Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!rutgers!bellcore!texbell!sugar!peter
From: peter@sugar.hackercorp.com (Peter da Silva)
Newsgroups: comp.sys.amiga
Subject: Re: Amiga to Mac?
Message-ID: <3702@sugar.hackercorp.com>
Date: 10 Apr 89 19:09:02 GMT
References: <7568@phoenix.Princeton.EDU> <2368@cps3xx.UUCP> <42377@tut.cis.ohio-state.edu>
Organization: Sugar Land Unix - Houston
Lines: 30

In article <42377@tut.cis.ohio-state.edu>, martens@shawnee.cis.ohio-state.edu (Jeff Martens) writes:
> In article <3696@sugar.hackercorp.com> peter@sugar.hackercorp.com (Peter da Silva) writes:
> ASCII defines two alternatives for the
> :'new-line' sequence: either CR-LF or (if a single character is to be used)
> :just plain LF.

> Actually, I'd rather see people use CR than linefeed, for a couple reasons:

> 1) Until Unix came along, just about everybody used CR for
> end-of-line.

Until UNIX came along people used 80-column card images, variable-length
records, Fortran carriage-controls, CR-LF, and LF. The only systems I know
of that just use a plain CR as a line-terminator are OS/9 and MacOS. One
of which is modelled on UNIX!

Some MS-DOS programs ignore the linefeeds and do not generate them, but an
MS-DOS text file is supposed to have a CRLF at the end of EVERY line.

> 2) Terminals use CR rather than LF to end the line, so somewhere along
> the way, to get a program to think a LF has been typed rather than a
> CR, CR has to be mapped to LF.

And sequences using DEL and/or BS have to be handled, and so on. The
amount of work you need to implement line-editing makes converting CR
to LF, to CRLF, to spaces to pad the line to 80-column card images,
and to a field-width for variable-length records, etc... trivial.
-- 
Peter "Have you hugged your wolf today" da Silva      `-_-'
...texbell!sugar!peter, or peter@sugar.hackercorp.com  'U`