On Mon, Mar 11, 2019 at 05:08:07PM -0500, Matt Zagrabelny wrote: > On Mon, Mar 11, 2019 at 2:12 PM Geert Stappers wrote: > > On Mon, Mar 11, 2019 at 01:50:49PM -0500, Matt Zagrabelny wrote: > > > > > > > > 2. What should the "master" branch be used for? > > > > Consider the string "master" a label for _your leading branch_ > > > > > > > I don't use the master branch with DEP-14. I believe the DEP is stating > > > that you'd use "master" for native packages - which from the sounds of it, > > > yours is not. Therefore, I'd not use "master". > > > > But you have your own leading branch > > > > When I am packaging someone else's software (upstream/latest) for inclusion > in Debian (debian/master), I don't feel like I have "my own" leading branch. > > What am I missing for using (or not using) a "master" branch?
Nothing (and nothing) The git term "clone" is a good term. Cloning is not just copying, is creating a child with the exact DNA of its parent. The new repository get its first branch named 'master'. Is it leading? It could. Git being a distributed VCS makes it confuesing for newcomers. We all start as child. There are parents. It takes a while before we become parents. With `git` it goes much faster. Upon clone there is a master. > > > > 3. When a new upstream tarball is released, where should it be imported? > > > > > > > > > > Assuming you have a remote named "github", I suppose you'd do something > > > like: > > > > > > git pull github upstream/latest > > > > > > I think it should be (be warned _not tested_ ) avoid that your > > current branch gets pollueted. > } git checkout upstream/latest > } git pull github > > > > What makes you believe that the current branch would get polluted? > > I believe a: > > git pull repo refspec > > is equivalent to: > > git checkout refspec > git pull repo > > Am I wrong? Don't know, _not tested_. But yes, It could work. I try (way too much) to be on the safe side. > Thanks for the dialog! YW MP Groeten Geert Stappers -- Leven en laten leven
