On Thu, Aug 28, 2014 at 3:31 PM, Marko Rauhamaa <ma...@pacujo.net> wrote: > I'd venture to say files are quite independent most of the time. That's > why such merges have been facilitated to the point that negates the > "feature" you mentioned. Nobody cares to take the trouble of analyzing > the validity of automatic merges.
Okay. Imagine a big C program (hundreds of .c files) and the latest build segfaults on certain input. You know that 60 commits ago it wasn't segfaulting. Now, you have two options: 1) Work with the repo as an entire whole, and bisect those 60 commits in about six steps, to find when the segfault began. 2) Look through which files have been changed, which might be a couple dozen, and treat each file's changes separately. I know which I'd rather do. The project is one thing, regardless of how many files it is, and a successful build might not even be possible if you get mismatched versions. I call it a feature. ChrisA -- https://mail.python.org/mailman/listinfo/python-list
