Carbon-txt

Added

- Add architecture docs for how the validator is built [7e39de2](https://github.com/thegreenwebfoundation/carbon-txt-validator/commit/7e39de25f96d439b03a4a907337a6110a5affd11)
- Add new Validator class that wraps the behaviour exposed to the CLI and API, making it more consistent [34](https://github.com/thegreenwebfoundation/carbon-txt-validator/pull/34)
- Add new logging behaviour in the validator [20](https://github.com/thegreenwebfoundation/carbon-txt-validator/issue/20)
- Add new API checker notebook [35](https://github.com/thegreenwebfoundation/carbon-txt-validator/pull/35)

Changed

- Refactored the Finder and TOML Parser classes, so that the Finder is always responsible for fetching files, and the Validator responsible for parsing strings and datastructures [34](https://github.com/thegreenwebfoundation/carbon-txt-validator/pull/34)

Fixed

- Fix handling of non TOML responses when carbon.txt files are served [33](https://github.com/thegreenwebfoundation/carbon-txt-validator/pull/33)

Added

- Extra details in the runbook section of the docs for operating the carbon.txt API server

Fixed

- Fix mismatch between API and CLI responses for certain lookups [31](https://github.com/thegreenwebfoundation/carbon-txt-validator/issues/30git)

Fixed

- Fix ordering of optional params when used with Granian webserver

Changed

- Switch to using more detailed exceptions

Added

- Add support for generating JSON Schema representation with CLI and API
- Add support for using Sentry to track exceptions, performance and so on.
- Add github action for automated publishing of releases to Pypi

Fixed

- Respect the provided `port` and `host` values with `carbon-txt serve` [27](https://github.com/thegreenwebfoundation/carbon-txt-validator/issues/27)

Added

- First release using the formal changelog