Contributions are more than welcome in the form of pull requests.
You can either pick one of the open issues to work on, or provide enhancements according to your own needs. In any case, we suggest getting in touch beforehand to make sure the contribution will be aligned with the current development status.
- Fork the repository and clone the fork to your local development environment
- Identify a modular contribution to the code (avoid too large contributions to simplify review)
- Create a branch in your repository where you tackle the "modular contributions"
- For multiple contributions tackling different functionalities, create different branches
- For all the new functionalities provide tests (see
run.shin the root to understand how tests can be run locally)
- When done, verify that all tests are passing
- If so, create a pull request against our repository (we cannot review pull requests with failing tests)
- Wait for the review
- Implement required changes
- Repeat until approval
- Done :) You can delete the branch in your repository.
Ultimately, contributing guides should keep aligned with those suggested by FIWARE (see here).
The development is mostly in python3 for now, and really in the early stages so things will change for sure. For now, you can get started with:
git clone https://github.com/smartsdk/ngsi-timeseries-api.git cd ngsi-timeseries-api pipenv install # if you want to set up a dev env to test everything locally, you'll need to... source setup_dev_env.sh
pytest is used as the testing framework,
but since most of QL's functionality is integration of components, you'll find
docker-compose.yml files in the test folders to be run as a setup for tests.
If you see
.travis.yml file you'll see how they are running today, but
probably at some point it's worth exploring pytest-docker plugins.
requirements.txt still needs to be split between testing and production,
that's also why the docker image is massive for now.
In the current project tree structure you can find:
docs: Holds documentation files.
docker: To hold docker-related files for the scope of the project.
experiments: Sandbox for quick manual tests to try some stuff and derive new test cases.
specification: Contains the OpenAPI definition that QL implements.
src: Source code folder.
geocoding: Holds the code for interacting with OSM and doing geo-related processing.
reporter: Modules acting as the receiver of the notifications and API requests. It "parses/validates" them before handling tasks to the translators.
translators: Specific translators for each time-series databases, responsible for interacting with the lower-level database details.
utils: Common shared stuff looking for a better place to live in.