Presently, in $SAGE_ROOT/local/share/jupyter/kernels/sagemath/kernel.json, we reference the executable $SAGE_ROOT/local/bin/sage If we were to reference $SAGE_ROOT/sage instead, we'd have the advantage that the kernel spec can also be used by a jupyter that isn't running in the sage python environment (the problem being that if the local/bin/sage runs outside of it, we get a complaint about $SAGE_ROOT not being set, whereas the sage script directly in $SAGE_ROOT goes through some length to set it.
If we would change the contents of kernel.json to refer to the $SAGE_ROOT/sage script, the $SAGE_ROOT/local/share/jupyter subdirectory would be suitable for a globally installed jupyter (which could make use of the files there by making a symlink or by editing the paths it uses). That makes it easier to integrate sage as a kernel in more general jupyter setups. Are there good reasons to refer to .../local/bin/sage rather than the .../sage script? If not, then it might be more flexible to change it. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
