Comments on: Squash-merging and other problems with GitHub https://blog.adamspiers.org/2017/08/16/squash-merging-and-other-problems-with-github/ because there's always something more interesting than what you should be doing Tue, 08 Oct 2019 23:48:06 +0000 hourly 1 https://wordpress.org/?v=7.0.2 By: Anne Archibald https://blog.adamspiers.org/2017/08/16/squash-merging-and-other-problems-with-github/#comment-1840829 Tue, 08 Oct 2019 23:48:06 +0000 http://blog.adamspiers.org/?p=1534#comment-1840829 One reason I end up with messy commits is that I am pushing PRs to projects that run on Windows and my only way of testing on Windows is through their CI systems. This is less terrible than it sounds but certainly far from ideal. More commonly I have something that passes tox on my local machine but fails to build on CI for some reason. Debugging via PR-to-CI is awful but I don’t have ready access to a better method for things that pass tox on the machines I have access to.

]]> By: Adam https://blog.adamspiers.org/2017/08/16/squash-merging-and-other-problems-with-github/#comment-1808843 Fri, 18 Aug 2017 15:31:27 +0000 http://blog.adamspiers.org/?p=1534#comment-1808843 In reply to Christof Damian.

That plus-pull tool looks cool, thanks! I’ll see if it makes sense for our team to use it.

I agree about having dirty history rather than none, although even with squash merge you still get *some* history 🙂

Also totally agree about force pushes to GitHub making it hard to figure out which bits need re-reviewing, this is SUPER annoying as I previously blogged about: https://blog.adamspiers.org/2015/03/24/why-and-how-to-correctly-amend-github-pull-requests/
However I think this could be largely addressed if https://github.com/isaacs/github/issues/999 was resolved, and as you can see from the comments there, it’s not necessarily that hard to implement a solution/workaround.

]]> By: Christof Damian https://blog.adamspiers.org/2017/08/16/squash-merging-and-other-problems-with-github/#comment-1808841 Fri, 18 Aug 2017 09:27:59 +0000 http://blog.adamspiers.org/?p=1534#comment-1808841 In our team we also have a desire for either squashes or forced push. I even implemented the support for different merge modes in the tool we use to merge pull requests: https://github.com/christofdamian/plus-pull

I am not a fan of either. I rather have a dirty history than no history at all.

The problem with forced pushes in pull requests for me is that I might have requested a change to a pull request. When this is then force pushed I have to read the whole patch again instead of just checking the most recent commit(s).

Somehow I don’t see a good solution for this anytime soon.

]]>