0.5.2

What's Changed

• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in #19

New Contributors

• @pre-commit-ci[bot] made their first contribution in #19

Full Changelog: 0.5.1...0.5.2

0.5.1

What's Changed

• feat(actor): Add key retrieval functionality to Actor class by @AmaseCocoa in #16

Full Changelog: 0.5.0...0.5.1

0.5.0 - next

apmodel 0.5.0 ("Next")

This release introduces a major migration to Pydantic, enhancing data validation and serialization. It also includes several new features, bug fixes, and significant improvements to the testing framework and documentation.

💥 Breaking Changes

• The underlying model implementation has been migrated from dataclasses to Pydantic. This is a significant change that may affect how models are instantiated and manipulated. Please
  review the documentation for details on the new Pydantic-based models.

✨ Features

• Pydantic Integration: All models are now based on Pydantic, providing robust data validation, serialization, and type hinting.
• Actor Endpoints: Added endpoints field to the Actor model to store a dictionary of endpoint URLs (e.g., proxyUrl, oauthTokenEndpoint).
• Likes and Shares: The base Object model now includes likes and shares fields, allowing for easy access to these common collections.
• LitePub Support: Added support for the EmojiReact object from the LitePub vocabulary.
• to_dict() Method: A new to_dict() method has been added to models for easy conversion to a dictionary.
• Performance Improvements:
  • Implemented lazy loading for the ModelRegistry to improve initial import times.
  • Core models are now preloaded for faster resolution.

🐛 Bug Fixes

• JSON-LD Context: The @context is now always serialized as a list to improve compatibility.
• Multibase Encoding: Fixed an issue with public key encoding format to use PKCS#8.
• Collection Handling: Correctly load OrderedCollectionPage and its items.
• Numerous type annotation fixes to improve static analysis and developer experience.

🧪 Testing

• Increased Coverage: Expanded test coverage, especially for model properties and serialization.
• Real-World Data: Added new tests using real-world data from Akkoma, Mastodon, and Misskey to ensure compatibility.
• Registry Tests: Implemented comprehensive tests for the ModelRegistry.

📚 Documentation

• New Documentation Site: A new documentation site has been created with mkdocs.
• Updated README: The project's README.md has been overhauled with more detailed information.
• API Documentation: Added detailed API documentation for models, functions, and the registry.

⚙️ Internal

• Developer Environment:
  • Introduced pre-commit hooks to automate dependency synchronization and code analysis with pyrefly.
  • CI workflows have been updated and improved.

0.4.5

• fix: PropertyValue is not implement to_json
• fix: True to keep_object as default value

0.4.4

🐛 Bug Fixes

• (text) Assert with str
• Object compression should be performed after conversion to dict
• Mastodon not provide type field of publicKey
• Remove CryptographicKey from legacy loader list

0.4.3

Full Changelog: 0.4.2...0.4.3

0.4.2

fix: re re fix: #2

to me: 2回ともForce Pushで消失してたぞ

0.4.1

🚀 Features

• (actor) Add extension field
• Auto convert to LDContext
• Auto convert actor/object to id

🐛 Bug Fixes

• Set SETUPTOOLS_SCM_PRETEND_VERSION env variable
• Add assertionMethod to dynamic_context
• Don't use Undefined for assertionMethod field
• Remove Union from assertionMethod

Full Changelog: 0.4.0...0.4.1

0.4.0

🚀 Features

• Implement core models
• Add test
• Nodeinfo parser
• Extra models
• Context parser
• Hashtag type
• Automatically add the URL to the context based on the value
• LCContext test
• Test cid
• Enforce keyword-only arguments for NodeinfoServices
• Add support for extra types
• Ruff rule
• Autochange version with vcs

🐛 Bug Fixes

• Setup uv before setup python
• Incorrect Link values
• Type hint
• Fix the code error causing the TypeError and use ctx.full_context instead of ctx.json.
• Use rmdir in windows
• Changes from gemini code assist's suggestions
• Check if the expected keys are a subset of the data's keys
• (actor) Resolve side effects and context handling in Actor.to_json
• (link) Fix context aggregation in Link.to_json
• (question) Prevent side effects in Question.to_json
• (tombstone) Prevent side effects in Tombstone.to_json
• Use ctx.full_context instead of ctx.json
• Initialize aggregated_context as a new LDContext instance, copied from self._context
• Use PrivateFormat.PKCS1 instead of PrivateFormat.Raw in rsa-priv
• (cid) Handle 'Z' timezone in DataIntegrityProof
• Fixes for violation of the Liskov Substitution Principle
• Use is instead of isinstance
• Use InvalidField instead of Exception
• (cid) Import InvalidField from exceptions.py
• (cid) Remove type: ignore
• Replace +00:00 with Z
• Use fromisoformat
• Include timezone information
• Use InvalidField instead of Exception
• Use InvalidField instead of Exception
• Use InvalidField instead of Exception
• Add Z in text end
• (ci) Set checkout branch
• (ci) Run in release branch

🚜 Refactor

• (loader) Refactor Nodeinfo detection logic
• Centralize JSON serialization logic
• (cid) Use central serializer for CID models
• Remove unused imports
• Remove whitespace in init.py
• Remove whitespace in init.py

⚙️ Miscellaneous Tasks

• Use context parser
• Context parser
• Context parser
• Styleguide
• Instructions for code review (likely, gemini code assist).
• Edit style guide
• Ignore git_commit_msg_tmp.txt
• Add classifiers
• Add dataclass decorator in ActivityPubModel
• Add styleguide
• Changelog auto generate
• Update changelog [skip ci]
• Update changelog [skip ci]

0.3.1

What's new

• feat: Activity.accept() can now generate an Accept activity for the activity and Activity.reject() can generate a Reject activity for the activity

Full Changelog: AmaseCocoa/apmodel@0.2.4...0.3.1