Community technical support mailing list was retired 2010 and replaced with a professional technical support team. For assistance please contact: Pre-sales Technical support via email to sales@march-hare.com.
On Tue, 31 May 2005 10:21:45 +0530, "Manjunath N.S." <manjunath.ns at gmail.com> wrote: >Hi, > >we have parallel development team, each works two different module, >each of these module has dependencies. So the main aim is if one >module breaks due compilation problem or some other thing, it should >not affect the other team. > >To acheive this what we have done is we take stable code from Main CVS >into our local CVS,and we start working this local CVS. If in between >from the Main CVS, if there is release is made, then those latest >changes from Main CVS we want to include into our Local CVS. > >hope this will be clear.. > >thanks >manju This is what branches are used for... Team1 works on branch Team1_FeatureX Team2 works on branch Team2_FeatureY Mainline contains only verified release code Whenever there is a verified release on any branch it gets merged to TRUNK and a release tag is set. At this time any of the teams that need to keep their code up-to-date with the releases merges the Release tag code back into their branch. So now they have a merge of their own work and the tested release code, which is their responsibility to verify and commit (on their branch). Why do you not consider this procedure??? It uses the functions of CVS that is there for this very purpose.... /Bo (Bo Berglund, developer in Sweden)