Skip to main content
9 events
when toggle format what by license comment
Oct 30, 2018 at 10:10 audit First posts
Oct 30, 2018 at 10:10
Oct 28, 2018 at 8:24 history edited Doc Brown CC BY-SA 4.0
added 525 characters in body
Oct 25, 2018 at 17:17 history edited Doc Brown CC BY-SA 4.0
added 210 characters in body
Oct 24, 2018 at 14:44 comment added cbojar In this instance, the missed branches could have been caught with a code coverage tool while developing the tests.
Oct 24, 2018 at 13:00 history edited Robbie Dee CC BY-SA 4.0
deleted 6 characters in body
Oct 24, 2018 at 10:30 comment added bdsl You can never be absolutely certain that the tests cover all possible edge cases, since it isn't feasible to test with all possible inputs. But there are lots of ways to gain more confience in tests. You could look into mutation testing, which is a way to test the effectiveness of the tests.
Oct 24, 2018 at 10:17 history edited Robbie Dee CC BY-SA 4.0
added 51 characters in body
Oct 24, 2018 at 9:20 comment added user200783 Thanks, that makes sense. So, if the ultimate solution to undesirable changes in behaviour is to have comprehensive tests, is there any way to be confident that tests cover all possible edge cases? For example, it would be possible to have 100% coverage of brie_tick while still never testing the problematic @days_remaining == 1 case by, for example, testing with @days_remaining set to 10 and -10.
Oct 24, 2018 at 5:56 history answered Doc Brown CC BY-SA 4.0