Re: https://github.com/wocommunity/wonder planned renaming "integration" to "master"
Kieran Kelleher <[email protected]> Tue, 24 Sep 2013 12:26:51 -0400
| Newsgroups | gmane.comp.web.webobjects.wonder-disc |
|---|---|
| Message-ID | <[email protected]> |
OTOH, should be sth like this...... # checkout a local branch based on upstream master git checkout -b up/master upstream/master # Replace your own master with the upstream ref git push -f origin up/master:master # Go into your own GH wonder and set master to be the default branch on your fork # Remove the dead branches in your own GH (only master will remain which is all you probably care about) git push origin :integration git push origin :patrice git push origin :Wonder_1_0_0_Branch git push origin :Wonder_1_0_1_Branch On Sep 24, 2013, at 10:17 AM, David LeBer <[email protected]> wrote: > Kieran, > > For us git noobs. If we have: > > local > origin > upstream > > With origin being a fork of the repo on Githup. It is now full of all of the old branches: > > -master > -integration > -patrice > -Wonder_1_0_0_Branch > -Wonder_1_0_1_Branch > ... > > Versus the current (upstream) repo > > -master > -Wonder_5_0_0_Legacy > -wonder_5 > > Is there an easy way to prune the old branches from our forks of wonder to match it's current state? > > D > > -- > David LeBer > Codeferous Software > > On 2013-09-24, at 9:42 AM, Kieran Kelleher <[email protected]> wrote: > >> It is done. >> >> Goodbye "integration" branch. >> >> If you are not sure what to do you can update your remote references with: >> >> $ git fetch --all --prune >> >> >> Regards, Kieran >> >> >> On Sep 13, 2013, at 9:05 PM, Paul Hoadley <[email protected]> wrote: >> >>> On 14/09/2013, at 3:15 AM, Ken Anderson <[email protected]> wrote: >>> >>>> Wasn't the purpose of integration to validate compatibility between changes before releasing to master? Or maybe I'm misunderstanding the statement… >>> >>> You're right—that was exactly the original intention of the integration and master branches. I think the reality turned out to be that we just don't have the community size to support that kind of model. It's only useful if a critical mass of people are able to build, run and test the integration branch as changes are made. The cost of the additional release engineering outweighed the benefit of the two branch workflow. And I'm saying all this as a _big_ fan of that workflow. I'm sad to see the integration branch go. But if it's not working, it's not working. >>> >>> >>> -- >>> Paul Hoadley >>> http://logicsquad.net/ >>> >>> >>> >> >> ------------------------------------------------------------------------------ >> October Webinars: Code for Performance >> Free Intel webinars can help you accelerate application performance. >> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from >> the latest Intel processors and coprocessors. See abstracts and register > >> http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk_______________________________________________ >> Wonder-disc mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/wonder-disc > ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk