There are roughly two roles in which you can contribute:
- Contributors who just want to add changes from time to time
- Maintainers who oversee the repo, code themselves and review contribution before they can be merged
The community can be found on discord. Nothing beats a good discussion about existing features, new features or ideas so jump right in.
We are all learners, some in the early stage some in the later. In a code review, we take the patience to imaginarily sit down together and explain other people why something is recommended differently or how things are usually done in software or python. This often seems tedious at first but growing together is important for any kind of project which wants to grow itself. So no fear in case of lack of experience but bring your learning spirit. Samewise to any experienced developer, have patience and explain things. Take the opportunity to sit down together on discord.
Tl;dr;
- Fork the open lasso python repository
- Clone the repo to your filesystem
- Install task
- Check out the
develop
branch - Set up the repo for development through
task setup
- Create a new branch from
develop
with the naming patternfeature/...
- Make changes, commit and push them
- Create a Pull Request in your for repo with target on the original repo
- Add as reviewer
open-lasso-python/developers
- Wait for review patiently but you may nudge us a bit 🫶
- Perform a Squash Merge and give a reasonable commit message as
<branch type>: <description>
wherebranch_type
is one of the categories below.
You can fork the repo (1) by clicking on for in the top-right of the original
repo.
Cloning the repo (2) is traditionally done with git then of course.
Task is required (3) since it is used to store complex commands such as testing,
linting, build docs, etc.
(4) All development activities originate from the develop
branch in which all
Pull Requests are finally merged again.
To create a branch choose a respective naming pattern following the angular
scheme: <branch type>/<issue nr if exists>-<rough description/name>
.
Typical branch types are:
feature
for new features or if you got no clue what it isbugfix
for 🐛-fixesci
for changes on the Continuous Integration pipelinedocs
for documentation related worksrefactor
if the PR just does cleanup 🧹 and improves the codetest
for solely test-related work
Don't take these too seriously but they ought to provide rough categories.
They are also used in the commit message when you squash merge a PR where it
is important!
Install all dependencies otherwise obviously you can't code (5).
After making changes and pushing your branch to your forked repo (7 & 8), open a
Pull Request but make the target not develop
in your fork but develop
in the
original repo (7).
If not done automatically, add the maintainer group as reviewers (9).
Enjoy a healthy code review but be a bit patient with time as people contribute
voluntarily and may privately be occupied (10).
After approval, perform a Squash Merge (11).
A Squash Merge tosses away all the little, dirty commits we all do during
development.
What stays is the final merge commit message and please pay attention here
to format it right.
Why is this important?
This is needed to automatically generate a reasonable changelog during releases.
Thanks for contributing at this point.
Go wild and have fun 🥳