Re: cvs update; merge

[James Youngman]

Jimm Grimm <address@hidden> writes:

>       I've been reading the mailing lists a bit more on the binary file
> issue, and it seems to come up a lot.  It seems to me like people have often
> merged two separate issues together:
> 
> 1) eol conversions
> 2) mergeable and nonmergeable files

I'm not sure it's only two...

a) eol conversions (at leat three types of these, neglecting
   fixed-record systems)
b) mergeable vs. non-mergeable files
c) files in which RCS keywords should be expanded vs. left as they
   were at checkin

The issue of whether checkouts should be exclusive is orthogonal to
these factors, though in the case of non-mergeable files eidted
concurrently on two branches, you may actually have to go and talk to
the other person who modified the file (oh, horrors!  interpersonal
communication?) to discover what the committed version should be on
any given branch (i.e. after any 'manual merge').  

Sometimes the answer of course is 'none of the above'; then solution
is to do some rework.  This is the case which is problematic with CVS;
you only find out when you attempt the merge that this situation
exists; having CVS tip you off to this situation earlier might have
meant that you spoke to your colleague ealier on, and avoided the
rework.  

Having said this, I avoid all this heartache by not checking binary
files into CVS.  Although I do use CVS to control web content, I have
no 'large' (i.e. many files & developers) web projects.  Maybe if I
did I would care about non-mergeable files more.  But then, maybe I
wouldn't be using CVS for that.

-- 
James Youngman
Manchester, UK.  +44 161 226 7339
PGP (GPG) key ID for <address@hidden> is 64A95EE5 (F1B83152).