Path: utzoo!attcan!uunet!amdahl!ames!ncar!tank!uwvax!puff!cat22!blochowi
From: blochowi@cat22.CS.WISC.EDU (Jason Blochowiak)
Newsgroups: comp.sys.apple
Subject: Re: The use of BINSCII
Summary: APW BinSCII will verify integrity
Keywords: APW, BinSCII
Message-ID: <2423@puff.cs.wisc.edu>
Date: 10 Mar 89 22:17:49 GMT
References: <2700143@ub.cc.umich.edu>
Sender: news@puff.cs.wisc.edu
Reply-To: blochowi@cat24.CS.WISC.EDU (Jason Blochowiak)
Organization: U of Wisconsin CS Dept
Lines: 20


	The version of BinSCII (for the //gs) will (attempt) to verify the
integrity of a decoded file. I'm writing the APW (shell utility) version of
BinSCII right now, and will port it to the desktop later on. At this point, it
properly encodes files, and is about 20% faster than BinSCII 8 (it's 30% faster
with a CRC lookup, but the lookup didn't work, so I went back to calculating
it - I'm going to try re-installing it after I get the whole schmeel done). I
just started on the decoding yesterday (got the framework in). I'm going to
try to guarantee integrity by keeping a list of the decoded segments - if
there are any gaps (if some of the segments are non-contiguous), then I'll
pretty much have to assume that it's non-integral, and warn the user (please
note that this list will be kept across input text files, so the alarm won't
go off if a particular file's segments are scattered across input files).

	By the way, I do think that the x/y (xth part out of y parts
total) is the best scheme to use.
 ------------------------------------------------------------------------------
		Jason Blochowiak (blochowi@garfield.cs.wisc.edu)
			"Not your average iconoclast..."
 ------------------------------------------------------------------------------