This is an experimental repo designed to highlight both the possibilities, and limitations, of using async await for monads other than Task like types.
You can find documentation on task like types at https://github.com/dotnet/roslyn/blob/master/docs/features/task-types.md. Whilst the extension points to allow a task to be awaited or used as the return type of an async method are designed around task based scenarios, they can be abused for a much larger selection of monads.
Option<T>Result<T>AwaitableEnumerable<T>
You can see example dummy programs here:
https://github.com/YairHalberstadt/awaitables/blob/master/Awaitables.Option.Example/Program.cs https://github.com/YairHalberstadt/awaitables/blob/master/Awaitables.Result.Example/Program.cs https://github.com/YairHalberstadt/awaitables/blob/master/Awaitables.Enumerable.Example/Program.cs
All three implementations have only limited testing and carry out only limited validation, as they were designed to highlight the possibilities of state machine based rewriting in C#, rather than to be used in production.
However all three of them almost work. Result<T> and Option<T> are allocation free and highly efficient, and AwaitableEnumerable<T> only allocates the enumerable required to store the result of the method. There is just one critical limitation they all share, and which the language currently provides no way around: they do not respect the semantics of try/catch/finally or using blocks.
The AwaitableEnumerable<T> returned from an async method is eager rather than lazy - currently making it lazy would require a significantly greater number of allocations, though would be possible.
Feel free to contribute your own awaitables, or to address any of the existing limitations!