🚀 go-pugleaf

In article <40f0bca6.14e07@gtephx.UUCP> covertr@gtephx.UUCP (Richard E. Covert) writes:
>ARC 5.21 has a serious bug which causes it make an ARC file larger then the
>sum of the individual files. ARC 5.12 doesn't have this problem. In fact,
>for a 400K ARCed file you can save about 50K by re-ARCing it with 5.12.
>
>so, don't use ARC 5.21 for new archives. Only use it if you have to de-ARC
>an IBM file arced with the IBM ARC 5.21.
>
>I only use ARC 5.12 myself!!!
>
>Richard (ARC 5.21 is BADDD!!) Covert

Y'know, I spent an awful lot of effort getting ARC 5.21 working. I also
included my e-mail address with the program docs. If you noticed a bug,
particularly a reproducible bug, you should have sent me a message about
it. Personally I don't see any difference between the encoding 5.12
produces vs 5.21. In fact, I relied heavily upon this fact when testing
5.21 prior to release. 5.21 is marginally slower than 5.12 because it
tests an additional compression method, but otherwise 5.21 produces
compression results either identical or superior to 5.12.

There was a bug in a pre-release version that sometimes caused zero length
files to be written incorrectly, producing enormous archive files. This
was fixed long ago.

Try to be a bit more objective, eh? Putting such blatantly opinionated
comments into your signatures is a little annoying, and doesn't reflect
well upon postings with statements that are already utterly wrong.
(You don't need ARC 5.21 to un-ARC files created by MSDOS ARC 5.21. You
*do* need the squashing code I added to un-ARC files created by the MSDOS
PKxxx utilities.)

--
  /
 /_ , ,_.                      Howard Chu
/ /(_/(__                University of Michigan
    /           Computing Center          College of LS&A
   '              Unix Project          Information Systems