthread_db enabled]
[New Thread -1220082000 (LWP 26077)]
[Switching to Thread -1220082000 (LWP 26077)]
Breakpoint 3 at 0x80d6f16: file xterm.c, line 8047.

Breakpoint 1, abort () at emacs.c:464
464       kill (getpid (), SIGABRT);
(gdb) xba
"image-search-load-path" (0x85ab41b)
"find-image" (0x8c6b47d)
"tool-bar-local-item-from-menu" (0x83917f1)
"apply" (0x886b341)
"tool-bar-add-item-from-menu" (0x83917f1)
"tool-bar-setup" (0x8)
"tool-bar-mode" (0x8)
"command-line" (0x8cc054b)
"normal-top-level" (0x83378c9)
(gdb) bt
#0  abort () at emacs.c:464
#1  0x08153e5c in mark_object (arg=140117575) at alloc.c:5793
#2  0x08153f8f in mark_object (arg=138141909) at alloc.c:5776
#3  0x081543ad in mark_object (arg=146743852) at alloc.c:5651
#4  0x08153f8f in mark_object (arg=138141269) at alloc.c:5776
#5  0x08153f8f in mark_object (arg=147227149) at alloc.c:5776
#6  0x08157a96 in Fgarbage_collect () at alloc.c:5107
#7  0x0816c5e9 in Ffuncall (nargs=2, args=0xbfa638f0) at eval.c:2931
#8  0x081980a9 in Fbyte_code (bytestr=136778675, vector=136778692,
maxdepth=40)
    at bytecode.c:679
#9  0x0816c2da in funcall_lambda (fun=136778628, nargs=1, 
    arg_vector=0xbfa63a24) at eval.c:3184
#10 0x0816c702 in Ffuncall (nargs=2, args=0xbfa63a20) at eval.c:3054
#11 0x081980a9 in Fbyte_code (bytestr=136778787, vector=136778804,
maxdepth=40)
    at bytecode.c:679
#12 0x0816c2da in funcall_lambda (fun=136778748, nargs=1, 
    arg_vector=0xbfa63b54) at eval.c:3184
#13 0x0816c702 in Ffuncall (nargs=2, args=0xbfa63b50) at eval.c:3054
#14 0x081980a9 in Fbyte_code (bytestr=136794795, vector=136794812,
maxdepth=64)
    at bytecode.c:679
#15 0x0816c2da in funcall_lambda (fun=136794708, nargs=6, 
    arg_vector=0xbfa63c84) at eval.c:3184
#16 0x0816c702 in Ffuncall (nargs=7, args=0xbfa63c80) at eval.c:3054
---Type <return> to continue, or q <return> to quit---
====================================================

Best,
Hongsheng.

P.S. Eli Zaretskii advice me to subscribe the emacs-devel@gnu.org, the
following is his email:

=======================================
> Date: Fri, 17 Nov 2006 23:45:49 +0800
> From: zhaohs <zhaohscas@yahoo.com.cn>
>
> Could I join the mailing list
> emacs-unicode@gnu.org and send bug reports here?

You don't need to join emacs-unicode to send bug reports.  You can
send them to emacs-devel@gnu.org.  Just make sure to mention that your
reports are for "Emacs 23" or "emacs-unicode-2 branch".

TIA
=======================================


__________________________________________________
¸Ï¿ì×¢²áÑÅ»¢³¬´óÈÝÁ¿Ãâ·ÑÓÊÏä?
http://cn.mail.yahoo.com

Re: A very strange thing about Emacs working with flyspell!

[Kenichi Handa]

In article <20061118025413.GB25925@localdomain>, zhaohs <zhaohscas@yahoo.com.cn> writes:

> On Fri, Nov 17, 2006 at 08:11:40PM +0900, Kenichi Handa wrote:
> > In article <20061116113746.GA5373@localdomain>, zhaohs <zhaohscas@yahoo.com.cn> writes:
> > 
> > > Though I can use the following command to run emacs and display full unicode chinese characters correctly:
> > 
> > > ======================
> > > emacs --enable-font-backend -fn "Dejavu Sans Mono-14"
> > > ====================== 
> > 
> > > But I find that if I use the above command, the *SCIM* input method won't be
> > > activated by *Ctrl+SPACE*,
> > 
> > Again, I can't reproduce that bug, but I suspect that your
> > emacs fails to create a fontset for XIC (XInputContext).
> > As I've just installed some workaround for such a case,
> > please try again with the latest code.

