Re: Questions on LaTeX Exporter

[Thomas S. Dye <tsd@tsdye.com>]

Aloha Rasmus,

On Mar 6, 2011, at 3:21 AM, Rasmus wrote:

> Hi,
>
>> This would make you an "early adopter."
>
> Well, to be fair the latex exporter have been there for years and I  
> have
> used it for non-important documents for years. But papers utilizing  
> more
> features have proven difficult, so far.
>

I didn't mean to imply any criticism of the Org-mode LaTeX exporter.   
It was designed to export notes and does a fantastic job of that.  The  
fact that we are wanting to use it for other kinds of more complex  
documents is testament to how good it is.

>> Of course, the paper doesn't have abbreviations that end  with  
>> dots ...
> The problem might be traced back to my less-than-stylish prose :)
>

No problem.  My conference paper is not very complex, but I am very  
happy with the LaTeX code produced by Org-mode.  The paper itself was  
easy to write in Org-mode and I am a big fan of having everything in  
one file.  I'd like to write larger, more complex documents in Org- 
mode, too.

>> Yes, and the overhead does get in the way, at least for my writing
>> projects.
>
> With folding, AUCTeX buffers become quite readable. But linking,  
> plannig
> and TODOs are nice. Previously, I have had a notes.org and several
> tex files.
>
>
>> On the other hand, I think it is the only way currently to get from
>> Org-mode to perfect LaTeX.
>
> ... And in that case AUCTeX provide a nicer environment in my
> oppinion. Org would be nice for text heavy documents, though.
>

Yes, AucTeX sets a high standard.

>>> Fixme in Org
>> If you figure this one out, please share
> I haven't. I think one of either of the following would be nice.
>
>  1) Have them fold like I do in AUCTeX (i.e. \fxnote{·} is replaced by
>     [fix] in the buffer)
>  2) Use special footnotes. This could be an org-centric system,  
> without
>     the need of fixme.sty. This would require that Org could tell the
>     difference between fn:x and fix:x, and further, one would need to
>     be able to specify which set(s) of footnotes were to be exported.
>
> *More questions, sorry*
>
> Org populates every section with a label. I would
> like to \ref or \vref these. I could predict \label's, but this a  
> rather
> fragile solution. When I use "Org-links" I get a text link suitable
> for e.g. html. I want to use \ref to get a number. One solution is
>
> ,----
> |   * section
> |   #+latex: \label{sec:sec}
> `----
>
> But there /must/ be a better way to this, eh?

Good point.  It would be great to leverage Org-links to resolve cross  
references to document sections.

>
> And other question, which should also be simple, but which I have  
> not been
> able to figure out.
>
> I have internalized word count in my org file using babel:
>
> ,----
> | ** Getting Word Count
> | #+srcname: wordcount
> | #+BEGIN_SRC sh
> |   #!/bin/bash
> |   alias calc='Rscript -e "cat( file=stdout(),  
> eval( parse( text=paste( commandArgs(TRUE), collapse=\"\"))),\"\n\")"'
> |   calc `texcount -inc -sum -relaxed -1 -q -total assignment.tex`/400
> | #+END_SRC
> |
> | #+results: wordcount
> | : 0.6425
> `----
>
> First, this should be evaluated post-export, but this is a trival  
> issue
> as I can export the document in question twice. However, I want to
> include the result in a \thanks{·}-node in the #+TITLE.
>
> I have tried varioues methods, but so far without luck. To given an
> example, I want to replace X by the results of my Babel-snip in the
> following:
>
> ,----
> | #+TITLE: \Large Education in Labor Markets With Asymmetric  
> Information\thanks{Approximatly X words using \TeX Count}
> `----
>

You are right.  Using an inline code block, e.g. src_sh[:noweb yes] 
{<<wordcount>>}, doesn't seem to work in #+TITLE.  I'm not sure how it  
might be made to work there.

All the best,
Tom

> Thanks in advance,
> Rasmus