Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!cbatt!ucbvax!MITRE-BEDFORD.ARPA!jhs
From: jhs@MITRE-BEDFORD.ARPA.UUCP
Newsgroups: comp.sys.atari.8bit
Subject: Re: Another try at uudecode for the 800...
Message-ID: <8701281547.AA22746@mitre-bedford.ARPA>
Date: Wed, 28-Jan-87 10:46:57 EST
Article-I.D.: mitre-be.8701281547.AA22746
Posted: Wed Jan 28 10:46:57 1987
Date-Received: Thu, 29-Jan-87 05:41:56 EST
References: <870128-070822-6550@Xerox>
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The ARPA Internet
Lines: 20

The ML routine pays no attention to the terminator character, but the BASIC
calling program is supposed to read lines and pass them to the ML routine.
If the terminator character is wrong, BASIC will never see an EOL so will
think the whole file is one long line.

As for the calling program setting the buffer length, the user is free to use
M as he/she sees fit, but I was trying to minimize the non-ML overhead in the
calling program.  I **HOPE** your concerns are unfounded, now that I have
gotten the logic of detecting BASIC present working right.  The variable
locations I used are well documented and should be fairly (!) standard in
all software that attempts to be compatible with Atari BASIC.

If you replace the POKE 6,1 in line 102 with POKE 1,0, that should convince
the subroutine that BASIC is not enabled, so it won't meddle with the length
parameter.  You might also try just removing this line, as it should not be
necessary.  It appears to be needed when running OMNIVIEW and switching BASIC
ina and out with a program I got from COMPUTE! awhile ago, but it probably is
not needed by most users.

-John S.