> I have done this, but the problem is still the same.

> So I Run emacs under gdb again, the following is the debuging results:

> ====================================================
> zhaohs@ubuntu-edgy:~/cvs-svn/emacs$ cd src/
> zhaohs@ubuntu-edgy:~/cvs-svn/emacs/src$ gdb emacs
> GNU gdb 6.4.90-debian
> Copyright (C) 2006 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you are
> welcome to change it and/or distribute copies of it under certain
> conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for details.
> This GDB was configured as "i486-linux-gnu"...Using host libthread_db
> library "/lib/tls/i686/cmov/libthread_db.so.1".

> DISPLAY = :0.0
> TERM = xterm
> Breakpoint 1 at 0x81035b6: file emacs.c, line 464.
> Breakpoint 2 at 0x811c3c6: file sysdep.c, line 1385.
> (gdb) run --enable-font-backend

No, no, the workaround I installed is against the problem of
activating SCIM input method.  So, please run Emacs as this

% emacs --enable-font-backend -fn "Dejavu Sans Mono-14"

and check if C-SPC activates SCIM.

---
Kenichi Handa
handa@m17n.org

Re: A very strange thing about Emacs working with flyspell!

[zhaohs]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 2661 bytes --]

On Sat, Nov 18, 2006 at 12:04:59PM +0900, Kenichi Handa wrote:
> In article <20061118025413.GB25925@localdomain>, zhaohs <zhaohscas@yahoo.com.cn> writes:
> 
> > On Fri, Nov 17, 2006 at 08:11:40PM +0900, Kenichi Handa wrote:
> > > In article <20061116113746.GA5373@localdomain>, zhaohs <zhaohscas@yahoo.com.cn> writes:
> > > 
> > > > Though I can use the following command to run emacs and display full unicode chinese characters correctly:
> > > 
> > > > ======================
> > > > emacs --enable-font-backend -fn "Dejavu Sans Mono-14"
> > > > ====================== 
> > > 
> > > > But I find that if I use the above command, the *SCIM* input method won't be
> > > > activated by *Ctrl+SPACE*,
> > > 
> > > Again, I can't reproduce that bug, but I suspect that your
> > > emacs fails to create a fontset for XIC (XInputContext).
> > > As I've just installed some workaround for such a case,
> > > please try again with the latest code.
> 
> > I have done this, but the problem is still the same.
> 
> > So I Run emacs under gdb again, the following is the debuging results:
> 
> > ====================================================
> > zhaohs@ubuntu-edgy:~/cvs-svn/emacs$ cd src/
> > zhaohs@ubuntu-edgy:~/cvs-svn/emacs/src$ gdb emacs
> > GNU gdb 6.4.90-debian
> > Copyright (C) 2006 Free Software Foundation, Inc.
> > GDB is free software, covered by the GNU General Public License, and you are
> > welcome to change it and/or distribute copies of it under certain
> > conditions.
> > Type "show copying" to see the conditions.
> > There is absolutely no warranty for GDB.  Type "show warranty" for details.
> > This GDB was configured as "i486-linux-gnu"...Using host libthread_db
> > library "/lib/tls/i686/cmov/libthread_db.so.1".
> 
> > DISPLAY = :0.0
> > TERM = xterm
> > Breakpoint 1 at 0x81035b6: file emacs.c, line 464.
> > Breakpoint 2 at 0x811c3c6: file sysdep.c, line 1385.
> > (gdb) run --enable-font-backend
> 
> No, no, the workaround I installed is against the problem of
> activating SCIM input method.  So, please run Emacs as this
> 
> % emacs --enable-font-backend -fn "Dejavu Sans Mono-14"
> 
> and check if C-SPC activates SCIM.

