SE-0311 (3rd review): Task-local values

[slashmo (Moritz Lang)]

+1 on the overall proposal. I've been using task-locals for distributed tracing since the first proposal where they prove to be invaluable for implicit context propagation and enjoy the new syntax introduced in the second revision.

@John_McCall

There's some discomfort over the verbosity and nesting of the with -style scoped API for installing a new task-local value. The author believes that scoping is an important part of the basic design. The Core Team agrees and believes that we can investigate ways of improving the syntax for this sort of scoped change to a value.

I agree that requiring the with-API for this use-case makes sense as it makes the scope of a task-local value very clear.

In Distributed Tracing, we hit a similar "issue" with indentation, where we also offer a withSpan API to ensure proper error handling and automatic ending of a span. At least for this use case, I think something like function wrappers could make it a bit easier and remove the extra level of indentation.