Metadata-Version: 2.4
Name: py-lamina
Version: 6.1.0
Summary: Add a new layer ("lâmina") to AWS lambda functions
License: MIT
License-File: LICENSE
Keywords: aws,lambda,decorator
Author: Chris Maillefaud
Author-email: chrismaille@users.noreply.github.com
Requires-Python: >=3.11,<4
Classifier: Development Status :: 4 - Beta
Classifier: Environment :: Plugins
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3.14
Requires-Dist: asgiref
Requires-Dist: case-converter
Requires-Dist: loguru
Requires-Dist: mistune
Requires-Dist: pydantic
Requires-Dist: python-magic
Project-URL: Repository, https://github.com/megalus/lamina
Description-Content-Type: text/markdown

<h1 align="center">Welcome to Lamina</h1>

<p align="center">
<a href="https://pypi.org/project/py-lamina/" target="_blank">
<img alt="PyPI" src="https://img.shields.io/pypi/v/py-lamina"/></a>
<a href="https://www.python.org" target="_blank">
<img alt="PyPI - Python Version" src="https://img.shields.io/pypi/pyversions/py-lamina"/>
</a>
<a href="https://github.com/megalus/lamina/blob/main/LICENSE" target="_blank">
<img alt="License: MIT" src="https://img.shields.io/github/license/megalus/lamina"/>
</a>
<a href="https://github.com/megalus/lamina/actions/workflows/tests.yml" target="_blank">
<img alt="Build Status" src="https://img.shields.io/github/actions/workflow/status/megalus/lamina/tests.yml?branch=main"/>
</a>
</p>

## Overview

Lamina (from Portuguese "lâmina", meaning "layer" or "blade") is a lightweight decorator library for AWS Lambda functions. It adds a powerful layer to your Lambda handlers, simplifying development by:

- Integrating both synchronous and asynchronous code in a single function
- Using Pydantic models for robust input and output data validation
- Handling errors gracefully with appropriate HTTP status codes
- Formatting responses according to AWS API Gateway expectations
- Supporting different content types (JSON, HTML, plain text)
- Providing convenient access to the original event and context objects

## Why use Lamina?

AWS Lambda functions often require repetitive boilerplate code for input validation, error handling, and response formatting. Lamina eliminates this boilerplate, allowing you to focus on your business logic while it handles:

- Input validation using Pydantic models
- Error handling with appropriate HTTP status codes
- Response formatting with content type control
- Support for both synchronous and asynchronous functions
- Custom headers support
- AWS Step Functions integration

## Installation

```shell
$ pip install py-lamina
```

Lamina requires Python 3.11 or later and has dependencies on:
- pydantic - For data validation
- asgiref - For async/sync conversion utilities
- loguru - For logging

## Usage

### Basic Example

Create the models for Input and Output data:

```python
# schemas.py
from pydantic import BaseModel

class ExampleInput(BaseModel):
    name: str
    age: int

class ExampleOutput(BaseModel):
    message: str
```

Create your AWS Lambda handler:

```python
# main.py
from typing import Any, Dict
from lamina import lamina, Request

@lamina(schema_in=ExampleInput, schema_out=ExampleOutput)
def handler(request: Request) -> Dict[str, Any]:
    response = {"message": f"Hello {request.data.name}, you are {request.data.age} years old!"}
    return response
```

#### Working with query parameters

You can also define Pydantic models for query parameters:

```python
# schemas.py
from pydantic import BaseModel
from typing import Any, Dict, Optional
from lamina import lamina, Request

class ExampleQueryParams(BaseModel):
    verbose: Optional[bool] = False

@lamina(params_in=ExampleQueryParams, schema_in=ExampleInput, schema_out=ExampleOutput)
def handler(request: Request) -> Dict[str, Any]:
    if request.query.verbose:
        response = {"message": f"Hello {request.data.name}, you are {request.data.age} years old!."}
    else:
        response = {"message": f"Hello World!"}
    return response
```


