Rain

Oh that's really interesting re not being able to find the libtest.so files. Wonder if we should just bundle them with the archive.

Yes, --exact is added by nextest in order to run that specified test, since nextest runs each test in its own process. See https://nexte.st/book/how-it-works.html#the-nextest-model for the execution model.

**Help wanted:** This is right at the boundary of "good first issue" and "not", and requires a particularly motivated contributor. That's because this isn't easy to do -- cargo's JSON...

Hi, thanks for the question! I agree that this would be good. The main challenge here is presenting this in the UI -- we need to make it clear that...

This is an artifact of how `cargo-nextest` is invoked: via a `cargo-nextest` binary that has a `nextest` subcommand. I guess it's worth fixing.

Sounds reasonable. Would you like to contribute a PR to achieve that?

I think we can just use the first line of the description returned here as the message: https://github.com/nextest-rs/nextest/blob/b3f8e565d6f9f06df15267432251ace40a717a60/nextest-runner/src/reporter/aggregator.rs#L273-L276 If it isn't present, something generic like "test failed" would be okay.

Going to mark this as "help wanted" and low-urgency since it looks like sonar has been fixed.

Hi there! Yes, we should have a reasonable depiction of skipped tests in the JUnit output. Would you like to work on this? I'd love to work out a design...

I think we do generally want machine-readable output to be stable, modulo bugs. This seems like a case where it would be pretty surprising to users to suddenly start seeing...