lidatong/dataclasses-json
Dataclasses Json
Dataclasses-json is a library designed to facilitate the bidirectional conversion between Python data classes and JSON formats. It serves as a serialization tool that populates structured objects from raw data and transforms object instances into JSON strings, utilizing Python type hints to automate the mapping process.
The library distinguishes itself by providing mechanisms for schema validation and runtime type coercion, ensuring that incoming data adheres to predefined model structures. It supports complex data requirements through recursive object traversal, which allows for the processing of nested hierarchies, and offers flexible field mapping to reconcile differences between internal naming conventions and external API requirements.
Beyond basic conversion, the library includes utilities for managing unexpected data, such as missing or extraneous fields, through configurable strategies. It also provides an extensible hook system that allows for custom serialization logic, enabling developers to override default transformation behaviors for specialized data types or formats.
Features
- JSON Serialization Libraries - Provides a library for transforming structured object instances into JSON formats and populating objects from raw data.
- Streaming-to-Object Conversions - Facilitates bidirectional conversion between structured Python objects and JSON formats, including support for nested hierarchies.
- Python Dataclass Serialization - Converts complex Python data classes into JSON strings and back into structured objects for storage or transmission.
- Class Instance JSON Serializers - Converts Python data classes to and from JSON strings while supporting custom field mapping and validation.
- Marshmallow Schema Integrations - Integrates with Marshmallow schemas to enforce strict type checking and validate incoming data against predefined models.
- Data Type Coercion - Automatically validates and converts raw input values into expected Python types based on model type hints.
- Struct Field Mappings - Supports custom field naming conventions to transform between internal snake_case and external camelCase or kebab-case formats.
- Database Schema Mappings - Decouples internal class attribute names from external JSON field names to support flexible data mapping.
- Serialization Logic Overrides - Enables precise formatting for specialized data types by overriding default encoding and decoding behaviors.
- Field Transform Hooks - Provides an extensible hook system to override default transformation logic for specialized data types.
- Type-Hint-Driven Mapping - Inspects Python type annotations to automatically map data fields to object attributes during serialization.
- Recursive Object Validation - Recursively traverses nested object hierarchies to process and validate complex data structures.
- Data Validation Schemas - Enforces schema constraints and manages data integrity during the serialization of complex models.
- Data Class Decorators - Uses decorators to store serialization rules and configuration directly within data class definitions.
- Expected and Unexpected Failure Separation - Manages missing or extraneous input data through configurable strategies like default values or catch-all fields.
- JSON Schema Validation - Enforces data integrity by validating incoming JSON payloads against predefined Python data models.
- Case-Conversion Mappings - Standardizes field naming between internal models and external API requirements during data exchange.
