Parse test results from tests running with --quiet

[narpfel]

Running tests with -q/--quiet outputs a . per test instead of the test name. When a test fails, the test name and FAILED is still printed, but the test at the beginning of the line is omitted:

# run without `--quiet`
test tests::failing_test ... FAILED
# run with `--quiet`
tests::failing_test --- FAILED

This PR changes the regex used to parse the output to parse both output formats.

[Canop]

Can you please build a test case which failed and is solved ?

[narpfel]

Here’s an example that shows this:

bacon.toml:

default_job = "test"

[jobs.test]
command = [
    "cargo", "test", "--color", "always",
    "--", "--color", "always", # see https://github.com/Canop/bacon/issues/124
]
need_stdout = true

[jobs.testquiet]
command = [
    "cargo", "test", "--color", "always",
    "--", "--color", "always", # see https://github.com/Canop/bacon/issues/124
    "--quiet",
]
need_stdout = true


[keybindings]
z = "job:testquiet"

src/lib.rs:

#[cfg(test)]
mod tests {
    #[test]
    fn first_test() {}

    #[test]
    fn failing_test() {
        assert!(false, "message for first failing test");
    }

    #[test]
    fn third_test() {}

    #[test]
    fn another_failing_test() {
        assert!(false, "message for second failing test");
    }
}

Output when running the test job:

 t   test    2 fails

 1  failed: tests::failing_test
 thread 'tests::failing_test' panicked at src/lib.rs:8:9:
 message for first failing test

 2  failed: tests::another_failing_test
 thread 'tests::another_failing_test' panicked at src/lib.rs:16:9:
 message for second failing test

Output when running the testquiet job with the bacon 2af07e7 (current main):

 t   testquiet   Command error code: 101

    Finished `test` profile [unoptimized + debuginfo] target(s) in 0.01s
     Running unittests src/lib.rs (target/debug/deps/t-32496683b92ff7c2)

running 4 tests
tests::another_failing_test --- FAILED
. 2/4
tests::failing_test --- FAILED
.
failures:

---- tests::another_failing_test stdout ----
thread 'tests::another_failing_test' panicked at src/lib.rs:16:9:
message for second failing test
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

---- tests::failing_test stdout ----
thread 'tests::failing_test' panicked at src/lib.rs:8:9:
message for first failing test


failures:
    tests::another_failing_test
    tests::failing_test

test result: FAILED. 2 passed; 2 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s

error: test failed, to rerun pass `--lib`

And with this PR:

 t   testquiet    2 fails

 1  failed: tests::failing_test
 thread 'tests::failing_test' panicked at src/lib.rs:8:9:
 message for first failing test

 2  failed: tests::another_failing_test
 thread 'tests::another_failing_test' panicked at src/lib.rs:16:9:
 message for second failing test

Or did you mean an automated test?

[Canop]

Or did you mean an automated test?

No, just enough to reproduce the problem.

[Canop]

Here's why I asked: it doesn't seem to be solved with your fix (in my test on linux, even with your test case).

I'll look deeper if necessary but not immediately, I can't dive into this problem until end of last week.

[narpfel]

it doesn't seem to be solved with your fix (in my test on linux, even with your test case).

I’m on nightly, and hadn’t realised that the output format was recently changed: rust-lang/rust#113518 So on stable (currently 1.77) this PR will not work (because the test name and FAILED are not printed), but on Rust 1.78+ it will work.

Sorry for the confusion!