The issue being that we depend upon tsplib95, which no longer supports python 3.10 or above, or networkx 3 or above. See rhgrant10/tsplib95#26 and https://github.com/ben-hudson/tsplib95.

At the moment, we are using rhgrant10 fork with the included fixes. But PyPi does not seem to support dependencies of GitHub repos.

Unless tsplib95 PyPi is updated, we cannot push to PyPi.

The only other way would be to somehow remove tsplib95.

OPLIb assume the problem is Orienteering Problem.

We should add a PRIZE_QUOTA that the salesperson must reach for PCTSP.

Argument should be an int rather than alpha type

OPLib is a GitHub repo of txt files. The only difference between OPLib and TSPLib is 3 new section headings, which are not necessary and diverge from the TSPLib standard.

One of the sections is a NODE_SCORE section. There isn't really a reason to have NODE_SCORE as well as a DEMANDS_SECTION.

The node scores can be generated by a function that takes a vertex ID and returns a node score (int). There are three generations of node scores to implement.