Open dockter-84 opened 8 years ago
As a work-around, I'm doing this after cloning the repository:
sed -i.$(date +'%s') \
-e s/"kernelName: 'python'"/"kernelName: 'python2-spark20'"/g \
-e s/"kernelName: 'scala'"/"kernelName: 'scala-spark20'"/g \
-e s/"kernelName: 'ir'"/"kernelName: 'r-spark20'"/g \
~/kernel_gateway_demos/node_client_example/src/client.js
@dockter-84 I'm happy to take a PR here that extends the demo scripts with the kernel-name
and source
parameters.
/cc @DTAIEB
Because there's a "hardcoded" relationship between input parameter, kernel name, and source code file, we are not able to run the python, scala, and R demos on kernels with different kernel names.
https://github.com/jupyter/kernel_gateway_demos/blob/b7fd80bda9e9b99f0c450c72ea5dc5aff55bd989/node_client_example/src/client.js#L15
Can you add parameters like
--kernel-name <kernel_name>
and--source <filename>
to the client.py and client.js demonstration programs? These parameters would supersede the values in the dictionary within the code.That way we could run: