You can read a PR and tell if it actually accomplishes what is says it does, right?
Mostly I can’t, not if there are subtle issues. Certainly I can look and see if any bugs jump out at me, or any areas look suspicious, but understanding a piece of code I didn’t write deeply enough tk execute it in my head usually takes longer than writing it myself.
What I can do is read a set of clearly-written functional or end-to-end tests, and see if they look like they should exercise the code written in the PR, and whether the assertions they make are the ones I’d expect, and whether there are any obvious cases that are missing. And, of course, I can look at CI and see whether said tests have passed.
Mostly I can’t, not if there are subtle issues. Certainly I can look and see if any bugs jump out at me, or any areas look suspicious, but understanding a piece of code I didn’t write deeply enough tk execute it in my head usually takes longer than writing it myself.
What I can do is read a set of clearly-written functional or end-to-end tests, and see if they look like they should exercise the code written in the PR, and whether the assertions they make are the ones I’d expect, and whether there are any obvious cases that are missing. And, of course, I can look at CI and see whether said tests have passed.
100%. A good test suite is worth its weight in gold.