Open gaurav opened 5 years ago
Note that some of these are covered by other equivalent class statements, so actually might have every combination in there after all.
One way of testing this would be to enumerate all possible topologies and test to make sure that every single one matches. This might be worth coding up as a unit test.
For example, a phyloreference with 5 specifiers should have 105 equivalent class expressions, but one example (Amphibia) has only 70. We need to figure out which combinations are being missed and/or rewrite the code for finding equivalent class expressions in a more explicitly tree-based fashion.