[bisq-network/bisq] (9/9) [PR COMMENTS] Split up P2PDataStoreTest (#3587)

[Julian Knutsen]

>From the author's side of things, each PR could have been independently merged and it would not have affected product stability. For cases like that AND if PRs are handled within a day or so then it allows iterative development and commits to master that can make smaller improvements that other devs will pull. You get some additional free "testing" although this should never be the sole reason to do it. The major benefit is any development on the same files or modules can be merged easier because you don't have 50 commits that may have merge conflicts.

If we expect longer cycles, I think just merging the last PR may make sense. This is similar to a feature branch merge, but you still have the benefit of going back to the second parent to see the individual commits in the event of bugs. With squash-merging, it all shows up as one huge changeset and it is hard to dig through.

After reading more about the different merge options available, I think I prefer merge-commits over squash-and-merge whenever possible. The ability to track a change back to the individual commits is useful for bisecting and understanding the original intention of a specific fix. If it is hard for devs to read `git log` I think `git log --first-parent --no-merges` will give you what you want. It just removes the second parent from the merge commits so you just see the individual PRs.

I'm interested to hear everyone else's thoughts on the benefits of squash-and-merge so I can see if that would make my workflow any better. There may be something I am missing since I don't have a ton of experience with Git.



-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/bisq-network/bisq/pull/3587#issuecomment-555127672
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bisq.network/pipermail/bisq-github/attachments/20191118/0c969a83/attachment.html>