Re: [O] [babel] Inconsistencies/bug with :results raw

[Eric Schulte]

Hi Rainer,

Rainer M Krug <address@hidden> writes:

> Hi
>
> When exporting this org document to pdf, I get the attached pdf. There
> are some inconsistencies in the export:
>
> 1) if I use :results once or twice makes a difference
> 2) when using :results twice, the result s not the same as the
> combined nor the second :result value.
> 3) when using :results raw, no results at all are exported (expected).
> 4) If each code block is previously evaluated, the exported pdf
> differs from the origial one,
>
> I do not assume, that this is caused by my configuration, but I attach
> my emacs.org anyway.
>
> Cheers,
>
> Rainer
>
> * Version 1
> This version only prints the org code for the table, but does not interprets 
> it.
> #+begin_src R :exports both :results output raw
>   pdf("pdf1.pdf")
>   plot(runif(100))
>   dev.off()
>   cat(
>       "\n|--|--|\n",
>       "|name|[[./pdf1.pdf]]|\n",
>       "|--|--|\n"
>       )
> #+end_src
>

If you remove the leading "\n" from before the table in your cat call,
then the output will be table-aligned.  Also, this will have the result
of allowing Babel to remove the existing table on subsequent
evaluations.

>
>
>
> * Version 2
> This version prints the table including the graph as expected, but
> :results is used twice as a header argument.
> #+begin_src R :results output :exports both :results raw
>   pdf("pdf1.pdf")
>   plot(runif(100))
>   dev.off()
>   cat(
>       "\n|--|--|\n",
>       "|name|[[./pdf1.pdf]]|\n",
>       "|--|--|\n"
>       )
> #+end_src
>

Oh, thanks for catching this behavior (the difference based on splitting
of :results across two header arguments).  I've just pushed up a fix.

>
>
>
> * Version 3
> Finally this version does only export the R code
> #+begin_src R :exports both :results raw
>   pdf("pdf1.pdf")
>   plot(runif(100))
>   dev.off()
>   cat(
>       "\n|--|--|\n",
>       "|name|[[./pdf1.pdf]]|\n",
>       "|--|--|\n"
>       )
> #+end_src

Best -- Eric