Metadata-Version: 2.1
Name: lazyfast
Version: 0.1.28
Summary: LazyFast = FastAPI + HTMX + Component-based approach + State management
Home-page: https://github.com/nikirg/lazyfast
Author: Nikita Irgashev
Author-email: nik.irg@yandex.ru
Requires-Python: >=3.11,<4.0
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Requires-Dist: asgiref (>=3.8.1,<4.0.0)
Requires-Dist: fastapi (>=0.111.0,<0.112.0)
Requires-Dist: python-multipart (>=0.0.9,<0.0.10)
Project-URL: Documentation, https://github.com/nikirg/lazyfast
Project-URL: Repository, https://github.com/nikirg/lazyfast
Description-Content-Type: text/markdown

# LazyFast

LazyFast is a lightweight Python library designed for building modern web interfaces using a component-based approach. It enables writing page logic on the server side in Python, integrating seamlessly with FastAPI. With LazyFast, interactive elements like inputs, buttons, and selects trigger component reloads that occur on the server, updating the component's state dynamically.

## What problems does LazyFast solve?
If you're a Python developer familiar with the basics of HTML, CSS, and JavaScript, but don’t want to dive into modern JavaScript frameworks, LazyFast lets you build web applications that meet modern standards—without needing to learn complex frontend frameworks.

## When is LazyFast a good fit?
1. **Low to Medium Traffic Projects**: LazyFast is ideal for projects with low to moderate traffic. Server-side rendering offloads work from the client’s machine, which can improve performance for users. However, for projects with high traffic volumes, server costs may increase due to the heavier backend load.
2. **Prototyping**: LazyFast was created to address the challenges I faced in my own work. Building prototypes and demos often needs to be fast and efficient, but involving frontend developers can slow things down and increase costs. I’ve worked extensively with Streamlit, but while it's quick, it has significant limitations and tends to produce applications that all look the same.

## Key Features

1. **Component-Based Server Rendering**: Build web interfaces using lazy loaded components that encapsulate logic, state, and presentation. 
2. **Server-Side Logic**: Handle interactions and state management on the server, reducing client-side complexity.
3. **FastAPI Integration**: Each component or page is a FastAPI endpoint, allowing for dependency injection and other FastAPI features.
4. **Lightweight**: The only dependencies are FastAPI for Python and HTMX for JavaScript, which can be included via CDN.
5. **State Management**: Utilize a state manager that can trigger component reloads, ensuring a reactive user experience.

## Installation

To install LazyFast, use pip:

```bash
pip install lazyfast
```
or
```bash
poetry add lazyfast
```

## Quick Start

Here's an example application to demonstrate how LazyFast works:

```python
from fastapi import FastAPI, Request
from lazyfast import LazyFastRouter, Component, tags


# LazyFastRouter inherits from FastAPI's APIRouter
router = LazyFastRouter()

# Define a lazy-loaded HTML component powered by HTMX
@router.component()
class MyComponent(Component):
    title: str

    async def view(self, request: Request) -> None:
        tags.h1(self.title, class_="my-class")

        with tags.div(style="border: 1px solid black"):
            tags.span(request.headers)

# Initialize the page dependencies for component rendering
# The page endpoint is also a FastAPI endpoint
@router.page("/{name}")
def root(name: str):
    with tags.div(class_="container mt-6"):
        MyComponent(title=f"Hello, World from {name}")

# Embed the router in a FastAPI app
app = FastAPI()
app.include_router(router)
```
If you use `uvicorn` instead as a server and want to reload on changes, use the following command:
```bash
uvicorn app:app --reload --timeout-graceful-shutdown 1
```

## Documentation
Documentation can be found [here](https://github.com/nikirg/lazyfast/blob/main/DOCS.md).

## Examples
You can find examples in [examples](https://github.com/nikirg/lazyfast/tree/main/examples).

## License

LazyFast is licensed under the [MIT License](https://github.com/nikirg/lazyfast/blob/main/LICENSE).

