[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Updating revision number on a branch.

From: Derek Robert Price
Subject: Re: Updating revision number on a branch.
Date: Mon, 20 Dec 2004 10:21:45 -0500
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7) Gecko/20040616

Hash: SHA1

Heather wrote:

>Heh, I think you misunderstood my second message.  I meant that we
>wanted to make the change for cosmetic reasons, not that it WAS a
>cosmetic change.  After reading on how the branches are created, I
>understand why you don't want to mess with those.  However, I don't
>understand what the problem is with updating the revision numbers on
>the main trunk.  Am I seriously missing something because I see people
>doing this all the time? :)  Or are all these people just treading into
>dangerous waters?  And yes, we know how to use symbolic tags and have
>been using them, so the issue isn't that we don't know how to do that.
>This is just a curiosity thing at this point in time.  

The basic problem is that you will may very well find out that CVS is
making assumptions about revision numbers fitting some 1.x format, as
revision numbers starting with anything other than `1.' are not tested
thoroughly in the regression test suite despite their being some
legacy support since, at least in theory, and RCS repository should be
importable into CVS.

Treading dangerous waters means that you may also very well find out
that most of CVS's functionality will still work correctly, but you
might encounter occasional bugs that, due to lack of interest from
others, you will likely be stuck fixing.  Patch submissions of fixes
for standard functionality that breaks when !1.x revisions are
encountered would likely be accepted by the CVS developers if they
don't break any other functionality, but you might also encounter some
edge cases like Alan Dayley pointed out from CVSNT where new revisions
are added with 1.x revisions again despite others at 2.x or otherwise,
and the CVS developers might have little interest in a fix for such an
issue, though I can imagine some simple patches for such an issue that
might be accepted.

Anyhow, if you have the resources to deal with these issues and work
with the CVS developers on patch submissions, by no means should you
let me discourage you and by no means would I want to discourage you.
I'm just trying to provide you with fair warning of what you might
encounter if you tried this.



- --

Email: address@hidden

Get CVS support at <>!
Version: GnuPG v1.2.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla -


reply via email to

[Prev in Thread] Current Thread [Next in Thread]