py-jsonl 1.3.18

A lightweight Python library for handling jsonlines files.

jsonl

About

jsonl is a lightweight Python library designed to simplify working with JSON Lines data, adhering to the jsonlines and ndjson specifications.

Features

• 🌎 Provides an API similar to Python's standard json module.
• 🚀 Supports custom (de)serialization via user-defined callbacks.
• 🗜️ Built-in support for gzip, bzip2, xz compression formats and ZIP or TAR archives.
• 🔧 Skips malformed lines during file loading.
• 📥 Loads from URLs directly.

Installation

To install jsonl using pip, run the following command:

pip install py-jsonl

Getting Started

Dumping data to a JSON Lines File

Use jsonl.dump to incrementally write an iterable of dictionaries to a JSON Lines file:

import jsonl

data = [
    {"name": "Gilbert", "wins": [["straight", "7♣"], ["one pair", "10♥"]]},
    {"name": "May", "wins": []},
]

jsonl.dump(data, "file.jsonl")

Loading data from a JSON Lines source

Use jsonl.load to incrementally load a JSON Lines source—such as a filename, URL, or file-like object—into as an iterator of dictionaries:

import jsonl

# Load data from a JSON Lines file
iterator = jsonl.load("file.jsonl")
print(tuple(iterator))

# Load data from a URL
iterator = jsonl.load("https://example.com/file.jsonl")
print(tuple(iterator))

Dump multiple JSON Lines Files into an Archive (ZIP or TAR)

Use jsonl.dump_archive to incrementally write structured data to multiple JSON Lines files, which are then stored in a ZIP or TAR archive.

import jsonl

data = [
    # Create `file1.jsonl` withing the archive
    ("file1.jsonl", [{"name": "Alice", "age": 30}, {"name": "Bob", "age": 25}]),
    # Create `file2.jsonl` within the archive
    ("path/to/file2.jsonl", [{"name": "Charlie", "age": 35}, {"name": "David", "age": 40}]),
    # Append to `file1.jsonl` within the archive
    ("file1.jsonl", [{"name": "Eve", "age": 28}]),
]
jsonl.dump_archive("archive.zip", data)

Load multiple JSON Lines Files from an Archive (ZIP or TAR)

Use jsonl.load_archive to incrementally load multiple JSON Lines files from a ZIP or TAR archive.

• It is possible to load the archive from a URL
• This function allows you to filter files using Unix shell-style wildcards.

import jsonl

# Load all JSON Lines files matching the pattern "*.jsonl" from a local archive
for filename, iterator in jsonl.load_archive("archive.zip"):
    print("Filename:", filename)
    print("Data:", tuple(iterator))

# Load all JSON Lines files matching the pattern "*.jsonl" from a remote archive
for filename, iterator in jsonl.load_archive("https://example.com/archive.zip"):
    print("Filename:", filename)
    print("Data:", tuple(iterator))

Dumping data to Multiple JSON Lines Files

Use jsonl.dump_fork to incrementally write structured data to multiple JSON Lines files, which can be useful when you want to separate data based on some criteria.

import jsonl

data = [
    # Create `file1.jsonl` or overwrite it if it exists
    ("file1.jsonl", [{"name": "Alice", "age": 30}, {"name": "Bob", "age": 25}]),
    # Create `file2.jsonl` or overwrite it if it exists
    ("file2.jsonl", [{"name": "Charlie", "age": 35}, {"name": "David", "age": 40}]),
    # Append to `file1.jsonl`
    ("file1.jsonl", [{"name": "Eve", "age": 28}]),
]
jsonl.dump_fork(data)

Documentation

For more detailed information and usage examples, refer to the project documentation

Development

To contribute to the project, you can run the following commands for testing and documentation:

First, ensure you have the latest version of pip:

python -m pip install --upgrade pip

Running Unit Tests

Install the development dependencies and run the tests:

pip install --group=test  # Install test dependencies
pytest tests/ # Run all tests
pytest --cov jsonl # Run tests with coverage

Running Linter

pip install --group=lint  # Install linter dependencies
ruff check . # Run linter

Building the Documentation

To build the documentation locally, use the following commands:

pip install --group=doc  # Install documentation dependencies
mkdocs serve # Start live-reloading docs server
mkdocs build # Build the documentation site

License

This project is licensed under the MIT license.