test: Log the number of the test account if there are multiple alices #7087
Conversation
src/test_utils.rs
Outdated
| Self::builder().configure_alice().build().await | ||
| Self::builder() | ||
| .configure_alice() | ||
| .build(&mut BTreeSet::new()) |
There was a problem hiding this comment.
Instead of creating fake argument to build, I think it should be optional and set with .with_used_names in the builder. Probably needs to be a reference-counted pointer. Otherwise instead of a builder we will eventually get a constructor with a lot of non-optional arguments that need to be faked.
There was a problem hiding this comment.
Probably needs to be a reference-counted pointer
It even needs to be Arc<Mutex>, because we need to mutate the BTreeSet (probably can't use Rc<RefCell> because our async runtime is multi-threaded). We may be able to solve this by putting lifetimes everywhere, but this will make the code hard to read esp. for Rust beginners. While possible, at this point I don't think it's worth it to make this abstraction perfect for a builder that is only used 9 times, all of which is in the same file. (and the TestContextBuilder exists since 3 years, so there was a lot of time to use it outside this file, but noone ever felt the need to do so).
I think moving a bit towards a constructor is fine.
But it makes sense to make this argument optional.
In order to debug some test failures wrt broadcast channels, I need to read the log of some tests that have an alice0 and an alice1. It's currently not possible to tell whether a line was logged by alice0 or alice1, so, I would like to change that with this PR.
Hocuri
force-pushed
the
hoc/log-testacc-number-in-tests
branch
from
42a12bf to
31cfcce
Compare
In order to debug some test failures wrt broadcast channels, I need to read the log of some tests that have an alice0 and an alice1.
It's currently not possible to tell whether a line was logged by alice0 or alice1, so, I would like to change that with this PR.
Edit: Turns out that there are more tests that call their profiles "alice1"/"alice2" (or "alice"/"alice2") than "alice0"/"alice1". So, I changed the logging to count "alice", "alice2", "alice3", ….
Not sure whether I should adapt old tests; it might save someone some confusion in the future, but I might also make a mistake and make it so that the test doesn't properly test anymore.