Skip to content

Latest commit

 

History

History
131 lines (95 loc) · 5.33 KB

CONTRIBUTING.md

File metadata and controls

131 lines (95 loc) · 5.33 KB
Raw

Contributing to TimescaleDB

We appreciate any help the community can provide to make TimescaleDB better!

You can help in different ways:

  • Open an issue with a bug report, build issue, feature request, suggestion, etc.

  • Fork this repository and submit a pull request

For any particular improvement you want to make, it can be beneficial to begin discussion on the GitHub issues page. This is the best place to discuss your proposed improvement (and its implementation) with the core development team.

Before we accept any code contributions, Timescale contributors need to sign the Contributor License Agreement (CLA). By signing a CLA, we can ensure that the community is free and confident in its ability to use your contributions.

Getting and building TimescaleDB

Please follow our README for instructions on installing from source.

Style guide

Before submitting any contributions, please ensure that it adheres to our Style Guide.

Code review workflow

  • Sign the Contributor License Agreement (CLA) if you're a new contributor.

  • Develop on your local branch:

    • Fork the repository and create a local feature branch to do work on, ideally on one thing at a time. Don't mix bug fixes with unrelated feature enhancements or stylistical changes.

    • Hack away. Add tests for non-trivial changes.

    • Run the test suite and make sure everything passes.

    • When committing, be sure to write good commit messages. Stylistically, we use commit message titles in the imperative tense, e.g., Add merge-append query optimization for time aggregate. In the case of non-trivial changes, include a longer description in the commit message body explaining and detailing the changes. That is, a commit message should have a short title, followed by a empty line, and then followed by the longer description.

  • Push your changes to an upstream branch:

    • Make sure that each commit in the pull request will represent a logical change to the code, will compile, and will pass tests.

    • Rebase your local feature branch against master (git fetch origin, then git rebase origin/master) to make sure you're submitting your changes on top of the newest version of our code.

    • When finalizing your PR (i.e., it has been approved for merging), aim for the fewest number of commits that make sense. That is, squash any "fix up" commits into the commit they fix rather than keep them separate. Each commit should represent a clean, logical change and include a descriptive commit message.

    • Push your commit to your upstream feature branch: git push -u <yourfork> my-feature-branch

  • Create and manage pull request:

    • Create a pull request using GitHub. If you know a core developer well suited to reviewing your pull request, either mention them (preferably by GitHub name) in the PR's body or assign them as a reviewer.

    • If you get a test failure in Travis CI, check them in the Travis CI build log.

    • Address feedback by amending your commit(s). If your change contains multiple commits, address each piece of feedback by amending that commit to which the particular feedback is aimed.

    • The PR is marked as accepted when the reviewer thinks it's ready to be merged. Most new contributors aren't allowed to merge themselves; in that case, we'll do it for you.

Testing

Every non-trivial change to the code base should be accompanied by a relevant addition to or modification of the test suite.

Please check that the full test suite (including your test additions or changes) passes successfully on your local machine before you open a pull request.

If you are running locally:

# Use Debug build mode for full battery of tests
./bootstrap -DCMAKE_BUILD_TYPE=Debug
cd build && make
make installcheck

All submitted pull requests are also automatically run against our test suite via Travis CI (that link shows the latest build status of the repository).

Advanced Topics

Testing on Windows

Currently our CI infrastructure only ensures that TimescaleDB builds on Windows, but does not run regression tests due to differences between Unix-based systems and Windows. We do run these tests before releases manually, and it would be a bonus if you could test at least non-trivial contributions for Windows. This involves setting up a remote Windows machine with TimescaleDB and a Unix-based (e.g., macOS or Linux) machine to serve as the client. To set up the Windows machine, build from source:

./bootstrap -DCMAKE_BUILD_TYPE=Debug
cmake --build ./build --config Debug
cmake --build ./build --config Debug --target install

Then on the client machine:

./bootstrap -DCMAKE_BUILD_TYPE=Debug -DTEST_PGHOST=ip_addr_of_Win_machine
cd build && make
make installchecklocal