Problem report: No syntax highlight for C

Problem report: No syntax highlight for C

[Jie Zhang]

Today I up to CVS HEAD Emacs, but C source code does not get syntax
highlighted. Other modes like lisp, shell look OK.

Jie

Re: Problem report: No syntax highlight for C

[David Kastrup]

"Jie Zhang" <jzhang918@gmail.com> writes:

> Today I up to CVS HEAD Emacs, but C source code does not get syntax
> highlighted. Other modes like lisp, shell look OK.

One likely cause is something in your .emacs intended to turn syntax
highlighting on for C mode, but which actually _toggles_ it.  And
since the default in CVS Emacs is _on_, you turn it off again.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Problem report: No syntax highlight for C

[Eli Zaretskii]

> Date: Sat, 3 Jun 2006 01:07:21 +0800
> From: "Jie Zhang" <jzhang918@gmail.com>
> 
> Today I up to CVS HEAD Emacs, but C source code does not get syntax
> highlighted.

Is this in "emacs -Q"?  If so, I cannot reproduce this.  Do you have
any messages in the *Messages* buffer that could shed a light on the
reason(s)?

Re: Problem report: No syntax highlight for C

[Eli Zaretskii]

[Please don't take the discussion off the list]

> Date: Sat, 3 Jun 2006 08:59:41 +0800
> From: "Jie Zhang" <jzhang918@gmail.com>
> 
> > Is this in "emacs -Q"?  If so, I cannot reproduce this.  Do you have
> > any messages in the *Messages* buffer that could shed a light on the
> > reason(s)?
> >
> I'm rather new to emacs. I don't know what exactly "emacs -Q" meas.
> Does it means passing -Q to emacs while invoking it?

Yes, it means invoke Emacs with the -Q command-line switch.

> The messages in the *Messages* buffer are:
> 
> (emacs t.c)
> For information about the GNU Project and its goals, type C-h C-p.
> t.c has auto save data; consider M-x recover-this-file
> Loading cc-mode...done
> File mode specification error: (void-variable c-font-lock-keywords-3)
> 
> The last line seems pointing out the cause of this issue.

Yes, it does.  But we have to find out what causes that error.

If using -Q doesn't make this problem go away, please tell how up to
date is your Emacs code.  If you checked out the CVS tree yourself,
please tell at what date you did that.  If you got the sources from
someone else, please tell what is the first entry that you see in the
file lisp/ChangeLog.

Re: Problem report: No syntax highlight for C

[Jie Zhang]

On 6/3/06, Eli Zaretskii <eliz@gnu.org> wrote:
> [Please don't take the discussion off the list]
>
> > Date: Sat, 3 Jun 2006 08:59:41 +0800
> > From: "Jie Zhang" <jzhang918@gmail.com>
> >
> > > Is this in "emacs -Q"?  If so, I cannot reproduce this.  Do you have
> > > any messages in the *Messages* buffer that could shed a light on the
> > > reason(s)?
> > >
> > I'm rather new to emacs. I don't know what exactly "emacs -Q" meas.
> > Does it means passing -Q to emacs while invoking it?
>
> Yes, it means invoke Emacs with the -Q command-line switch.
>
> > The messages in the *Messages* buffer are:
> >
> > (emacs t.c)
> > For information about the GNU Project and its goals, type C-h C-p.
> > t.c has auto save data; consider M-x recover-this-file
> > Loading cc-mode...done
> > File mode specification error: (void-variable c-font-lock-keywords-3)
> >
> > The last line seems pointing out the cause of this issue.
>
> Yes, it does.  But we have to find out what causes that error.
>
> If using -Q doesn't make this problem go away, please tell how up to
> date is your Emacs code.  If you checked out the CVS tree yourself,
> please tell at what date you did that.  If you got the sources from
> someone else, please tell what is the first entry that you see in the
> file lisp/ChangeLog.
>

I have narrowed down the changes which cause this issue. The CVS on
20060524 has no such issue, but 20060525 does. There are two and half
entries which were added between these two dates. I add their owners
to the CC list. Hope this help.


