I totally agree, Arkady.  That's why I propose using branches.

Arkady Kanevsky (arkady_kanev...@dell.com) wrote:
> We have a long term dependency on existing CB-1.x code base structure and 
> tools.
> All the products released to date depends on it and must be reproducible.
> 
> 
> From: crowbar-bounces On Behalf Of Terpstra, John
> Sent: Wednesday, December 18, 2013 9:29 AM
> To: aspi...@suse.com; crowbar
> Subject: Re: [Crowbar] Rationalizing Crowbar Project Infrastructure for our 
> new code base
> 
> Unfortunately, our internal infrastructure makes use of DevTool to build and 
> test CB1.x - if we make the changes Adam suggests will take only 60 sec, it 
> will take several engineers from our team to update/change/fix the DevTool - 
> and all the Jenkins build systems we use, and will impose a load on our QA 
> team that we do not have resources to handle.
> By splitting off the CB2.0 code into a new repo, we have zero impact on the 
> CB1.x tree upon which we will remain dependent for a number of years into the 
> future.  We do not want to impact any responsibilities we have to meet change 
> and build requirements in the CB1.x code base.
> Having CB2.0 code totally separate from CB1.x means that we all have a green 
> field to work from.
> 
> -          John T.
> 
> -----Original Message-----
> From: crowbar-bounces On Behalf Of Adam Spiers
> Sent: Wednesday, December 18, 2013 9:16 AM
> To: crowbar
> Subject: Re: [Crowbar] Rationalizing Crowbar Project Infrastructure for our 
> new code base
> 
> Victor Lowther (victor.lowt...@gmail.com<mailto:victor.lowt...@gmail.com>) 
> wrote:
> > I do not want to go back to having branches in the top-level repo -- I
> > strongly prefer that it only hold release-independent (and therefore
> > non-branch-needing) things, and pull in other repos that are branched
> > based on release.
> 
> We agree (as we already did in Austin) that the long term solution is to 
> split release-dependent and release-independent code into different 
> repositories.  But this is useless as a short term solution - as has already 
> been pointed out, there is not spare resource or energy to do that any time 
> soon.
> 
> Fortunately the workaround is ridiculously easy, which is why I have been 
> proposing it again and again until I am blue in the face:
> 
> *immediately* branch the old releases in the main repo.  I estimate I could 
> do this in about 60 seconds:
> 
>     cd ~/crowbar
>     for release in pebbles roxy whatever else we care about; do
>       git checkout -b release/$release/master
>       git push crowbar HEAD
>     done
> 
> This would allow people who only care about 2.0 to hack away to their heart's 
> content in the master branch of the main repo, without the slightest fear of 
> breaking *anything* for the people who care about 1.x.
> 
> Then any release-dependent code can be moved from master to other 
> repositories at our leisure, with no time pressure, and none of the awful 
> problems which the lack of branches in the main repo has been causing for the 
> last ~13 months.
> 
> _______________________________________________
> Crowbar mailing list
> Crowbar@dell.com<mailto:Crowbar@dell.com>
> https://lists.us.dell.com/mailman/listinfo/crowbar
> For more information: http://crowbar.github.com/

_______________________________________________
Crowbar mailing list
Crowbar@dell.com
https://lists.us.dell.com/mailman/listinfo/crowbar
For more information: http://crowbar.github.com/

Reply via email to