M-x compile re-uses name of old compilation buffer, contrary to documentation

M-x compile re-uses name of old compilation buffer, contrary to documentation

[Edward Welbourne]

I opened the top-level directory of a source tree, to make it current directory.
M-x compile
added cvs up; to start of make command-line, hit return
Killed directory buffer.  Switched to *compilation* and did
M-x rename-buffer
Supplied name *myprog update build*, left it to finish building.
After it'd finished I fixed a compile error and, with cursor in *myprog update build*, did
M-x compile
removed the cvs up; from the command-line, hit return.

The new compilation re-used *myprog update build* as its buffer,
instead of creating a new buffer called *compilation* as clearly
explained in the documentation for M-x compile, specifically when
explaining how to avoid conflict between compilations (albeit ones
running in parallel).

The problem with this is that I actually wanted to keep the ouptut
from the original build - it recorded what files I've got locally
changed, and contained compiler warnings that I wanted to fix !  I now
have to do another cvs up to recover the list of local changes; and
I'll need to make clean and make in order to recover the list of
warnings.

compilation-buffer-name-function is nil

This is probably just a documentation bug.
I can see the given behaviour might even be intended, it's just not
what I'm used to, so it caught me out.


In GNU Emacs 22.1.1 (i486-pc-linux-gnu, X toolkit, Xaw3d scroll bars)
 of 2007-07-07 on raven, modified by Debian
Windowing system distributor `The X.Org Foundation', version 11.0.10300000
configured using `configure  '--build=i486-linux-gnu' '--host=i486-linux-gnu' '--prefix=/usr' '--sharedstatedir=/var/lib' '--libexecdir=/usr/lib' '--localstatedir=/var/lib' '--infodir=/usr/share/info' '--mandir=/usr/share/man' '--with-pop=yes' '--enable-locallisppath=/etc/emacs22:/etc/emacs:/usr/local/share/emacs/22.1/site-lisp:/usr/local/share/emacs/site-lisp:/usr/share/emacs/22.1/site-lisp:/usr/share/emacs/site-lisp:/usr/share/emacs/22.1/leim' '--with-x=yes' '--with-x-toolkit=athena' '--with-toolkit-scroll-bars' 'build_alias=i486-linux-gnu' 'host_alias=i486-linux-gnu' 'CFLAGS=-DDEBIAN -g -O2''

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: en_GB.ISO-8859-15
  locale-coding-system: iso-8859-15
  default-enable-multibyte-characters: t

Major mode: Lisp Interaction

Minor modes in effect:
  tooltip-mode: t
  mouse-wheel-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  unify-8859-on-encoding-mode: t
  utf-translate-cjk-mode: t
  auto-compression-mode: t
  line-number-mode: t

Recent input:
C-e C-p M-b M-b M-b M-b C-SPC M-f M-f M-f M-f M-w C-x 
o C-x o C-h C-g C-x C-x M-w C-h v C-y <return> q C-x 
3 C-x k <return> M-x c o m p i l e <return> C-a M-f 
M-f M-f M-f M-f M-b <M-backspace> <M-backspace> <M-backspace> 
<M-backspace> <return> M-< C-x o q C-x k <return> C-x 
b * m a i <tab> <return> M-{ M-{ M-{ C-n M-> M-{ C-x 
C-x C-w C-b M-b a g a i n SPC C-n <return> a l s o 
: SPC w h a t SPC i s SPC t h e SPC b e n f <backspace> 
e f i t SPC f SPC <backspace> <backspace> o f SPC a 
d d i n g SPC a n SPC e l s e SPC t o SPC t h e s e 
SPC f u n c t i o n s SPC ? C-n M-> C-c C-c C-x o C-n 
C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n 
C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n 
C-n C-n C-n C-n C-n C-n C-n C-n C-n C-n C-p C-p C-p 
C-p C-p C-p C-p C-p C-x C-f a d j <tab> <return> M-! 
c v s SPC u p SPC - r SPC p e r e g r i n e <return> 
C-x k <return> C-x C-f <backspace> <return> M-x c o 
m p i l e <return> <return> C-x k <return> C-x o M-x 
M-p <return> <return> C-x o C-x k <return> C-x o C-x 
o M-x e m a <tab> b u <tab> <M-backspace> r e p <tab> 
<M-backspace> <M-backspace> b u <tab> g <tab> <M-backspace> 
r e p <tab> o <tab> r <tab> <return>

Recent messages:
(No changes need to be saved)
Mark set [2 times]
Sending...done
? quick/setup/munge
P m2/src/import/EudoraImporter.cpp
M quick/quick-tweaks.h
M quick/windows/DocumentDesktopWindow.cpp
(No files need saving) [2 times]
Making completion list... [3 times]
Loading emacsbug...done

Re: M-x compile re-uses name of old compilation buffer, contrary to documentation

[Richard Stallman]

I will clarify the doc.  Thanks.

Re: M-x compile re-uses name of old compilation buffer, contrary to documentation

[Edward Welbourne]

> I will clarify the doc.

OK.  I'm warming to the new functionality, now that I've been using it
for a few days :-)

I also notice that emcas 22's HTML-mode (the one provided by
sgml-mode) is treating /* and */ as comment-delimiters - at least as
far as font-lock-mode is concerned - possibly in support of comments
in embedded CSS style sheets, but maybe a bug in some mode-related
code borrowing off C by mistake.  Very tiresome in documents that talk
about files, e.g. "see source/*.mk for details."

And, on a wildly separate topic, I owe you some updates to the GNU
make manual: see
  http://www.chaos.org.uk/~eddy/craft/make.html
for a talk in which I tried out my changes on the Bergen LUG, to see
if (a) anyone could spot any mistakes I was making and (b) whether it
was intelligible.  Even documentation deserves testing ...
Main obstacle to supplying a patch (aside from the usual scarcity of
time) is that I'm not info-literate, although I can probably remember
enough TeX to work it out,

	Eddy.

Re: M-x compile re-uses name of old compilation buffer, contrary to documentation

[martin rudalics]

> I also notice that emcas 22's HTML-mode (the one provided by
> sgml-mode) is treating /* and */ as comment-delimiters - at least as
> far as font-lock-mode is concerned - possibly in support of comments
> in embedded CSS style sheets, but maybe a bug in some mode-related
> code borrowing off C by mistake.  Very tiresome in documents that talk
> about files, e.g. "see source/*.mk for details."

Could you please give a recipe to reproduce this with emacs -Q?

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

>> I also notice that emcas 22's HTML-mode (the one provided by
>> sgml-mode) is treating /* and */ as comment-delimiters
> Could you please give a recipe to reproduce this with emacs -Q?

I'm guessing you mean -q but can't seem to reproduce it with a new
session even without it.  The running session from which I'm sending
this e-mail still exhibits the problem, but a fresh session doesn't.

Strange.
But that probably means you can ignore this problem !