Jie



$ cvs -z9 -q diff -u -p -N -D20060524 -D20060525 ChangeLog
Index: ChangeLog
===================================================================
RCS file: /sources/emacs/emacs/src/ChangeLog,v
retrieving revision 1.5083
retrieving revision 1.5086
diff -u -p -r1.5083 -r1.5086
--- ChangeLog   23 May 2006 08:12:13 -0000      1.5083
+++ ChangeLog   24 May 2006 13:51:11 -0000      1.5086
@@ -1,3 +1,61 @@
+2006-05-24  Alan Mackenzie  <acm@muc.de>
+
+       * lread.c (Vload_history): Enhance doc-string to say that the file
+       is the absolute truename of the loaded file.
+
+       * lread.c (Vafter_load_alist): doc-string: state that an element
+       now has a regexp to match file names, not a file name as such.
+
+       * lread.c (readevalloop): Call file-truename on the name for
+       load-history, except at preloading time.
+
+       * lread.c (Fload): At preloading time, preserve the extension of
+       the filename which goes into load-history.  New variable
+       hist_file_name.
+
+       * lread.c (Fload): Do eval-after-load stuff by calling the lisp
+       function do-after-load-evaluation.
+
+2006-05-24  YAMAMOTO Mitsuharu  <mituharu@math.s.chiba-u.ac.jp>
+
+       * mac.c (ae_attr_table): New variable.
+       (syms_of_mac): Intern and staticpro its elements.
+       (mac_aelist_to_lisp): Also convert Apple event attributes.
+       (mac_ae_put_lisp): New function.
+       (create_apple_event_from_event_ref) [MAC_OSX]: Use typeUTF8Text.
+
+       * macfns.c (Fx_server_version): Use gestaltSystemVersionMajor etc.
+
+       * macselect.c (Qemacs_suspension_id): New variable.
+       (syms_of_macselect): Intern and staticpro it.
+       (struct suspended_ae_info): New struct.
+       (deferred_apple_events, defer_apple_events)
+       (Fmac_process_deferred_apple_events): Use it.
+       (suspended_apple_events): New variable.
+       (mac_handle_apple_event_1): New function.
+       (mac_handle_apple_event): Use it.  Don't process previously
+       suspended events.
+       (cleanup_suspended_apple_events, get_suspension_id)n
+       (cleanup_all_suspended_apple_events): New functions.
+       (init_apple_event_handler): Call cleanup_all_suspended_apple_events
+       at exit.
+       (Fmac_cleanup_expired_apple_events, Fmac_ae_set_reply_parameter)
+       (Fmac_resume_apple_event): New defuns.
+       (syms_of_macselect): Defsubr them.
+
+       * macterm.c (fn_keycode_to_keycode_table, XTread_socket) [MAC_OSX]:
+       Fix last change.  Don't map `fn' modifier if pressed with F1 ... F12.
+
+       * macterm.h (TYPE_FILE_NAME): Change from macro to enumerator.
+       (KEY_EMACS_SUSPENSION_ID_ATTR): New enumerator.
+       (keyReplyRequestedAttr) [MAC_OS_X_VERSION_MAX_ALLOWED < 1030]:
Likewise.+       (gestaltSystemVersionMajor,
gestaltSystemVersionMinor)
+       (gestaltSystemVersionBugFix) [MAC_OS_X_VERSION_MAX_ALLOWED < 1040]:
+       Likewise.
+       (typeUTF8Text, kEventParamWindowMouseLocation)
+       [MAC_OSX && MAC_OS_X_VERSION_MAX_ALLOWED < 1020]: Likewise.
+       (x_get_focus_frame, mac_ae_put_lisp): Add externs.
+
 2006-05-23  YAMAMOTO Mitsuharu  <mituharu@math.s.chiba-u.ac.jp>

        * macterm.c (fn_keycode_to_xkeysym_table, convert_fn_keycode): Remove.
@@ -18,6 +76,9 @@

 2006-05-21  Richard Stallman  <rms@gnu.org>

+       * xfaces.c (best_matching_font): Abort for best == NULL
+       before we start to use it.
+
        * buffer.c (syms_of_buffer, Fmake_overlay): Doc fixes.

