unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
* ANN: Emacs.app 9.0-pre1
@ 2005-11-05 20:51 Adrian Robert
  2005-11-06 22:47 ` Stefan Monnier
  0 siblings, 1 reply; 10+ messages in thread
From: Adrian Robert @ 2005-11-05 20:51 UTC (permalink / raw)


Hi all,

I'm happy to announce an initial release of the GNUstep / Cocoa port 
based on the unicode-2 branch:

http://emacs-app.sf.net/

It's available (see the site) both as a patch to 2005/11/05 CVS (+ 
mini-tarball) or a monolithic tarball.  It should build and run on Mac 
OS X 10.3 and 10.4, and GNUstep on at least Linux (that's the only 
system I have for testing).


What works:

- all the basics
- fringes
- proportional text
- images (except XPM)
- boxes/borders, etc.


What doesn't work:

- toolbars
- tooltips


What has glitches:

- non-ascii text rendering and keyboard handling
- font and font size selection
- ctrl-g handling (interrupt input is not (yet) used)
- image backgrounds (e.g., gnus)

For the last, note that images are displayed via compositing, so the 
problem only comes up for images without an alpha layer whose 
background does not match the local face.


Why this Port?

Although there is already an existing port of GNU Emacs to Mac OS X, 
the "Emacs.app" port offers three advantages over it:

1) it runs in addition on the free GNUstep implementation of the Mac's 
Cocoa APIs (which are based on NeXT/OpenStep);

2) it makes a more concerted attempt to take advantage of the desktop / 
UI facilities both OS X and GNUstep offer for font and color selection, 
setting preferences, and so on, without compromising emacs's 
functionality;

3) because Cocoa is a higher-level API than Carbon, the windowing code 
is about half the size, and hopefully correspondingly easier to 
maintain.

On the other hand, the Carbon port supports OS X 10.1 and 10.2, plus 
Mac OS 8 and 9, and it is currently more complete, bug-free, and 
probably more stable than Emacs.app.


Why the 9.0 version number?

Emacs.app actually dates back to the early 1990s when Emacs 19 was 
ported to the NeXT computer.  The NeXTstep windowing APIs used on that 
machine became the basis for first OpenStep, then Rhapsody, then MacOS 
X.  Emacs.app has been kept updated by various people for most of that 
time.


What about Emacs 22?

I believe Emacs.app could trivially be converted to work off of the CVS 
Head codebase, however without some additional Mule <-> UTF conversion 
code I embedded into Emacs on Aqua, it will not render non-ASCII 
characters.  This code was written by Otried Cheong 
(http://tclab.kaist.ac.kr/~otfried/Mule/) and I have not asked him what 
the copyright status of that code is.  In addition, there is not nearly 
sufficient time for the implementation to be finished, debugged, and 
integrated before the initial 22 release.


Note on the implementation

I mostly followed the Carbon port in updating the Makefiles and 
configuration.  However MAC_OSX and MAC_OS are UNdefined when compiling 
for NS due to these being used for Carbon-specific things.  Unexec 
seems to work perfectly using Andrew Choi's unexmacosx.c, however I 
haven't been able to get dumped images to run under GNUstep.


regards,
Adrian Robert

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
  2005-11-05 20:51 Adrian Robert
@ 2005-11-06 22:47 ` Stefan Monnier
  0 siblings, 0 replies; 10+ messages in thread
From: Stefan Monnier @ 2005-11-06 22:47 UTC (permalink / raw)
  Cc: emacs-devel

> I'm happy to announce an initial release of the GNUstep / Cocoa port based
> on the unicode-2 branch:

> http://emacs-app.sf.net/

I think if we all agree that this is intended to be included in some (not
too distant) future release of Emacs, it would be good to put it on a branch
in the CVS.  The only reason not to do that yet is if some legal paperwork
is still pending.

Hopefully this branch can "quickly" be merged into the unicode branch
(AFAICT the main criteria for this would be that the patch doesn't
introduce any bug in the other ports, including the Carbon port).


        Stefan

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
@ 2005-11-07  8:17 Sun Yijiang
  2005-11-07 13:02 ` Adrian Robert
  0 siblings, 1 reply; 10+ messages in thread
From: Sun Yijiang @ 2005-11-07  8:17 UTC (permalink / raw)
  Cc: emacs-devel


[-- Attachment #1.1: Type: text/plain, Size: 6242 bytes --]

I have GNUStep installed on my GNU/Linux, but got this error message:
# make
gcc -c -D_BSD_SOURCE -Demacs -DHAVE_CONFIG_H -I. -I/home/sun/sandbox/emacs-
23.0.0_NS-9.0pre1/src -D_BSD_SOURCE -g -O2 nsterm.m
nsterm.m: In function `ns_init_paths':
nsterm.m:292: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:292: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:298: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:301: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:311: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:311: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:311: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:313: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:332: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:332: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:334: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:349: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:364: error: cannot find interface declaration for
`NXConstantString'
nsterm.m: In function `ns_update_end':
nsterm.m:616: warning: `NSView' may not respond to `-unlockFocusNeedsFlush:'
nsterm.m:616: warning: (Messages without a matching method signature
nsterm.m:616: warning: will be assumed to return `id' and accept
nsterm.m:616: warning: `...' as arguments.)
nsterm.m: In function `ns_get_color':
nsterm.m:1129: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:1246: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:1248: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:1250: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:1252: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:1254: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:1272: error: cannot find interface declaration for
`NXConstantString'
nsterm.m: In function `ns_term_init':
nsterm.m:3046: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:3048: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:3051: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:3058: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:3082: error: cannot find interface declaration for
`NXConstantString'
nsterm.m: In function `-[EmacsView menuDown:]':
nsterm.m:4200: warning: multiple methods named `-representedObject' found
/usr/GNUstep/System/Library/Headers/AppKit/NSCell.h:351: warning: using
`-(id)representedObject'
/usr/GNUstep/System/Library/Headers/AppKit/NSRulerMarker.h:76: warning: also
found `-(id <NSCopying>)representedObject'
nsterm.m:4202: warning: multiple methods named `-representedObject' found
/usr/GNUstep/System/Library/Headers/AppKit/NSCell.h:351: warning: using
`-(id)representedObject'
/usr/GNUstep/System/Library/Headers/AppKit/NSRulerMarker.h:76: warning: also
found `-(id <NSCopying>)representedObject'
nsterm.m: In function `-[EmacsPrefsController init]':
nsterm.m:4781: error: cannot find interface declaration for
`NXConstantString'
nsterm.m: In function `ns_dumpglyphs':
nsterm.m:5001: warning: comparison between pointer and integer
nsterm.m:5087: warning: comparison between pointer and integer
nsterm.m:5093: warning: comparison between pointer and integer
nsterm.m: In function `removeSynthItal':
nsterm.m:5340: error: cannot find interface declaration for
`NXConstantString'
nsterm.m: In function `ns_load_font':
nsterm.m:5372: warning: assignment discards qualifiers from pointer target
type
nsterm.m:5423: warning: assignment discards qualifiers from pointer target
type
nsterm.m:5448: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:5582: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:5583: error: cannot find interface declaration for
`NXConstantString'
nsterm.m: In function `ns_list_fonts':
nsterm.m:5679: warning: assignment discards qualifiers from pointer target
type
nsterm.m:5688: warning: assignment makes pointer from integer without a cast
nsterm.m:5699: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:5702: warning: assignment discards qualifiers from pointer target
type
nsterm.m:5717: warning: assignment discards qualifiers from pointer target
type
nsterm.m:5724: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:5725: warning: assignment discards qualifiers from pointer target
type
nsterm.m:5732: warning: assignment discards qualifiers from pointer target
type
nsterm.m: In function `ns_font_to_xlfd':
nsterm.m:5794: warning: initialization discards qualifiers from pointer
target type
nsterm.m:5800: error: cannot find interface declaration for
`NXConstantString'
nsterm.m:5813: error: cannot find interface declaration for
`NXConstantString'
nsterm.m: In function `ns_fontname_to_xlfd':
nsterm.m:5847: error: cannot find interface declaration for
`NXConstantString'
make: *** [nsterm.o] Error 1

I add "#include <Foundation/Foundation.h>" to the head of nsterm.m, but got
same error messages.
This is what configure outputs:
What operating system and machine description files should Emacs use?
`s/gnu-linux.h' and `m/intel386.h'
What compiler should emacs be built with? gcc -g -O2
Should Emacs use the GNU version of malloc? yes
(Using Doug Lea's new malloc from the GNU C Library.)
Should Emacs use a relocating allocator for buffers? yes
Should Emacs use mmap(2) for buffer allocation? no
What window system should Emacs use? nextstep
What toolkit should Emacs use? none
Where do we find X Windows header files? NONE
Where do we find X Windows libraries? NONE
Does Emacs use -lXaw3d? no
Does Emacs use -lXpm? no
Does Emacs use -ljpeg? no
Does Emacs use -ltiff? no
Does Emacs use -lungif? no
Does Emacs use -lpng? no
Does Emacs use X toolkit scroll bars? yes

[-- Attachment #1.2: Type: text/html, Size: 8817 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

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

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
  2005-11-07  8:17 ANN: Emacs.app 9.0-pre1 Sun Yijiang
@ 2005-11-07 13:02 ` Adrian Robert
  2005-11-08  2:52   ` Sun Yijiang
  0 siblings, 1 reply; 10+ messages in thread
From: Adrian Robert @ 2005-11-07 13:02 UTC (permalink / raw)
  Cc: emacs-devel



On Nov 7, 2005, at 3:17 AM, Sun Yijiang wrote:

> I have GNUStep installed on my GNU/Linux, but got this error message:
> # make
> gcc -c -D_BSD_SOURCE   -Demacs -DHAVE_CONFIG_H   -I. -I/home/sun/ 
> sandbox/emacs-23.0.0_NS-9.0pre1/src -D_BSD_SOURCE -g -O2 nsterm.m
> nsterm.m: In function `ns_init_paths':
> nsterm.m:292: error: cannot find interface declaration for  
> `NXConstantString'

Try editing nextstep/compile and add

  -fconstant-string-class=NSConstantString

to the CFLAGS variable setting for the GNUstep case.  Then rm ../ 
config.log and invoke ./compile again.  (The other warnings are "OK"..)

At one point I needed that flag when compiling it under GNUstep, but  
then I stopped needing it and I'd assumed that it was because of some  
other flag settings rendering it unnecessary -- but it looks like I  
might have been wrong.  What OS and gcc version are you running?   
What version of gnustep?

Adrian

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
  2005-11-07 13:02 ` Adrian Robert
@ 2005-11-08  2:52   ` Sun Yijiang
  2005-11-08  3:53     ` Adrian Robert
  0 siblings, 1 reply; 10+ messages in thread
From: Sun Yijiang @ 2005-11-08  2:52 UTC (permalink / raw)
  Cc: emacs-devel


[-- Attachment #1.1: Type: text/plain, Size: 3319 bytes --]

Thank you Adrian. This is my env, generally speaking, it's Gentoo
2005.1with gcc
3.4.4 and gnustep from `ACCEPT_KEYWORDS="~x86" emerge gnustep-env'.
===============================================
# uname -a
Linux SUN 2.6.13-gentoo-r5 #3 Sun Oct 30 21:32:52 CST 2005 i686 Intel(R)
Pentium(R) M processor 1.86GHz GenuineIntel GNU/Linux