	Eddy.

Re: C-comments in sgml-mode's html-mode

[Lennart Borgman (gmail)]

Edward Welbourne wrote:
>>> I also notice that emcas 22's HTML-mode (the one provided by
>>> sgml-mode) is treating /* and */ as comment-delimiters
>> Could you please give a recipe to reproduce this with emacs -Q?
> 
> I'm guessing you mean -q but can't seem to reproduce it with a new
> session even without it.  The running session from which I'm sending
> this e-mail still exhibits the problem, but a fresh session doesn't.
> 
> Strange.
> But that probably means you can ignore this problem !


Could you please use "Help - Send bug report" from the menus in that 
running session? This will give more information to us.

PS: -Q is new in Emacs 22.

Re: C-comments in sgml-mode's html-mode

[martin rudalics]

 >>>I also notice that emcas 22's HTML-mode (the one provided by
 >>>sgml-mode) is treating /* and */ as comment-delimiters
 >>
 >>Could you please give a recipe to reproduce this with emacs -Q?
 >
 >
 > I'm guessing you mean -q but can't seem to reproduce it with a new
 > session even without it.  The running session from which I'm sending
 > this e-mail still exhibits the problem, but a fresh session doesn't.
 >
 > Strange.
 > But that probably means you can ignore this problem !

Could you please, in the running session, move the cursor before
instances of the involved characters (in all four positions), do M-x
describe-char, and post the results here.  In addition, please tell us
the values (with C-h v) for comment-start and comment-end in the
affected buffer.

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> PS: -Q is new in Emacs 22.

but absent from its info documentation !

> Could you please use "Help - Send bug report" from the menus in that 
> running session?

hmm ... yes, I can turn menus back on ... ah, it's interactive and
wants to nuke this *mail* buffer ... worked round by renaming.  This
appears to be just like using M-x report-emacs-bug, like I did when
originally submitting this - but here it is anyway.  Point was in an
aflicted HTML buffer when I invoked the menu.  My .emacs executes:

(autoload 'html-mode "sgml-mode" "Edit HTML docs" t)
(setq auto-mode-alist (cons (cons "\\.x?html?$" 'html-mode) auto-mode-alist))
(global-font-lock-mode 1)
(setq font-lock-maximum-decoration t
      font-lock-support-mode 'lazy-lock-mode)

among other things.

A .html file containing <file>

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
<html lang="en"><head><title>Test page for HTML-mode /* comment issue </title><!-- */ sanity
--></head><body><h1>Test page for HTML-mode /* comment issue </h1><!-- */ sanity --></body></html>

</file> displays each text from /* to */ in comment-face; and each
"sanity" as plain text.  However, syntactic movement commands such as
C-M-f correctly understand the SGML comments as comments, unconfused
by the /* ... */.

This only happens in my current session: I have started up a fresh
session and seen the file displayed perfectly sensibly (i.e. treating
/* ... */ as ordinary text).


In GNU Emacs 22.1.1 (i486-pc-linux-gnu, X toolkit, Xaw3d scroll bars)
 of 2007-07-07 on raven, modified by Debian
Windowing system distributor `The X.Org Foundation', version 11.0.10300000
configured using `configure  '--build=i486-linux-gnu' '--host=i486-linux-gnu' '--prefix=/usr' '--sharedstatedir=/var/lib' '--libexecdir=/usr/lib' '--localstatedir=/var/lib' '--infodir=/usr/share/info' '--mandir=/usr/share/man' '--with-pop=yes' '--enable-locallisppath=/etc/emacs22:/etc/emacs:/usr/local/share/emacs/22.1/site-lisp:/usr/local/share/emacs/site-lisp:/usr/share/emacs/22.1/site-lisp:/usr/share/emacs/site-lisp:/usr/share/emacs/22.1/leim' '--with-x=yes' '--with-x-toolkit=athena' '--with-toolkit-scroll-bars' 'build_alias=i486-linux-gnu' 'host_alias=i486-linux-gnu' 'CFLAGS=-DDEBIAN -g -O2''

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: en_GB.ISO-8859-15
  locale-coding-system: iso-8859-15
  default-enable-multibyte-characters: t

Major mode: HTML

Minor modes in effect:
  tooltip-mode: t
  mouse-wheel-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  unify-8859-on-encoding-mode: t
  utf-translate-cjk-mode: t
  auto-compression-mode: t
  line-number-mode: t

Recent input:
. SPC SPC H o w e v e r , C-a <return> M-q <backspace> 
<return> C-e SPC s y n t a c t i c SPC m o v e m e 
n t SPC o <backspace> c o m m a n d s SPC ( C - M - 
M-b M-b <backspace> s u c h SPC a s SPC C-e f <return> 
c o r r e c t l y SPC u n d e r s t a n d SPC t h e 
SPC S G M L SPC c o m m e n t s SPC a s SPC c o m m 
e n t s C-x o C-p C-e SPC c o m <backspace> <backspace> 
<backspace> s a n i t y C-n C-M-b C-M-b C-b C-b C-b 
C-b SPC s a n i t y C-x o C-p M-f M-f C-p C-p C-p C-e 
SPC s a n i t y C-n M-b M-b M-b M-f C-f C-f C-f C-f 
C-f C-f C-f C-f SPC s a n i t y C-n C-n C-n C-p M-b 
M-b M-b M-b M-f ; SPC a n d SPC e a c h SPC u s e SPC 
o f SPC t h e SPC w o r d <M-backspace> <M-backspace> 
<M-backspace> <M-backspace> " s a n i t y " SPC a s 
SPC p l a i n SPC t e x t M-q C-n C-e C-x o C-x C-s 
C-x o M-{ M-{ C-f C-p M-} M-} M-{ C-SPC M-} M-} C-x 
C-x M-w C-x k <return> C-x <return> C-x <return> C-g 
C-x k <return> <help-echo> <help-echo> <help-echo> 
<help-echo> <help-echo> <help-echo> <help-echo> <help-echo> 
<help-echo> <help-echo> <help-echo> C-x o <help-echo> 
<help-echo> <help-echo> <help-echo> <help-echo> <help-echo> 
<menu-bar> <help-menu> <report-emacs-bug>

Recent messages:
Auto-saving...done
Undo! [4 times]
undo-more: No further undo information
Undo! [3 times]
Redo!
Undo! [2 times]
Redo! [2 times]
Auto-saving...done
Wrote /disk/home/eddy/work/tmp/comment.html
Mark set
Quit

Re: C-comments in sgml-mode's html-mode

[martin rudalics]

 >>PS: -Q is new in Emacs 22.
 >
 >
 > but absent from its info documentation !
 >

 From the Emacs manual:

`-Q'
`--quick'
      Start emacs with minimum customizations.  This is like using `-q'
      and `--no-site-file', but also disables the startup screen.

