[Orgmode] Re: Debugger entered--Lisp error: (void-function -mode)

[Sébastien Vauban]

Hi David,

David Maus wrote:
> Sébastien Vauban wrote:

Thanks a lot for looking at this, really!

>> When having the following code in any Org page, I now have an error 
>> generated.
>> 
>> - If the code is in a =begin_src html= block, no problem.
>> - If between a =begin_html=, see what I get.
>> 
>> * Some code
>> 
>> #+begin_html
>> <script type="text/javascript" src="jquery-1.4.4.min.js"></script>
>> <script type="text/javascript">
>> $(document).ready(function () {
>> $('<a class="backtotop" href="#top">^</a>')
>>     .insertAfter('h2');
>> $('<a id="top"></a>')
>>     .prependTo('body');
>> });
>> </script>
>> #+end_html
>> 
>> * Debug info
>> 
>> #+begin_src emacs-lisp
>> Debugger entered--Lisp error: (void-function -mode)
>>   -mode()
>>   funcall(-mode)
>>   (if (eq major-mode lang-mode) nil (funcall lang-mode))
>>   (unless (eq major-mode lang-mode) (funcall lang-mode))
>>   (save-current-buffer (set-buffer (get-buffer-create ...)) (delete-region 
>> (point-min) (point-max)) (insert string) (unless (eq major-mode lang-mode) 
>> (funcall lang-mode)) (font-lock-fontify-buffer) (setq pos (point-min)) 
>> (while (setq next ...) (put-text-property ... ... ... ... org-buffer) (setq 
>> pos next)))
>>   (with-current-buffer (get-buffer-create (concat " org-src-fontification:" 
>> ...)) (delete-region (point-min) (point-max)) (insert string) (unless (eq 
>> major-mode lang-mode) (funcall lang-mode)) (font-lock-fontify-buffer) (setq 
>> pos (point-min)) (while (setq next ...) (put-text-property ... ... ... ... 
>> org-buffer) (setq pos next)))
>>   (let* ((lang-mode ...) (string ...) (modified ...) (org-buffer ...) pos 
>> next) (remove-text-properties start end (quote ...)) (with-current-buffer 
>> (get-buffer-create ...) (delete-region ... ...) (insert string) (unless ... 
>> ...) (font-lock-fontify-buffer) (setq pos ...) (while ... ... ...)) 
>> (add-text-properties start end (quote ...)) (set-buffer-modified-p modified))
>>   org-src-font-lock-fontify-block("" 27 287)
>> #+end_src
>> 
>> I don't understand why. Maybe it will ring a bell to you?
>
> I cannot reproduce this on
>
>
> Org-mode version 7.3 (release_7.3.123.g950be)
>
> GNU Emacs 23.2.1 (i486-pc-linux-gnu, GTK+ Version 2.20.1) of
>  2010-11-03 on potassium, modified by Debian
>
> It looks like some problem with the font-locking: I suspect you have
> `org-fontify-natively' set to a non-nil value and for whatever reason
> Org tries to fontify the #+begin_html as a source block with the
> language beeing the empty string.

I was testing some patches both from Eric and Dan (about native
fontification)...

I've pulled a new version, temporarily setting the patches aside, as I needed
to be able to export. Since then, I don't have the problem anymore.


> I have no idea about how font-locking works, but one thing for sure:
> The regexp in `org-fontify-meta-lines-in-blocks' does match
> a #+begin_html statement and the matching group used to determine the
> language (it's 7) returns an empty string.

That looks like the explanation of what I observed. Do I understand you
correctly if I read that the regexp should be patched to be more restrictive?

Best regards,
  Seb

-- 
Sébastien Vauban