 2006-05-20  Kim F. Storm  <storm@cua.dk>

Re: Problem report: No syntax highlight for C

[Jie Zhang]

On 6/3/06, Alan Mackenzie <acm@muc.de> wrote:
> On Sat, Jun 03, 2006 at 10:59:01AM +0300, Eli Zaretskii wrote:
> > [Please don't take the discussion off the list]
>
> Yes, please don't!  Otherwise, I can't see it, and can't reply to it.
> :-(
>
Oops! This was not what I meant to do. It must be an accident.

> > > Date: Sat, 3 Jun 2006 08:59:41 +0800
> > > From: "Jie Zhang" <jzhang918@gmail.com>
> > > (emacs t.c)
> > > For information about the GNU Project and its goals, type C-h C-p.
> > > t.c has auto save data; consider M-x recover-this-file
> > > Loading cc-mode...done
> > > File mode specification error: (void-variable c-font-lock-keywords-3)
>
> > > The last line seems pointing out the cause of this issue.
>
> > Yes, it does.  But we have to find out what causes that error.
>
> This is (probably) the problem with eval-after-load that I was trying to
> fix with my recent amendments.  The following form in cc-mode.el:
>
>     (eval-after-load "font-lock"
>       '(require 'cc-fonts))
>
> fails to trigger if Emacs's directory (e.g. ~acm/emacs/emacs/) is
> actually a symbolic link to another directory.
>
> Jie, is this the way your Emacs directory is set up?  Are you actually

No.

> running the Emacs CVS?  (Try M-x emacs-version.  The CVS will tell you
> something like "GNU Emacs 22.0.50 ...." whereas the released version will
> be "GNU Emacs 21.4.1 .....".
>
Yes.

> As a workaround, please try putting the following into your .emacs:
>
> (require 'cc-mode)
> (require 'cc-fonts)
>
> Does this make the problem go away?
>
Yes.

Jie

Re: Problem report: No syntax highlight for C

[Eli Zaretskii]

> Date: Sat, 3 Jun 2006 19:59:10 +0800
> From: "Jie Zhang" <jzhang918@gmail.com>
> Cc: emacs-devel@gnu.org, acm@muc.de, mituharu@math.s.chiba-u.ac.jp, rms@gnu.org
> 
> > > > Is this in "emacs -Q"?  If so, I cannot reproduce this.  Do you have
> > > > any messages in the *Messages* buffer that could shed a light on the
> > > > reason(s)?
> > > >
> > > I'm rather new to emacs. I don't know what exactly "emacs -Q" meas.
> > > Does it means passing -Q to emacs while invoking it?
> >
> > Yes, it means invoke Emacs with the -Q command-line switch.

You didn't say if the problem happens with -Q.  Please do.

> > If using -Q doesn't make this problem go away, please tell how up to
> > date is your Emacs code.  If you checked out the CVS tree yourself,
> > please tell at what date you did that.  If you got the sources from
> > someone else, please tell what is the first entry that you see in the
> > file lisp/ChangeLog.

You didn't reply to this.  Please do.

> I have narrowed down the changes which cause this issue. The CVS on
> 20060524 has no such issue, but 20060525 does. There are two and half
> entries which were added between these two dates. I add their owners
> to the CC list. Hope this help.

This doesn't explain why I don't see the problem on my machine, with
today's CVS.  Do you have this problem with every .c file, or just
with some?  If the latter, please post the smallest .c file which
causes the problem.

TIA

Re: Problem report: No syntax highlight for C

[Jie Zhang]

On 6/3/06, Eli Zaretskii <eliz@gnu.org> wrote:
> > Date: Sat, 3 Jun 2006 19:59:10 +0800
> > From: "Jie Zhang" <jzhang918@gmail.com>
> > Cc: emacs-devel@gnu.org, acm@muc.de, mituharu@math.s.chiba-u.ac.jp, rms@gnu.org
> >
> > > > > Is this in "emacs -Q"?  If so, I cannot reproduce this.  Do you have
> > > > > any messages in the *Messages* buffer that could shed a light on the
> > > > > reason(s)?
> > > > >
> > > > I'm rather new to emacs. I don't know what exactly "emacs -Q" meas.
> > > > Does it means passing -Q to emacs while invoking it?
> > >
> > > Yes, it means invoke Emacs with the -Q command-line switch.
>
> You didn't say if the problem happens with -Q.  Please do.
>
It happens with -Q.

