7 comments

[ 2.8 ms ] story [ 24.2 ms ] thread
Why subtree over submodule?
Most of it is a personal preference. I'd rather pull the full repo of my dependencies in rather than "deal" with submodules. Submodules add some extra necessary workflow steps that others contributing to your repo might not be expecting. Subtree simplifies a lot of that.

As well, I think it helps to solidify the "reproducible build" concept by taking submodules 1 step further.

Have you ever tried using submodules, in a repository that multiple people commit to?
Yeah, it was extremely frustrating for actively changing code and not an issue for third party dependencies. I wonder what the trade offs between submodules and subtrees are for this use case.
submodule can only track a revision, not a branch.

(at least in the git versions distributed with centos/redhat6; I believe later versions can)

That actually seems like a benefit to me. Pinned versions are helpful for build/test reliability.