Hi

Am 22.07.2011 um 09:34 schrieb Rainer M Krug:

For tangling: you could put   from __future__ imports into the :shebang and use padline ":padline no", i.e.:

#+source: the_test

#+begin_src python :var x=3 :tangle test.py :results output :shebang from __future__ imports :padline no

print x

#+end_src

which results in

from __future__ imports

print x

Keep in mind, that I have NEVER used python (although I should…).

While I'd not call that a clean solution to the problem it will keep me going for now. 

Thank you very much for pointing that possibility out, I haven't thought about trying to move the import statement. 

As for python: its a nice language worth trying in my opinion ;)  

With this i could do the table calculation manually by inserting 100-1000 call statements (in the real use case I need for my masters-thesis) but it would be really nice if i could use a table cell as argument for code-blocks. 

A second problem I have at the moment lies with the execution of source-blocks in tables. What I'd like to do:

| argument | result |
|        1 | #ERROR |
|          |        |
#+TBLFM: $2=call_the_test(x=$<)

I guess I'm just doing something wrong here. Executing the #+Tblfm results in the error: "reference $< not found in buffer". How do I do the reference correctly in this case?

You can test both cases in the attached org-file.

best regards,

Dirk