> > > If using -Q doesn't make this problem go away, please tell how up to
> > > date is your Emacs code.  If you checked out the CVS tree yourself,
> > > please tell at what date you did that.  If you got the sources from
> > > someone else, please tell what is the first entry that you see in the
> > > file lisp/ChangeLog.
>
> You didn't reply to this.  Please do.
>
I'm using the CVS code checked out today. Actually, this problem
exists for me since the code checked out using -D20060525 or later.
The first entry in lisp/ChangeLog in my code is:

2006-06-03  Vivek Dasmohapatra  <vivek@etla.org>

        * progmodes/sh-script.el (sh-quoted-exec): New face for quoted
        exec constructs like `foo bar`.
        (sh-quoted-subshell): New helper function to search for a possibly
        nested subshell (like `` or $()) within a "" quoted string.
        (sh-font-lock-keywords-var): Add sh-quoted-exec for Bash.
        (sh-apply-quoted-subshell): Flag quote characters inside a
        subshell, which is itself already in a quoted region, as
        punctuation, since this is the closest to what they actually are.
        (sh-font-lock-syntactic-keywords): Add sh-quoted-subshell and
        sh-apply-quoted-subshell.
        (sh-font-lock-syntactic-face-function): Apply the new face for
        text inside `` instead of the old font-lock-string-face.

> > I have narrowed down the changes which cause this issue. The CVS on
> > 20060524 has no such issue, but 20060525 does. There are two and half
> > entries which were added between these two dates. I add their owners
> > to the CC list. Hope this help.
>
> This doesn't explain why I don't see the problem on my machine, with
> today's CVS.  Do you have this problem with every .c file, or just
> with some?  If the latter, please post the smallest .c file which
> causes the problem.
>
I have this problem in every .c file I opens. I'm using Ubuntu Dapper,
which is upgraded everyday. I checked out emacs from
":pserver:anonymous@cvs.savannah.gnu.org:/sources/emacs".

Jie

Re: Problem report: No syntax highlight for C

[Eli Zaretskii]

> Date: Sun, 4 Jun 2006 01:30:31 +0800
> From: "Jie Zhang" <jzhang918@gmail.com>
> Cc: emacs-devel@gnu.org, acm@muc.de, mituharu@math.s.chiba-u.ac.jp
> 
> I have this problem in every .c file I opens. I'm using Ubuntu Dapper,
> which is upgraded everyday. I checked out emacs from
> ":pserver:anonymous@cvs.savannah.gnu.org:/sources/emacs".

Then maybe the difference is with how you configured and built Emacs.
Could you please tell what commands you used to configure and build,
and which options, if any, you used with those commands?  The
``usual'' build is to type "configure ; make bootstrap"; anything else
is ``unusual'', so please describe that.

TIA

Re: Problem report: No syntax highlight for C

[Jie Zhang]

On 6/4/06, Eli Zaretskii <eliz@gnu.org> wrote:
> > Date: Sun, 4 Jun 2006 01:30:31 +0800
> > From: "Jie Zhang" <jzhang918@gmail.com>
> > Cc: emacs-devel@gnu.org, acm@muc.de, mituharu@math.s.chiba-u.ac.jp
> >
> > I have this problem in every .c file I opens. I'm using Ubuntu Dapper,
> > which is upgraded everyday. I checked out emacs from
> > ":pserver:anonymous@cvs.savannah.gnu.org:/sources/emacs".
>
> Then maybe the difference is with how you configured and built Emacs.
> Could you please tell what commands you used to configure and build,
> and which options, if any, you used with those commands?  The
> ``usual'' build is to type "configure ; make bootstrap"; anything else
> is ``unusual'', so please describe that.
>
I configured emacs using:

$ ../emacs/configure --prefix=/home/jie/installs/emacs --with-gtk

then do

$ make; make install

After reading your email, I tried "make bootstrap; make install" and
found then the highlighting works for C mode. Then I tried "make; make
install" again, however highlighting still works. Even "rm -rf
/home/jie/installs/emacs; make; make install" works now. It's rather
strange.

