The Docker-Image source for martinsaurer/jlang:jkernel
If you want to test the Jupyter Notebook/Lab J integration, without installing Anaconda or Miniconda or J, just pull the prconfigured Docker-Image:
docker pull martinsaurer/jlang:jkernel
and run it:
docker run -p 127.0.0.1:8000:8000 martinsaurer/jlang:jkernel
then use your browser to connect to:
http://127.0.0.1:8000/?token=<the token that is displayed when running the docker image>
You may take this repository as a starting point:
-
Clone this repository: git clone https://github.com/martin-saurer/jkernel-docker.git
-
Get Miniconda from: https://docs.conda.io/en/latest/miniconda.html
-
Install Miniconda to (e.g.) /home/<user>/miniconda3
-
Add Jupyter Lab to Miniconda: conda install -c conda-forge jupyterlab
-
Get J from: https://code.jsoftware.com/wiki/System/Installation
-
Install J to (e.g.) /home/<user>/J807
-
Ensure that the J user directory is under the J807 base installation folder (Edit: J807/bin/profilex.ijs)
-
Clone the jkernel repository: git clone https://github.com/martin-saurer/jkernel.git
-
Install the jkernel into your current Miniconda installation: python setup.py install
-
Test the installation from command line: jupyter lab
-
If everything goes well, copy miniconda3 from your home directory to the Docker-Image directory
-
Also copy J807 from your home directory to the Docker-Image directory
-
Your Docker-Image directory should contain two directories J807, and miniconda3, as well as the Dockerfile, and run.sh
-
Optionally you may copy some example notebooks to the Docker-Image directory
-
Go to the Docker-Image directory and perform the following command: docker build --tag=jkernel .
-
If everything goes well, you have successfully built your own Docker-Image including J, and Jupyter Lab
-
Enjoy