If all the #[wasm_bindgen_test]s pass, then we shouldn't spam the console with the console.log output.
Would this include console.log calls from the user? If so that might lead to some confusion.
@chinedufn I think that's pretty standard. If I recall correctly, Jest doesn't show console.log calls unless the test actually failed.
I think showing unrelated console.logs from passing tests might be more confusing.
@Pauan In both JS (I don't use Jest I use tape) and Rust I'll frequently be TDD'ing and want to print something out real quick and see what's going on.
I could see some friction if whether or not I saw my quick little prints was determined by whether or not my tests were passing at that time.
However I admittedly haven't used wasm-bindgen-test very much yet so I'll defer here!
FWIW, cargo test won't show println!s from passing tests.
Not by default but you can pass a --nocapture flag in to turn it on (that鈥檚 what I do). If that鈥檚 feasible could be nice here
Not by default but you can pass a
--nocaptureflag in to turn it on (that鈥檚 what I do). If that鈥檚 feasible could be nice here
Oh, yeah that seems very reasonable!
I could see some friction if whether or not I saw my quick little prints was determined by whether or not my tests were passing at that time.
You can always add an assert!(false) to a test to make it fail.
Although the --nocapture approach is much nicer.
I'd very much welcome console log output capturing. If you log a lot, it's very hard to see what's going on.
You can always add an
assert!(false)to a test to make it fail.
Sorry not to bikeshed here ........ but this wouldn't work for me as I'll sometimes toss a quick log in just to see if a code path was followed so if I forgot to add a false assertion I'd wrongly think that it wasn't until I remembered that my logs were hidden at the moment because tests were passing.
--nocapture sounds like it works for everyone though!
Most helpful comment
FWIW,
cargo testwon't showprintln!s from passing tests.