Re: [babel] suggestion: wrap creation of graphics into try() block

[Eric Schulte <schulte.eric@gmail.com>]

Rainer M Krug <Rainer@krugs.de> writes:

> On 01/20/14, 15:00 , Andreas Leha wrote:
>> Rainer M Krug <Rainer@krugs.de> writes:
>>
>>> Hi
>>>
>>> I have two suggestions which are liked to each other. They are
>>> based on R code blocks, but the should be useful for other
>>> languages as well.
>>>
>>> I have a document in which I use :session for creating R graphs.
>>> These look like:
>>>
>>> #+begin_src R :file Correlation_1.pdf :results graphics IFN.mean
>>> <- load.IFN.mean() grid <- load.grid.CASTANEA.average()
>>> image(IFN.mean) #+end_src
>>>
>>> Everything works fine, unless there is an error in one graph.
>>>
>>> In this case, the device remains open, which leaves open devices
>>> at the end of the export of the document.
>>
>> I experience the same problem.  And would love to see that
>> addressed.
>
> Just checked in the code from ob-R.el, and it should not be to difficult:
>
> The functions
>   org-babel-R-construct-graphics-device-call
> and
>   org-babel-expand-body:R
>
> would be affected.
>
> I changed :
>
>
>
> ... Trying it out ...
>
> This seems to be working - I just changed it and it closes the device.
> But further tests are needed. I will leave it to see if anything
> unexpected happens.
>

Sounds great.  When you're confident that this is working please submit
a patch and I'll be happy to apply it.

Thanks,

>
> If somebody could look if this makes sense?
> (defun org-babel-expand-body:R (body params &optional graphics-file)
>   "Expand BODY according to PARAMS, return the expanded body."
>   (let ((graphics-file
> 	 (or graphics-file (org-babel-R-graphical-output-file params))))
>     (mapconcat
>      #'identity
>      (let ((inside
>             (append
>              (when (cdr (assoc :prologue params))
>                (list (cdr (assoc :prologue params))))
>              (org-babel-variable-assignments:R params)
>              (list body)
>              (when (cdr (assoc :epilogue params))
>                (list (cdr (assoc :epilogue params)))))))
>        (if graphics-file
>            (append
>             (list (org-babel-R-construct-graphics-device-call
>                    graphics-file params))
>             inside
> ;; my edits
>             (list "},error=function(e){plot(-1:1, -1:1, type='n');
> text(0,0,'DUMMY')}); dev.off()"))
>          inside))
> ;; end
>      "\n")))
>
>
>
> and
>
>     (format "%s(%s=\"%s\"%s%s%s); tryCatch({"
> 	    device filearg out-file args
> 	    (if extra-args "," "") (or extra-args ""))))
>
> in org-babel-R-construct-graphics-device-call
>
> It is working with the Dummy.
>
> Could somebody please check if this is working?
>
> I am leaving my changes and will see during my work if it is fine.
>
> Cheers,
>
> Rainer
>
>
>>
>>>
>>> If the code block which should create the graph would be wrapped
>>> into a try() block, so that it would look like the following:
>>>
>>> try( { pdf("./Correlation_1.pdf") IFN.mean <- load.IFN.mean()
>>> grid <- load.grid.CASTANEA.average() image(IFN.mean) } )
>>> dev.off()
>>>
>>> The device would be closed even if an error occurred during the
>>> execution of the code.
>>>
>>> But still, when  exporting to pdf, the call to pdflatex fails as
>>> only a empty pdf is available. So it becomes necessary to go
>>> through the pdf log to identify the graphs which failed.
>>>
>>> In this case it would be useful, to have a placeholder grah in
>>> the final pdf, so that one can see which graphs did not work.
>>>
>>> In addition, this could be used as placeholders (well - they are
>>>  placeholders) for to be created graphs, while the text has
>>> already been written.
>>>
>>> So my second suggestion would be to include a placeholder image,
>>> which would be used if the generation of the actual graph fails.
>>> It would be great (but not necessary) if the actual error
>>> message would be in the image.
>>
>> Such an place holder image would be great, indeed.
>>
>>
>> Thanks for bringing this up!
>>
>> Regards, Andreas
>>
>>
>
> --
> Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation
> Biology, UCT), Dipl. Phys. (Germany)
>
> Centre of Excellence for Invasion Biology
> Stellenbosch University
> South Africa
>
> Tel :       +33 - (0)9 53 10 27 44
> Cell:       +33 - (0)6 85 62 59 98
> Fax :       +33 - (0)9 58 10 27 44
>
> Fax (D):    +49 - (0)3 21 21 25 22 44
>
> email:      Rainer@krugs.de
>
> Skype:      RMkrug

-- 
Eric Schulte
https://cs.unm.edu/~eschulte
PGP: 0x614CA05D