3 ms·
Wanted to add one thing I noticed after I've been on the reviewing side for a while. Add a section in the description of the MR describing how the changes were
by foobarian 6y ago
Wanted to add one thing I noticed after I've been on the reviewing side for a while. Add a section in the description of the MR describing how the changes were tested. If this is detailed enough and covers all the edge cases I worry about, sometimes I don't even need to look at the code to be comfortable with approving the change. Conversely, it's frustrating when I get a MR with a blank or generic description without any notes on purpose, approach, how it was tested, etc.
- mtlynch 6y agoOh, that's interesting. I'd really only want to do that for things that are hard to test automatically. I have 100x more confidence in a test where I can see the source for it and watch it run in CI. But I agree that for exceptional cases where it's hard to write test automation, it's a good idea to explain how the author verified the changes.