Re: [O] Non-interactive export

[Simon Thum]

Sorry, I was being imprecise.


I do have maxima and shell and other source bocks since the dawn of org, 
approximately. I did not have any gnuplot blocks until before two weeks. 
The gnuplot blocks break my setup, currently.

So, what is weird is that maxima and shell do well but gnuplot wants its 
interactive things.

I noticed the same behaviour when exporting interactively - babel  asks 
me for the gnuplot stuff. Could it be that the export defaults vary per 
backend?

My org-confirm-babel-evaluate is t. However on export, only gnuplot 
triggers this question. I now notice that maxima is executed on export 
without asking first; I guess shell is the same.

Things are getting stranger...

Cheers,

Simon

On 02/21/2014 12:38 AM, John Hendy wrote:
> On Thu, Feb 20, 2014 at 4:12 PM, Simon Thum <address@hidden> wrote:
> > Hi all,
> >
> > I have a server over which I sync my Org files and provide drops for other
> > clients, chiefly iCal and vCard. Therefore I need non-interactive export.
> >
> > Things recently stopped working however, and although I acknowledge it is
> > likely not Org's fault maybe here I can find somewone more into lisp to help
> > me out a bit.
> >
> > In my logs I find, since it stopped working:
> >
> > > Evaluate this gnuplot code block on your system? (yes or no)
> >
> >
> > Which stops the show, waiting for input. Emacs is started with -batch and is
> > given an export script. This so far avoided such questions (from org-babel?)
> > successfully. I have maxima blocks since ages and do not remember having
> > seen this behaviour.
> >
> > Does naybody have an idea what could be going on?
> >
> > The machine does not even have gnuplot.
>
> I don't think the machine matters so much as the source of Org/babel
> thinking gnuplot *code* exists in one of your files.
>
> Can you try something like this from the directory containing the
> exported files?
>
> grep -r "#+begin_src gnuplot" *
>
> I would have suggested just tweaking the variable
> =org-confirm-babel-evaluate=, but I think it's more important to track
> the source of this down, as turning the inquiries off implies that we
> all know exactly what's going to be evaluated, and this sounds like an
> instance of something not only unknown, but a change on what you
> believe to be a rather unchanged system, which is definitely
> concerning.
>
> Good luck,
> John
>
> > Thanks in advance,
> >
> > Simon