# gcc -v
Reading specs from /usr/lib/gcc/i686-pc-linux-gnu/3.4.4/specs
Configured with: /var/tmp/portage/gcc-3.4.4-r1/work/gcc-3.4.4/configure
--prefix=/usr --bindir=/usr/i686-pc-linux-gnu/gcc-bin/3.4.4
--includedir=/usr/lib/gcc/i686-pc-linux-gnu/3.4.4/include
--datadir=/usr/share/gcc-data/i686-pc-linux-gnu/3.4.4
--mandir=/usr/share/gcc-data/i686-pc-linux-gnu/3.4.4/man
--infodir=/usr/share/gcc-data/i686-pc-linux-gnu/3.4.4/info
--with-gxx-include-dir=/usr/lib/gcc/i686-pc-linux-gnu/3.4.4/include/g++-v3
--host=i686-pc-linux-gnu --build=i686-pc-linux-gnu --disable-altivec
--enable-nls --without-included-gettext --with-system-zlib
--disable-checking --disable-werror --disable-libunwind-exceptions
--disable-multilib --disable-libgcj --enable-languages=c,c++,objc,f77
--enable-shared --enable-threads=posix --enable-__cxa_atexit
--enable-clocale=gnu
Thread model: posix
gcc version 3.4.4 (Gentoo 3.4.4-r1, ssp-3.4.4-1.0, pie-8.7.8)

# emerge -s gnustep
* gnustep-base/gnustep-back-art
Latest version installed: 0.9.6-r1

* gnustep-base/gnustep-back-xlib
Latest version installed: [ Not Installed ]

* gnustep-base/gnustep-base
Latest version installed: 1.10.3-r1

* gnustep-base/gnustep-env
Latest version installed: 0.1.6-r1

* gnustep-base/gnustep-gui
Latest version installed: 0.9.5

* gnustep-base/gnustep-make
Latest version installed: 1.10.0-r2
===============================================

I googled for a while after that reply, and found the same suggestion as you
gave. It works for compiling *.m to *.o, but there're still error messages
when gcc links *.o to `temacs', all are "undefined reference to
whatwhatwhat". I added `-lobjc' to the gcc link command line, some errors
gone, but still many like "undefined reference to NSxxx" and "undefined
reference to __objc_class_name_NSxxx". I searched the GNUStep directory,
found no *.a but only *.so. I'm stuck here.

2005/11/7, Adrian Robert <arobert@cogsci.ucsd.edu>:
>
>
>
> On Nov 7, 2005, at 3:17 AM, Sun Yijiang wrote:
>
> > I have GNUStep installed on my GNU/Linux, but got this error message:
> > # make
> > gcc -c -D_BSD_SOURCE -Demacs -DHAVE_CONFIG_H -I. -I/home/sun/
> > sandbox/emacs-23.0.0_NS-9.0pre1/src -D_BSD_SOURCE -g -O2 nsterm.m
> > nsterm.m: In function `ns_init_paths':
> > nsterm.m:292: error: cannot find interface declaration for
> > `NXConstantString'
>
> Try editing nextstep/compile and add
>
> -fconstant-string-class=NSConstantString
>
> to the CFLAGS variable setting for the GNUstep case. Then rm ../
> config.log and invoke ./compile again. (The other warnings are "OK"..)
>
> At one point I needed that flag when compiling it under GNUstep, but
> then I stopped needing it and I'd assumed that it was because of some
> other flag settings rendering it unnecessary -- but it looks like I
> might have been wrong. What OS and gcc version are you running?
> What version of gnustep?
>
> Adrian
>
>
>
>

[-- Attachment #1.2: Type: text/html, Size: 4096 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

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

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
  2005-11-08  2:52   ` Sun Yijiang
@ 2005-11-08  3:53     ` Adrian Robert
  2005-11-08  6:54       ` Sun Yijiang
  0 siblings, 1 reply; 10+ messages in thread
From: Adrian Robert @ 2005-11-08  3:53 UTC (permalink / raw)
  Cc: emacs-devel


On Nov 7, 2005, at 9:52 PM, Sun Yijiang wrote:

> Thank you Adrian.  This is my env, generally speaking, it's Gentoo  
> 2005.1 with gcc 3.4.4 and gnustep from `ACCEPT_KEYWORDS="~x86"  
> emerge gnustep-env'.
>
>
> # emerge -s gnustep
> *  gnustep-base/gnustep-back-art
>       Latest version installed: 0.9.6-r1
> ...

