org-mode: Alias for code block identifier

[fleimgruber]

When users want to migrate from their existing org files which use

#+BEGIN_SRC python
1 + 1
#+END_SRC

to

#+BEGIN_SRC jupyter-python
1 + 1
#+END_SRC

the question arose whether it is possible to let org-mode know that a source block identifier python
should really use the machinery of the identifier jupyter-python - so no changes to existing org files are necessary.
For major mode discrimination there is org-src-lang-modes, but is there a similar thing for choosing the eval backend?

[non-Jedi]

Here's a snippet from my config that does what you're asking for:

  (with-eval-after-load 'ob-jupyter
    (defalias 'org-babel-execute:julia 'org-babel-execute:jupyter-julia)
    (defalias 'org-babel-julia-initiate-session
      'org-babel-jupyter-julia-initiate-session)
    (setq org-babel-default-header-args:julia '((:session . "ijulia")
                                                (:kernel . "julia-1.1"))))

[fleimgruber]

Thanks for the snippet. I adapted it for Python and it works. We could put this into the README, @DZoP what do you think?

[nnicandro]

Sounds like a good idea, but there are more aliases that would also be useful, e.g. to setup an org-edit-src buffer connected to the same session as the src-block.

What we can do is create a new function along the lines of org-babel-jupyter-make-language-alias and then give an example on how to use that function to override the default org-mode languages in the README. Note, all of the aliases in that function can be used except for expand-body and variable-assignments.