Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!unmvax!pprg.unm.edu!hc!lanl!cmcl2!yale!root
From: root@yale.UUCP (Celray Stalk)
Newsgroups: comp.sys.atari.st
Subject: Re: Error in Dumas' uud
Keywords: uud, bug
Message-ID: <54230@yale-celray.yale.UUCP>
Date: 20 Mar 89 20:42:55 GMT
References: <1207@laura.UUCP>
Reply-To: fischer-michael@CS.YALE.EDU (Michael Fischer)
Organization: Yale University Computer Science Dept, New Haven CT  06520-2158
Lines: 34

In article <1207@laura.UUCP> klute%trillian.irb@unido.UUCP (Rainer Klute) writes:
>Dumas' uud failed to decode quikview.uue which was just posted
>to comp.binaries.atari.st. With the -d flag on uud gave the
>following output:
>
>Begin uudecoding: quikview.arc
>No sequence check (1).
>Non uuencoded char <^>, line 40 in file: quikview.uue
>Bad line =MDH,E^1H2V,1O8   2EH!(&$&A9 \ PL^B0B  [P7.09N8/&#!P D3H)P>:($
>
>Then I tried to decode the stuff with the regular Unix uudecode
>and succeeded; a subsequent test of the generated arc also ran
>ok. So I deduced that the bug must be in uud.

The problem seems to be in the quikview.uue file itself.  In the copy
that reached me, a tilde (~) appared in the table that begins the uuencoded
file (4th character after Z), whereas the uuencoded file itself contained
carets (^):

    table
     !"#$%&'()*+,-./0123456789:;<=>?
    @ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]~_
    begin 644 quikview.arc

Changing that one occurrence of tilde to a caret allowed it to decode
and dearc properly, using Dumas' "uud".  I believe the Unix "uudecode"
does not know about the table or look at it, which would explain why
it wouldn't be bothered by the error.
==================================================
| Michael Fischer                                |
|    Arpanet:    <fischer-michael@cs.yale.edu>   |
|    Bitnet:     <fischer-michael@yalecs.bitnet> |
|    UUCP:       <fischer-michael@yale.UUCP>     |
==================================================