apmodel

apmodel is a Python library that provides model implementations for various decentralized social web protocols.

It is designed to easily parse and handle JSON data from sources like Mastodon, Misskey, and other Fediverse software.

Features

• Automatic Model Resolution: The apmodel.load function automatically deserializes a JSON object into the appropriate Python object by reading its type field. If no matching model is found, the original dictionary is returned.
• Flexible Deserialization: Any properties in the JSON that do not have a corresponding field in the model are collected into an model_extra dictionary. This ensures no data is lost.
• Type Hinting: Fully type-hinted for a better development experience and robust static analysis.

Installation

pip install apmodel

# uv
uv add apmodel

Usage

Here is a basic example of how to parse an ActivityStreams 2.0 Note object from a JSON string.

import apmodel

# Example JSON from a Fediverse server
json_data = {
  "@context": "https://www.w3.org/ns/activitystreams",
  "id": "https://example.com/users/alice/statuses/1",
  "type": "Note",
  "published": "2023-12-25T12:00:00Z",
  "attributedTo": "https://example.com/users/alice",
  "content": "<p>Hello, world!</p>",
  "to": ["https://www.w3.org/ns/activitystreams#Public"],
  "cc": ["https://example.com/users/alice/followers"]
}

# Load the JSON into an apmodel object
obj = apmodel.load(json_data)

# Now you can access properties with type safety
if isinstance(obj, apmodel.vocab.Note):
    print(f"Type: {obj.type}")
    print(f"Content: {obj.content}")
    print(f"Published: {obj.published}")

# >> Type: Note
# >> Content: <p>Hello, world!</p>
# >> Published: 2023-12-25 12:00:00+00:00

Documentation

https://fedi-libs.github.io/apmodel

Supported Specifications

apmodel provides models for the following specifications:

• Activity Streams 2.0: Core types like Object, Activity, Collection, and Link.
• Activity Vocabulary: All types.
• Security Vocabulary v1: CryptographicKey (Key).
• Controlled Identifiers v1.0: Multikey and DataIntegrityProof.
• schema.org: PropertyValue.
• NodeInfo 2.0/2.1
• Litepub: EmojiReact
• Others: Emoji and Hashtag

Development

This project uses Task for running scripts. You need to install it first.

• Setup dev environment:

  task
  

• Run tests with Pytest:

  task test
  

• Build packages for PyPI:

  task build
  

• Development Document:

  task docs:dev
  

License

MIT License