I run emacs as the following:

=======================================================
% emacs --enable-font-backend -fn "Dejavu Sans Mono-14"
=======================================================

Though the emacs won't crash, but I can't activete the SCIM still, when
I stroke C-SPC first, the emacs's minibuffer say: Mark set; the second time,
it will say: Mark activated.

Best,
Hongsheng.

__________________________________________________
¸Ï¿ì×¢²áÑÅ»¢³¬´óÈÝÁ¿Ãâ·ÑÓÊÏä?
http://cn.mail.yahoo.com

Re: A very strange thing about Emacs working with flyspell!

[Kenichi Handa]

In article <20061118032327.GA27684@localdomain>, zhaohs <zhaohscas@yahoo.com.cn> writes:

> > No, no, the workaround I installed is against the problem of
> > activating SCIM input method.  So, please run Emacs as this
> > 
> > % emacs --enable-font-backend -fn "Dejavu Sans Mono-14"
> > 
> > and check if C-SPC activates SCIM.

> I run emacs as the following:

> =======================================================
> % emacs --enable-font-backend -fn "Dejavu Sans Mono-14"
> =======================================================

> Though the emacs won't crash, but I can't activete the SCIM still, when
> I stroke C-SPC first, the emacs's minibuffer say: Mark set; the second time,
> it will say: Mark activated.

Hmmm, then please start Emacs as this:

% cd EMACS_BUILD_DIR/src
% gdb emacs
(gdb) br xfns.c:2120
(gdb) run --enable-font-backend -fn "Dejavu Sans Mono-14"

Emacs will stop at the break point as below.

Breakpoint 3, xic_create_xfontset (f=0x89ecb10, 
    base_fontname=0x88186a8 "-etl-*-medium-r-normal-*-24-*-*-*-*-*-iso8859-1")
    at xfns.c:2120
2120	  if (!xfs)
(gdb) 

The value of base_fontname will be different in your case.

Please type "n" (for next) and RET to execute the code one
line by line, and check each value of the second argument to
XCreateFontSet.  Does it return non-NULL value to `xfs' at
some point?  Don't you see any warning message from X
library?

If XCreateFontSet alwasy return NULL, the next thing is to
invoke "run" command with no argument and check the second
argument to XCreateFontSet at the successful call as this.

(gdb) set args
(gdb) run

---
Kenichi Handa
handa@m17n.org

Re: A very strange thing about Emacs working with flyspell!

[zhaohs]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 2385 bytes --]

> Hmmm, then please start Emacs as this:
> 
> % cd EMACS_BUILD_DIR/src
> % gdb emacs
> (gdb) br xfns.c:2120
> (gdb) run --enable-font-backend -fn "Dejavu Sans Mono-14"
> 
> Emacs will stop at the break point as below.
> 
> Breakpoint 3, xic_create_xfontset (f=0x89ecb10, 
>     base_fontname=0x88186a8 "-etl-*-medium-r-normal-*-24-*-*-*-*-*-iso8859-1")
>     at xfns.c:2120
> 2120	  if (!xfs)
> (gdb) 

-- Below this line is the debug information.

zhaohs@ubuntu-edgy:~/cvs-svn/emacs/src$ gdb emacs
GNU gdb 6.4.90-debian
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i486-linux-gnu"...Using host libthread_db
library "/lib/tls/i686/cmov/libthread_db.so.1".

DISPLAY = :0.0
TERM = xterm
Breakpoint 1 at 0x81035b6: file emacs.c, line 464.
Breakpoint 2 at 0x811c3c6: file sysdep.c, line 1385.
(gdb) br xfns.c:2120
Breakpoint 3 at 0x80e2073: file xfns.c, line 2120.
(gdb) run --enable-font-backend -fn "Dejavu Sans Mono-14"
Starting program: /media/hdb6/MyHomeFiles/cvs-svn/emacs/src/emacs
--enable-font-backend -fn "Dejavu Sans Mono-14"
[Thread debugging using libthread_db enabled]
[New Thread -1219627344 (LWP 5936)]
[Switching to Thread -1219627344 (LWP 5936)]
Breakpoint 4 at 0x80d6f16: file xterm.c, line 8047.

