Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!apple!bloom-beacon!think!ephraim
From: ephraim@think.COM (Ephraim Vishniac)
Newsgroups: comp.sys.mac
Subject: Re: BinHex4 bug?
Message-ID: <22574@news.Think.COM>
Date: 19 Jun 89 13:00:46 GMT
References: <610@sys.uea.ac.uk>
Sender: news@Think.COM
Reply-To: ephraim@vidar.think.com.UUCP (Ephraim Vishniac)
Organization: Thinking Machines Corporation, Cambridge MA, USA
Lines: 38

In article <610@sys.uea.ac.uk> jrk@uea-sys.UUCP (Richard Kennaway) writes:

>I recently had problems decoding Font Harmony Updater, Font & Sound Valet
>Updater, and Suitcase II Updater that were recently posted to
>comp.binaries.mac.

Note the relatively long file names.

>All three gave CRC errors from BinHex4.

Yup, they'll do that.

>So how did he decode it?  With the Decode BinHex feature of Stuffit.  I
>tried Stuffit, and all three files decoded and unpacked successfully.

>Then I tried reencoding the .SIT files with BinHex4, and decoding the
>resulting files with BinHex4.  Result: each encoded successfully, and
>produced a CRC error on decoding.  Looks like a bug in BinHex4.

It is.

>Is this a known bug?  Is there a fix?

There is a known bug in BinHex 4.0, which causes it to overrun an
internal buffer when decoding files with long names.  (Over 28
characters, I think.)  There's no fixed version of BinHex 4, but there
are several alternatives.  One is to use StuffIt, another is to use
xbin on your friendly :-) local UNIX box.

The source code for BinHex 4.0, including some notes about this bug,
is available via anonymous ftp from sumex-aim.stanford.edu, under the
info-mac directory.  So, the ultimate alternative is to fix the source
code and release a new version.

Ephraim Vishniac                        One of the flaws in the anarchic
Thinking Machines Corporation           bopper society was the ease with
245 First Street, Cambridge, MA 02142-1214      which such crazed rumors
Internet: ephraim@think.com   AppleLink: ThinkingCorp      could spread.