The PR mostly corrects the behaviours of several translation algorithms.
By adding tests for each translation algorithm specifically, their validity is confirmed (although there are still some ways to go for correctness testing). The PSO option in particular used outdated networkX calls and handled DOP incorrectly.
In future, re-implementing the entire translator should be considered.
Coverage increased (+0.5%) to 82.646% when pulling 484ec0a1cedfc15e8257c538bacaba1fae699fff on LIU-303 into 30bc92a9147a86d7fdc2af06123ccbe5b9b790b0 on master.
The PR mostly corrects the behaviours of several translation algorithms. By adding tests for each translation algorithm specifically, their validity is confirmed (although there are still some ways to go for correctness testing). The PSO option in particular used outdated networkX calls and handled DOP incorrectly. In future, re-implementing the entire translator should be considered.