Timeline for Should I write and commit unit tests for minor changes and bug fixes?
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 23, 2024 at 23:06 | comment | added | Flater | It is possible for OP's current task to be listed as a bugfix. The distinction between a bugfix and a change (i.e. whether the current behavior was once intended (change) or not (bugfix)) has no bearing on whether it should be backed by a test. Bugfixes also tend to need new tests, as the existence of the bug indicates that the test suite failed to cover this gap in the first place (else it wouldn't have been released with the bug). You've already made the mistake (bug) once, so refusing to write a test effectively opens you up to regressions on a mistake you've already made before. | |
| Jul 23, 2024 at 7:51 | comment | added | jwenting | @Flater I explicitly covered that scenario | |
| Jul 23, 2024 at 7:14 | comment | added | Flater | This argument only applies from a position where all necessary tests have already been written. It does not account for a scenario whereby a developer discovers that a test that should've been written long ago was never added to the test suite. While not wrong, per se, it is idealistic as a baseline for judging whether a test should be written today or not. | |
| Jul 23, 2024 at 3:21 | history | answered | jwenting | CC BY-SA 4.0 |