-- The debug information ends here.

As you can see, Emacs won't crash at all in my case according to your
advice, so the (gdb) won't appear and I can't debug the following things you
have posted to me.

> 
> The value of base_fontname will be different in your case.
> 
> Please type "n" (for next) and RET to execute the code one
> line by line, and check each value of the second argument to
> XCreateFontSet.  Does it return non-NULL value to `xfs' at
> some point?  Don't you see any warning message from X
> library?
> 
> If XCreateFontSet alwasy return NULL, the next thing is to
> invoke "run" command with no argument and check the second
> argument to XCreateFontSet at the successful call as this.
> 
> (gdb) set args
> (gdb) run

Best,
Hongsheng.
__________________________________________________
¸Ï¿ì×¢²áÑÅ»¢³¬´óÈÝÁ¿Ãâ·ÑÓÊÏä?
http://cn.mail.yahoo.com

Re: A very strange thing about Emacs working with flyspell!

[Kenichi Handa]

In article <20061120030401.GB5498@localdomain>, zhaohs <zhaohscas@yahoo.com.cn> writes:

> As you can see, Emacs won't crash at all in my case according to your
> advice, so the (gdb) won't appear and I can't debug the following things you
> have posted to me.

Emacs doesn't have to crash to stop at the break point, but
ummm, I have no idea why Emacs doesn't reach the break
point.

How about setting break point as this instead?

(gdb) br xic_create_fontset

---
Kenichi Handa
handa@m17n.org

Re: A very strange thing about Emacs working with flyspell!

[Hongsheng]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 1629 bytes --]

> Emacs doesn't have to crash to stop at the break point, but
> ummm, I have no idea why Emacs doesn't reach the break
> point.
> 
> How about setting break point as this instead?
> 
> (gdb) br xic_create_fontset

--- Well, see the following, Emacs won't stop at the break point as well.

zhaohs@ubuntu-edgy:~/cvs-svn/emacs/src$ gdb emacs
GNU gdb 6.4.90-debian
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i486-linux-gnu"...Using host libthread_db
library "/lib/tls/i686/cmov/libthread_db.so.1".

DISPLAY = :0.0
TERM = xterm
Breakpoint 1 at 0x81035b6: file emacs.c, line 464.
Breakpoint 2 at 0x811c3c6: file sysdep.c, line 1385.
(gdb) br xic_create_fontset
Function "xic_create_fontset" not defined.
Make breakpoint pending on future shared library load? (y or [n]) y

Breakpoint 3 (xic_create_fontset) pending.
(gdb) run --enable-font-backend -fn "Dejavu Sans Mono-14"
Starting program: /media/hdb6/MyHomeFiles/cvs-svn/emacs/src/emacs
--enable-font-backend -fn "Dejavu Sans Mono-14"
[Thread debugging using libthread_db enabled]
[New Thread -1219778896 (LWP 7688)]
[Switching to Thread -1219778896 (LWP 7688)]
Breakpoint 4 at 0x80d6f16: file xterm.c, line 8047.

--- Debug information ends here.

Best,
Hongsheng.

__________________________________________________
¸Ï¿ì×¢²áÑÅ»¢³¬´óÈÝÁ¿Ãâ·ÑÓÊÏä?
http://cn.mail.yahoo.com

Re: A very strange thing about Emacs working with flyspell!

[Kenichi Handa]

In article <20061120040501.GA7539@localdomain>, Hongsheng <zhaohscas@yahoo.com.cn> writes:

> (gdb) br xic_create_fontset
> Function "xic_create_fontset" not defined.
> Make breakpoint pending on future shared library load? (y or [n]) y

??? This means that xic_create_fontset is not compiled into
your Emacs, and that means that HAVE_X_I18N is not defined
in your src/config.h.  Can you use SCIM if you run Emacs
without --enable-font-backend?  If no, the problem is not
related to font-backend code.

