[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Sockets Layer Counter Proposal

This page is part of the web mail archives of SRFI 106 from before July 7th, 2015. The new archives for SRFI 106 contain all messages, not just those from before July 7th, 2015.



(2012/10/07 21:27), Aaron W. Hsu wrote:
If you are talking about Scheme implementations, then I am not sure I get
your point here. Implementations with an FFI will not have any trouble,
while implementations that support what you have specified here directly
certainly seem capable of support the approach that I have taken as well,
any other implementation will not likely be able to support either api.
I didn't think about FFI. That's true. However using FFI might limits the supporting platform. As far as I know, Mosh and Ypslion supports only X86 and X64 for their FFI and they already have own socket libraries.

My API is not much more complicated than what you have provided here.
I disagree with this. At least your APIs have at least three times more procedures than I provided. And it needs to have addrinfo structure to handle all BSD style socket APIs.

I suggest that they be merged simply because I don't think the differences
are very great, and I think having two SRFIs for this would only make
things more confusing.
SRFI already have such example, SRFI-28 and 48. Both specifies 'format' procedure and users can choose how much they need. If future SRFI supports complete BSD-style socket APIs, I wouldn't think it's confusing but different layer of APIs. If you need more control, you can use lower layer of socket.

--
_/_/
Takashi Kato
E-mail: ktakashi@xxxxxxxxx