json-tools-rs

What it is and what it does

json-tools-rs is a Rust-backed JSON transformation library that flattens nested structures into flat key-value pairs (e.g., `{"user": {"name": "John"}}` becomes `{"user.name": "John"}`) and unfolds them back with perfect roundtrip fidelity. It uses SIMD-accelerated parsing and optional Rayon-based parallelism for batch operations, making it suitable for processing large volumes of JSON documents or deeply nested structures. The library ships with both Rust and Python bindings and integrates natively with Pandas, Polars, PyArrow, and PySpark DataFrames.

The builder-pattern API lets you chain configuration methods—separator customization, key/value filtering and replacement (literal or regex), type conversion, collision handling, and date normalization—before executing the transformation. Python bindings preserve input type (dict in, dict out; DataFrame in, DataFrame out), and the package depends only on orjson for JSON serialization. It's in Beta status, actively maintained, and carries no known security vulnerabilities.

Use it for:

• Flatten API responses or user-uploaded JSON for database insertion or CSV export.
• Normalize and clean messy JSON data from external sources by removing nulls, empty strings, and applying regex-based key/value replacements.
• Batch-process large collections of nested JSON documents in parallel for analytics or ETL pipelines.
• Convert nested JSON to wide-format DataFrames (Pandas, Polars, PyArrow) for data analysis without manual schema mapping.
• Roundtrip JSON through flatten and unflatten cycles while applying transformations (type conversion, filtering) to preserve data fidelity.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Flattens and unflattens nested JSON structures with SIMD-accelerated parsing and optional parallel processing, supporting both Rust and Python with DataFrame integration.

Yes, if you need to flatten or unflatten JSON at scale with minimal dependencies and cross-language support. The Rust backend and parallelism make it well-suited for batch processing and large nested structures. Install friction is moderate due to compiled wheels, but pre-built binaries cover common platforms. No security issues and active maintenance are positive signals. Consider it especially if you're already using Pandas, Polars, or PyArrow and want native DataFrame integration.

Install

json-tools-rs on PyPI

pip install json-tools-rs

uv add json-tools-rs

poetry add json-tools-rs

Installing json-tools-rs

Before you install

Medium install friction due to compiled Rust bindings, but wheels are pre-built for common platforms (Linux x86_64, ARM, macOS, Windows). Active maintenance with release 5 days ago and no known vulnerabilities.

License in practice

Dual-licensed MIT OR Apache-2.0, both permissive; you may choose either license terms for your use.

Quickstart

pip install json-tools-rs

import json_tools_rs as jt

result = jt.JSONTools().flatten().execute({"user": {"name": "John", "age": 30}})
print(result)  # {'user.name': 'John', 'user.age': 30}

Requires Python 3.9 or later; pre-built wheels available for Linux (x86_64, ARM, i686, ppc64le), macOS (x86_64, ARM64), and Windows (x86_64).

Verify before relying

• Whether the parallel processing (Rayon-based work-stealing pool) provides measurable speedup for typical batch sizes in production workloads.
• Performance comparison with other JSON flattening libraries to validate the SIMD and parallelism claims.
• Whether DataFrame/Series support (Pandas, Polars, PyArrow, PySpark) is feature-complete or has known limitations.

Package facts

Evidence: json_tools_rs-0.9.30-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl; json_tools_rs-0.9.30-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl; json_tools_rs-0.9.30-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl; json_tools_rs-0.9.30-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl; json_tools_rs-0.9.30-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl; json_tools_rs-0.9.30-cp310-cp310-musllinux_1_2_aarch64.whl; json_tools_rs-0.9.30-cp310-cp310-musllinux_1_2_armv7l.whl; json_tools_rs-0.9.30-cp310-cp310-musllinux_1_2_i686.whl; json_tools_rs-0.9.30-cp310-cp310-musllinux_1_2_x86_64.whl; json_tools_rs-0.9.30-cp310-cp310-win_amd64.whl; json_tools_rs-0.9.30-cp311-cp311-macosx_10_12_x86_64.whl; json_tools_rs-0.9.30-cp311-cp311-macosx_11_0_arm64.whl; json_tools_rs-0.9.30-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl; json_tools_rs-0.9.30-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl; json_tools_rs-0.9.30-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl; json_tools_rs-0.9.30-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl; json_tools_rs-0.9.30-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl; json_tools_rs-0.9.30-cp311-cp311-musllinux_1_2_aarch64.whl; json_tools_rs-0.9.30-cp311-cp311-musllinux_1_2_armv7l.whl; json_tools_rs-0.9.30-cp311-cp311-musllinux_1_2_i686.whl

Keywords: json, flatten, manipulation, parsing, rust, simd, performance, dataframe

Tags

More Python Modules packages