info-cvs
[Top][All Lists]
Advanced

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

Re: Multilevel vendor branch import


From: Mark D. Baushke
Subject: Re: Multilevel vendor branch import
Date: Sun, 26 Oct 2003 12:59:48 -0800

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Rodolfo,

I have not tried to do it, but you might be able to use multiple vendor
branches by using the -b switch to import and then have the vendor be
LINUX_0_01, LINUX_0_10

That is, 

     cvs import -b1.1.1 module LINUX_0_01 LINUX-0_01_x_RELEASE
     cvs import -b1.1.3 module LINUX_0_10 LINUX-0_10_x_RELEASE
     cvs import -b1.1.5 module LINUX_0_10 LINUX-0_10_x_RELEASE
  ...
     cvs import -b1.1.127 module LINUX_2_7 LINUX-2_7_x_RELEASE

Do remember that the -b switch needs a list of odd numbers and the final
digit must also be odd. I suppose you could make it deeper for sub-releases
of each major release

     cvs import -b1.1.1.1.1 module LINUX_0_01 LINUX-0_01_0_RELEASE
     cvs import -b1.1.1.1.3 module LINUX_0_10 LINUX-0_10_1_RELEASE
     cvs import -b1.1.1.1.5 module LINUX_0_10 LINUX-0_10_2_RELEASE
...
     cvs import -b1.1.3.1.1 module LINUX_1_0 LINUX-1_0_0_RELEASE
...
     cvs import -b1.1.3.1.19 module LINUX_1_0 LINUX-1_0_19_RELEASE

It might work, but I can not guarantee it...

        Good luck,
        -- Mark

Rodolfo Schulz de Lima <address@hidden> writes:

> One of my uses of CVS is to record various releases of software, to see
> changes between them. For instance, is kind of interesting and educative to
> have all releases of linux kernel since linux-0.01 in a CVS repository so
> you can see how the kernel evolved through time. I try to keep the original
> dates, but the only way to do this is via import -d. The next problem: how
> to arrange the repository so that when a new release comes out, it's easy to
> import it in? I've come with the following arrangement:
> 
> LINUX    LINUX-x.x-BRANCH LINUX-x.x.x-BRANCH
> BRANCH
> 
> 0.01.x
> 0.10.x
> 0.11.x
>   .      
>   .
> 1.0.x     -    1.0.0
>                1.0.1
>                1.0.2
>                  .
>                  .
>                1.0.10
> 1.1.x
>   .
>   .
> 2.0.x
>   .
> 2.4.x    -     2.4.0
>                2.4.1
>                2.4.2     -   2.4.2-pre1
>                              2.4.2-pre2
>                              2.4.2-pre3
>                                  .
>                                  .
>                                2.4.2
> 
> ... and so forth. Just *after* thinking about all this, I discovered that I
> cannot import a version as a branch of one specific revision. All I wanted
> is something like this: cvs import -r LINUX-2_0_0 linux LINUX
> LINUX-2_0-BRANCH LINUX-2_0_1. This would create a branch off revision
> LINUX-2_0_0 named LINUX-2_0-BRANCH, and import the sources to there. In case
> of new files, dead revisions would be created as 1.1 and 1.1.1.1 (trunk
> 1.1.1 named LINUX) and the revision itself as 1.1.1.1.1.1 (trunk 1.1.1.1.1
> named LINUX-2_0-BRANCH), tagged as LINUX-2_0_1.
> 
> Questions: is there a better way to do it without altering the source code?
> Is my issue general enough to be in the next cvs' feature release? Creating
> dead revisions is really the best way of handling new files inside a
> sub-branch of a vendor branch?
> 
> I really am interested in getting this feature in cvs, and, being a C
> programmer, would like to help coding it if needed, as I feel kind of
> comfortable with how's cvs code structured.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (FreeBSD)

iD8DBQE/nDXE3x41pRYZE/gRAu3OAKC1Xyu2ZFqLe2bYLeGN2tQK0q2wdwCfRUe6
f6CmxlxW/v5zv00ENfiFSvo=
=sFQY
-----END PGP SIGNATURE-----




reply via email to

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