On 16.04.19 17:26, Nils Bruin wrote: > Do you want to test that you get *that* string representation back? > Or do you want to check that you are getting string representations of > that shape, but with possibly differently order frozensets? > Or do you want to check that you get a list back with transitions that > have certain properties? > > Unless this is about the `repr` method, I expect that you don't actually > care about the string reps. Can you just explicitly construct the list > of transitions that you want to compare with and then use comparison on > actual, meaningful objects? Or perhaps you can just test the transitions > in the list to have the desired properties? I guess they're maps, so you > could test some of the images of known inputs.
No, it is not about the repr method, but that is one typical output of a doctest. It can be read quite easily, so one can check it mathematically and understand what happens (after all its in the documentation). But the doctests should pass as well and testing such an output is one major feature of our doctesting. So I certainly want to keep it that way (and this discussion is not new). There have been attempts to stabilize the displayed output before... -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