Should be OK.


> ===============================================
>
> I googled for a while after that reply, and found the same  
> suggestion as you gave.  It works for compiling *.m to *.o, but  
> there're still error messages when gcc links *.o to `temacs', all  
> are "undefined reference to whatwhatwhat".  I added `-lobjc' to the  
> gcc link command line, some errors gone, but still many like  
> "undefined reference to NSxxx" and "undefined reference to  
> __objc_class_name_NSxxx".  I searched the GNUStep directory, found  
> no *.a but only *.so.  I'm stuck here.

OK, I'd forgotten that -fconstant-string-class=NSConstantString  
should have been set up in the Makefile already.  (That's why it  
wasn't needed in the config anymore.)  Take a look at src/Makefile.in  
and search for 'GNUSTEP'.  If the .m.o rule there wasn't making it  
into your actual Makefile, or was being ignored for some reason,  
that's the source of that problem.  The linking issue could be  
related, because those flags are also keyed off of #ifdef GNUSTEP, in  
src/Makefile.in and src/config.h.

I'm not sure what's happening in your case, but for the heck of it,  
you could try adding '-DGNUSTEP' to the CFLAGS in nextstep/compile --  
maybe there's something different on your system regarding how CFLAGS  
and CPPFLAGS are processed.

(I'm more than open to suggestions for improving any of this, if you  
can find something that works on your setup..)

thanks,
Adrian

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
  2005-11-08  3:53     ` Adrian Robert
@ 2005-11-08  6:54       ` Sun Yijiang
  2005-11-08 12:05         ` Adrian Robert
  0 siblings, 1 reply; 10+ messages in thread
From: Sun Yijiang @ 2005-11-08  6:54 UTC (permalink / raw)
  Cc: emacs-devel


[-- Attachment #1.1: Type: text/plain, Size: 2778 bytes --]

I can now get temacs and emacs built in src, after some work:
1) setup env variables:
export GNUSTEP_SYSTEM_ROOT=/usr/GNUstep/System
export GNUSTEP_MAKEFILES=$GNUSTEP_SYSTEM_ROOT/Library/Makefiles
export CPATH=$GNUSTEP_SYSTEM_ROOT/Library/Headers:$CPATH
export
LD_LIBRARY_PATH=/usr/GNUstep/System/Library/Libraries:$LD_LIBRARY_PATH

2) modify src/Makefile.in:
+#define GNUSTEP
#define THIS_IS_MAKEFILE
#define NO_SHORTNAMES

than configure and build:
# ./configure --prefix=xxx --with-ns --without-x
# make

I think it's better to define `GNUSTEP' in ./configure process.

But there's still problems! After src/emacs is built, the build process
proceeds to byte-compile elisp codes, and that makes a Segmentation fault:

make[1]: Entering directory `/home/sun/sandbox/emacs-23.0.0_NS-9.0pre1/leim'
EMACSLOADPATH=/home/sun/sandbox/emacs-23.0.0_NS-9.0pre1/leim/../lispLC_ALL=C
../src/emacs -batch --no-init-file --no-site-file --multibyte -l
/home/sun/sandbox/emacs-23.0.0_NS-9.0pre1/leim/../lisp/international/quail \
-f batch-byte-compile-if-not-done quail/CCDOSPY.el quail/Punct.el
quail/QJ.el quail/SW.el quail/TONEPY.el quail/4Corner.el quail/ARRAY30.el
quail/ECDICT.el quail/ETZY.el quail/Punct-b5.el quail/PY-b5.el quail/QJ-
b5.el quail/ZOZY.el quail/tsang-b5.el quail/quick-b5.el
quail/tsang-cns.elquail/quick-
cns.el quail/PY.el quail/ZIRANMA.el quail/CTLau.el quail/CTLau-b5.el
make[1]: *** [leim-list.el] Segmentation fault
make[1]: Leaving directory `/home/sun/sandbox/emacs-23.0.0_NS-9.0pre1/leim'
make: *** [leim] Error 2

