Re: GS Shrinkit 1.1 problem... anybody else?

[dempson@actrix.gen.nz (David Empson)]

In article <32m9jd$l53@hermes.louisville.edu>,
Thomas Dorris <twdorr01@starbase.spd.louisville.edu> wrote:
> I recently downloaded the gs-shrinit11.bsc file off ftp.apple.com and
> decoded it to produce the .sea file, but when I ran it, it died with some
> error (I really should have written the error code down, I know).

Boy, this is an old one.  Almost everyone ran into this problem when
GSHK 1.1 was released.

The problem is that there is a bug in the BinSCII extraction of GSCII+
and SSCII.  They get the file length wrong (adding some zero bytes on
the end of the output file to round it up to the next multiple of 48
bytes in size).

A self-extracting archive is a GS/OS application, and the IIgs System
Loader _requires_ that the file length be exactly right (if there are
any extra bytes, it thinks there is a corrupted segment header at the
end of the file, and refuses to load it).

This isn't a problem for ShrinkIt files or most other file types, only
GS/OS applications and SEA files which have been BinSCII'ed without a
Binary II header (which would require an extra extraction stage anyway).

To avoid the problem, use a different BinSCII extractor, such as
BINSCII (ProDOS-8) or my DESCII utility (requires ORCA or GNO shell).
You can also use SCIIBIN on a Unix system, as long as you download the
file using YModem or ZModem (so that the file size is preserved), but
you'll have to set the filetype to $B3 after downloading before you
can launch the SEA.
-- 
David Empson
dempson@actrix.gen.nz
Snail mail: P.O. Box 27-103, Wellington, New Zealand