We welcome contributions to sentry-python
by the community.
This file outlines the process to contribute to the SDK itself. For contributing to the documentation, please see the Contributing to Docs page.
Please search the issue tracker before creating a new issue (a problem or an improvement request). Please also ask in our Sentry Community on Discord before submitting a new issue. There are a ton of great people in our Discord community ready to help you!
- Fork the
sentry-python
repo and prepare your changes. - Add tests for your changes to
tests/
. - Run tests and make sure all of them pass.
- Submit a pull request, referencing any issues your changes address. Please follow our commit message format when naming your pull request.
We will review your pull request as soon as possible. Thank you for contributing!
Make sure that you have Python 3 installed. Version 3.7 or higher is required to run style checkers on pre-commit.
On macOS, we recommend using brew
to install Python. For Windows, we recommend an official python.org release.
Before you can contribute, you will need to fork the sentry-python
repository. Then, clone the forked repository to your local development environment.
To keep your Python development environment and packages separate from the ones used by your operation system, create a virtual environment:
cd sentry-python
python -m venv .venv
Then, activate your virtual environment with the following command. You will need to repeat this step every time you wish to work on your changes for sentry-python
.
source .venv/bin/activate
Install sentry-python
in editable mode. This will make any changes you make to the SDK code locally immediately effective without you having to reinstall or copy anything.
pip install -e .
Hint: Sometimes you need a sample project to run your new changes to sentry-python
. In this case install the sample project in the same virtualenv and you should be good to go.
This will make sure that your commits will have the correct coding style.
cd sentry-python
pip install -r requirements-devenv.txt
pip install pre-commit
pre-commit install
That's it. You should be ready to make changes, run tests, and make commits! If you experience any problems, please don't hesitate to ping us in our Discord Community.
You can run all tests with the following command:
pytest tests/
If you would like to run the tests for a specific integration, use a command similar to the one below:
pytest -rs tests/integrations/flask/ # Replace "flask" with the specific integration you wish to test
Hint: Tests of integrations need additional dependencies. The switch -rs
will show you why tests were skipped and what dependencies you need to install for the tests to run. (You can also consult the tox.ini file to see what dependencies are installed for each integration)
-
Write the integration.
-
Instrument all application instances by default. Prefer global signals/patches instead of configuring a specific instance. Don't make the user pass anything to your integration for anything to work. Aim for zero configuration.
-
Everybody monkeypatches. That means:
-
Make sure to think about conflicts with other monkeypatches when monkeypatching.
-
You don't need to feel bad about it.
-
-
Make sure your changes don't break end user contracts. The SDK should never alter the expected behavior of the underlying library or framework from the user's perspective and it shouldn't have any side effects.
-
Avoid modifying the hub, registering a new client or the like. The user drives the client, and the client owns integrations.
-
Allow the user to turn off the integration by changing the client. Check
Hub.current.get_integration(MyIntegration)
from within your signal handlers to see if your integration is still active before you do anything impactful (such as sending an event).
-
-
Write tests.
-
Consider the minimum versions supported, and test each version in a separate env in
tox.ini
. -
Create a new folder in
tests/integrations/
, with an__init__
file that skips the entire suite if the package is not installed.
-
-
Update package metadata.
-
We use
extras_require
insetup.py
to communicate minimum version requirements for integrations. People can use this in combination with tools like Poetry or Pipenv to detect conflicts between our supported versions and their used versions programmatically.Do not set upper bounds on version requirements as people are often faster in adopting new versions of a web framework than we are in adding them to the test matrix or our package metadata.
-
-
Write the docs. Follow the structure of existing integration docs. And, please make sure to add your integration to the table in
python/integrations/index.md
(people often forget this step 🙂). -
Merge docs after new version has been released. The docs are built and deployed after each merge, so your changes should go live in a few minutes.
-
(optional, if possible) Update data in
sdk_updates.py
to give users in-app suggestions to use your integration. This step will only apply to some integrations.
(only relevant for Sentry employees)
- All the changes that should be released must be on the
master
branch. - Every commit should follow the Commit Message Format convention.
- CHANGELOG.md is updated automatically. No human intervention is necessary, but you might want to consider polishing the changelog by hand to make it more user friendly by grouping related things together, adding small code snippets and links to docs, etc.
- On GitHub in the
sentry-python
repository, go to "Actions" and select the "Release" workflow. - Click on "Run workflow" on the right side, and make sure the
master
branch is selected. - Set the "Version to release" input field. Here you decide if it is a major, minor or patch release. (See "Versioning Policy" below)
- Click "Run Workflow".
This will trigger Craft to prepare everything needed for a release. (For more information, see craft prepare.) At the end of this process a release issue is created in the Publish repository. (Example release issue: getsentry/publish#815)
Now one of the persons with release privileges (most probably your engineering manager) will review this issue and then add the accepted
label to the issue.
There are always two persons involved in a release.
If you are in a hurry and the release should be out immediately, there is a Slack channel called #proj-release-approval
where you can see your release issue and where you can ping people to please have a look immediately.
When the release issue is labeled accepted
, Craft is triggered again to publish the release to all the right platforms. (See craft publish for more information.) At the end of this process the release issue on GitHub will be closed and the release is completed! Congratulations!
There is a sequence diagram visualizing all this in the README.md of the Publish
repository.
This project follows semver, with three additions:
-
Semver says that major version
0
can include breaking changes at any time. Still, it is common practice to assume that only0.x
releases (minor versions) can contain breaking changes while0.x.y
releases (patch versions) are used for backwards-compatible changes (bugfixes and features). This project also follows that practice. -
All undocumented APIs are considered internal. They are not part of this contract.
-
Certain features (e.g. integrations) may be explicitly called out as "experimental" or "unstable" in the documentation. They come with their own versioning policy described in the documentation.
We recommend to pin your version requirements against 2.x.*
or 2.x.y
.
Either one of the following is fine:
sentry-sdk>=2.0.0,<3.0.0
sentry-sdk==2.4.0
A major release N
implies the previous release N-1
will no longer receive updates. We generally do not backport bugfixes to older versions unless they are security relevant. However, feel free to ask for backports of specific commits on the bugtracker.