search

gap-packages / JupyterKernel

Native Jupyter kernel for GAP
https://gap-packages.github.io/JupyterKernel/
BSD 3-Clause "New" or "Revised" License
19 stars 12 forks source link

Installation under Anaconda #117

Closed olexandr-konovalov closed 3 years ago

olexandr-konovalov commented 3 years ago

We have tried to install this package (release JupyterKernel 1.3 included in the GAP 4.11.1 distribution) to use with Anaconda Python installation of @xpan-eileen. The obstacle was that it had GAP 4.11.0 already (apparently installed with some SageMath installation). First called conda remove gap, then tried to install this kernel with python setup.py install and it complained about permissions. @xpan-eileen, I have some ideas what to try next:

a) python setup.py install --user may allow that due to the --user option

b) if not, then try to replace JupyterKernel 1.3 by a clone of this repository, and install it using

pip install .

or, if that also fails, using

pip install . --user

(this was introduced in https://github.com/gap-packages/JupyterKernel/pull/106 by @isuruf).

Pinging @isuruf who may have more relevant advise for Anaconda.

isuruf commented 3 years ago

either b or pip install git+https://github.com/gap-packages/JupyterKernel should work

xpan-eileen commented 3 years ago

Thank you @alex-konovalov and @isuruf! I ran isuruf's suggestion directly in JupyterLab and restarted it, now it works perfectly in both JupyterLab and JupyterNotebook.

olexandr-konovalov commented 3 years ago

Wonderful, thanks for confirming @xpan-eileen! So looks like no special instructions for Anaconda are needed for the newest state of the package on GitHub, and so things will become better from the next JupyterKernel release.