An Investigation of Quality Control and Code Metrics in Astropy by Juan Cabanela

[Pey Lian Lim]

Reposting from Slack, as requested. From Juan E. Cabanela:

Hello all.  I have been an astropy user for a while and have lurked around here.  I am an astrophysics professor of 20+ years who was recently 'retrained' to be a CompSci professor by their university.  As part of my retraining, I have had to take graduate level courses in computer science.  One of them was Software Design which required a final project.  For my final project I choose to try to understand how AstroPy (the organization) approaches quality control in astropy (the package) and what the code metrics generally suggest about the code quality.

Here is a version of that final paper, in case anyone is interest in quantitative measurements of code quality via code metrics.  Keep in mind this was NOT a professional/publication quality paper (although I tried to keep it close to that).  I'd appreciate any feedback.

Also, if someone wants to post this paper to the astropy-dev mailing list, that would be fine.  Just tell them to email me directly with their comments.

I want to thank pllim and Matt Craig for their help with this project.