Re: How does this work?

[Larry Jones]

address@hidden writes:
> 
> Lets say I IMPORT an old version of a code.
> I make some changes.
> Then I IMPORT a newer version over the top of it with new release_tag.

That's not how it works.  When you import a version, it goes on the
vendor branch.  If you then make changes, you would typically commit
those changes on the trunk.  You now have two separate lines of
development: the vendor branch (which will track vendor changes) and the
trunk (which will track local changes).  When you import a new version,
it also goes on to the vendor branch.  CVS notices that you have local
changes on the trunk and advises you to merge the vendor's changes into
your local version -- this requires separate CVS commands and may
require you to merge changes.  Typically, though, you just do special
form of checkout which causes CVS to merge the vendors changes into your
local version automatically, then you commit the local version just as
if you have changed it manually.

So, the key point is that you never change OLD or NEW -- you change your
local version that is based on them.

-Larry Jones

If I was being raised in a better environment, I wouldn't
do things like that. -- Calvin