Releases

First stable release of csv-to-json: a lightweight, zero-dependency CSV-to-JSON converter for developer and agent workflows.

What it does

Converts CSV data into clean JSON using only the Python standard library (3.10+). Works as a CLI tool or an importable library.

Features

• Type inference — numeric strings become ints/floats, true/yes/false/no become booleans, empty cells become null (opt out with --no-type-inference)
• Nested key expansion — dot-notation headers like user.address.city expand into nested JSON objects (--expand-keys)
• Streaming mode — stream_csv_to_json() yields records one at a time, so large files never need to fit in memory
• Flexible input — file paths, raw CSV strings, or stdin (-)
• Custom delimiters — handle TSV and other delimited formats via -d
• Compact or pretty output — indented by default, minified with --compact

Quick start

python src/csv_converter.py data.csv -o output.json
cat data.csv | python src/csv_converter.py - --expand-keys

from src.csv_converter import csv_to_json
print(csv_to_json("name,age\nAlice,30"))

MIT licensed. See README.md for full CLI reference and type-inference rules.