Thanks,
Jie

Re: Problem report: No syntax highlight for C

[Eli Zaretskii]

> Date: Sun, 4 Jun 2006 23:04:50 +0800
> From: "Jie Zhang" <jzhang918@gmail.com>
> Cc: emacs-devel@gnu.org, acm@muc.de, mituharu@math.s.chiba-u.ac.jp
> 
> After reading your email, I tried "make bootstrap; make install" and
> found then the highlighting works for C mode. Then I tried "make; make
> install" again, however highlighting still works. Even "rm -rf
> /home/jie/installs/emacs; make; make install" works now. It's rather
> strange.

I'm guessing that you had some stale *.elc files in your tree, and
"make bootstrap" cured that, because it removes all *.elc files and
recompiles them anew.

Re: Problem report: No syntax highlight for C

[Jie Zhang]

On 6/5/06, Eli Zaretskii <eliz@gnu.org> wrote:
> > Date: Sun, 4 Jun 2006 23:04:50 +0800
> > From: "Jie Zhang" <jzhang918@gmail.com>
> > Cc: emacs-devel@gnu.org, acm@muc.de, mituharu@math.s.chiba-u.ac.jp
> >
> > After reading your email, I tried "make bootstrap; make install" and
> > found then the highlighting works for C mode. Then I tried "make; make
> > install" again, however highlighting still works. Even "rm -rf
> > /home/jie/installs/emacs; make; make install" works now. It's rather
> > strange.
>
> I'm guessing that you had some stale *.elc files in your tree, and
> "make bootstrap" cured that, because it removes all *.elc files and
> recompiles them anew.
>
Is it possible that the *.elc in my previous installed emacs caused
this issue? I found that I was prompted to "make bootstrap" when there
is no emacs installed. But I don't have to do "make boostrap" when
there is an emacs in stalled. How about to make "make bootstrap" the
default when doing "make"?

Jie

Re: Problem report: No syntax highlight for C

[Eli Zaretskii]

> Date: Mon, 5 Jun 2006 12:44:41 +0800
> From: "Jie Zhang" <jzhang918@gmail.com>
> Cc: emacs-devel@gnu.org, acm@muc.de, mituharu@math.s.chiba-u.ac.jp
> 
> > I'm guessing that you had some stale *.elc files in your tree, and
> > "make bootstrap" cured that, because it removes all *.elc files and
> > recompiles them anew.
> >
> Is it possible that the *.elc in my previous installed emacs caused
> this issue?

Yes, that's what I tried to explain above.

> But I don't have to do "make boostrap" when there is an emacs in
> stalled.

Sometimes, you need to do that, especially when some unexpected and
strange problems happen.  The build procedure is not smart enough to
detect all of those situations.

> How about to make "make bootstrap" the default when doing "make"?

That would not be very wise: bootstrap takes much more time, and is
most of the time unnecessary if one resync's with the CVS frequently
enough.  See INSTALL.CVS for more about this.  If bootstrap becomes
the default, many users will be unnecessarily punished.

Re: Problem report: No syntax highlight for C

[Jie Zhang]

On 6/6/06, Eli Zaretskii <eliz@gnu.org> wrote:
> > Date: Mon, 5 Jun 2006 12:44:41 +0800
> > From: "Jie Zhang" <jzhang918@gmail.com>
> > Cc: emacs-devel@gnu.org, acm@muc.de, mituharu@math.s.chiba-u.ac.jp
> >
> > > I'm guessing that you had some stale *.elc files in your tree, and
> > > "make bootstrap" cured that, because it removes all *.elc files and
> > > recompiles them anew.
> > >
> > Is it possible that the *.elc in my previous installed emacs caused
> > this issue?
>
> Yes, that's what I tried to explain above.
>
> > But I don't have to do "make boostrap" when there is an emacs in
> > stalled.
>
> Sometimes, you need to do that, especially when some unexpected and
> strange problems happen.  The build procedure is not smart enough to
> detect all of those situations.
>
> > How about to make "make bootstrap" the default when doing "make"?
>
> That would not be very wise: bootstrap takes much more time, and is
> most of the time unnecessary if one resync's with the CVS frequently
> enough.  See INSTALL.CVS for more about this.  If bootstrap becomes
> the default, many users will be unnecessarily punished.
>
Thanks for helping me solve this issue and explain this to me!

Jie