### Asynchronous Handlers

Lamina seamlessly supports both synchronous and asynchronous handlers:

```python
# main.py
import asyncio
from typing import Any, Dict
from lamina import lamina, Request

@lamina(schema_in=ExampleInput, schema_out=ExampleOutput)
async def handler(request: Request) -> Dict[str, Any]:
    # Perform async operations
    await asyncio.sleep(1)
    response = {"message": f"Hello {request.data.name}, you are {request.data.age} years old!"}
    return response
```

### Customizing Responses

#### Status Codes

The default status code is 200. You can customize it by returning a tuple:

```python
from typing import Any, Dict
from lamina import lamina, Request

@lamina(schema_in=ExampleInput, schema_out=ExampleOutput)
def handler(request: Request):
    response = {"message": f"Hello {request.data.name}, you are {request.data.age} years old!"}
    return response, 201  # Created status code
```

#### Content Types

Lamina autodiscovers the content-type based on the return type:

```python
from lamina import lamina, Request

@lamina(schema_in=ExampleInput)
def handler(request: Request):
    html = f"""
        <html>
            <head><title>User Profile</title></head>
            <body>
                <h1>Hello {request.data.name}!</h1>
                <p>You are {request.data.age} years old.</p>
            </body>
        </html>
    """
    return html
```

You can explicitly set the content type using the `content_type` parameter:

```python
@lamina(schema_in=ExampleInput, content_type="text/plain; charset=utf-8")
def handler(request: Request):
    return f"Hello {request.data.name}, you are {request.data.age} years old!"
```

#### Custom Headers

You can add custom headers by returning them as the third element in the response tuple:

```python
@lamina(schema_in=ExampleInput)
def handler(request: Request):
    response = {"message": f"Hello {request.data.name}!"}
    return response, 200, {
        "Cache-Control": "max-age=3600",
        "X-Custom-Header": "custom-value"
    }
```

## Hooks

Lamina provides four extensibility points executed around your handler.

Configuration (pyproject.toml):

```toml
[tool.lamina]
pre_parse_callback = "lamina.hooks.pre_parse"
pre_execute_callback = "lamina.hooks.pre_execute"
pos_execute_callback = "lamina.hooks.pos_execute"
pre_response_callback = "lamina.hooks.pre_response"
```

Environment variables override these values at runtime:
- LAMINA_PRE_PARSE_CALLBACK
- LAMINA_PRE_EXECUTE_CALLBACK
- LAMINA_POS_EXECUTE_CALLBACK
- LAMINA_PRE_RESPONSE_CALLBACK

Hook signatures and responsibilities:
- pre_parse(event, context) -> event
- pre_execute(request, event, context) -> request
- pos_execute(response, request) -> response
- pre_response(body, request) -> body

### The Request Object

The `Request` object provides access to:

- `data`: The validated input data (as a Pydantic model if schema_in is provided)
- `event`: The original AWS Lambda event
- `context`: The original AWS Lambda context
- `query`: Query parameters from the event (as a Pydantic model if params_in is provided)
- `headers`: Headers from the AWS Lambda event

### Using Without Schemas

You can use Lamina without schemas for more flexibility:

```python
import json
from lamina import lamina, Request

@lamina()
def handler(request: Request):
    # Parse the body manually
    body = json.loads(request.event["body"])
    name = body.get("name", "Guest")
    age = body.get("age", "unknown")

    return {
        "message": f"Hello {name}, you are {age} years old!"
    }
```

> **Note**: Without a schema_in, the `request.data` attribute contains the raw body string from the event. You'll need to parse and validate it manually.

