Re: Using R in Org mode with several sourced files

From: "Thomas S. Dye" <tsd@tsdye.com>
To: Rainer M Krug <r.m.krug@gmail.com>
Cc: Dan Davison <davison@stats.ox.ac.uk>, emacs-orgmode@gnu.org
Subject: Re: Using R in Org mode with several sourced files
Date: Tue, 2 Feb 2010 08:03:42 -1000	[thread overview]
Message-ID: <B2BD11F4-0AEE-4715-92B3-0C7C4931BC7A@tsdye.com> (raw)
In-Reply-To: <fb7c7e871002020535y32bf7bcw4451a18f6dc75e04@mail.gmail.com>

[-- Attachment #1.1: Type: text/plain, Size: 9145 bytes --]

On Feb 2, 2010, at 3:35 AM, Rainer M Krug wrote:

> Hi
>
>  Following a discussion about how one could create a TODO list in  
> ESS (see below), Thomas Dye and Dan Davison suggested to look into  
> org-mode and org-babel and it looks really interesting. I just have  
> one aspect, which would be more difficult with my workflow. Let me  
> explain. I am writing fairly complex simulations, and I prefer to  
> keep functions for different aspects of the simulation in different  
> files. Therefore I have several files, each one containing R code,  
> which is sourced from a main file, which then sets the simulation  
> up, which I then can start with calling simulateIt() in R. I could  
> use .org files and tangle the code, but, as I am not that organised,  
> I even forget sometimes to save before I source the file and  
> therefore still work with the old version - this would be more  
> difficult, if I had to tangle each file in addition to the saving  
> (Ctrl-x s saves all files - very useful for me!).
>
> One option would be to combine all actual files into a single one,  
> and then tangle in different .R files, but I somehow do not like the  
> idea of having one HUGE file, with all the code, which then will be  
> tangled into several .R files.
> So I would like to have several .org files, where each one will be  
> tangled into a separate .R file, which then will be sourced from a  
> final main file. As I understand org-mode, I could still have a  
> single TODO list, based on the agenda in org-mode.
>
> So my question is, if this process could be automated. One idea was  
> a kind of a "make file", in which I specify:
> tangle file1.org
> tangle file2.org
> tangle file3.org
> tangle fileAll.org
>
> These would then result in the files attached, which then can be  
> used from fileAll.R
>
> Another idea was to have emacs automatically tangle the file after  
> saving.
>
> But I am sure there are also other possibilities.
>
> I have to say, I am absolutely new to org-mode, and know nearly  
> nothing about lisp.
>
> Cheers,
>
> Rainer
>
>
> ---------- Forwarded message ----------
> From: Dan Davison <davison@stats.ox.ac.uk>
> Date: Mon, Feb 1, 2010 at 9:59 PM
> Subject: Re: Fwd: [ESS] generating TODO list?
> To: Rainer M Krug <r.m.krug@gmail.com>
> Cc: "Thomas S. Dye" <tsd@tsdye.com>
>
>
> [...]
> >     Sure, when you change the code in the org file, you have to re- 
> run
> >     org-babel-tangle before you run your new R code.
> >
> >
> > Ant this is the critical point: I can save all files which need to  
> be saved in
> > ESS by Ctrl-x s (or was it Ctrl-x Ctrl-s ? I always do both). But  
> having
> > manually to go through all files, is something I am likely to  
> forget. If I
> > could tangle all files with one command, that wouls work (I have  
> about 10 files
> > for the project which are then sourced...). Could I combine the  
> save-all with a
> > tangle-all, or have the saving action automatically followed by a  
> tangling (if
> > it is an org file)?
> >
> >
> >     But that's just one possibility. Another would be simply to  
> tangle all
> >     the blocks of R code to the same file. Then when you run that  
> file, you
> >     know that your function definitions will be up to date.
> >
> >
> > It is quite strange, that there is no possibility to script emacs  
> from org mode
> > like all the other languages? because that would be perfect for  
> this.
>
>
> I am certain that there is going to be a good solution for all your
> concerns! I think you want to be thinking along the lines of  
> maintaining
> a _single_ org file for your project, and either including the code in
> different subtrees of that file, or referring to external files from
> within that file. But if the right solution doesn't doesn't already
> exist, and it's a good idea, then someone will cook it up in
> elisp. Could we move this discussion to the org-mode list? If you
> describe what you want to do, with a toy example if that's helpful,  
> then
> I believe you will see that problems do not remain without solution  
> for
> very long on emacs-orgmode@gnu.org.
>
> Meanwhile, have a look at this
>
> http://doc.norang.ca/org-mode.html
>
> to see some of what org-mode can do (the website itself is made in
> org-mode of course).
>
> Dan
>
>
> >
> >
> >
> >     By the way, have you tried out the export facilities of org- 
> mode? Have a
> >     look at the html export of the above org file [C-c C-e (org- 