I can run temacs with `./temacs -q -nw' in src, but `./emacs -q -nw' failed
with the same "Segmentation fault". Maybe there's something wrong in the
dumping process? I tried to build temacs with -O0 and -O1, but none woks.

2005/11/8, Adrian Robert <arobert@cogsci.ucsd.edu>:

>
> OK, I'd forgotten that -fconstant-string-class=NSConstantString
> should have been set up in the Makefile already. (That's why it
> wasn't needed in the config anymore.) Take a look at src/Makefile.in
> and search for 'GNUSTEP'. If the .m.o rule there wasn't making it
> into your actual Makefile, or was being ignored for some reason,
> that's the source of that problem. The linking issue could be
> related, because those flags are also keyed off of #ifdef GNUSTEP, in
> src/Makefile.in and src/config.h.
>
> I'm not sure what's happening in your case, but for the heck of it,
> you could try adding '-DGNUSTEP' to the CFLAGS in nextstep/compile --
> maybe there's something different on your system regarding how CFLAGS
> and CPPFLAGS are processed.
>
> (I'm more than open to suggestions for improving any of this, if you
> can find something that works on your setup..)
>
> thanks,
> Adrian
>

[-- Attachment #1.2: Type: text/html, Size: 3246 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

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

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
  2005-11-08  6:54       ` Sun Yijiang
@ 2005-11-08 12:05         ` Adrian Robert
  2005-11-08 16:05           ` Sun Yijiang
  0 siblings, 1 reply; 10+ messages in thread
From: Adrian Robert @ 2005-11-08 12:05 UTC (permalink / raw)
  Cc: Developer GNUstep, emacs-devel


On Nov 8, 2005, at 1:54 AM, Sun Yijiang wrote:

> I can now get temacs and emacs built in src, after some work:
> 1) setup env variables:
> export GNUSTEP_SYSTEM_ROOT=/usr/GNUstep/System
> export GNUSTEP_MAKEFILES=$GNUSTEP_SYSTEM_ROOT/Library/Makefiles
> export CPATH=$GNUSTEP_SYSTEM_ROOT/Library/Headers:$CPATH
> export LD_LIBRARY_PATH=/usr/GNUstep/System/Library/Libraries: 
> $LD_LIBRARY_PATH

I think this relates to gnustep's recent config change to stop using  
environment variables.  I'll look into it..


> 2) modify src/Makefile.in:
> +#define GNUSTEP
>  #define THIS_IS_MAKEFILE
>  #define NO_SHORTNAMES
>
> than configure and build:
> # ./configure --prefix=xxx --with-ns --without-x
> # make
>
> I think it's better to define `GNUSTEP' in ./configure process.

Not sure if this is causing your particular segfault, but make sure  
you add -DCANNOT_DUMP to C/CPPFLAGS (see nextstep/compile).  And  
GNUSTEP _should_ be being defined if you use the compile script.   
(But probably in the long run GNUSTEP / COCOA should be defined in  
the appropriate place in configure.in as you say.)

