|
From: | Magnus Achim Deininger |
Subject: | Re: [Chicken-hackers] blob-literal read syntax |
Date: | Fri, 22 Oct 2010 16:28:17 +0200 |
User-agent: | Opera Mail/10.63 (Win32) |
Hallo, On Mon, Oct 18, 2010 at 3:45 PM, Alaric Snell-Pym <address@hidden> wrote:Why not? Sure, to read/write them we have to pick *some* kind of model (an even-lengthed string of nibbles, in this case) but it's arbitrary; it doesn't matter as long as it can represent all strings of bytes (the eight-bit byte having become a de-facto standard, we can at least rely on that), and it's the same at both ends.Why can't this model be a R6RS-style bytevector?
they kind of have the look and feel of vectors, while blobs and vectors are different things...
What about yEnc instead of a sequence of numbers? That would greatly reduce the encoded size of the blobs, too, at least compared to a list of decimal numbers or the hex encoding. It's really just roughly 1-2% bigger than the actual data content of the blob, compared to hex (2x the size) or decimal (2x-4x counting the space, depending on the byte).
[Prev in Thread] | Current Thread | [Next in Thread] |