Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!know!zaphod.mps.ohio-state.edu!sdd.hp.com!hp-pcd!hplsla!richw
From: richw@hplsla.HP.COM (Rich Wilson)
Newsgroups: comp.binaries.ibm.pc.d
Subject: Re: DOS RCS
Message-ID: <11230001@hplsla.HP.COM>
Date: 21 Aug 90 15:11:32 GMT
References: <1990Aug17.183814.11958@athena.mit.edu>
Organization: HP Lake Stevens, WA
Lines: 32

/ hplsla:comp.binaries.ibm.pc.d / lfk@athena.mit.edu (Lee F Kolakowski) / 11:38 am  Aug 17, 1990 /

There are so many parts to this, I have also placed it as a complete
compressed tar file in mbio.med.upenn.edu (128.91.18.14) for a while.

-Frank

---------------------

I played around with this last night, and can offer a couple pointers:

The compressed tar file above should probably be uncompressed before you move
it to your PC, since the MKS uncompress can't handle its 16-bit
conpressed format. The un-taring can be done by MKS.

This DOS version stores rcs files in like-named files in an 'rcs'
subdirectory. (no ,v stuff is used).

I had problems (at least with the MKS binaries) if I didn't have
the environment variable LOGNAME defined. Doing a 'ci', I ended up with an
author name of '(null)' in the rcs file, and subsequent operations
on the file resulted in error messages. Defining LOGNAME helped.

Now for the questions: Can anyone show how to integrate this with
Borland make? Will the make I just saw posted integrate with RCS any
better? My desire would be that the make process would, (for example
if you were making the file t.c) look first for t.c and check for
t.obj being up to date; if t.c was not present, it should check
for rcs/t.c, and if t.obj was out of date with respect to it, would
do a co, cc, rm sequence.

Rich Wilson