Re: Wrong type argument: characterp, 134217782 (??)

["Charles C. Berry" <ccberry@ucsd.edu>]

On Sat, 23 Jan 2016, Charles C. Berry wrote:

> On Sat, 23 Jan 2016, Eric S Fraga wrote:
>
>> On Saturday, 23 Jan 2016 at 14:36, vendo.libri@libero.it wrote:
>> 
>> [...]
>> 
>>> Then I discovered that the backquote symbol was not
>>> recognized. Everytime I press it, I get the 'Wrong type argument:
>>> characterp, 134217782' message in the mini-buffer.
>> 
>> Maybe try
>>
>>  M-x toggle-debug-on-error
>> 
>> and see where that error is arising to see if that gives us a hint as to
>> what may be wrong.  I am not sure if the cdlatex code you downloaded has
>> been updated for later versions of emacs...
>
>
> Funny, I get `apply: Wrong type argument: characterp, 134217785'
>
> (not `134217782'), which is what M-9 is recognized as:
>
> #+BEGIN_SRC emacs-lisp
> (read-char) ; type C-c C-c y M-9 *here*
> #+END_SRC
>
> #+RESULTS:
> : 134217785
>
> on my MacBook and that makes some sense as cdlatex.el uses `(read-char)' once 
> it is triggered by a back tic. And I guess that `read-char' is immune to 
> keymapping for good reason.
>
> For the record, this happens with
>
> #+BEGIN_SRC elisp
>  (define-key key-translation-map (kbd "M-9") (kbd "`"))
> #+END_SRC
>
>
> set in org-cdlatex-mode when the first back tic (or M-9) is followed by a M-9 
> and it also happens for other key events that do not map to characters like 
> `C-S-s' after the back tic.
>
> So the easiest fix is *don't do that*.
>
> I suppose this counts as a bug in cdlatex, but it seems easy enough to work 
> around.

Just to be clear on what such a workaround would amount to:

With M-9 defined as above, customize `cdlatex-math-symbol-prefix' to any 
of the unused symbols.

Suppose it is ':'. Then in a buffer in which org-cdlatex-mode is 
operative,

M-9 : : : : ...

cycles thru the different levels until the user selects another 
letter/symbol in the menu.

HTH,

Chuck