Codecov Report

Attention: Patch coverage is 63.63636% with 4 lines in your changes missing coverage. Please review.

Project coverage is 76.87%. Comparing base (6291747) to head (be811e2).
Report is 5 commits behind head on main.

@@            Coverage Diff             @@
##             main    #1578      +/-   ##
==========================================
+ Coverage   76.86%   76.87%   +0.01%     
==========================================
  Files         394      394              
  Lines       63246    63248       +2     
  Branches    10360    10361       +1     
==========================================
+ Hits        48611    48622      +11     
+ Misses      12187    12186       -1     
+ Partials     2448     2440       -8     

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

I only have one question on whether we should handle the scenario where num_VLE > 1 by raising an exception or at least logging a warning. It may be the case that num_VLE can never be >1 because it is caught elsewhere, but I didn't do a deep enough to dive to check that.

It's possible, in theory, for there to be multiple VLE if we have a vapor/liquid/liquid system. That's why the exception isn't raised. Such a system isn't explicitly supported at the moment, but a user who knows what they're doing could probably set one up.

I only have one question on whether we should handle the scenario where num_VLE > 1 by raising an exception or at least logging a warning. It may be the case that num_VLE can never be >1 because it is caught elsewhere, but I didn't do a deep enough to dive to check that.

It's possible, in theory, for there to be multiple VLE if we have a vapor/liquid/liquid system. That's why the exception isn't raised. Such a system isn't explicitly supported at the moment, but a user who knows what they're doing could probably set one up.

In that case, we should at least log a warning (or possibly raise a NotImplementedError that an advanced user could bypass with a try-except clause).

At this point, I'm inclined to merge this with or without Codecov's approval. I first created a warning to implement @adam-a-a 's suggestion, but I found I couldn't test it because initialization throws an exception if constraints have large residuals at the end. Before that, however, I had to correct an issue in which results was undefined in the initialization routine for the tortured model I had to use to test this issue.