On 2015-05-31 Sun 01:40, Nicolas Goaziou wrote: > Titus von der Malsburg <[email protected]> writes: > >> Huh, ~code~ is translated to \verb and =verbatim= is translated to >> \texttt (plus escaping of some special characters)? > > Correct. > >> If \verb is used at all, shouldn’t it be generated by =verbatim= >> instead of by ~code~? > > Why is that? Because \verb reminds =verbatim=? Really, both "code" and > "verbatim" syntax produces verbatim contents. Therefore both are > eligible for "\verb".
Well, \verb is short for verbatim. So, yeah, it is pretty unexpected
that =verbatim= is translated to testtt and ~code~ to \verb.
>>> Note that we could do better anyway and switch command depending on
>>> context.
>>
>> Yes, I suppose anything is better than producing malformed LaTeX.
>
> Could you give the list of all contexts requiring such a switch, and
> what the result should be in each case?
Below is a list of things that I tested. This list of test cases is
probably not exhaustive.
Things that don't work with \verb (produce malformed LaTeX):
#+TITLE: ~test~
#+CAPTION: ~test~
[[/tmp/img.png]]
* ~test~
** ~test~
…
These two did not produce the intended results (sub and super script
don’t work) but the LaTeX is not malformed:
test^~test~
test_~test~
Things that did work as expected with \verb:
~test~
| ~test~ |
[[http://www.fsf.org/][~test~]]
- ~test~ :: this is a ~test~
[fn:1]
[fn:1] ~test~
*~test~*
=~test~=
/~test~/
+~test~+
_~test~_
--
Dr Titus von der Malsburg
Feodor Lynen Research Fellow
Dept. Psychology & Dept. Linguistics
University of California, San Diego
http://pages.ucsd.edu/~tvondermalsburg/
signature.asc
Description: PGP signature