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
  2005-11-08 12:05         ` Adrian Robert
@ 2005-11-08 16:05           ` Sun Yijiang
  2005-11-08 16:57             ` Fabien VALLON
  0 siblings, 1 reply; 10+ messages in thread
From: Sun Yijiang @ 2005-11-08 16:05 UTC (permalink / raw)
  Cc: Developer GNUstep, emacs-devel


[-- Attachment #1.1: Type: text/plain, Size: 1332 bytes --]

2005/11/8, Adrian Robert <arobert@cogsci.ucsd.edu>:

>
> Not sure if this is causing your particular segfault, but make sure
> you add -DCANNOT_DUMP to C/CPPFLAGS (see nextstep/compile). And
> GNUSTEP _should_ be being defined if you use the compile script.
> (But probably in the long run GNUSTEP / COCOA should be defined in
> the appropriate place in configure.in <http://configure.in> as you say.)
>

Yes, -DCANNOT_DUMP solves the segfault problem, I can now "make install" : )
I did not use the nextstep/compile script, but performed its tasks one by
one manually.

But when I cd into nextstep/build/Emacs.app and run "./Emacs -q", I met
these error messages:

ns_init_paths: '/home/sun/sandbox/emacs-
23.0.0_NS-9.0pre1/nextstep/build/Emacs.app'
/home/sun/sandbox/emacs-23.0.0_NS-9.0pre1/nextstep/build/Emacs.app/Resources
2005-11-08 23:56:16.000 Emacs[13820] NSApplication.m:203 Assertion failed in
initialize_gnustep_backend. Unable to find backend libgnustep-back.bundle
./Emacs: Uncaught exception NSInternalInconsistencyException, reason:
NSApplication.m:203 Assertion failed in initialize_gnustep_backend. Unable
to find backend libgnustep-back.bundle

It seems there's something wrong with my GNUStep env, maybe the "backend"
part. I'm new to GNUStep, what's that mean? and what should I do?

[-- Attachment #1.2: Type: text/html, Size: 1708 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

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

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: ANN: Emacs.app 9.0-pre1
  2005-11-08 16:05           ` Sun Yijiang
@ 2005-11-08 16:57             ` Fabien VALLON
  0 siblings, 0 replies; 10+ messages in thread
From: Fabien VALLON @ 2005-11-08 16:57 UTC (permalink / raw)
  Cc: Adrian Robert, emacs-devel, Developer GNUstep

On Mar 8 novembre 2005 17:05, Sun Yijiang wrote:
> Yes, -DCANNOT_DUMP solves the segfault problem, I can now "make install" :
/home/sun/sandbox/emacs-23.0.0_NS-9.0pre1/nextstep/build/Emacs.app/Resources
> 2005-11-08 23:56:16.000 Emacs[13820] NSApplication.m:203 Assertion failed
> in
> initialize_gnustep_backend. Unable to find backend libgnustep-back.bundle
> ./Emacs: Uncaught exception NSInternalInconsistencyException, reason:
> NSApplication.m:203 Assertion failed in initialize_gnustep_backend. Unable
> to find backend libgnustep-back.bundle
>
> It seems there's something wrong with my GNUStep env, maybe the "backend"
> part. I'm new to GNUStep, what's that mean? and what should I do?

Yes Jeff Teunissen and I had the same problem.
After straced it, it looks like that the app seeach the backend in wrong
paths.

Maybye introduced here :
2005-07-17 * Source/NSApplication.m (gnustep_backend_path,
	gnustep_backend_framework, gnustep_backend_bundle): New functions.
	(initialize_gnustep_backend): Support versioning of backend bundle.


but not sure ..


Fabien

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2005-11-08 16:57 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-11-07  8:17 ANN: Emacs.app 9.0-pre1 Sun Yijiang
2005-11-07 13:02 ` Adrian Robert
2005-11-08  2:52   ` Sun Yijiang
2005-11-08  3:53     ` Adrian Robert
2005-11-08  6:54       ` Sun Yijiang
2005-11-08 12:05         ` Adrian Robert
2005-11-08 16:05           ` Sun Yijiang
2005-11-08 16:57             ` Fabien VALLON
  -- strict thread matches above, loose matches on Subject: below --
2005-11-05 20:51 Adrian Robert
2005-11-06 22:47 ` Stefan Monnier

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).