Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!samsung!uakari.primate.wisc.edu!sdd.hp.com!elroy.jpl.nasa.gov!ames!ads.com!sparkyfs!zwicky
From: zwicky@sparkyfs.istc.sri.com (Elizabeth Zwicky)
Newsgroups: comp.lang.postscript
Subject: Re: Printing Barcodes
Keywords: barcode code39 3 of 9
Message-ID: <32107@sparkyfs.istc.sri.com>
Date: 7 Jun 90 19:16:15 GMT
References: <9208@teda.UUCP> <1004@io.tegra.COM> <1303@chinacat.Unicom.COM>
Reply-To: zwicky@stegosaur.itstd.sri.com.UUCP (Elizabeth Zwicky)
Organization: SRI International, Menlo Park, CA 94025
Lines: 23

In article <1303@chinacat.Unicom.COM> woody@chinacat.Unicom.COM (Woody Baker @ Eagle Signal) writes:
>> In article <9208@teda.UUCP> bob@teda.UUCP (Bob Armstrong) writes:
>> >  Note that code39 is not the same as UPC barcodes (the ones you see in
>> >grocery stores)...

>Here is a UPC barcode routine for what it is worth.
>Please don't use it to create stick on barcodes for products down at your
>local grocery store. 

Aside from the fact that he specifically stated that he didn't want a
UPC font, this doesn't look to me like it could possibly produce
readable UPC. UPC is not like Code-39; characters are not mapped
1-1 into sets of lines. There are 4 codes per character (right-even,
right-odd, left-even, and left-odd), not to mention start and stop
codes. Your supermarket reader is going to want not only a right
half with all rights and a left half with all lefts, but probably
also a check digit encoded by whether characters are printed even or
odd.  You're missing a bit over 75% of the necessary characters to
produce barcode that a reader will accept, plus the fact that without
the human readable and bars of various lengths it won't pass a
human...

	Elizabeth Zwicky.