On 11/13/2015 20:20, Thomas Gazagnaire wrote:
> Good read, thanks. I am not sure how you can deal with blocking
> events (timer, read/write access to the disk). In that case, you are
> often "forced" to use "''a Lwt.t" function. For instance, in [1] when
> reading the Git pack file I need to pass
On 11/13/2015 17:37, Anil Madhavapeddy wrote:
> One of the design goals of Cohttp was to allow multiple, completely
> independent network stacks (including sockets and direct stacks) to
> run within the same program.
Uhm, how does this work? Like a transparent HTTP proxy using a sockets
> On 11/13/2015 20:20, Thomas Gazagnaire wrote:
> > Good read, thanks. I am not sure how you can deal with blocking
> > events (timer, read/write access to the disk). In that case, you are
> > often "forced" to use "''a Lwt.t" function. For instance, in [1] when
> > reading the Git pack file I
On 11/16/2015 11:08, Daniel Bünzli wrote:
> The foxnet project did that: http://www.cs.cmu.edu/~fox/foxnet.html
Thanks for reminding me to read up on Foxnet
(http://www.cs.cmu.edu/~rwh/papers/foxnet/final.pdf). I think their
PROTOCOL signature (and NETWORK_PROTOCOL) is very sensible (apart from
Hannes could you expand on what's meant by "Abstraction should be used
where needed"?
I kinda agree with the second part of the quoted snippet.
For one thing, chosen abstractions are partly a matter of taste -- one
person's abstraction is another person's abomination...
For another,
