json-tabulate 0.1.1

Python library and CLI app that translates arbitrarily-nested JSON into CSV

json-tabulate

json-tabulate is a Python library and CLI app that you can use to translate an arbitrarily-nested JSON object—or an array of those objects—into a flat CSV table.

graph LR
  json_obj[JSON<br>object] --> app_1[json-tabulate] --> csv_single[CSV<br>1 header row + 1 data row]
  json_list[JSON<br>array of N objects] --> app_2[json-tabulate] --> csv_multi[CSV<br>1 header row + N data rows]

Example

It translates this (arbitrarily-nested JSON object):

{
  "a": 1,
  "b": {
    "d": "f",
    "e": ["g", "h", "i"]
  },
  "c": 2
}

Into this (CSV table):

Show/hide CSV string

$.a,$.b.d,$.b.e[0],$.b.e[1],$.b.e[2],$.c
1,f,g,h,i,2

$.a	$.b.d	$.b.e[0]	$.b.e[1]	$.b.e[2]	$.c
1	f	g	h	i	2

Each column names is a JSONPath expression indicating where the values in that column came from.

Usage

Python library

pip install json-tabulate

>>> from json_tabulate.core import translate_json
>>> translate_json(r'{"name": "Ken", "age": 26}')
'$.age,$.name\n26,Ken\n'

CLI app

Here's the usage string displayed by the CLI app:

json-tabulate --help

 Usage: json-tabulate [OPTIONS] [JSON_STRING]

 Translate JSON into CSV.

 Usage examples:

  • json-tabulate '{"name": "Ken", "age": 26}' (specify JSON via argument)
  • echo '{"name": "Ken", "age": 26}' | json-tabulate (specify JSON via STDIN)
  • cat input.json | json-tabulate > output.csv (write CSV to file)

╭─ Arguments ──────────────────────────────────────────────────────────────────╮
│   json_string      [JSON_STRING]  JSON string to translate. If not provided, │
│                                   program will read from STDIN.              │
╰──────────────────────────────────────────────────────────────────────────────╯
╭─ Options ────────────────────────────────────────────────────────────────────╮
│ --version          Show version number and exit.                             │
│ --help             Show this message and exit.                               │
╰──────────────────────────────────────────────────────────────────────────────╯

Development

Show/hide developer documentation

Using VS Code? The file, .vscode/tasks.json, contains VS Code task definitions for several of the commands shown below. You can invoke those tasks via the command palette, or—if you have the Task Runner extension installed—via the "Task Runner" panel.

Setup Python virtual environment

Here's how you can create a Python virtual environment and install the Python dependencies within it:

uv sync

Lint Python source code

uv run ruff check --fix

# Other option: Do a dry run.
uv run ruff check

Format Python source code

uv run ruff format

# Other option: Do a dry run.
uv run ruff format --diff

Check data types

uv run mypy

The default configuration is defined in pyproject.toml.

Run tests

uv run pytest

# Other option: Run tests and measure code coverage.
uv run pytest --cov

The default configuration is defined in pyproject.toml.

Build distributable package

uv build

The build artifacts will be in the dist/ directory.