Expect error 101 from nextest when build fails?

[sourcefrog]

 WARN nextest process exited with unexpected code (not TEST_RUN_FAILED) code=101
unviable src/process.rs:78:9: replace Process::start -> Result<Process> with Ok(Default::default()) in 1.6s build

Probably this is the normal effect of the build failing, and the warning is wrong?

[jaskij]

According to nextest-metadata it does indicate a failed build.

In fact, if I'm reading the README correctly, nextest-metadata was made specifically for other tools that wish to wrap cargo-nextest and you should be using it.

[sourcefrog]

Yes, the very small bug here is to match that interpretation and not emit a warning

[jaskij]

Speaking of, I also get a warning on nextest exiting with 4 - which seems to indicate no tests run.