Unfortunately, our best answer currently is "whatever our particular
markdown parser supports". We know that is a painful situation to be
in. It is our intention to document the syntax supported in the
classic notebook and then begin to make changes to support commonmark.
The repo that Matthias pointed was a beginning attempt to document how
our markdown compares the commonmark, but the differences were
substantial in the small details.
> To view this discussion on the web visit
https://groups.google.com/d/msgid/jupyter/CANJQusWM6x0G2ugUeaVLmKosWjNB9SjdvM9%2Bq6KBNZVBW0anRw%40mail.gmail.com.
--
Brian E. Granger
Associate Professor of Physics and Data Science
Cal Poly State University, San Luis Obispo
@ellisonbg on Twitter and GitHub
bgra...@calpoly.edu and
elli...@gmail.com