Instructions, not link.
Let me try again, here's the usage from the Windows port of LhA:
C:\>lha
LHarc for UNIX V 1.02 Copyright(C) 1989 Y.Tagawa
LHx for MSDOS V C2.01 Copyright(C) 1990 H.Yoshizaki
LHx(arc) for OSK V 2.01 Modified 1990 Momozou
LHa for UNIX V 1.00 Copyright(C) 1992 Masaru Oki
LHa for UNIX V 1.14 Modified 1995 Nobutaka Watazaki
LHa for UNIX V 1.14i Modified 2000 Tsugio Okamoto
Autoconfiscated 2001-2005 Koji Arai
usage: lha [-]<commands>[<options>] [-<options> ...] archive_file [file...]
commands: [axelvudmcpt]
options: [q[012]vnfto[567]dizg012e[w=<dir>|x=<pattern>]]
long options: --system-kanji-code={euc,sjis,utf8,cap}
--archive-kanji-code={euc,sjis,utf8,cap}
--extract-broken-archive
--help
--version
commands: options:
a Add(or replace) to archive q{num} quiet (num:quiet mode)
x,e EXtract from archive v verbose
l,v List / Verbose List n not execute
u Update newer files to archive f force (over write at extract)
d Delete from archive t FILES are TEXT file
m Move to archive (means 'ad') o[567] compression method (a/u/c)
c re-Construct new archive d delete FILES after (a/u/c)
p Print to STDOUT from archive i ignore directory path (x/e)
t Test file CRC in archive z files not compress (a/u/c)
g Generic format (for compatibility)
or not convert case when extracting
0/1/2 header level (a/u/c)
e TEXT code convert from/to EUC
w=<dir> specify extract directory (x/e)
x=<pattern> eXclude files (a/u/c)
C:\>
We're interested in the compression method, thus:
o[567] compression method (a/u/c)
...is the interesting part. -o5 will generate -lh5- archives.