Could you run configure script again and show me the all
output and the contents of generated src/config.h.

---
Kenichi Handa
handa@m17n.org

Re: A very strange thing about Emacs working with flyspell!

[Hongsheng]

[-- Attachment #1: Type: text/plain, Size: 1080 bytes --]

> > (gdb) br xic_create_fontset
> > Function "xic_create_fontset" not defined.
> > Make breakpoint pending on future shared library load? (y or [n]) y
> 
> ??? This means that xic_create_fontset is not compiled into
> your Emacs, and that means that HAVE_X_I18N is not defined
> in your src/config.h.  Can you use SCIM if you run Emacs
> without --enable-font-backend?

Yes, I can.

BTW, when compiling the Emacs, I configured the Emacs with the following arguments:

===============
./configure --prefix=/usr \
--x-includes=/usr/include/X11 --x-libraries=/usr/lib/X11 \
--enable-font-backend --enable-asserts \
--enable-locallisppath=/usr/share/emacs/site-lisp:/etc/emacs-snapshot/site-start.d
\
--enable-largefile --with-pop --with-sound \
--with-x=yes --with-x-toolkit=gtk --with-toolkit-scroll-bars \
--with-xpm --with-jpeg --with-tiff --with-gif --with-png \
--with-freetype  --with-xft --with-xim
===============

> Could you run configure script again and show me the all
> output and the contents of generated src/config.h.

See the attachments, please.

Best,
Hongsheng.

[-- Attachment #2: config.h.bz2 --]
[-- Type: application/octet-stream, Size: 8646 bytes --]

[-- Attachment #3: output-of-configure.bz2 --]
[-- Type: application/octet-stream, Size: 3496 bytes --]

[-- Attachment #4: Type: text/plain, Size: 142 bytes --]

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

Re: A very strange thing about Emacs working with flyspell!

[Kenichi Handa]

In article <20061120053240.GB12383@localdomain>, Hongsheng <zhaohscas@yahoo.com.cn> writes:

> [1  <text/plain; us-ascii (7bit)>]
> > > (gdb) br xic_create_fontset
> > > Function "xic_create_fontset" not defined.
> > > Make breakpoint pending on future shared library load? (y or [n]) y
> > 
> > ??? This means that xic_create_fontset is not compiled into
> > your Emacs, and that means that HAVE_X_I18N is not defined
> > in your src/config.h.  Can you use SCIM if you run Emacs
> > without --enable-font-backend?

> Yes, I can.

> BTW, when compiling the Emacs, I configured the Emacs with the following arguments:

> ===============
> ./configure --prefix=/usr \
> --x-includes=/usr/include/X11 --x-libraries=/usr/lib/X11 \
> --enable-font-backend --enable-asserts \
> --enable-locallisppath=/usr/share/emacs/site-lisp:/etc/emacs-snapshot/site-start.d
> \
> --enable-largefile --with-pop --with-sound \
> --with-x=yes --with-x-toolkit=gtk --with-toolkit-scroll-bars \
> --with-xpm --with-jpeg --with-tiff --with-gif --with-png \
> --with-freetype  --with-xft --with-xim
> ===============

> > Could you run configure script again and show me the all
> > output and the contents of generated src/config.h.

> See the attachments, please.

I couldn't find any problem with them.  Sigh...

Ok, then how about setting break point as this:

(gdb) br create_frame_xic

and use "s" command to step into the function
xic_create_fontset.

---
Kenichi Handa
handa@m17n.org

Re: A very strange thing about Emacs working with flyspell!

[Hongsheng]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 6619 bytes --]

> Ok, then how about setting break point as this:
> 
> (gdb) br create_frame_xic
> 
> and use "s" command to step into the function
> xic_create_fontset.

I use "s" and "n" commands alternatively, the following is my results.

BTW, I'm a layman about GDB.

==========
zhaohs@ubuntu-edgy:~$ cd cvs-svn/emacs/
zhaohs@ubuntu-edgy:~/cvs-svn/emacs$ cd src/
zhaohs@ubuntu-edgy:~/cvs-svn/emacs/src$ gdb emacs
GNU gdb 6.4.90-debian
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i486-linux-gnu"...Using host libthread_db
library "/lib/tls/i686/cmov/libthread_db.so.1".

