Some (stupid) services don’t answer the same way when getting requested with HEAD or GET.
Like GET -> 200, HEAD -> 405.
https://github.com/h2non/imaginary is an example of such a service.
- The web interface is exposed at /, and the api
at /api.
- Include picocss for a minimal CSS framework
- Added some queries and models.Task properties
to access the latest results
- Adjusted test cases in test_api.py
- Updated test_db fixture to delete the database between each test.
- Consolidated test cases for tasks retrieval and results
- Restructured server module to separate the application creation and configuration.
- Moved code dealing with SQLAlchemy database
setup and teardown to the main application file.
- Moved functions related to configuration file loading to `argos.server.settings`.
- Fixed SQLAchemy expressions in `argos.server.queries`.
- Implemented a more granular system of setting checks' schedule on the server.
- Introduced frequency scheduling on per-website basis in the YAML config.
- Introduced Pytest fixtures for handling test database and authorized HTTP client in `tests/conftest.py`.
- Included a first test for the api
- Implemented changes to models to accommodate changes to task scheduling.
- Fixed errors concerning database concurrency arising from changes to the application setup.
- Service endpoints now require an authentication
- Changed the location of the commands, which are now in `commands.py`
- There is now only one "argos" command which can start the server or an agent.
- Refactored the `get_check_by_name` method to `get_registered_check` in the
BaseCheck class and added a `get_registered_checks` method to get all the
registered checks.
- Added a validation in the `WebsitePath` class to ensure that a check exists
when parsing the configuration file.
- Updated the existing test for parsing SSL duration and added new tests to
validate path parsing and check existence validation.