Repository navigation
Development Workflow
remonds edited this page Apr 11, 2025
·
3 revisions
We follow a three-level branching strategy:
-
mainbranch: Contains stable, production-ready code. Each release is tagged here using semantic versioning. -
devbranch: Where the active development of the next version happens. -
feature/*branches: Created for individual features or tasks. Once a feature is complete, it is merged back intodev.
-
Start a New Feature
- Create a feature branch from
dev. - Naming convention:
feature/<short-description> - Example:
feature/lognormal-true-model
- Create a feature branch from
-
Development
- Work on the feature in the feature branch.
- Regular commits help to track progress and keep the work isolated from the main development.
-
Testing
- Test the feature thoroughly in the feature branch before merging.
- When the feature is complete and passes all tests, proceed to the next step.
-
Merge to
dev- Once the feature is complete, reviewed, and tested, it should be merged into the
devbranch. - Ensure that the merge is done after resolving any conflicts and that the integration is smooth.
- Once the feature is complete, reviewed, and tested, it should be merged into the
-
Release Preparation
- When development in
devis complete and the new version is ready, mergedevintomain. - Tag the release with an appropriate version number, following semantic versioning (e.g.,
v1.0.0-additive-on-obs).
- When development in
-
Tag a Release
- After merging
devtomain, tag the release to mark it as a stable version. - The release tag is used for versioning purposes and to make the code accessible for production.
- After merging
-
Publish the Release
- Publish the release on GitHub, and update relevant documentation (e.g., in the README.md).
- Regularly sync your feature branches with
devto minimize merge conflicts. - Commit often with meaningful commit messages.
- Perform code reviews before merging a feature into
dev. - Ensure the
devbranch is stable and tested before tagging a release.
See our README for versioning and release information.
- Development workflow
- Measurement uncertainty models
- Additive uncertainty on observed values
- Lognormal uncertainty on true values