 > aflicted HTML buffer when I invoked the menu.  My .emacs executes:
 >
 > (autoload 'html-mode "sgml-mode" "Edit HTML docs" t)
 > (setq auto-mode-alist (cons (cons "\\.x?html?$" 'html-mode) auto-mode-alist))
 > (global-font-lock-mode 1)
 > (setq font-lock-maximum-decoration t
 >       font-lock-support-mode 'lazy-lock-mode)
 >
 > among other things.

On my system I have to require lazy-lock explicitly before doing

(setq font-lock-support-mode 'lazy-lock-mode)

However none of these explain the behavior you observe.  Please post the
results of describe-char as I asked you in my last mail.

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> Could you please, in the running session, move the cursor before
> instances of the involved characters (in all four positions), do M-x
> describe-char, and post the results here.  In addition, please tell us
> the values (with C-h v) for comment-start and comment-end in the
> affected buffer.

comment-start is a variable defined in `newcomment.el'.
Its value is "<!-- "
Local in buffer comment.html; global value is nil

This variable is safe as a file local variable if its value
satisfies the predicate `string-or-null-p'.

Documentation:
*String to insert to start a new comment, or nil if no comment syntax.


comment-end is a variable defined in `newcomment.el'.
Its value is " -->"
Local in buffer comment.html; global value is ""

This variable is safe as a file local variable if its value
satisfies the predicate `string-or-null-p'.

Documentation:
*String to insert to end a new comment.
Should be an empty string if comments are terminated by end-of-line.


Not sure what you mean by "in all four positions", but here are the
results for each of the four characters in " /* " in the title:

  character: SPC (32, #o40, #x20, U+0020)
    charset: ascii (ASCII (ISO646 IRV))
 code point: #x20
     syntax:   	which means: whitespace
   category: a:ASCII graphic characters 32-126 (ISO646 IRV:1983[4/0]) l:Latin
buffer code: #x20
  file code: #x20 (encoded by coding system undecided-unix)
    display: by this font (glyph code)
     -Misc-Fixed-Medium-R-Normal--8-80-75-75-C-50-ISO8859-1 (#x20)

There are text properties here:
  face                 (bold underline)
  fontified            t


  character: / (47, #o57, #x2f, U+002F)
    charset: ascii (ASCII (ISO646 IRV))
 code point: #x2F
     syntax: . 14	which means: punctuation,
	  is the first character of a comment-start sequence,
	  is the second character of a comment-end sequence
   category: a:ASCII graphic characters 32-126 (ISO646 IRV:1983[4/0]) l:Latin
buffer code: #x2F
  file code: #x2F (encoded by coding system undecided-unix)
    display: by this font (glyph code)
     -Misc-Fixed-Medium-R-Normal--8-80-75-75-C-50-ISO8859-1 (#x2F)

There are text properties here:
  face                 (bold underline font-lock-comment-face)
  fontified            t


  character: * (42, #o52, #x2a, U+002A)
    charset: ascii (ASCII (ISO646 IRV))
 code point: #x2A
     syntax: . 23	which means: punctuation,
	  is the second character of a comment-start sequence,
	  is the first character of a comment-end sequence
   category: a:ASCII graphic characters 32-126 (ISO646 IRV:1983[4/0]) l:Latin
buffer code: #x2A
  file code: #x2A (encoded by coding system undecided-unix)
    display: by this font (glyph code)
     -Misc-Fixed-Medium-R-Normal--8-80-75-75-C-50-ISO8859-1 (#x2A)

There are text properties here:
  face                 (bold underline font-lock-comment-face)
  fontified            t


  character: SPC (32, #o40, #x20, U+0020)
    charset: ascii (ASCII (ISO646 IRV))
 code point: #x20
     syntax:   	which means: whitespace
   category: a:ASCII graphic characters 32-126 (ISO646 IRV:1983[4/0]) l:Latin
buffer code: #x20
  file code: #x20 (encoded by coding system undecided-unix)
    display: by this font (glyph code)
     -Misc-Fixed-Medium-R-Normal--8-80-75-75-C-50-ISO8859-1 (#x20)

There are text properties here:
  face                 (bold underline font-lock-comment-face)
  fontified            t


Similarly, for " */ " in the comment after the title:


  character: SPC (32, #o40, #x20, U+0020)
    charset: ascii (ASCII (ISO646 IRV))
 code point: #x20
     syntax:   	which means: whitespace
   category: a:ASCII graphic characters 32-126 (ISO646 IRV:1983[4/0]) l:Latin
buffer code: #x20
  file code: #x20 (encoded by coding system undecided-unix)
    display: by this font (glyph code)
     -Misc-Fixed-Medium-R-Normal--8-80-75-75-C-50-ISO8859-1 (#x20)

There are text properties here:
  face                 font-lock-comment-face
  fontified            t


  character: * (42, #o52, #x2a, U+002A)
    charset: ascii (ASCII (ISO646 IRV))
 code point: #x2A
     syntax: . 23	which means: punctuation,
	  is the second character of a comment-start sequence,
	  is the first character of a comment-end sequence
   category: a:ASCII graphic characters 32-126 (ISO646 IRV:1983[4/0]) l:Latin
buffer code: #x2A
  file code: #x2A (encoded by coding system undecided-unix)
    display: by this font (glyph code)
     -Misc-Fixed-Medium-R-Normal--8-80-75-75-C-50-ISO8859-1 (#x2A)

There are text properties here:
  face                 font-lock-comment-face
  fontified            t


  character: / (47, #o57, #x2f, U+002F)
    charset: ascii (ASCII (ISO646 IRV))
 code point: #x2F
     syntax: . 14	which means: punctuation,
	  is the first character of a comment-start sequence,
	  is the second character of a comment-end sequence
   category: a:ASCII graphic characters 32-126 (ISO646 IRV:1983[4/0]) l:Latin
buffer code: #x2F
  file code: #x2F (encoded by coding system undecided-unix)
    display: by this font (glyph code)
     -Misc-Fixed-Medium-R-Normal--8-80-75-75-C-50-ISO8859-1 (#x2F)

There are text properties here:
  face                 font-lock-comment-face
  fontified            t


  character: SPC (32, #o40, #x20, U+0020)
    charset: ascii (ASCII (ISO646 IRV))
 code point: #x20
     syntax:   	which means: whitespace
   category: a:ASCII graphic characters 32-126 (ISO646 IRV:1983[4/0]) l:Latin
buffer code: #x20
  file code: #x20 (encoded by coding system undecided-unix)
    display: by this font (glyph code)
     -Misc-Fixed-Medium-R-Normal--8-80-75-75-C-50-ISO8859-1 (#x20)

There are text properties here:
  fontified            t


Is that what you wanted ?

	Eddy.

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> >>PS: -Q is new in Emacs 22.
> >
> >
> > but absent from its info documentation !
> >

>  From the Emacs manual:

> `-Q'
> `--quick'
>       Start emacs with minimum customizations.  This is like using `-q'
>       and `--no-site-file', but also disables the startup screen.