DISPLAY = :0.0
TERM = xterm
Breakpoint 1 at 0x81035b6: file emacs.c, line 464.
Breakpoint 2 at 0x811c3c6: file sysdep.c, line 1385.
(gdb) br create_frame_xic
Breakpoint 3 at 0x80e2319: file xfns.c, line 2238.
(gdb) run --enable-font-backend
Starting program: /media/hdb6/MyHomeFiles/cvs-svn/emacs/src/emacs
--enable-font-backend
[Thread debugging using libthread_db enabled]
[New Thread -1219709264 (LWP 20555)]
[Switching to Thread -1219709264 (LWP 20555)]
Breakpoint 4 at 0x80d6f16: file xterm.c, line 8047.

Breakpoint 3, create_frame_xic (f=0x8bbfe28) at xfns.c:2238
2238    {
(gdb) s
2243      if (FRAME_XIC (f))
(gdb) s
2247      xfs = xic_create_xfontset
(gdb) s
fontset_ascii (id=2) at fontset.c:917
917     {
(gdb) s
921       elt = FONTSET_ASCII (fontset);
(gdb) s
917     {
(gdb) s
921       elt = FONTSET_ASCII (fontset);
(gdb) s
923       if (CONSP (elt))
(gdb) s
929     }
(gdb) s
xic_create_xfontset (f=0x8bbfe28, 
    base_fontname=0x876da60
"-Adobe-Courier-Medium-R-Normal--14-140-75-75-M-90-ISO8859-1") at
xfns.c:2103
2103      if (!base_fontname)
(gdb) s
2096    {
(gdb) s
2098      char **missing_list = NULL;
(gdb) s
2103      if (!base_fontname)
(gdb) s
2107      FOR_EACH_FRAME (rest, frame)
(gdb) s
2109          struct frame *cf = XFRAME (frame);
(gdb) s
2110          if (cf != f && FRAME_LIVE_P (f) && FRAME_X_P (cf)
(gdb) s
2122          char *fontsetname = xic_create_fontsetname (base_fontname,
False);
(gdb) s
xic_create_fontsetname (
    base_fontname=0x876da60
"-Adobe-Courier-Medium-R-Normal--14-140-75-75-M-90-ISO8859-1", motif=0) at
xfns.c:1972
1972      const char *sep = motif ? ";" : ",";
(gdb) s
1971    {
(gdb) s
1972      const char *sep = motif ? ";" : ",";
(gdb) s
1971    {
(gdb) s
1972      const char *sep = motif ? ";" : ",";
(gdb) s
1976      if (xic_defaut_fontset == base_fontname)
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1981          strcpy (fontsetname, base_fontname);
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) s
1993          for (i = 0; *p; p++)
(gdb) s
1994            if (*p == '-') i++;
(gdb) set args
(gdb) run --enable-font-backend -fn "Dejavu Sans Mono-14"
The program being debugged has been started already.
Start it from the beginning? (y or n) y

Starting program: /media/hdb6/MyHomeFiles/cvs-svn/emacs/src/emacs
--enable-font-backend -fn "Dejavu Sans Mono-14"
[Thread debugging using libthread_db enabled]
[New Thread -1220417872 (LWP 20590)]
[Switching to Thread -1220417872 (LWP 20590)]

