Closed ben-heil closed 5 years ago
At this point it may make sense to also move the py2neo imports out the top-level module so we don't burden users with this dependency unless they need it. Perhaps tqdm as well.
Sorry for the commit naming weirdness, I squashed too many commits together, which caused me to have to merge with upstream despite not having new changes there. Is it better to leave all the commits as individuals when creating the pull request, then squash them at the end?
I reverted my master branch to only contain the changes related to py2neo, and added the rest of the requested changes in the latest pull request.
Added the requested changes, sorry that I missed the whitespace changes in git diff earlier
@ben-heil great. So now I've merged this, you should configure hetio/hetio
as the upstream
remote for your local repository. You should then change your master branch to be identical to hetio/hetio:master
. For future pull requests, make a new branch with your commits rather than making your commits on your master branch. For each PR, you should have a different local branch. Make sure the PRs are build on top of commit https://github.com/hetio/hetio/commit/7280c5804a809354ba13efdb153bffdfa2361ac2 (see git log
).
@ben-heil great. So now I've merged this, you should configure
hetio/hetio
as theupstream
remote for your local repository. You should then change your master branch to be identical tohetio/hetio:master
. For future pull requests, make a new branch with your commits rather than making your commits on your master branch. For each PR, you should have a different local branch. Make sure the PRs are build on top of commit 7280c58 (seegit log
).
That makes sense, will do
Addressing #31
This should make it possible to import neo4j.py regardless of which py2neo version you're using. I looked at the documentation for version 4 and the functions that actually use py2neo should still work fine. I'd write tests for them to make sure that's the case, but they all modify or delete information in existing neo4j databases. It should be possible to spin up a neo4j database at test time if we want to take that route.