I consistently underestimate the probability that apparent bugs in my software are in the test code rather than what it's testing.
Show this thread
-
Probably this is because I think of the test code as the good guy (finding bugs) and the actual program as the bad guy (containing them).
Show this thread
Whereas the test code is generally much more sloppily written and thus more likely to have bugs.
-
-
Do you find that the test code is much simpler than the code under test?
-
In oo and procedural languages I've used, that's almost always true, but less so in functional ones. Related: have you tried quickcheck?
- Show replies
New conversation -
-
-
Test code is written sloppily most of the times. Try to avoid doing that by writing test code in advance of development. Later always
#fails -
Furthermore, my personal way of testing is via code reviews and it is important for team leads & peers to dedicate considerable time to it
- Show replies
New conversation -
-
-
Writing good tests can be hard, especially for people new to it. Tests that pass / fail for reasons you don't expect can be a problem
-
Tests can also collect tech debt - a bit of refactoring love and attention can go a long way.
End of conversation
New conversation -
-
-
If a test fails I usually run debugger on the test. It always pays.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.