Matthias, thank you for your willingness to try.

I did contact the author, but he feels that script is out of date and is 
not interested in GDL, as he has access to IDL.

I have both practical and fundamental interest in it, however.  I am 
planning to write a very similar interface to another "engine" using a 
similar principle: text commands in to CLI, a mixture of text/png graphics 
in return. I thought I would start with something known to works in the 
same way.

You are right, I am confused by the relationships between kernels and 
extensions.  Your message is the first time I see a "1 notebook == 1 
kernel" principle stated plainly. I do not mind; but the idea of one 
notebook, many languages must prevail, or else the notebook format is 
useless to me. I find it somewhat distressing that everyone continues to 
refer to Jupyter as IPython: while I appreciate the origin of the tool, the 
reason I am interested has nothing to do with python itself. I guess, my 
expectation was one cell = one kernel; the language in which such a tool 
(Jupyter) itself is written is of no consequence. 

I guess, a broader question is: if a kernel "abc" is installed and known to 
the notebook, shouldn't %%abc enable the execution of its code in any cell?

I also have not been able to make (some) extensions work.  For example, 
this works fine:
  %load_ext oct2py.ipython
  %%octave -f svg
 ...octave commands...
but this (after pip install 
git+git://github.com/mkrphys/ipython-tikzmagic.git)
  %load_ext tikzmagic
  %tikz \draw (0,0) rectangle (1,1);
fails instantly with a LaTeX error "! LaTeX Error: Missing 
\begin{document}." Since this is the very basic part of functionality, I 
assume it's not a bug in the extension code but in my use of it.

GDL installation is rather simple (unlike endless variations of pythonic 
installations - I use pip), and it tests standalone in minutes:
  $ gdl
  GDL> x = findgen(5)
  GDL> y = x^2.
  GDL> print, mean(y)
  GDL> plot, x,y
would check both text and graphics output, and the same commands in the 
%%idl cell should do exactly the same. If I can make it work, all I need is 
to change the name of the external executable, and the syntax of the 
commands passed to it, to "write" a new notebook extension.

Do you see any fundamental faults with my understanding of the process?

-- 
You received this message because you are subscribed to the Google Groups 
"Project Jupyter" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jupyter/456918d7-3e9c-4002-adef-b504d0b820bd%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to