Re: branching / cvs question

[Kaz Kylheku]

On Mon, 7 Apr 2003, LN Cisneros wrote:

> 1) There is an xf-3_2A-branch and a xf-3_3-branch.  The both seem to
> be branched from the 1.1.1.1 version of the X11R6 branch.  Why do they
> both have their changes versioned the exact same, i.e. they both have a
> "magic" branch number of 1.1.1.1.0.4?  Shouldn't one of them be
> 1.1.1.1.0.6?

That is correct; each branch should have its own branch number within
each file. The last digit of the branch number should be even (except
for vendor branches). Typically the numbers start at 2 and then
increment from there through the remaining even positive integers.

Assuming that the duplication of 1.1.1.1.0.4 was not caused by some bug
in CVS, the only possible reason can be that someone used cvs admin
to create an alias symbol.

The two symbols identify the same branch, and consequently if two
sandboxes each check out these branches, they get the same thing, and
can see each other's commits---at least for this file!

> 2)  Is the XF11R6 a vendor branch?
> 
> 3)  What do all those "symbols" mean?  How do you tell if they are a
> branch name or not? Will a branch always point to a "magic" branch
> number?

A symbol names a branch revision if that revision ends with .0.x  . In
some contexts, the .0. component is omitted; in that case a branch
number is distinguished by having an odd number of components, like
1.3.2 (odd number of digits) which is the same as 1.3.0.2 (zero in the
second-to-last position).

> 3)  If changes are merged from a branch back into main, do cvs track this?

No; CVS users must come up with a tracking convention and use explicit
CVS commands to record it. The manual suggests some ideas.

-- 
Meta-CVS: directory structure versioning; versioned symbolic links;
versioned execute permission; versioned property lists; easy branching
and merging and third party code tracking; all implemented over the
standard CVS command line client -- http://freshmeat.net/projects/mcvs