-
Notifications
You must be signed in to change notification settings - Fork 88
Contributing
We follow Ansible Code of Conduct in all our contributions and interactions within this repository.
Whether you are looking for an opportunity to contribute or you found a bug and already know how to solve it, please go to the issue tracker. There you can find feature ideas to implement, reports about bugs to solve, or submit an issue to discuss your idea before implementing it which can help choose a right direction initially and potentially save a lot of time and effort. Also somebody may already have started discussing or working on implementing the same or a similar idea, so you can cooperate to create a better solution together.
Look through currently open pull requests. You can help by reviewing them. Reviews help move pull requests to merge state. Some good pull requests cannot be merged only due to a lack of reviews. Also, consider taking up a valuable, reviewed, but abandoned pull request which you could politely ask the original authors to complete yourself. And it is always worth saying that good reviews are often more valuable than pull requests themselves.
For open questions, broad suggestions and other comments that won't typically fit in the scope of an issue or pull request, discussions are available. That section provides a place to have a more open and informal conversation about any and all things related to this collection, included but not limited to future development plans, functionality explanations and feature proposals that are not yet fleshed out enough for an issue of their own.
First, see the paragraphs above.
If you came up with a new feature, it is always worth creating an issue before starting to write code to discuss the idea with the community first.
We assume that you use Linux as a work environment (you can use a virtual machine as well) and have git
installed.
-
If possible, make sure that you have installed and started
docker
. While you can also run tests without docker, this makes it a lot easier since you do not have to install the precise requirements, and tests are running properly isolated and in the exact same environments as in CI. You often can also usepodman
with thedocker
executable shim, so if you have that you probably do not need to installdocker
. -
Clone the ansible-core repository:
git clone https://github.com/ansible/ansible.git
Instead of installing ansible-core from source, you can also work with an already existing installation of Ansible, ansible-base or ansible-core. Simply skip steps 2 and 3 in that case.
- Go to the cloned repository and prepare the environment:
cd ansible && source hacking/env-setup
pip install -r requirements.txt
cd ~
- Create the following directories in your home directory:
mkdir -p ~/ansible_collections/community/mysql
-
Fork the
community.mysql
repository through the GitHub web interface. -
Clone the forked repository from your profile to the created path:
git clone https://github.com/YOURACC/community.mysql.git ~/ansible_collections/community/mysql
If you prefer to use the SSH protocol:
git clone [email protected]:YOURACC/community.mysql.git ~/ansible_collections/community/mysql
- Go to your new cloned repository.
cd ~/ansible_collections/community/mysql
- Be sure you are in the main branch:
git status
- Show remotes. There should be the
origin
repository only:
git remote -v
- Add the
upstream
repository:
git remote add upstream https://github.com/ansible-collections/community.mysql.git
- Update your local
main
branch:
git fetch upstream
git rebase upstream/main
- Create a branch for your changes:
git checkout -b name_of_my_branch
- We recommend you start with writing integration tests if applicable.
Note: If there are any difficulties with writing the tests or you are not sure if the case can be covered, feel free to skip this step. If needed, other contributors can help you with it later.
All integration tests are stored in tests/integration/targets
subdirectories.
Go to the subdirectory containing the name of module you are going to change.
For example, if you are fixing the mysql_user
module, its tests are in tests/integration/targets/test_mysql_user/tasks
.
The main.yml
file holds test tasks and includes other test files.
Look for a suitable test file to integrate your tests or create and include a dedicated test file.
You can use one of the existing test files as a draft.
When fixing a bug, write a task which reproduces the bug from the issue.
Put the reported case in the tests, then run integration tests with the following command:
ansible-test integration test_mysql_user --docker -vvv
If the tests do not want to run, first, check you complete step 3 of this guide.
If the tests ran successfully, there are usually two possible outcomes: a) If the bug has not appeared and the tests have passed successfully, ask the reporter to provide more details. The bug can be not a bug actually or can relate to a particular software version used or specifics of local environment configuration.
b) The bug has appeared and the tests has failed as expected showing the reported symptoms.
-
Fix the bug.
-
Run
flake8
against a changed file. If it isplugins/modules/mysql_user.py
:
flake8 plugins/modules/mysql_user.py
It is worth installing and running flake8
against the changed file(s) first.
It shows unused imports, which is not shown by sanity tests (see the next step), as well as other common issues.
Optionally, you can use the --max-line-length=160
command-line argument.
- Run sanity tests:
ansible-test sanity plugins/modules/mysql_user.py --docker
If they failed, look at the output carefully - it is usually very informative and helps to identify a problem line quickly. Sanity failings usually relate to wrong code and documentation formatting.
- Run integration tests:
ansible-test integration test_mysql_user --docker -vvv
There are two possible outcomes: a) They have failed. Look at the output of the command. Fix the problem place in the code and run again. Repeat the cycle until the tests pass.
b) They have passed. Remember they failed originally? Our congratulations! You have fixed the bug.
- Commit your changes with an informative but short commit message:
git add /path/to/changed/file
git commit -m "mysql_user: fix crash when ..."
- Push the branch to the
origin
(your fork):
git push origin name_of_my_branch
-
Go to the
upstream
(http://github.com/ansible-collections/community.mysql). -
Go to
Pull requests
tab and create a pull request.
GitHub is tracking your fork, so it should see the new branch in it and automatically offer
to create a pull request. Sometimes GitHub does not do it and you should click the New pull request
button yourself.
Then choose compare across forks
under the Compare changes
title.
Choose your repository and the new branch you pushed in the right drop-down list.
Confirm. Fill out the pull request template with all information you want to mention.
Put "Fixes + link to the issue" in the pull request's description.
Put "[WIP] + short description" in the pull request's title. It's often a good idea to mention the name of the module/plugin you are modifying at the beginning of the description.
Click Create pull request
.
- Add a changelog fragment to the
changelog/fragments
directory. It will be published in release notes, so users will know about the fix.
Commit and push it:
git add changelog/fragments/myfragment.yml
git commit -m "Add changelog fragment"
git push origin name_of_my_branch
- The CI tests will run automatically on Red Hat infrastructure after every commit.
You will see the CI status in the bottom of your pull request. If they are green, remove "[WIP]" from the title. Mention the issue reporter in a comment and let contributors know that the pull request is "Ready for review".
-
Wait for reviews. You can also ask for review on IRC in the #ansible-community channel.
-
If the pull request looks good to the community, committers will merge it.
For details, refer to the Ansible developer guide.
If you find any inconsistencies or places in this document which can be improved, feel free to edit this wiki page or raise an issue to fix it.