Multiple patch queues

[Bela Babik]

Another solution can be:

- you can divide you series file into sections (and use guards):
http://hgbook.red-bean.com/hgbookch13.html#x17-31000013.7

- use a central repo for patches, and use directories in the patch repo:
http://hgbook.red-bean.com/hgbookch13.html#x17-31300013.9

"hg qselect" does have a --pop option, which could remove all testing
patches for you.


On Sun, May 4, 2008 at 11:18 PM, Paul Moore <p.f.moore at gmail.com> wrote:
> 2008/5/4 Peter Arrenbrecht <peter.arrenbrecht at gmail.com>:
>
> >  Maybe it would work a little better if you cloned the windows queue as
>  >  your queue:
>  >
>  >   hg clone /the/windows/queue .hg/patches
>  >
>  >  and then added a guard "win" to all the patches in that queue (easiest
>  >  by editing series directly). Then you start adding your own patches on
>  >  top.
>
>  Yes, that sounds plausible.
>
>  What I've started doing is having 2 repos - dev and test. I manage my
>  patches in dev, and in test I have the windows testing patches
>  applied. Then when I want to test, I apply all my patches in dev, cd
>  test, hg qpop -a, hg pull ../dev, hg qpush -a. The only thing I need
>  to worry about then is when I edit a patch - in that case I need to hg
>  strip in test to remove the old version, before I pull.
>
>  Once again, I find that Mercurial offers me the pieces I need to build
>  lots of different ways of working, but leaves the choices up to me.
>  This isn't a complaint at all - it's one of the things I like about
>  Mercurial - but maybe a "best practices" page on the Wiki would be
>  useful to capture some of these patterns? (Off I go to add that to my
>  ever-growing list of things to do :-))
>
>  Thanks,
>  Paul.
>