Path: utzoo!mnetor!uunet!husc6!tut.cis.ohio-state.edu!uwmcsd1!leah!itsgw!batcomputer!pyramid!csg
From: csg@pyramid.pyramid.com (Carl S. Gutekunst)
Newsgroups: comp.mail.uucp
Subject: Re: protocol negotiation in 4.3 uucp
Message-ID: <17275@pyramid.pyramid.com>
Date: 21 Mar 88 02:56:01 GMT
References: <3ac9315f.c6e5@delrio.cc.umich.edu>
Organization: Pyramid Technology Corp., Mountain View, CA
Lines: 24

In article <3ac9315f.c6e5@delrio.cc.umich.edu> emv@umix.cc.umich.edu (Edward Vielmetti) writes:
>With HoneyDanber it's quite easy to stick it into f protocol with
>a flag in the Devices file.  What I can't see in 4.3 uucp is how to
>do the same.  Nothing in the L.sys line indicates protocol selection,
>and everything in cntrl.c in 4.3 makes me think that it tries t if 
>it's TCP, f if it's PAD, and defaults to g.  

That's exactly right. In 4.3BSD, the protocol type is directly tied to the
caller, so if you want to talk 'f' protocol, you'll have to pretend that the
device is a PAD. PAD works exactly the same as DIR (there is no PAD driver),
so this may be the solution for you. Note that, unfortunately, you cannot put
a chat script in L-devices on a DIR or PAD device, so you'll have to put all
the rubbish in L.sys. 

The 4.3+ version that Rick and I are developing supports the "ACU,f"-style
notation of HDB; it also has genuine PAD support; and it lets you put a chat
script on any line in L-devices. So we nipped all the bugs in the bud. :-)

(This note will immediately draw a reply from Bob Page, who'll want to know
why the bloody hell I haven't shipped him a copy of the new UUCP yet. Well, I
really did copy it to uunet, I just forgot to tell you about it. :-( Pyramid
has been shipping it for six months now.)

<csg>