You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There is a problem that old layouts start to give different results after a while with new Sverchok versions.
I see several ways to handle this:
Create tests which would insure that attribute names of nodes are not changes, that socket names are unchanged etc. But it's not clear how such tests can be created (I guess git should be used to find difference with main branch). It's not clear which set of rules will insure the output persistence. And such test won't be able to catch all cases.
To create tests for every single node in every single mode. First of all it's huge amount of work and second it also won't be able to catch all errors.
And last variant is to test our existent examples. It should be enough to catch data of last (output) nodes in the layouts and use it as standard during testing new changes. Also it requires writing coverage tests. They should show which nodes in which modes are not tested.
The text was updated successfully, but these errors were encountered:
Problem statement
There is a problem that old layouts start to give different results after a while with new Sverchok versions.
I see several ways to handle this:
git
should be used to find difference with main branch). It's not clear which set of rules will insure the output persistence. And such test won't be able to catch all cases.The text was updated successfully, but these errors were encountered: