"Sebastien Vauban" <wxhgmqzgw...@spammotel.com> writes: > Hi Thomas, > > Thomas S. Dye wrote: >> I want to use one of the dot utilities, tred, in a way that preserves my >> ability to distribute the Org-mode file as reproducible research, i.e., >> intermediate results should end up in the Org-mode file or be passed >> along in a chain. >> >> Right now I have a python routine that reads a couple of Org-mode tables >> and outputs a valid dot graph. I'm saving this to a file, dot-temp.gv, >> and then running some shell commands, like this: >> >> #+begin_src sh >> tred dot-temp.gv > test.gv >> dot -o test.pdf -Tpdf test.gv >> open test.pdf >> #+end_src >> >> The graph is just what I want, but I can't see how to automate the >> process in Org-mode. In particular, the tred step trips me up. Do I >> have to make babel, tred-aware? > > Sorry if I misunderstand, but could you be more explicit (for me) on what the > problem is? Why is it a problem for you to have that sh block inside your > Org. It is executed automagically upon exporting, no? If yes, why is that not > good enough? > > Best regards, > Seb
Aloha Seb, Yes, babel is working fine and the sh block does its job. The problem is that I like to see intermediate results in the Org-mode file and the sh block works instead with files that reside outside Org-mode. In this particular case, I'd prefer to have dot-temp.gv in a results block in the Org-mode file, and then pass this results block into a source code block where tred can manipulate it, and which returns its results to the Org-mode file. That way, the skeptical inquirer need only consult the Org-mode file to be convinced that tred removed the transitive relations from the digraph. It would be an easy matter to compare the two results blocks. All the best, Tom -- Thomas S. Dye http://www.tsdye.com
