Thank you for your interest in contributing to our project! 💛
Whether it's a bug report, new feature, correction, or additional documentation, we greatly value feedback and contributions from our community. Please read through these guidelines carefully before submitting a PR or issue and let us know if it's not up-to-date (or even better, submit a PR with your proposed corrections 😉).
- Getting Started
- Pull Requests
- Bug Reports
- Commits
- Tests
- Debugging
- Code Style
- Finding Contributions
- Community
- Code of Conduct
- Security Issue Reporting
- Licensing
Our work is done directly on Github and PR's are sent to the github repo by core team members and contributors. Everyone undergoes the same review process to get their changes into the repo.
This section should get you running with Amplify CLI and get you familiar with the basics of the codebase.
-
Ensure you have Node.js 18 installed, which comes bundled with
npm
. Use it to install or upgradeyarn
:npm install --global yarn
If you are using Yarn v2, run
yarn set version classic
to change to Yarn Classic.Ensure that Yarn global bin is added to your PATH. For example, add
export PATH="$(yarn global bin):$PATH"
to your shell profile file on Linux or macOS. -
Ensure you have Java installed and
java
command is available in your system. This is required for DynamoDB emulator. -
Ensure you are using the npm registry, even with yarn by running
yarn config set registry https://registry.npmjs.org
-
Start by forking the dev branch of amplify-cli. Then clone it to your machine to work with it locally using one of the following methods:
# HTTPS git clone https://github.com/[username]/amplify-cli.git # SSH git clone [email protected]:[username]/amplify-cli.git # GitHub CLI gh repo clone [username]/amplify-cli
-
Move into your project folder:
cd amplify-cli
-
Then, you can run the
setup-dev
script, which installs dependencies and performs initial configuration:# Linux / macOS yarn setup-dev # Windows yarn setup-dev-win ## Must be run in Powershell
NOTE: Make sure to always sync your fork with dev branch of amplify-cli
Amplify CLI is a monorepo built with Yarn Workspaces and Lerna. All the categories of Amplify live within the packages/
directory in the root. Each category inside packages has its own src/
and package.json
.
Packages inside Amplify CLI Monorepo
- To make changes with respect to a specific category, go into
packages/[category]
. - Make changes to required file.
- Write unit tests
- Yarn build
- Run test suite
- Test in sample app using amplify-dev
- Submit a PR
For a long time, the codebase had relatively lax lint checking. We have now added more strict rules but decided that it wasn't feasible to update all the code to adhere to the new rules at once. Instead we have opted for an iterative approach where lint errors are fixed as files are touched. If you are the first person to touch a file since the rules have been inforced we ask that you try your best to address the lint errors in that file. If addressing an error would significantly increase the scope of the change, it is okay to add a lint disable comment and a justification in the PR description.
To get lint warnings as you type, configure the ESLint VSCode plugin. Alternatively, run yarn lint-fix
to auto-fix errors where possible
and print out errors that need manual attention.
Pull requests are welcome!
You should open an issue to discuss your pull request, unless it's a trivial change. It's best to ensure that your proposed change would be accepted so that you don't waste your own time. If you would like to implement support for a significant feature that is not yet available, please talk to us beforehand to avoid any duplication of effort. Additionally, please be mindful of the length of the pull request - if your change requires more than 12 file changes, consider breaking the change down into smaller, non-dependent changes. This includes any changes that may be added as a result of the linter.
Pull requests should be opened against dev.
Don't include any build files i.e. dist/
in your PR. These will be built upon publish to npm and when a release is created on GitHub.
If the change is a breaking change (as defined by semantic versioning), please add the semver-major
label to your pull request on GiHub.
- Go through the Local Environment Setup
- Within your local fork, create a new branch based on the issue you're addressing - e.g.
git checkout -b category-auth/admin-auth-support
- Use grouping tokens at the beginning of the branch names. For e.g, if you are working on changes specific to
amplify-category-auth
, then you could start the branch name ascategory-auth/...
- Use slashes to separate parts of branch names
- Use grouping tokens at the beginning of the branch names. For e.g, if you are working on changes specific to
- Once your work is committed and you're ready to share, run
yarn test
. Manually test your changes in a sample app with different edge cases and also test across different platforms if possible. - Run
yarn lint-fix
to find and fix any linting errors - Run
yarn prettify:changes
to fix styling issues - Then, push your branch:
git push origin HEAD
(pushes the current branch to origin remote) - Open GitHub to create a PR from your newly published branch. Fill out the PR template and submit a PR.
- Finally, the Amplify CLI team will review your PR. Add reviewers based on the core member who is tracking the issue with you or code owners. In the meantime, address any automated check that fail (such as linting, unit tests, etc. in CI)
Bug reports and feature suggestions are always welcome. Good bug reports are extremely helpful, so thanks in advance!
When filing a bug, please try to be as detailed as possible. In addition to the bug report form information, details like these are incredibly useful:
- A reproducible test case or series of steps
- The date/commit/version(s) of the code you're running
- Any modifications you've made relevant to the bug
- Anything unusual about your environment or deployment
Guidelines for bug reports:
- Check to see if a duplicate or closed issue already exists!
