Fix namespaces CURIE prefix & suffix extraction

[vincentkelleher]

Motivation Behind This

While working on the LinkML JSON-LD Context generator, I chose to use the namespaces.py prefix_suffix(..) method to build @id attributes.

Then, I started having issues with the test_curie_prefix_matching.py test class (BTW thanks to the developer of this class 👍 ) which has very specific CURIEs like p1:/c/c1 for example.

Solution

I went through the CURIE specification to find which types of CURIEs should be supported and I added them to the test_namespaces.py test class to make sure specific cases were managed.

The solution I propose is straightforward but as this method is used in many places, the impacts can be big. I have a pull request coming up to update the LinkML project with my JSON-LD Context generator developments and test updates.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 62.91%. Comparing base (27b9158) to head (d409e57).
Report is 3 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #308      +/-   ##
==========================================
+ Coverage   62.88%   62.91%   +0.02%     
==========================================
  Files          62       62              
  Lines        8528     8528              
  Branches     2436     2436              
==========================================
+ Hits         5363     5365       +2     
+ Misses       2554     2553       -1     
+ Partials      611      610       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[vincentkelleher]

Thanks @cmungall 🙏