This adds support for Python 3.12.
This makes the library compatible with mypy strict checking, and slightly cleans up the release workflow.
This makes the entire library compatible with PEP-561, so that it can now be used with a static
type checker like mypy. The codebase has been reformatted with black and isort, and the tags
field of a PlanQuery now takes a list of strings, rather than a single string containing
the tags separated by commas and spaces. pdf is now the only return format which returns bytes;
native returns a dataclass and all other formats return a UTF-8 string.
The changelog has also been updated to include the changes of version 0.5.0 and earlier. A pre-commit file has been added to ensure all checks will pass before committing.
This fixes a bug in the core API interface where HTTP headers were being passed into requests as parameters.
This is a minor update, which adds support for Python 3.11 and moves the package configuration from setup.py to pyproject.toml. No breaking changes have been introduced. A bug has been fixed which was causing aiohttp to crash on null parameters, and a bug in the quickstart example has been fixed.
This is another complete rewrite of the library, in which it is entirely converted to async. This should mean faster execution of parallel requests, and no blocking when called from another async library. Support for Python 3.7 has been dropped in this release. Python 3.11 is not yet supported as aiohttp does not yet support Python 3.11 at the time of release.
This is a complete rewrite of the library, which moves functions out of classes. This does have the side effect of requiring a key to be passed into every authenticated request, instead of being passed into a class once on initialisation. The rewrite also incorporates several small bugfixes, and changes the test environment from unittest to pytest. Python 3.10 is now supported.
This adds support for the OM, MM, and IM navaid types, fixing issue #14. include_route is
made into a function argument rather than a dataclass field, fixing issue #13. Parts of the
code are also refactored to use keyword arguments instead of positional arguments to help
reduce bugs.
This documents the return format options for plan fetching, and differentiates between a
default native option where the returned json is unpacked into an appropriate dataclass
and a json option where it is returned as json.
This adds a dark theme to the sphinx documentation. Exceeding of the API limit now raises
a dedicated TooManyRequestsException. Additionally, all Union[<type>, None] type hints have been
replaced by Optional[<type>]
This updates the documentation, and fixes some incorrect type hints.
This splits the codebase up into separate submodules. Custom exception classes have been written to handle different HTTP errors. Additionally, unit tests have been written, and Github workflows have been added to run tests and lint the codebase on a push and upload to pip on a version release.
This changes the wrapper file into an installable Python package, moves the documentation from the readme to a Sphinx project on readthedocs, and adds docstrings for all functions. All data is now handled via dataclasses. The project now uses semantic versioning.
This adds functions for all remaining API endpoints which have not yet been wrapped. Error handling has also been added, and the readme has been expanded.
This is the initial, incomplete release of this wrapper. Many functions are not yet implemented, and the wrapper is highly unstable.