Re: [boost] Boost.Fiber mini-review May 23 - June 1

Subject: Re: [boost] Boost.Fiber mini-review May 23 - June 1
From: Bjorn Reese (breese_at_[hidden])
Date: 2016-05-28 13:14:07

On 05/23/2016 06:20 PM, Nat Goodspeed wrote:

> Please do remember to state in your review whether you think the
> library should be accepted as a Boost library.

Yes, I vote for unconditional acceptance of Boost.Fiber.

I did not participate in the formal review (due to time constraints),
so I am not going to address the individual issues.

I happened to have studied Boost.Fiber recently because I came across
limitations of the inter-coroutine interaction in Boost.Coroutine, and
Boost.Fiber addresses these limitations rather nicely. I especially
like the message passing channels, which makes Boost.Fiber an obvious
choice for building a coroutine-based actor framework.

The only finding I have is that several Boost.Fiber headers include
detail headers from Boost.Context. While this is not a problem per se
(because Oliver is the maintainer of both libraries), it may be an
idea to consider if these detail headers should be made public.