Path: utzoo!attcan!telly!druid!darcy
From: darcy@druid.uucp (D'Arcy J.M. Cain)
Newsgroups: comp.unix.sysv386
Subject: Re: ESIX compress(1) broken (was: Re: Using Interactive TPWB...)
Keywords: ESIX compress
Message-ID: <1991Jan16.233354.4201@druid.uucp>
Date: 16 Jan 91 23:33:54 GMT
References: <1991Jan6.225758.27472@portia.Stanford.EDU> <1991Jan7.154351.16652@jwt.UUCP> <1991Jan10.082959.18602@hawkmoon.MN.ORG>
Organization: D'Arcy Cain Consulting, West Hill, Ontario
Lines: 30

In article <1991Jan10.082959.18602@hawkmoon.MN.ORG> Derek E. Terveer writes:
>>didn't work.  It turns out that "compress -d", which is used by news
>>software to uncompress batches, does not work with the ESIX-supplied
>>compress.
>
>Yes, this is true.  I had to use a public domain compress to get news
>unbatching to work in Cnews.  Very strange; one would have thought that it
>would have been tested...?

Both C News and and Esix' compress work.  The problem is in the way they
work together.  C News pipes the output of "compress -d" into a shell
variable and tests that variable for zero length.  It assumes that compress
like all good Unix programs, keeps quiet when everything is OK.  This
is usually true but the Esix version prints a message telling the user that
it was successful.  It is this success message that C News chokes on.  In
C News, no news is good news and any news is bad news and news of success
is news to C News.

Since I paid for Esix and C News was free I argue with Everex that their
code is wrong, not Henry's.  However I have suggested to Henry that a
change from "compress -d" to "uncompress" (which does work, even though
it uses the same binary) is a cheap fix to a nagging problem but I
understand his reluctance to fix the OS in his application.  It just
ain't the right way to do it.

-- 
D'Arcy J.M. Cain (darcy@druid)     |
D'Arcy Cain Consulting             |   There's no government
West Hill, Ontario, Canada         |   like no government!
+1 416 281 6094                    |