Fast JSON schema inference with support for Polars DataFrames.

This workspace contains multiple interconnected crates, most people will probably want the Python package:

• polars-genson-py/ - Python bindings and Polars plugin

pip install polars-genson[polars]

import polars as pl
import polars_genson

df = pl.DataFrame({
    "json_data": [
        '{"name": "Alice", "age": 30, "scores": [95, 87]}',
        '{"name": "Bob", "age": 25, "city": "NYC", "active": true}'
    ]
})

json_schema = df.genson.infer_json_schema("json_data")
polars_schema = df.genson.infer_polars_schema("json_data")

• genson-core/ - Core JSON schema inference library
• polars-jsonschema-bridge/ - JSON Schema ↔ Polars type conversion
• genson-cli/ - Command-line schema inference tool

• Fast schema inference from JSON strings in Polars columns
• Dual output formats: JSON Schema and native Polars schemas
• Complex type support: nested objects, arrays, mixed types
• Multiple JSON formats: single objects, arrays, NDJSON
• Rust performance with Python convenience

Each component has detailed documentation:

• Python users: See polars-genson-py/README.md
• Rust developers: See individual crate READMEs
• Development: See DEVELOPMENT.md

Licensed under the MIT License. See LICENSE for details.