First of all, thank you for your interest in Puppeteer! We'd love to accept your patches and contributions!
Contributions to this project must be accompanied by a Contributor License Agreement. You (or your employer) retain the copyright to your contribution, this simply gives us permission to use and redistribute your contributions as part of the project. Head over to https://cla.developers.google.com/ to see your current agreements on file or to sign a new one.
You generally only need to submit a CLA once, so if you've already submitted one (even if it was for a different project), you probably don't need to do it again.
- Clone this repository
git clone https://github.com/GoogleChrome/puppeteer
cd puppeteer
- Install dependencies
npm install
- Run Puppeteer tests locally. For more information about tests, read Running & Writing Tests.
npm run unit
All submissions, including submissions by project members, require review. We use GitHub pull requests for this purpose. Consult GitHub Help for more information on using pull requests.
- Coding style is fully defined in .eslintrc
- Code should be annotated with closure annotations.
- Comments should be generally avoided. If the code would not be understood without comments, consider re-writing the code to make it self-explanatory.
To run code linter, use:
npm run lint
When authoring new API methods, consider the following:
- Expose as little information as needed. When in doubt, don’t expose new information.
- Methods are used in favor of getters/setters.
- The only exception is namespaces, e.g.
page.keyboard
andpage.coverage
- The only exception is namespaces, e.g.
- All string literals must be small case. This includes event names and option values.
- Avoid adding "sugar" API (API that is trivially implementable in user-space) unless they're very demanded.
Commit messages should follow the Semantic Commit Messages format:
label(namespace): title
description
footer
- label is one of the following:
fix
- puppeteer bug fixes.feat
- puppeteer features.docs
- changes to docs, e.g.docs(api.md): ..
to change documentation.test
- changes to puppeteer tests infrastructure.style
- puppeteer code style: spaces/alignment/wrapping etc.chore
- build-related work, e.g. doclint changes / travis / appveyor.
- namespace is put in parenthesis after label and is optional. Must be lowercase.
- title is a brief summary of changes.
- description is optional, new-line separated from title and is in present tense.
- footer is optional, new-line separated from description and contains "fixes" / "references" attribution to github issues.
- footer should also include "BREAKING CHANGE" if current API clients will break due to this change. It should explain what changed and how to get the old behavior.
Example:
fix(page): fix page.pizza method
This patch fixes page.pizza so that it works with iframes.
Fixes #123, Fixes #234
BREAKING CHANGE: page.pizza now delivers pizza at home by default.
To deliver to a different location, use "deliver" option:
`page.pizza({deliver: 'work'})`.
All public API should have a descriptive entry in docs/api.md
. There's a documentation linter which makes sure documentation is aligned with the codebase.
To run the documentation linter, use:
npm run doc
For all dependencies (both installation and development):
- Do not add a dependency if the desired functionality is easily implementable.
- If adding a dependency, it should be well-maintained and trustworthy.
A barrier for introducing new installation dependencies is especially high:
- Do not add installation dependency unless it's critical to project success.
- Every feature should be accompanied by a test.
- Every public api event/method should be accompanied by a test.
- Tests should be hermetic. Tests should not depend on external services.
- Tests should work on all three platforms: Mac, Linux and Win. This is especially important for screenshot tests.
Puppeteer tests are located in test/test.js
and are written with a TestRunner framework.
Despite being named 'unit', these are integration tests, making sure public API methods and events work as expected.
- To run all tests:
npm run unit
- To run tests in parallel, use
-j
flag:
npm run unit -- -j 4
- To run tests in "verbose" mode or to stop testrunner on first failure:
npm run unit -- --verbose
npm run unit -- --break-on-failure
- To run a specific test, substitute the
it
withfit
(mnemonic rule: 'focus it'):
...
// Using "fit" to run specific test
fit('should work', async function({server, page}) {
const response = await page.goto(server.EMPTY_PAGE);
expect(response.ok).toBe(true);
});
- To disable a specific test, substitute the
it
withxit
(mnemonic rule: 'cross it'):
...
// Using "xit" to skip specific test
xit('should work', async function({server, page}) {
const response = await page.goto(server.EMPTY_PAGE);
expect(response.ok).toBe(true);
});
- To run tests in non-headless mode:
HEADLESS=false npm run unit
- To run tests with custom Chromium executable:
CHROME=<path-to-executable> npm run unit
- To run tests in slow-mode:
HEADLESS=false SLOW_MO=500 npm run unit
- To debug a test, "focus" a test first and then run:
node --inspect-brk test/test.js
Every public API method or event should be called at least once in tests. To ensure this, there's a coverage
command which tracks calls to public API and reports back if some methods/events were not called.
Run coverage:
npm run coverage
See Debugging Tips in the readme.
Releasing to npm consists of the following phases:
- Source Code: mark a release.
- Bump
package.json
version following the SEMVER rules, runnpm run doc
to update the docs accordingly, and send a PR titled'chore: mark version vXXX.YYY.ZZZ'
(example). - Make sure the PR passes all checks.
- WHY: there are linters in place that help to avoid unnecessary errors, e.g. like this
- Merge the PR.
- Once merged, publish the release notes using GitHub's "draft new release tag" option.
- NOTE: tag names are prefixed with
'v'
, e.g. for version1.4.0
the tag isv1.4.0
. - For the "raw notes" section, use
git log --pretty="%h - %s" v1.19.0..HEAD
.
- NOTE: tag names are prefixed with
- Update the “Releases per Chromium Version” list in
docs/api.md
to include the new version.
- Bump
- Publish
puppeteer
to npm.- On your local machine, pull from upstream and make sure the last commit is the one just merged.
- Run
git status
and make sure there are no untracked files.- WHY: this is to avoid adding unnecessary files to the npm package.
- Run
npx pkgfiles
to make sure you don't publish anything unnecessary. - Run
npm publish
. This publishes thepuppeteer
package.
- Publish
puppeteer-core
to npm.- Run
./utils/prepare_puppeteer_core.js
. The script changes the name insidepackage.json
topuppeteer-core
. - Run
npm publish
. This publishes thepuppeteer-core
package. - Run
git reset --hard
to reset the changes topackage.json
.
- Run
- Source Code: mark post-release.
- Bump
package.json
version to-post
version and send a PR titled'chore: bump version to vXXX.YYY.ZZZ-post'
(example)- NOTE: make sure to update the "released APIs" section in the top of
docs/api.md
by runningnpm run doc
. - NOTE: no other commits should be landed in-between release commit and bump commit.
- NOTE: make sure to update the "released APIs" section in the top of
- Bump
For both puppeteer
and puppeteer-core
we maintain the following npm tags:
chrome-*
tags, e.g.chrome-75
and so on. These tags match the Puppeteer version that corresponds to thechrome-*
release.chrome-stable
tag. This tag points to the Puppeteer version that works with the current Chrome stable release.
These tags are updated on every Puppeteer release.
NOTE: due to Chrome's rolling release, we take omahaproxy's linux stable version as stable.
Managing tags 101:
# list tags
$ npm dist-tag ls puppeteer
# Removing a tag
$ npm dist-tag rm puppeteer-core chrome-stable
# Adding a tag
$ npm dist-tag add [email protected] chrome-stable