A Dart workspace containing JSON parsing tools and applications.

MIT

json.org

This workspace contains four packages:

• console_app: A command-line JSON parser that reads from stdin and pretty-prints parse trees
• shelf_app: A server application using the shelf package and Docker
• shared_lib: A Dart library for parsing JSON with token-based representation and pretty-printing
• dart_frog_app: A server application using the dart_frog framework that exposes the same JSON parsing API

• Dart SDK >= 3.9.2
• Docker
• Melos

# From the root directory
dart pub get

Format all Dart code in the workspace:

dart format .

Run static analysis on all packages:

dart analyze .

This workspace uses Melos to manage multi-package tasks. Prefer the Melos scripts below for day-to-day development and CI.

Bootstrap the workspace (install package dependencies for all packages):

# From the repo root
dart pub get
dart run melos bootstrap

Build workspace executables (replaces bin/build_all):

dart run melos run build

Run all package tests:

dart run melos exec -- "dart test -r expanded --chain-stack-traces"

Run the CI test script to produce a combined LCOV report at coverage/lcov.info:

dart run melos run test:ci

The console app provides a command-line JSON parser.

# Enter JSON into the console app
./out/console_app

# Example: pipe JSON to the console app
echo '{"name": "John", "age": 30}' | ./out/console_app

The server apps provide a REST API for JSON parsing. You can test them using the REST Client VS Code extension.

1. Install the REST Client extension for VS Code

Start the Shelf-based server (existing implementation):

dart run melos run run:shelf

Start the Dart Frog dev server (development-friendly):

dart run melos run run:dart_frog

1. Create .rest files in the testdata/ folder at the root of the workspace

2. Write HTTP requests in the following format (both servers expose the same endpoint):

POST http://localhost:8000/api/v1/parse
Content-Type: text/plain

{"key": "value", "number": 42}

3. Click the "Send Request" link above the request to execute it

POST http://localhost:8000/api/v1/parse
Content-Type: text/plain

{"name": "John", "age": 30, "items": [1, 2, 3]}

POST http://localhost:8000/api/v1/parse
Content-Type: text/plain

[1, "hello", null, true, {"nested": "object"}]

POST http://localhost:8000/api/v1/parse
Content-Type: text/plain

{"invalid": json}

POST http://localhost:8000/api/v1/parse
Content-Type: application/json

{"key": "value"}

• Success (200): Returns the parsed JSON as application/json
• Parse Error (400): Returns {"code": 400, "message": "error details"}
• Unsupported Media Type (415): Returns {"code": 415, "message": "Unsupported Media Type"}