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
Just came across this very interesting article about testing challenges in the interactive programming game Factorio. There the author lays out reasons for why having tests depend on each other makes refactorings easier.
Test A -> tests feature A
Test B -> tests feature B
# No need to run this test if "Test A" or "Test B" fails
Test C -> tests how feature A + B work together
The idea is that one usually has e2e test which are a bit more low level, like test feature A and others which usually exercise multiple features at once and how they work togther. So in that situation it's not useful to run tests which depend on A to work, when A is broken. It's totally fine to just run the tests for A and skip all further tests that depend on feature A to work.
The text was updated successfully, but these errors were encountered:
Just came across this very interesting article about testing challenges in the interactive programming game Factorio. There the author lays out reasons for why having tests depend on each other makes refactorings easier.
Test A -> tests feature A Test B -> tests feature B # No need to run this test if "Test A" or "Test B" fails Test C -> tests how feature A + B work together
The idea is that one usually has e2e test which are a bit more low level, like test feature
A
and others which usually exercise multiple features at once and how they work togther. So in that situation it's not useful to run tests which depend onA
to work, whenA
is broken. It's totally fine to just run the tests forA
and skip all further tests that depend on featureA
to work.The text was updated successfully, but these errors were encountered: