Jeff Walker
Jeff Walker
The doc comments for `AsyncMonitor.Pulse()` and `AsyncMonitor.PulseAll()` say "The monitor MUST already be entered when calling this method." It doesn't explain what will actually happen if that isn't done. Testing...
I understand and agree with your concerns about ["Try" being ambiguous](https://github.com/StephenCleary/AsyncEx/issues/141#issuecomment-409287646). When I first saw the approach of [using already canceled tokens to do atomic waits](https://github.com/StephenCleary/AsyncEx/blob/master/doc/Cancellation.md#atomic-wait), I thought it was...
### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior When including all prerelease versions, a tilde range like `~1.2.*` does...
### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior A version range like `1.x.5` is valid and treated as equivalent...
### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior A range like `>=1.x.x-rc` appears to ignore the `-rc` and be...
### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior A version range like `>1.x.x+build` is valid, but `>1.x+build` and `>1+build`...
Please add the option of selecting less versions: - 2.0.0 - 2.1.0 - 2.1.1 - 2.1.2 - 2.2.0 - 2.3.0 - 2.3.1 - 2.4.0 - 2.5.0
Since these may only terminate blocks or replace blocks, they aren't really expressions. They are more like statements.
Right now they compare identity for references. Make them always compare value.
Can you have a variable of type `ref ref T` or does that always collapse to `ref T`? What about `ref var ref var T`? It seems like `ref ref...