Jonathan Grynspan
Jonathan Grynspan
Rather, our string here is ideally the same as the one produced by `swift --version` (minus the hashes, probably). swift-driver has enough information at compile time to construct that string...
If CMake can see the branch we're building from is `release/*`, that may be sufficient info for us to drop `-dev`.
@swift-ci please test
I think we have an open issue tracking out-of-process parallelization? @stmontgomery
The nuances of the `--num-workers` argument itself are up to SwiftPM, please note.
We do mention it in [Parallelization.md](https://github.com/swiftlang/swift-testing/blob/main/Sources/Testing/Testing.docc/Parallelization.md), but a nod here wouldn't be a bad idea.
This constraint exists because, in order to correctly compute availability on a suite, we need to be able to see the availability of all its containing types (or if a...
Ah, right, now I remember. It's not just about syntax, because we need some way to determine the fully-qualified name of the suite type and that information isn't available if...
Tracked internally as rdar://111430835.
We have a lot of the moving parts already. What we don't have is a way to hook into the event stream arbitrarily. Currently, there is support for streaming JSON...