Open jvdwetering opened 1 month ago
I have been able to reproduce the issue with the following simplest graph. Fusion does not seem necessary; first decompose hadamard and then move the rightmost x spider. The pi/2 is actually not disembodied. Somehow it becomes the phase of the boundary node.
Start with following graph:
Apply 'decompose hadamard' and then fuse the left Z-spider to get:
Now move one of the X-spiders. A disembodied pi/2 phase will appear and the following exception is thrown: