[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[MIT-Scheme-devel] Efficient reading from a network socket
From: |
Matt Birkholz |
Subject: |
[MIT-Scheme-devel] Efficient reading from a network socket |
Date: |
Tue, 1 May 2012 13:21:25 -0700 |
> From: <address@hidden>
> Date: Tue, 01 May 2012 11:03:23 +0200
>
> Hello fellow Schemers!
>
> Another question, what is an efficient way to read from a network (or
> unix domain) socket? After reading the manual, I'd guess READ-SUBSTRING!
> with an aptly allocated string and then using VECTOR-8B-REF etc. to
> access the actual values. Is this correct?
Yes, that seems to be the fashion, consistent with recently written
stuff like runtime/blowfish.scm, though input-port/read-substring! is
also popular, avoiding an optional argument initialization. (I take
my cues from the canon.)
> Is there any further support for processing network packages that
> might be split up and fragmented?
Out-of-order UDP network packets? Never heard of them. :-}