jsonstate

Manage global App state in a JSON (Python dictionary).

Installation

• pip install jsonstate

Usage

from jsonstate import State

state = State({
    "title": "State Example",
    "profile": {
        "name": "Foo",
    },
    "products": [
        {"name": "Foo", "description": "Foo spam"},
        {"name": "Bar", "description": "Bar spam"},
    ]
})
print_event = lambda **kwargs: print("Event", kwargs)
state.callbacks(key="title").append(print_event)
state['profile'].callbacks(key="name").append(print_event)
state['products'].callbacks().append(print_event)

# This statement updates the state and also invokes on_change callback:
state["title"] = "Eggs"
# Event {'new_value': 'Eggs', 'old_value': 'State Example', 'action': 'update'}

state["profile"]["name"] = "Spam"
# Event {'new_value': 'Spam', 'old_value': 'Foo', 'action': 'update'}

# This statement also updates the state and invokes on_change callback:
state["products"].append({"name": "Eggs", "description": "Eggs spam"})
# Event {'new_value': {'name': 'Eggs', 'description': 'Eggs spam'}, 'action': 'add'}

Development

Commonly used commands for package development:

• make check - run unit tests and linters.
• make fix - format code and fix detected fixable issues.
• make publish - publishes current package version to pypi.org.
• make compile - bump and freeze dependency versions in requirements*.txt files
• make sync - upgrade installed dependencies in Virtual Environment (executed after make compile)

Toolset

This package uses these cutting edge tools:

• ruff - for linting and code formatting
• mypy - for type checking
• pip-audit - for known vulnerability detection in dependencies
• deadcode - for unused code detection
• pytest - for collecting and running unit tests
• coverage - for code coverage by unit tests
• hatch - for publishing package to pypi.org
• uv - for Python virtual environment and dependency management
• pyproject.toml - configuration file for all tools
• Makefile - aliases for commonly used command line commands