Not visible in the emacs manual I get when I C-h i d m emacs RET - but
a closer look through the C-h i d list shows me that I have two of
each Emacs manual, presumably because I've got Emacs 22 and Emacs 21
installed side-by-side.  Each resolves to the Emacs 21.3 version.
There might be some benefit to marking Emacs manuals with a version
number, in the info listing !

	Eddy.

Re: C-comments in sgml-mode's html-mode

[martin rudalics]

>   character: / (47, #o57, #x2f, U+002F)
>     charset: ascii (ASCII (ISO646 IRV))
>  code point: #x2F
>      syntax: . 14	which means: punctuation,
> 	  is the first character of a comment-start sequence,
> 	  is the second character of a comment-end sequence

Some css-mode I presume.

> There are text properties here:
>   face                 (bold underline font-lock-comment-face)
>   fontified            t

Looks like a weird mixture of html- and css-mode.  Are you using
a css-mode or html-helper-mode?

> Is that what you wanted ?

Yes, thank you very much.

BTW, I recommend you remove the

(global-font-lock-mode 1)
(setq font-lock-maximum-decoration t
       font-lock-support-mode 'lazy-lock-mode)

lines from your .emacs.  jit-lock-mode supersedes lazy-lock-mode for
some time already and the other settings are default.  These settings
are, however, not responsible for the behavior you describe.

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> Some css-mode I presume.

That was my guess, too (see initial report).

>> There are text properties here:
>>   face                 (bold underline font-lock-comment-face)
>>   fontified            t

> Looks like a weird mixture of html- and css-mode.  Are you using
> a css-mode or html-helper-mode?

However, C-h m gives me a buffer full of mode information in which
searching for css gets no hits, as does searching for helper.  It
starts with <quote>

Enabled minor modes: Auto-Compression Blink-Cursor File-Name-Shadow
Font-Lock Global-Font-Lock Line-Number Mouse-Wheel Tooltip
Unify-8859-On-Encoding Utf-Translate-Cjk

</quote> (and I've no idea what's putting in File-Name-Shadow, Tooltip
or the two encoding-related ones on the last line), introducing the
major mode as: <quote>

HTML mode:
Major mode based on SGML mode for editing HTML documents.

</quote>

	Eddy.

Re: C-comments in sgml-mode's html-mode

[martin rudalics]

 >>Some css-mode I presume.
 >
 >
 > That was my guess, too (see initial report).

I was aware of your guess.  By looking at the text properties I wanted
to exclude your guess about C.

 > However, C-h m gives me a buffer full of mode information in which
 > searching for css gets no hits, as does searching for helper.

C-h m won't tell about any remnants left by a mode in a buffer after the
mode has been turned off.  In your case someone has set up the syntax of
/ and * much as a css-mode would do.  What does C-h f css-mode say on
your system?

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> > However, C-h m gives me a buffer full of mode information in which
> > searching for css gets no hits, as does searching for helper.

> C-h m won't tell about any remnants left by a mode in a buffer after the
> mode has been turned off.

However, everything I've reported has been based on freshly-opened
buffers looking at the simple test file from my Help -> Report a bug
menu report.  Although there seems to be something odd in my *session*
the problem isn't due to anything odd in any particular *buffer*, or
its history.  Presumably a data structure associated with html-mode
has been mangled.

> In your case someone has set up the syntax of / and * much as a
> css-mode would do.  What does C-h f css-mode say on your system?

<quote src="C-h f css-mode RET">

css-mode is an interactive compiled Lisp function in `css-mode.el'.
(css-mode)

Major mode for editing CSS style sheets.
key             binding
---             -------

C-c		Prefix Command
ESC		Prefix Command
"		cssm-insert-quotes
(		cssm-insert-parenthesises
[		cssm-insert-brackets
{		cssm-insert-curlies
}		cssm-insert-right-brace-and-indent

M-TAB		cssm-complete-property

C-c C-c		cssm-insert-comment
C-c C-u		cssm-insert-url

</quote> and I don't believe I've used it directly ('though I may have
looked at a .css file at some point, which would cause it to load
automagically).  However, when I open up comment.html in a fresh clean
emacs -Q, it doesn't believe in css-mode at all.  I just get a bell
(and [No match] in minibuffer) when I try C-h f css-mode RET.  No
script in my usual .emacs (a minimal .emacs which adds an entry to
load-path and loads a bunch of little fragments from a directory of
mine) mentions css-mode.

In an emacs -q session, it knows it would autoload css-mode if asked
for: loading a css file, I now find I am able to reproduce the bug in
a new session :-)

So, steps to reproduce: in an emacs -q session evaluate <quote>
(autoload 'html-mode "sgml-mode" "Edit HTML docs" t)
(setq auto-mode-alist (cons (cons "\\.x?html?$" 'html-mode) auto-mode-alist))
</quote>, visit a .css file (it need not exist / can be empty),
open a .html file containing: <quote>
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
<html lang="en"><head><title>Test page for HTML-mode /* comment issue </title><!-- */ sanity
--></head><body><h1>Test page for HTML-mode /* comment issue </h1><!-- */ sanity --></body></html>
</quote>

It would appear that css-mode is hacking html-mode !

	Eddy.

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

Further to successful reproduction, I've now simplified reproducing
this even further - css-mode is even hacking GNUMakefile-mode !

run: emacs -q dummy.css dummy.mk
In dummy.mk's buffer, type <quote>
/* I look like a comment
but I don't
</quote> and you'll see it's true !

	Eddy.

Re: C-comments in sgml-mode's html-mode

[Lennart Borgman (gmail)]

Edward Welbourne wrote:
>> In your case someone has set up the syntax of / and * much as a
>> css-mode would do.  What does C-h f css-mode say on your system?
> 
> <quote src="C-h f css-mode RET">
> 
> css-mode is an interactive compiled Lisp function in `css-mode.el'.
> (css-mode)
> 
> Major mode for editing CSS style sheets.
> key             binding
> ---             -------
> 
> C-c		Prefix Command
> ESC		Prefix Command
> "		cssm-insert-quotes
> (		cssm-insert-parenthesises
> [		cssm-insert-brackets
> {		cssm-insert-curlies
> }		cssm-insert-right-brace-and-indent
> 
> M-TAB		cssm-complete-property
> 
> C-c C-c		cssm-insert-comment
> C-c C-u		cssm-insert-url
> 
> </quote>

If you put point in the help buffer at `css-mode.el' and press RET it 
will open the file css-mode.el. Could you please look at the beginning 
of that file and show us what it says (version, author etc)? If you in 
this file do

   C-h v buffer-file-name RET

you can see where that file is located.


