home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.questions,cs.questions
- Path: sparky!uunet!psinntp!dg-rtp!webo!dg-webo!pds
- From: pds@lemming.webo.dg.com (Paul D. Smith)
- Subject: Re: RCS usage
- Sender: usenet@webo.dg.com (Usenet Administration)
- Message-ID: <PDS.92Dec10173836@lemming.webo.dg.com>
- In-Reply-To: tk@dcs.ed.ac.uk's message of 9 Dec 92 22:37:03 GMT
- Date: 10 Dec 92 22:38:36 GMT
- To: tk@dcs.ed.ac.uk (Tommy Kelly), ajw.dcs.ed.ac.uk (Amanda Welsh)
- Lines: 76
- References: <Bz0K5r.5K5@dcs.ed.ac.uk>
- Organization: NSDD/ONSD, Data General Corp., Westboro, MA
- Lines: 76
-
- [] Regarding RCS usage;
- [] tk@dcs.ed.ac.uk (Tommy Kelly) writes:
-
- tk> Suppose I have five source files, A, B, C, D and E which make
- tk> up an application. As I develop the application, I want to
- tk> keep track of which revisions of each file combine to create a
- tk> particular version of my application.
-
- tk> At the moment I am keeping a shellscript also under RCS which
- tk> checks out all required revisions (along with relevant README,
- tk> Makefile, etc). I update that shellscript each time I am
- tk> satisfied with a new version. Every time I want a particular
- tk> version of my application, I check out that revision of my
- tk> shellscript and run it.
-
- tk> Question: Is there an easier way to do this, perhaps using RCS
- tk> itself?
-
- These are the kinds of things one uses source control *for*, so you
- don't need to do all that mess.
-
- There are at *least* two easy ways to do this.
-
- (a) The most obvious and natural is to update the RCS major revision
- each time you create a version you want to keep track of.
-
- So after you get to a certain point you decide you're done with
- this "version", which has RCS major rev 1 (files have versions
- like 1.1, 1.2, 1.3, etc.), co -l all your files, then
- use something like:
-
- ci -f -m"Updating to new vesion 2.0" -r2 <files>
-
- This will check in all your files with a new major rev number,
- starting with 2.1, then 2.2, etc.
-
- If you want the "version 1" files back, just say:
-
- co -r1 <files>
-
- Note that if you don't give a minor version number, RCS
- automatically substitutes the latest rev. So that checks out the
- last version of each file in <files> with major revision 1.
-
- This is definitely the best way to keep track of "major stopping
- points", like versions released to others.
-
- (b) Better for "your own information" version numbers, where you just
- might want to get back these files but it's not a "real" stopping
- point, is to name each revision: just say:
-
- rcs -nMY_REV:1 <files>
-
- This will associate the symbolic name MY_REV with the latest
- version of major revision 1 in each file in <files>.
-
- MY_REV can be used in place of any revision on any RCS command, so
- you could then say:
-
- co -rMY_REV <files>
-
- To get all the files so marked.
-
- You should really spend some time reading the ci(1), co(1), and rcs(1)
- man pages and familiarize yourself with the features of RCS. They are
- many and powerful :)
- --
-
- paul
- -----
- ------------------------------------------------------------------
- | Paul D. Smith | paul_smith@dg.com |
- | Data General Corp. | pds@lemming.webo.dg.com |
- | Network Systems Development Division | |
- | Open Network Systems Development | "Pretty Damn S..." |
- ------------------------------------------------------------------
-