Breakpoint 3, create_frame_xic (f=0x8bbfe28) at xfns.c:2238
2238    {
(gdb) n
2243      if (FRAME_XIC (f))
(gdb) n
2247      xfs = xic_create_xfontset
(gdb) n
2251      xim = FRAME_X_XIM (f);
(gdb) n
2252      if (xim)
(gdb) n
2251      xim = FRAME_X_XIM (f);
(gdb) n
2252      if (xim)
(gdb) n
2251      xim = FRAME_X_XIM (f);
(gdb) n
2252      if (xim)
(gdb) n
2263          if (xic_style == 0)
(gdb) n
2259          s_area.x = 0; s_area.y = 0; s_area.width = 1; s_area.height =
1;
(gdb) n
2263          if (xic_style == 0)
(gdb) n
2259          s_area.x = 0; s_area.y = 0; s_area.width = 1; s_area.height =
1;
(gdb) n
2260          spot.x = 0; spot.y = 1;
(gdb) n
2263          if (xic_style == 0)
(gdb) n
2269              xic_style = best_xim_style (&supported_list,
(gdb) n
2223        for (j = 0; j < xim->count_styles; ++j)
(gdb) n
2224          if (user->supported_styles[i] == xim->supported_styles[j])
(gdb) n
2223        for (j = 0; j < xim->count_styles; ++j)
(gdb) n
2224          if (user->supported_styles[i] == xim->supported_styles[j])
(gdb) s
2223        for (j = 0; j < xim->count_styles; ++j)
(gdb) s
2224          if (user->supported_styles[i] == xim->supported_styles[j])
(gdb) s
2223        for (j = 0; j < xim->count_styles; ++j)
(gdb) s
2224          if (user->supported_styles[i] == xim->supported_styles[j])
(gdb) s
2223        for (j = 0; j < xim->count_styles; ++j)
(gdb) s
2224          if (user->supported_styles[i] == xim->supported_styles[j])
(gdb) s
2223        for (j = 0; j < xim->count_styles; ++j)
(gdb) s
2224          if (user->supported_styles[i] == xim->supported_styles[j])
(gdb) s
2223        for (j = 0; j < xim->count_styles; ++j)
(gdb) s
2222      for (i = 0; i < user->count_styles; ++i)
(gdb) s
2223        for (j = 0; j < xim->count_styles; ++j)
(gdb) s
2224          if (user->supported_styles[i] == xim->supported_styles[j])
(gdb) s
2269              xic_style = best_xim_style (&supported_list,
(gdb) s
2273          preedit_attr = XVaCreateNestedList (0,
(gdb) s
s
2284          status_attr = XVaCreateNestedList (0,
(gdb) s
2273          preedit_attr = XVaCreateNestedList (0,
(gdb) s
2284          status_attr = XVaCreateNestedList (0,
(gdb) s
2295          xic = XCreateIC (xim,
(gdb) s
s
^[2302        XFree (preedit_attr);
(gdb) s
2303          XFree (status_attr);
(gdb) n
2306      FRAME_XIC (f) = xic;
(gdb) n
2307      FRAME_XIC_STYLE (f) = xic_style;
(gdb) n
2308      FRAME_XIC_FONTSET (f) = xfs;
(gdb) n
2309    }
(gdb) 
==========

Best,
Hongsheng.
__________________________________________________
¸Ï¿ì×¢²áÑÅ»¢³¬´óÈÝÁ¿Ãâ·ÑÓÊÏä?
http://cn.mail.yahoo.com

Re: A very strange thing about Emacs working with flyspell!

[Kenichi Handa]

I wrote:
>> Ummm, then, please apply the attached patch to src/xfns and
>> rebuild emacs as this:

Oops, I forgot to attach the patch, sorry.  Here it is.