> and I don't believe I've used it directly ('though I may have
> looked at a .css file at some point, which would cause it to load
> automagically).  However, when I open up comment.html in a fresh clean
> emacs -Q, it doesn't believe in css-mode at all.  I just get a bell
> (and [No match] in minibuffer) when I try C-h f css-mode RET.  No
> script in my usual .emacs (a minimal .emacs which adds an entry to
> load-path and loads a bunch of little fragments from a directory of
> mine) mentions css-mode.
> 
> In an emacs -q session, it knows it would autoload css-mode if asked
> for: loading a css file, I now find I am able to reproduce the bug in
> a new session :-)
> 
> So, steps to reproduce: in an emacs -q session evaluate <quote>
> (autoload 'html-mode "sgml-mode" "Edit HTML docs" t)
> (setq auto-mode-alist (cons (cons "\\.x?html?$" 'html-mode) auto-mode-alist))
> </quote>, visit a .css file (it need not exist / can be empty),
> open a .html file containing: <quote>
> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
> <html lang="en"><head><title>Test page for HTML-mode /* comment issue </title><!-- */ sanity
> --></head><body><h1>Test page for HTML-mode /* comment issue </h1><!-- */ sanity --></body></html>
> </quote>
> 
> It would appear that css-mode is hacking html-mode !


This does not happen for me when starting with emacs -q. Nor does it 
happen for you with -Q. Using -q site-start.el is run, but it is not run 
when using -Q. Can you please do

   M-x find-library RET site-start RET

and show us the file?

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> If you put point in the help buffer at `css-mode.el' and press RET it 
> will open the file css-mode.el. Could you please look at the beginning 
> of that file and show us what it says (version, author etc)?

(defvar cssm-version "0.11.zap1"
  "The current version number of css-mode.")
;;; copyright (c) 1998 Lars Marius Garshol, larsga@ifi.uio.no
;;; $Id: css-mode.el,v 1.9 2000/01/05 21:21:56 larsga Exp $
;;; Modified by John Zaitseff, 3rd November 2004.

and I can confirm <quote src="C-h v cssm-version RET">

cssm-version is a variable defined in `css-mode.el'.
Its value is "0.11.zap1"


Documentation:
The current version number of css-mode.

</quote> and this coincides with what I find early in
/usr/share/emacs22/site-lisp/css-mode/css-mode.elc
shortly after

;;; Compiled by root@opera.com on Wed Jul 18 10:39:53 2007
;;; from file /usr/share/emacs22/site-lisp/css-mode/css-mode.el
;;; in Emacs version 22.1.1
;;; with all optimizations.

which is odd, because the .el file it mentions doesn't exist.  In
load-path, "/usr/share/emacs22/site-lisp/css-mode" appears before
"/usr/share/emacs/site-lisp/css-mode".

> If you in this file do

>    C-h v buffer-file-name RET

> you can see where that file is located.

buffer-file-name is a variable defined in `C source code'.
Its value is 
"/usr/share/emacs/site-lisp/css-mode/css-mode.el"
Local in buffer css-mode.el; global value is nil
Automatically becomes buffer-local when set in any fashion.


Documentation:
Name of file visited in current buffer, or nil if not visiting a file.

> This does not happen for me when starting with emacs -q. Nor does it 
> happen for you with -Q.

confirmed, also for .mk; and also using --no-site-file instead of -q
or -Q (so my .emacs is innocent :-)

> Can you please do

>    M-x find-library RET site-start RET

> and show us the file?

<quote src="/etc/emacs/site-start.el">

;; Emacsen independent startup file.  All of the various installed
;; flavors of emacs (emacs 19, emacs 20, xemacs) will load this file
;; at startup.  Make sure any code you put here is emacs flavor
;; independent.

;; Package maintainers: do not have Debian packages edit this file.
;; See the policy manual for the proper way to handle Emacs package
;; initialization code.


</quote> containing *no* code.  The policy document referenced is
/usr/share/doc/emacsen-common/debian-emacs-policy.gz
and makes clear that this file is "the sole domain of the local system
admin" which would fit with it being empty.  I configure ~/.emacs
rather than the site ...

	Eddy.

Re: C-comments in sgml-mode's html-mode

[martin rudalics]

 > So, steps to reproduce: in an emacs -q session evaluate <quote>
 > (autoload 'html-mode "sgml-mode" "Edit HTML docs" t)
 > (setq auto-mode-alist (cons (cons "\\.x?html?$" 'html-mode) auto-mode-alist))
 > </quote>, visit a .css file (it need not exist / can be empty),
 > open a .html file containing: <quote>
 > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
 > <html lang="en"><head><title>Test page for HTML-mode /* comment issue </title><!-- */ sanity
 > --></head><body><h1>Test page for HTML-mode /* comment issue </h1><!-- */ sanity --></body></html>
 > </quote>
 >
 > It would appear that css-mode is hacking html-mode !

But this is not the css-mode that comes with Emacs 22.2.

Re: C-comments in sgml-mode's html-mode

[martin rudalics]

 > (defvar cssm-version "0.11.zap1"
 >   "The current version number of css-mode.")
 > ;;; copyright (c) 1998 Lars Marius Garshol, larsga@ifi.uio.no
 > ;;; $Id: css-mode.el,v 1.9 2000/01/05 21:21:56 larsga Exp $
 > ;;; Modified by John Zaitseff, 3rd November 2004.

css-mode by Garshol, modified by Zaitseff, and distributed by Debian.
Apparently, Zaitseff's modifications are too persistent but I can't tell
anything because I don't have the sources.  In any case this should be
reported to the Debian bug tracking system since probably more people
might be affected by this.

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> In any case this should be reported to the Debian bug tracking
> system since probably more people might be affected by this.

Debian bug submitted.
I'll follow up with bug number and URL once the robots answer.

> Apparently, Zaitseff's modifications are too persistent but I can't tell
> anything because I don't have the sources.

Source follows,

	Eddy.
-- 

;;;; A major mode for editing CSS.

;;; Adds font locking, some rather primitive indentation handling and
;;; some typing help.
;;;
(defvar cssm-version "0.11.zap1"
  "The current version number of css-mode.")
;;; copyright (c) 1998 Lars Marius Garshol, larsga@ifi.uio.no
;;; $Id: css-mode.el,v 1.9 2000/01/05 21:21:56 larsga Exp $
;;; Modified by John Zaitseff, 3rd November 2004.

;;; css-mode is free software; you can redistribute it and/or modify
;;; it under the terms of the GNU General Public License as published
;;; by the Free Software Foundation; either version 2, or (at your
;;; option) any later version.
;;;
;;; css-mode is distributed in the hope that it will be useful, but
;;; WITHOUT ANY WARRANTY; without even the implied warranty of
;;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
;;; General Public License for more details.
;;;
;;; You should have received a copy of the GNU General Public License
;;; along with GNU Emacs; see the file COPYING.  If not, write to the
;;; Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.

; Send me an email if you want new features (or if you add them yourself).
; I will do my best to preserve the API to functions not explicitly marked
; as internal and variables shown as customizable. I make no promises about
; the rest.