- Provide a short and descriptive issue title
- Remove any sensitive data from your examples or snippets
- Format any code snippets using Markdown syntax
- If you're not using the latest version of the CLI, see if the issue still persists after upgrading - this helps to isolate regressions!
Finally, thank you for taking the time to read this, and taking the time to write a good bug report.
Commit messages should follow the conventional commits specification. For example:
git commit -m 'docs(cli): correct spelling of CHANGELOG'
Valid commit types are as follows:
build
chore
ci
docs
feat
fix
perf
refactor
style
test
You will notice the extra actions carried out when you run the git commit
or git push
commands on this monorepo, that's because the following git hooks are configured using husky (you can see them in the root package.json file):
"husky": {
"hooks": {
"commit-msg": "commitlint -E HUSKY_GIT_PARAMS",
"pre-push": "yarn verify-api-extract && yarn build-tests-changed && yarn split-e2e-tests",
"pre-commit": "yarn verify-commit"
}
}
NOTE: To ensure those git hooks properly execute, run
yarn
ornpm install
at the root of this monorepo to install the necessary dev dependency packages.
The "commit-msg" hook ensures the commit message follows the Conventional Commits convention, so that proper CHANGELOG.md files and package versions are maintained.
The "pre-commit" hook runs the verify-commit script and runs eslint of changed files.
The "pre-push" hook will build test files and run the split-e2e-tests
script to ensure the correct configuration file is generated for our CI/CD workflow.
Please ensure that your change still passes unit tests. It's OK if you're still working on tests at the time that you submit, but be prepared to be asked about them. Pull requests should contain tests as appropriate. Bugfixes should contain tests that exercise the corrected behavior (i.e., the test should fail without the bugfix and pass with it), and new features should be accompanied by tests exercising the feature.
Unit tests should be located under .../src/__tests__/
with the expectation that the directory tree under __tests__
mirrors that of src
. In general it is expected that unit tests take the name of the file they test. For example:
File:
amplify-category-function/src/provider-utils/service-walkthroughs/execPermissionsWalkthough.ts
Unit Tests:
amplify-category-function/src/__tests__/provider-utils/service-walkthroughs/execPermissionsWalkthrough.test.ts
To run the test suite:
yarn test
Or, to run the tests for a specific package,
cd packages/amplify-category-function
yarn test
A preferred workflow is to watch tests while writing code. For example, you can open a terminal in the directory of the package you are updating and watch tests in that package.
cd packages/amplify-category-function
yarn test --watch
Using the watch
option, the unit tests will re-run every time a change is made to the code.
Amplify CLI uses Jest for testing. See the Jest Documentation for more information.
End-to-end tests can be found in packages/amplify-e2e-tests
. It is not recommended to run all of these tests at a time but to, instead, run the tests in a single file in order to debug, fix, or update it.
You can run an end to end test with the following:
cd packages/amplify-e2e-tests
yarn e2e src/__tests__/some_e2e_test.test.ts
You can also run a specific test from a file using:
yarn e2e src/__tests__/some_e2e_test.test.ts -t name-of-test
End to end tests generally provision real resources and attempt to tear them down again after the test has run. If tests are interrupted due to manual intervention or some other issue, resources may stick around and require manual removal.
Code coverage can be seen by running all tests, yarn test
, then running yarn coverage:collect
.
The coverage is collected in the coverage/
file at the root of the project.
The coverage can be viewed in a browser by opening coverage/lcov-report/index.html
,
or in IDE tools utilizing coverage/lcov.info
.
To test your category in sample application, do the following:
cd <your-test-front-end-project>
amplify-dev init
amplify-dev <your-category> <subcommand>
Sometimes issues can be solved by doing a clean and fresh build. To start from a clean project state:
-
Removes ./lib, tsconfig.tsbuildinfo, and node_modules from all packages and run their clean script:
yarn clean
-
Remove all unstaged changes and everything listed in the .gitignore file:
git clean -fdx
-
Reset dev branch to that of origin/dev:
git fetch origin && git checkout --track origin/dev -B dev
-
Then, run the
setup-dev
script:# Linux / macOS yarn setup-dev # Windows yarn setup-dev-win
If you are using Visual Studio Code, the built-in Javascript Debug Terminal is useful for performing runtime debugging.
In order to use the terminal, build the application. Then, execute the built binary, amplify-dev
, from the Javascript Debug Terminal. See VSCode's documentation
for more information.
Generally, match the style of the surrounding code. Please ensure your changes don't wildly deviate from those rules. You can run yarn lint-fix
to identify and automatically fix most style issues.
Looking at the existing issues is a great way to find something to contribute on. As our projects, by default, use the default GitHub issue labels (enhancement/bug/duplicate/help wanted/invalid/question/wontfix), looking at any help-wanted
or good first issue
is a great place to start.
You could also contribute by reporting bugs, reproduction of bugs with sample code, documentation and test improvements.
Join the Discord Server. If it's your first time contributing, checkout the #contributing
channel.
This project has adopted the Amazon Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact [email protected] with any additional questions or comments.
If you discover a potential security issue in this project we ask that you notify AWS/Amazon Security via our vulnerability reporting page. Please do not create a public GitHub issue.
AWS Amplify CLI is Apache 2.0-licensed. Contributions you submit will be released under that license.
We may ask you to sign a Contributor License Agreement (CLA) for larger changes.