> export)
> >     brings up the export options]. You can create links in the org  
> file to
> >     graphical output from R and these graphs will be included in the
> >     exported HTML/LaTeX.
> >
> >
> > No - I just discovered the org-mode yesterday.
> >
> > Cheers,
> >
> > Rainer
> >
> >
> >
> >     Dan
> >
> >     >
> >     > If that could be done, it would be a perfect example, how  
> one could use
> >     > org-babel in real R programming. The other option, would be  
> a "make file"
> >     > equivalent, which contains all the "tangle file ..."  
> commands, and which
> >     could
> >     > be executed in the block above.
> >     >
> >     >
> >     >
> >     >     And it's also worth looking at the noweb-style  
> references for
> >     assembling
> >     >
> >     >     different code blocks into a single code block. So with  
> reference to
> >     the
> >     >     distinction between "source is real" vs. "R objects are  
> real", from
> >     >     the org-babel point of view the org-mode files are real  
> and contain
> >     the
> >     >     source code along with essential project development and  
> planning
> >     notes,
> >     >     and the .R files are an ephemeral product of the org- 
> mode source.
> >     >
> >     >
> >     > As I said, it makes perfect sense, but the problem is the  
> additional step
> >     > needed. If this could be avoided (e.g. "make file") I would  
> very likely
> >     use the
> >     > org-mode.
> >     >
> >     > Thanks for your very helpful comments,
> >     >
> >     > Rainer
> >     >
> >     >
> >     >
> >     >     Dan
> >     >
> >     >     http://orgmode.org/worg/org-contrib/babel/intro.php#
> >     literate-programming
> >     >     http://orgmode.org/worg/org-contrib/babel/reference.php#
> >     >     function-org-babel-tangle
> >     >
> >     >
> >     >
> >     >
> >     >     >
> >     >     >     I will be looking into it anyway, because it  
> REALLY looks very
> >     >     useful.
> >     >     >
> >     >     >
> >     >     >
> >     >     >         You might want to join the Org-mode mailing  
> list.  The
> >     authors of
> >     >     >         Org-babel, Dan Davison and Eric Schulte, both  
> respond to
> >     >     questions
> >     >     >         there.  Also, there are several Org-babel  
> users there who
> >     are
> >     >     much
> >     >     >         better than I am at figuring out why things  
> aren't working
> >     right.
> >     >     >
> >     >     >         What happens when you try Ctrl-C ' while in a  
> code block?
> >      If
> >     >     Org-babel
> >     >     >         is initialized then this should take you to a  
> new window
> >     with ESS
> >     >     as
> >     >     >         the editing mode.
> >     >     >
> >     >     >
> >     >     >     Well - I used Ctrl-c ` instead of Ctrl-c ' --  
> working now.
> >     >     >
> >     >     >
> >     >     >     Thanks a lot,
> >     >     >
> >     >     >     Rainer
> >     >     >
> >     >     >
> >     >     >
> >     >     >
> >     >     >         All the best,
> >     >     >         Tom
> >     >     >
> >     >     >
> >     >     >
>
>
> -- 
> NEW GERMAN FAX NUMBER!!!
>
> Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation  
> Biology, UCT), Dipl. Phys. (Germany)
>
> Centre of Excellence for Invasion Biology
> Natural Sciences Building
> Office Suite 2039
> Stellenbosch University
> Main Campus, Merriman Avenue
> Stellenbosch
> South Africa
>
> Cell:           +27 - (0)83 9479 042
> Fax:            +27 - (0)86 516 2782
> Fax:            +49 - (0)321 2125 2244
> email:          Rainer@krugs.de
>
> Skype:          RMkrug
> Google:         R.M.Krug@gmail.com
>
> <file1.R><file2.R><file3.R><fileAll.R>

Aloha Rainer,

I'm interested to know why you're adverse to one HUGE file.  I like  
one HUGE file for the simplicity it brings to the often onerous task  
of organizing a project.  In a paper on Statistical Analyses and  
Reproducible Research,
Robert Gentleman and Duncan Temple Lang develop the idea of a  
compendium and argue for its utility in what appears to be your kind  
of application (http://www.bepress.com/bioconductor/paper2).  To my  
mind, one of the strong features of Org-babel/Org-mode is that it  
almost effortlessly creates a compendium.  Also, once you become  
accustomed to using Org-mode and get the hang of folding and unfolding  
headings, it very quickly maps onto the old division of material into  
separate files.

All the best,
Tom

Thomas S. Dye, Ph.D.
T. S. Dye & Colleagues, Archaeologists, Inc.
Phone: (808) 529-0866 Fax: (808) 529-0884
http://www.tsdye.com

[-- Attachment #1.2: Type: text/html, Size: 15458 bytes --]

[-- Attachment #2: Type: text/plain, Size: 201 bytes --]

_______________________________________________
Emacs-orgmode mailing list
Please use `Reply All' to send replies to the list.
Emacs-orgmode@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-orgmode