; Bug reports are very welcome. New versions of the package will appear at
; http://www.stud.ifi.uio.no/~larsga/download/css-mode.html
; You can register at the same address if you want to be notified when a
; new version appears.

; Thanks to Philippe Le Hegaret, Kjetil Kjernsmo, Alf-Ivar Holm and
; Alfred Correira for much useful feedback. Alf-Ivar Holm also contributed
; patches.

; To install, put this in your .emacs:
;
; (autoload 'css-mode "css-mode")
; (setq auto-mode-alist       
;      (cons '("\\.css\\'" . css-mode) auto-mode-alist))

;; Todo:

; - must not color URL file name extensions as class selectors (*.css)
; - color [] and url() constructs correctly, even if quoted strings present
; - disregard anything inside strings

;; Possible later additions:
;
; - forward/backward style/@media rule commands
; - more complete syntax table

;; Required modules

(require 'apropos)
(require 'font-lock)
(require 'cl)

;;; The code itself

; Customizable variables:

(defvar cssm-indent-level 2 "The indentation level inside rules.")
(defvar cssm-mirror-mode t
  "Whether brackets, quotes etc should be mirrored automatically on
  insertion.")
(defvar cssm-newline-before-closing-bracket nil
  "In mirror-mode, controls whether a newline should be inserted before the
closing bracket or not.")
(defvar cssm-indent-function #'cssm-old-style-indenter
  "Which function to use when deciding which column to indent to. To get
C-style indentation, use cssm-c-style-indenter. To get old-style indentation,
use cssm-old-style-indenter.")

; The rest of the code:

(defvar cssm-properties
  '("font-family" "font-style" "font-variant" "font-weight" "font-size"
    "font" "background-color" "background-image" "background-repeat"
    "background-attachment" "background-position" "background" "color"
    "word-spacing" "letter-spacing" "text-decoration" "vertical-align"
    "text-transform" "text-align" "text-indent" "line-height" "margin-top"
    "margin-right" "margin-bottom" "margin-left" "margin" "padding-top"
    "padding-right" "padding-bottom" "padding-left" "padding"
    "border-top-width" "border-right-width" "border-left-width"
    "border-bottom-width" "border-width" "border-color" "border-style"
    "border-top" "border-right" "border-bottom" "border-left" "border"
    "width" "height" "float" "clear" "display" "white-space"
    "list-style-type" "list-style-image" "list-style-position"
    "list-style"

    ; CSS level 2.1:

    "border-bottom-color" "border-bottom-style" "border-collapse"
    "border-left-color" "border-left-style" "border-right-color"
    "border-right-style" "border-spacing" "border-top-color"
    "border-top-style" "bottom" "caption-side" "clip" "content"
    "counter-increment" "counter-reset" "cursor" "direction" "empty-cells"
    "left" "max-height" "max-width" "min-height" "min-width" "orphans"
    "outline" "outline-color" "outline-style" "outline-width" "overflow"
    "page-break-after" "page-break-before" "page-break-inside" "position"
    "quotes" "right" "table-layout" "top" "unicode-bidi" "visibility"
    "widows" "z-index"

    ; CSS level 2 (dropped from 2.1):

    "azimuth" "cue" "cue-after" "cue-before" "elevation"
    "font-size-adjust" "font-stretch" "marker-offset" "marks" "page"
    "pause" "pause-after" "pause-before" "pitch" "pitch-range"
    "play-during" "richness" "size" "speak" "speak-header" "speak-numeral"
    "speak-punctuation" "speech-rate" "stress" "text-shadow"
    "voice-family" "volume"

    ; Non-standard properties

    "cell-spacing" "column-span" "row-span" "speak-date" "speak-time")
  "A list of all CSS properties.")

(defvar cssm-properties-alist
  (mapcar (lambda(prop)
	    (cons (concat prop ":") nil)) cssm-properties)
  "An association list of the CSS properties for completion use.")

(defvar cssm-keywords 
  (append '("!\\s-*important"
    
	  ; CSS level 2:

	    "@media" "@import" "@page" "@font-face")
	  (mapcar (lambda(property)
		    (concat property "\\s-*:"))
		  cssm-properties))
  "A list of all CSS keywords.")

(defvar cssm-pseudos
  '("link" "visited" "active" "first-line" "first-letter"

    ; CSS level 2
    "first-child" "before" "after" "hover" "focus" "lang")
  "A list of all CSS pseudo-classes.")

; internal
(defun cssm-list-2-regexp(altlist)
  "Takes a list and returns the regexp \\(elem1\\|elem2\\|...\\)"
  (let ((regexp "\\("))
    (mapcar (lambda(elem)
	      (setq regexp (concat regexp elem "\\|")))
	    altlist)
    (concat (substring regexp 0 -2) ; cutting the last "\\|"
	    "\\)")
    ))

(defvar cssm-font-lock-keywords
  (list
   (cons (cssm-list-2-regexp cssm-keywords) font-lock-keyword-face)
   (cons "\\.[a-zA-Z][-a-zA-Z0-9.]+" font-lock-variable-name-face)
   (cons (concat ":" (cssm-list-2-regexp cssm-pseudos))
	 font-lock-variable-name-face)
   (cons "#[a-fA-F0-9][a-fA-F0-9][a-fA-F0-9]\\([a-fA-F0-9][a-fA-F0-9][a-fA-F0-9]\\)?"
	 font-lock-reference-face)
   (cons "\\[.*\\]" font-lock-variable-name-face)
   (cons "#[-a-zA-Z0-9]*" font-lock-function-name-face)
   (cons "rgb(\\s-*[0-9]+\\(\\.[0-9]+\\s-*%\\s-*\\)?\\s-*,\\s-*[0-9]+\\(\\.[0-9]+\\s-*%\\s-*\\)?\\s-*,\\s-*[0-9]+\\(\\.[0-9]+\\s-*%\\s-*\\)?\\s-*)"
	 font-lock-reference-face)
   )
  "Rules for highlighting CSS style sheets.")

