alan-turing-institute / rds-course

Materials for Turing's Research Data Science course
https://alan-turing-institute.github.io/rds-course/
31 stars 13 forks source link

Consider whether Poetry is the best choice #112

Open jack89roberts opened 2 years ago

jack89roberts commented 2 years ago

Feel it complicated setup for students (and I don't think debugging Poetry setup issues is relevant for the message of this course) and also doesn't play nicely with Binder/Thebe (#87 ). Maybe just a requirements.txt with links/instructions for various virtual env options would be better. We actually already have dependencies in both Poetry and requirements.txt to try and (half-successfully) workaround the Binder/Thebe issue.

AoifeHughes commented 1 year ago

Would a conda enviornment yml file be easier? One of the easiest ways to install python at all is via anaconda + things like graphviz require additional binaries to run. They could be included in the conda setup file whereas requirements.txt can't really do this?

jack89roberts commented 1 year ago

I'd support that, but I'd lean towards installing as much as possible with pip/requirements.txt to be flexible for anyone that doesn't want to use conda (you can have an environment.yml file with a pip section).

jack89roberts commented 1 year ago

The RSE course now has a setup using poetry to keep dependencies up to date but creating a requirements.txt for participants to use (setup by James R I think) - maybe this course should do the same.