Async flow control

[danielmarbach]

Proposed Changes

Addresses #1644

Types of Changes

What types of changes does your code introduce to this project?
Put an x in the boxes that apply

• Bug fix (non-breaking change which fixes issue #NNNN)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause an observable behavior change in existing systems)
• Documentation improvements (corrections, new content, etc)
• Cosmetic change (whitespace, formatting, etc)

Checklist

Put an x in the boxes that apply. You can also fill these out after creating
the PR. If you're unsure about any of them, don't hesitate to ask on the
mailing list. We're here to help! This is simply a reminder of what we are
going to look for before merging your code.

• I have read the CONTRIBUTING.md document
• I have signed the CA (see https://cla.pivotal.io/sign/rabbitmq)
• All tests pass locally with my changes
• I have added tests that prove my fix is effective or that my feature works
• I have added necessary documentation (if appropriate)
• Any dependent changes have been merged and published in related repositories

Further Comments

If this is a relatively large or complex change, kick off the discussion by
explaining why you chose the solution you did and what alternatives you
considered, etc.

[danielmarbach]

@stebet @bollhals @paulomorgado would appreciate a look since this is some non trivial magic.

[lukebakken]

Thank you @danielmarbach

[lukebakken]

Should we worry about the scenario where flow control is active and a publisher just keeps on going? I'm assuming this will "pile up" tasks/continuations/"under the hood magic" behind the scenes. Should there be a limit to the number of publishes that can happen while blocked by flow control?

[danielmarbach]

Should we worry about the scenario where flow control is active and a publisher just keeps on going? I'm assuming this will "pile up" tasks/continuations/"under the hood magic" behind the scenes. Should there be a limit to the number of publishes that can happen while blocked by flow control?

Assuming this is actually a problem that requires to be solved isn't this already an issue before this change?

[danielmarbach]

With these changes in we will have also the door open to implement IAsyncDisposable across both TFMs.

[lukebakken]

Assuming this is actually a problem that requires to be solved isn't this already an issue before this change?

Ah, ok, this will only be a problem if someone is not immediately await-ing the BasicPublishAsync result, and batching them up. I guess it's on the user in that case 😉

[danielmarbach]

FYI I did not setup the commit signing on my temporary device while my main macbook is on repair. When I rebased on my temporary device the commit signing got lost

[lukebakken]

No worries, I'm not too worried about anyone doing an xz-utils-style exploit via this library 😉 😆

[bollhals]

@stebet @bollhals @paulomorgado would appreciate a look since this is some non trivial magic.

Looks good to me, but since I'm currently on vacation, it's a bit difficult to judge the new class. Is it inspired from somewhere? I think I only used a similar class once so far.

[danielmarbach]

@bollhals I took the original version from Stephen's blog post b52c3d1 and modernized it a bit according to my best knowledge

Enjoy your vacation!

[bollhals]

Then I the implementation will be good 👍🏼

[paulomorgado]

@bollhals I took the original version from Stephen's blog post b52c3d1 and modernized it a bit according to my best knowledge

Enjoy your vacation!

We should keep a link to the articles. Here's the new home:

• Building Async Coordination Primitives, Part 1: AsyncManualResetEvent

Just in case, here are the others:

• Building Async Coordination Primitives, Part 2: AsyncAutoResetEvent
• Building Async Coordination Primitives, Part 3: AsyncCountdownEvent
• Building Async Coordination Primitives, Part 4: AsyncBarrier
• Building Async Coordination Primitives, Part 5: AsyncSemaphore
• Building Async Coordination Primitives, Part 6: AsyncLock
• Building Async Coordination Primitives, Part 7: AsyncReaderWriterLock

[danielmarbach]

Having here in the PR is more then enough especially considering the implementation I provided is significantly different due to modernizations applied