---
Kenichi Handa
handa@m17n.org


*** xfns.c	29 Nov 2006 11:31:06 +0900	1.590.2.52
--- xfns.c	29 Nov 2006 11:34:03 +0900	
***************
*** 2125,2130 ****
--- 2125,2137 ----
        xfs = XCreateFontSet (FRAME_X_DISPLAY (f),
  			    fontsetname, &missing_list,
  			    &missing_count, &def_string);
+ #ifdef DEBUG_XIC_FONTSET
+       if (xfs)
+ 	printf ("Fontset created: %s\n", fontsetname);
+       else
+ 	printf ("Fontset fail: %s\n", fontsetname);
+ #endif
+ 
        if (missing_list)
  	XFreeStringList (missing_list);
        if (! xfs)
***************
*** 2143,2148 ****
--- 2150,2161 ----
  	      xfs = XCreateFontSet (FRAME_X_DISPLAY (f),
  				    p0, &missing_list,
  				    &missing_count, &def_string);
+ #ifdef DEBUG_XIC_FONTSET
+ 	      if (xfs)
+ 		printf ("Fontset created: %s\n", p0);
+ 	      else
+ 		printf ("Fontset fail: %s\n", p0);
+ #endif
  	      if (missing_list)
  		XFreeStringList (missing_list);
  	      if (xfs)
***************
*** 2158,2163 ****
--- 2171,2182 ----
  	  xfs = XCreateFontSet (FRAME_X_DISPLAY (f),
  				fontsetname, &missing_list,
  				&missing_count, &def_string);
+ #ifdef DEBUG_XIC_FONTSET
+ 	  if (xfs)
+ 	    printf ("Fontset created: %s\n", fontsetname);
+ 	  else
+ 	    printf ("Fontset fail: %s\n", fontsetname);
+ #endif
  	  if (missing_list)
  	    XFreeStringList (missing_list);
  	  xfree (fontsetname);

Re: A very strange thing about Emacs working with flyspell!

[Hongsheng]

On Wed, Nov 29, 2006 at 11:34:54AM +0900, Kenichi Handa wrote:
> I wrote:
> >> Ummm, then, please apply the attached patch to src/xfns and
> >> rebuild emacs as this:
> 
> Oops, I forgot to attach the patch, sorry.  Here it is.
> 
> ---
> Kenichi Handa
> handa@m17n.org
> 
> 
> *** xfns.c	29 Nov 2006 11:31:06 +0900	1.590.2.52
> --- xfns.c	29 Nov 2006 11:34:03 +0900	
> ***************
> *** 2125,2130 ****
> --- 2125,2137 ----
>         xfs = XCreateFontSet (FRAME_X_DISPLAY (f),
>   			    fontsetname, &missing_list,
>   			    &missing_count, &def_string);
> + #ifdef DEBUG_XIC_FONTSET
> +       if (xfs)
> + 	printf ("Fontset created: %s\n", fontsetname);
> +       else
> + 	printf ("Fontset fail: %s\n", fontsetname);
> + #endif
> + 
>         if (missing_list)
>   	XFreeStringList (missing_list);
>         if (! xfs)
> ***************
> *** 2143,2148 ****
> --- 2150,2161 ----
>   	      xfs = XCreateFontSet (FRAME_X_DISPLAY (f),
>   				    p0, &missing_list,
>   				    &missing_count, &def_string);
> + #ifdef DEBUG_XIC_FONTSET
> + 	      if (xfs)
> + 		printf ("Fontset created: %s\n", p0);
> + 	      else
> + 		printf ("Fontset fail: %s\n", p0);
> + #endif
>   	      if (missing_list)
>   		XFreeStringList (missing_list);
>   	      if (xfs)
> ***************
> *** 2158,2163 ****
> --- 2171,2182 ----
>   	  xfs = XCreateFontSet (FRAME_X_DISPLAY (f),
>   				fontsetname, &missing_list,
>   				&missing_count, &def_string);
> + #ifdef DEBUG_XIC_FONTSET
> + 	  if (xfs)
> + 	    printf ("Fontset created: %s\n", fontsetname);
> + 	  else
> + 	    printf ("Fontset fail: %s\n", fontsetname);
> + #endif
>   	  if (missing_list)
>   	    XFreeStringList (missing_list);
>   	  xfree (fontsetname);

Now, I redownload all the emacs source-code from its cvs repository and the
patch above has been used.

I can run emacs and display CJK-EXTB correctly by: 

% emacs --enable-font-backend

But the Ctrl+SPC can't activate the SCIM yet, the minibuffer will say Mark
set and Mark activated.

-- 
Hongyi Zhao
赵红生（弘毅）
--

题目:《山中寄招叶秀才》
作者:林逋（９６７－１０２８）
夜鹤晓猿时复闻，寥寥长似耿离群。
月中未要恨丹桂，岭上且来看白云。
棋子不妨临水着，诗题兼好共僧分。
新忧他日荣名后，难得幽栖事静君。