Lukas, I'm just summizing in this issue the contents of my email on 13 September.
Progress in the 'issue8' branch towards symmetric limits has created a layout issue, with the following output being produced:
The output from the solver contains these degeneracies. However, these diagrams were laid out fine before the symmetric limit changes, and the solver does not seem to make any use of the Limit class, so I guess the problem is not in the solver. The only remaining conclusion is that the diagram objects being constructed are somehow invalid, but they look fine to me, and the simplest diagram exhibiting this issue (an endomorphism of an endomorphism of a 0-cell) is very simple.
Lukas, I'm just summizing in this issue the contents of my email on 13 September.
Progress in the 'issue8' branch towards symmetric limits has created a layout issue, with the following output being produced:
The output from the solver contains these degeneracies. However, these diagrams were laid out fine before the symmetric limit changes, and the solver does not seem to make any use of the Limit class, so I guess the problem is not in the solver. The only remaining conclusion is that the diagram objects being constructed are somehow invalid, but they look fine to me, and the simplest diagram exhibiting this issue (an endomorphism of an endomorphism of a 0-cell) is very simple.