On Tue, May 20, 2014 at 7:00 PM, Pierre-François CLEMENT <lik...@gmail.com> wrote: > Okay then, in that case what you need is the --squash option of git-merge. > >> --squash, --no-squash
Interesting idea, but I don't think it will work, since I don't want any of the build products from the last commit on the deploy branch. Imagine master starts with a.c and b.c, then a.c is removed, and d.c is added: master: A (a.c, b.c) -> B(b.c, d.c) What I want is for a deploy to look like: deploy: A1 (a.c,b.c) -> A2 (a.c, b.c, a.o, b.o) -> B1 (b.c,d.c) -> B2(b.c, d.c, b.o, d.o) or preferably just: deploy: A* (a.c, b.c, a.o, b.o) -> B*(b.c, d.c, b.o, d.o) (above is easy, its just a squash of the pre-built source with the post-built source). If I squash merged B into A*, I think the files ADDed into the deploy branch A* would not get removed, I'd have: bad-deploy: A2 -> B*(b.c, d.c, a.o, b.o) Where a.o is build file for a source file that doesn't exit anymore. I don't want anything from the previous commit, and everything from the next. I think I'll read carefully through this: http://git-scm.com/book/en/Git-Internals-Git-Objects it looks like I made be able to read/write commits constructed as I wish. Sam Btw, thanks for the suggestions, I appreciate it. -- You received this message because you are subscribed to the Google Groups "Git for human beings" group. To unsubscribe from this group and stop receiving emails from it, send an email to git-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