(defvar cssm-mode-map ()
  "Keymap used in CSS mode.")
(when (not cssm-mode-map)
  (setq cssm-mode-map (make-sparse-keymap))
  (define-key cssm-mode-map (read-kbd-macro "C-c C-c") 'cssm-insert-comment)
  (define-key cssm-mode-map (read-kbd-macro "C-c C-u") 'cssm-insert-url)
  (define-key cssm-mode-map (read-kbd-macro "}") 'cssm-insert-right-brace-and-indent)
  (define-key cssm-mode-map (read-kbd-macro "M-TAB") 'cssm-complete-property))

;;; Cross-version compatibility layer

(when (not (or (apropos-macrop 'kbd)
	     (fboundp 'kbd)))
    (defmacro kbd (keys)
      "Convert KEYS to the internal Emacs key representation.
KEYS should be a string constant in the format used for
saving keyboard macros (see `insert-kbd-macro')."
      (read-kbd-macro keys)))

;;; Auto-indentation support

; internal
(defun cssm-insert-right-brace-and-indent()
  (interactive)
  (insert "}")
  (cssm-indent-line))

; internal
(defun cssm-inside-atmedia-rule()
  "Decides if point is currently inside an @media rule."
  (let ((orig-pos (point))
	(atmedia (re-search-backward "@media" 0 t))
	(balance 1)   ; used to keep the {} balance, 1 because we start on a {
	)
     ; Going to the accompanying {
    (re-search-forward "{" (point-max) t)
    (if (null atmedia)
	nil  ; no @media before this point => not inside
      (while (and (< (point) orig-pos)
		  (< 0 balance))
	(if (null (re-search-forward "[{}]" (point-max) 0))
	    (goto-char (point-max)) ; break
	  (setq balance
		(if (string= (match-string 0) "{")
		    (+ balance 1)
		  (- balance 1)))))
      (= balance 1))
    ))

; internal
(defun cssm-rule-is-atmedia()
  "Decides if point is currently on the { of an @media or ordinary style rule."
  (let ((result (re-search-backward "[@}{]" 0 t)))
    (if (null result)
	nil
      (string= (match-string 0) "@"))))

; internal
(defun cssm-find-column(first-char)
  "Find which column to indent to." 

  ; Find out where to indent to by looking at previous lines
  ; spinning backwards over comments
  (let (pos)
    (while (and (setq pos (re-search-backward (cssm-list-2-regexp
					       '("/\\*" "\\*/" "{" "}"))
					      (point-min) t))
		(string= (match-string 0) "*/"))
      (search-backward "/*" (point-min) t))

    ; did the last search find anything?
    (if pos
	(save-excursion
	  (let ((construct      (match-string 0))
		(column         (current-column)))
	    (apply cssm-indent-function
		   (list (cond
			  ((string= construct "{")
			   (cond
			    ((cssm-rule-is-atmedia)
			     'inside-atmedia)
			    ((cssm-inside-atmedia-rule)
			     'inside-rule-and-atmedia)
			    (t
			     'inside-rule)))
			  ((string= construct "/*")
			   'inside-comment)
			  ((string= construct "}")
			   (if (cssm-inside-atmedia-rule)
			       'inside-atmedia
			     'outside))
			  (t 'outside))
			 column
			 first-char))))
      
      (apply cssm-indent-function
	     (list 'outside
		   (current-column)
		   first-char)))))

(defun cssm-indent-line()
  "Indents the current line."
  (interactive)
  (beginning-of-line)
  (let* ((beg-of-line (point))
	 (pos (re-search-forward "[]@#a-zA-Z0-9;,.\"{}/*\n:[]" (point-max) t))
	 (first (match-string 0))
	 (start (match-beginning 0)))

    (goto-char beg-of-line)

    (let ((indent-column (cssm-find-column first)))
      (goto-char beg-of-line)

      ; Remove all leading whitespace on this line (
      (if (not (or (null pos)
		   (= beg-of-line start)))
	  (kill-region beg-of-line start))

      (goto-char beg-of-line)
    
      ; Indent
      (while (< 0 indent-column)
	(insert " ")
	(setq indent-column (- indent-column 1))))))

;;; Indent-style functions

(defun cssm-old-style-indenter(position column first-char-on-line)
  (cond
   ((eq position 'inside-atmedia)
    (if (string= "}" first-char-on-line)
	0
      cssm-indent-level))
   
   ((eq position 'inside-rule)
    (+ column 2))

   ((eq position 'inside-rule-and-atmedia)
    (+ column 2))

   ((eq position 'inside-comment)
    (+ column 3))

   ((eq position 'outside)
    0)))

(defun cssm-c-style-indenter(position column first-char-on-line)
  (cond
   ((or (eq position 'inside-atmedia)
	(eq position 'inside-rule))
    (if (string= "}" first-char-on-line)
	0
      cssm-indent-level))

   ((eq position 'inside-rule-and-atmedia)
    (if (string= "}" first-char-on-line)
	cssm-indent-level
      (* 2 cssm-indent-level)))

   ((eq position 'inside-comment)
    (+ column 3))

   ((eq position 'outside)
    0)))

;;; Typing shortcuts

(define-skeleton cssm-insert-curlies
  "Inserts a pair of matching curly parenthesises." nil
  "{"
  (if cssm-newline-before-closing-bracket "\n" " ")
  (if cssm-newline-before-closing-bracket '>)
  _
  (if cssm-newline-before-closing-bracket "\n" " ")
  "}"
  (if cssm-newline-before-closing-bracket '>))

(define-skeleton cssm-insert-quotes
  "Inserts a pair of matching quotes." nil
  "\"" _ "\"")

(define-skeleton cssm-insert-parenthesises
  "Inserts a pair of matching parenthesises." nil
  "(" _ ")")

(define-skeleton cssm-insert-comment
  "Inserts a full comment." nil
  "/* " _ " */")

(define-skeleton cssm-insert-url
  "Inserts a URL." nil
  "url(" _ ")")

(define-skeleton cssm-insert-brackets
  "Inserts a pair of matching brackets." nil
  "[" _ "]")

(defun cssm-enter-mirror-mode()
  "Turns on mirror mode, where quotes, brackets etc are mirrored automatically
  on insertion."
  (interactive)
  (define-key cssm-mode-map (read-kbd-macro "{")  'cssm-insert-curlies)
  (define-key cssm-mode-map (read-kbd-macro "\"") 'cssm-insert-quotes)
  (define-key cssm-mode-map (read-kbd-macro "(")  'cssm-insert-parenthesises)
  (define-key cssm-mode-map (read-kbd-macro "[")  'cssm-insert-brackets))

(defun cssm-leave-mirror-mode()
  "Turns off mirror mode."
  (interactive)
  (define-key cssm-mode-map (read-kbd-macro "{")  'self-insert-command)
  (define-key cssm-mode-map (read-kbd-macro "\"") 'self-insert-command)
  (define-key cssm-mode-map (read-kbd-macro "(")  'self-insert-command)
  (define-key cssm-mode-map (read-kbd-macro "[")  'self-insert-command))

;;; Property completion

(defun cssm-property-at-point()
  "If point is at the end of a property name: returns it."
  (let ((end (point))
	(start (+ (re-search-backward "[^-A-Za-z]") 1)))
    (goto-char end)
    (buffer-substring start end)))

; internal
(defun cssm-maximum-common(alt1 alt2)
  "Returns the maximum common starting substring of alt1 and alt2."
  (let* ((maxlen (min (length alt1) (length alt2)))
	 (alt1 (substring alt1 0 maxlen))
	 (alt2 (substring alt2 0 maxlen)))
    (while (not (string= (substring alt1 0 maxlen)
			 (substring alt2 0 maxlen)))
      (setq maxlen (- maxlen 1)))
    (substring alt1 0 maxlen)))

