Open mathstuf opened 2 weeks ago
As a bonus, CI could also mutate just the diff and run tests in that direction too.
This is similar to #24.
But, maybe here you're saying we could remember which specific tests failed due to each specific mutation?
But, maybe here you're saying we could remember which specific tests failed due to each specific mutation?
Yes. When discussing with colleages, they suggested doing it by coverage, but I'm not aware of any coverage tools that save such information in a per-test bucket (especially in-process testing like Rust's #[test]
or pytest
).
Yes, I think it's an interesting idea. I think it's more closely connected to a coverage tool than to cargo-mutants, and it would probably require running each test separately like Nextest does.
It would be nice to be able to generate some kind of database that records what tests cover which source lines. This could then be consumed by something like
nextest
in conjunction withgit diff
to only run tests that are affected by the changes in a patch to avoid running the entire test suite all the time.