; internal
(defun cssm-common-beginning(alts)
  "Returns the maximum common starting substring of all alts elements."
  (let ((common (car alts)))
    (dolist (alt (cdr alts) common)
      (setq common (cssm-maximum-common alt common)))))

(defun cssm-complete-property-frame(completions)
  ; This code stolen from message.el. Kudos to larsi.
  (let ((cur (current-buffer)))
    (pop-to-buffer "*Completions*")
    (buffer-disable-undo (current-buffer))
    (let ((buffer-read-only nil))
      (erase-buffer)
      (let ((standard-output (current-buffer)))
	(display-completion-list (sort completions 'string<)))
      (goto-char (point-min))
      (pop-to-buffer cur))))

(defun cssm-complete-property()
  "Completes the CSS property being typed at point."
  (interactive)
  (let* ((prop   (cssm-property-at-point))
	 (alts   (all-completions prop cssm-properties-alist))
	 (proplen (length prop)))
    (if (= (length alts) 1)
	(insert (substring (car alts) proplen))
      (let ((beg (cssm-common-beginning alts)))
	(if (not (string= beg prop))
	    (insert (substring beg proplen))
	  (insert (substring
		   (completing-read "Property: " cssm-properties-alist nil
				    nil prop)
		   proplen)))))))

(defun css-mode()
  "Major mode for editing CSS style sheets.
\\{cssm-mode-map}"
  (interactive)

  ; Initializing
  (kill-all-local-variables)

  ; Setting up indentation handling
  (make-local-variable 'indent-line-function)
  (setq indent-line-function 'cssm-indent-line)
  
  ; Setting up font-locking
  (make-local-variable 'font-lock-defaults)
  (setq font-lock-defaults '(cssm-font-lock-keywords nil t nil nil))

  ; Setting up typing shortcuts
  (make-local-variable 'skeleton-end-hook)
  (setq skeleton-end-hook nil)
  
  (when cssm-mirror-mode
    (cssm-enter-mirror-mode))
  
  (use-local-map cssm-mode-map)
  
  ; Setting up syntax recognition
  (make-local-variable 'comment-start)
  (make-local-variable 'comment-end)
  (make-local-variable 'comment-start-skip)

  (setq comment-start "/* "
	comment-end " */"
	comment-start-skip "/\\*[ \n\t]+")

  ; Setting up syntax table
  (modify-syntax-entry ?* ". 23")
  (modify-syntax-entry ?/ ". 14")
  
  ; Final stuff, then we're done
  (setq mode-name "CSS"
	major-mode 'css-mode)
  (run-hooks 'css-mode-hook))

(provide 'css-mode)

;; CSS-mode ends here

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> In any case this should be reported to the Debian bug tracking
> system since probably more people might be affected by this.

Bug address: 434556@bugs.debian.org
URL: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=434556

	Eddy.

Re: C-comments in sgml-mode's html-mode

[Lennart Borgman (gmail)]

martin rudalics wrote:
>  > (defvar cssm-version "0.11.zap1"
>  >   "The current version number of css-mode.")
>  > ;;; copyright (c) 1998 Lars Marius Garshol, larsga@ifi.uio.no
>  > ;;; $Id: css-mode.el,v 1.9 2000/01/05 21:21:56 larsga Exp $
>  > ;;; Modified by John Zaitseff, 3rd November 2004.
> 
> css-mode by Garshol, modified by Zaitseff, and distributed by Debian.
> Apparently, Zaitseff's modifications are too persistent but I can't tell
> anything because I don't have the sources.  In any case this should be
> reported to the Debian bug tracking system since probably more people
> might be affected by this.

And Stefan Monnier's css-mode.el ought to be included in Emacs as soon 
as possible I believe.

Re: C-comments in sgml-mode's html-mode

[Dan Nicolaescu]

"Lennart Borgman (gmail)" <lennart.borgman@gmail.com> writes:

  > martin rudalics wrote:
  > >  > (defvar cssm-version "0.11.zap1"
  > >  >   "The current version number of css-mode.")
  > >  > ;;; copyright (c) 1998 Lars Marius Garshol, larsga@ifi.uio.no
  > >  > ;;; $Id: css-mode.el,v 1.9 2000/01/05 21:21:56 larsga Exp $
  > >  > ;;; Modified by John Zaitseff, 3rd November 2004.
  > >
  > > css-mode by Garshol, modified by Zaitseff, and distributed by Debian.
  > > Apparently, Zaitseff's modifications are too persistent but I can't tell
  > > anything because I don't have the sources.  In any case this should be
  > > reported to the Debian bug tracking system since probably more people
  > > might be affected by this.
  > 
  > And Stefan Monnier's css-mode.el ought to be included in Emacs as soon
  > as possible I believe.

It already is, both CVS HEAD and 22.2 branch.

Re: C-comments in sgml-mode's html-mode

[martin rudalics]

 >>Apparently, Zaitseff's modifications are too persistent but I can't tell
 >>anything because I don't have the sources.
 >
 >
 > Source follows,

Thanks.  Basically, the problem is here

(defun css-mode()
   ...
   ; Setting up syntax table
   (modify-syntax-entry ?* ". 23")
   (modify-syntax-entry ?/ ". 14")
   ...

This will modify the syntax-table of the current buffer, which can be
any syntax table.  You can, for example, in *scratch* do M-x css-mode
and you will get C-like comments in all your Elisp buffers.  Weird.

Maybe this stuff works in a mmm-like environment where a buffer may have
multiple major modes.  As it stands, the code above is highly dangerous
and should be either fixed (by making a new syntax table) or removed.

In any case it was useful to do these investigations.  Thank you for the
collaboration.

Re: C-comments in sgml-mode's html-mode

[martin rudalics]

 >   > And Stefan Monnier's css-mode.el ought to be included in Emacs as soon
 >   > as possible I believe.
 >
 > It already is, both CVS HEAD and 22.2 branch.

This won't help in the OP's case, because site-lisp's css-mode might be
loaded after Stefan's.  The problem must be fixed by Debian first.

Re: C-comments in sgml-mode's html-mode

[Edward Welbourne]

> >   > And Stefan Monnier's css-mode.el ought to be included in Emacs as soon
> >   > as possible I believe.
> >
> > It already is, both CVS HEAD and 22.2 branch.

> This won't help in the OP's case, because site-lisp's css-mode might be
> loaded after Stefan's.  The problem must be fixed by Debian first.

well, once Emacs22 is released with its own css-mode built in, one of
the ways Debian can fix this is by saying Emacs22 Conflicts: css-mode,
since it supplies its own and css-mode is broken ...

I've now tested with Emacs21: although I never saw this problem before
upgrading to Emacs22, I see that the bug is just as present in Emacs21.
I'll tell the Debian bug that it should be against css-mode !

	Eddy.