### AWS Step Functions Integration

Lamina supports AWS Step Functions with the `step_functions` parameter:

```python
@lamina(schema_in=ExampleInput, schema_out=ExampleOutput, step_functions=True)
def handler(request: Request):
    # For Step Functions, the input is directly available as the event
    # No need to parse from event["body"]
    return {
        "message": f"Step function processed for {request.data.name}"
    }
```

### Error Handling

Lamina automatically handles common errors:

- **Validation Errors**: Returns 400 Bad Request with detailed validation messages
- **Type Errors**: Returns 400 Bad Request when input cannot be parsed
- **Serialization Errors**: Returns 500 Internal Server Error when output cannot be serialized
- **Unhandled Exceptions**: Returns 500 Internal Server Error with the error message

All errors are logged using the loguru library for easier debugging.

## OpenAPI (Swagger) 3.1 Generation

Lamina can generate an OpenAPI 3.1 document by inspecting your decorated handlers and the metadata you place inside decorator or in your Pydantic models using `json_schema_extra`.

### Define Path:
- You can  pass the path directly in the decorator: `@lamina(path="/items" ...)`.
- If `path` is omitted, Lamina will derive it from the function, method or package name in kebab-case (e.g., `foo_bar` -> `/foo-bar`).
- To define which object to use for path derivation, set the environment variable `LAMINA_USE_OBJECT_NAME` to one of: `function`, `method`, or `package`. The default is `function`.
- You can also define `use_object_name` in pyproject.toml under `[tool.lamina]`.

### Define Methods:
- You can pass accepted HTTP methods via the decorator: `@lamina(..., methods=["get", "post"])`.
- If omitted, the default is `POST` (API Gateway typical default).

### Define Models:
- Use Pydantic models for `schema_in`, `schema_out`, and `params_in` in the decorator.

### Define Responses:
- All views automatically include `400` and `500` responses that reflect Lamina's built-in error handling.
- Default return code for the `schema_out` is `200`. You can change it defining `LAMINA_DEFAULT_SUCCESS_STATUS_CODE` or `default_success_status_code` in pyproject.toml, under `[tool.lamina]`.
- Custom responses can be declared in Pydantic and added in the decorator via `responses={404: {"schema": ErrorOut}}`. These responses will override any existing status code.

### Define Authentication:
- Authentication settings can be provided to `get_openapi_spec`.
- If not provided, the default is API Key in header `Authorization`. You can change this header name by setting `LAMINA_DEFAULT_AUTH_HEADER_NAME` or `default_auth_header_name` in pyproject.toml, under `[tool.lamina]`.

### Define Summary, Description, and Tags:
- Operation summary/description are derived from the handler docstring when present
- The first line is used as summary
- The following free-text (until Args/Returns/etc.) as description.
- If no docstring is present, the generator falls back to json_schema_extra values; if neither exists, the summary becomes the function name in title case (e.g., foo_bar -> Foo Bar) and the description is empty.

### Adding/Remove the Handler from the Spec:
- You can exclude a handler from the generated spec by setting `add_to_spec=False` in the decorator (useful for HTML endpoints or internal views).
- Handlers without schemas (e.g., `@lamina()` with no `schema_in`/`schema_out`) are ignored by the spec generator unless sufficient metadata is available via models.

### Using json_schema_extra
You can add OpenAPI metadata to your Pydantic models using the `json_schema_extra` config:
```python
from pydantic import BaseModel, ConfigDict
class CreateItemIn(BaseModel):
    model_config = ConfigDict(
        json_schema_extra={
            "method": "post",  # HTTP method
            "summary": "Create an item",  # Operation summary
            "tags": ["items"],  # Tags for grouping
        }
    )
    name: str
```
Preference order is environment variables > decorator > model extras > defaults.

### Generating the OpenAPI Document
Call `get_openapi_spec(...)` to receive a Python dict ready to be dumped as JSON.

Example:

```python
from typing import Any, Dict
from pydantic import BaseModel, ConfigDict
from lamina import lamina, Request, get_openapi_spec


class CreateItemIn(BaseModel):
    model_config = ConfigDict(
        json_schema_extra={
            "method": "post",
            "summary": "Create an item",
            "tags": ["items"],
        }
    )
    name: str


class CreateItemOut(BaseModel):
    model_config = ConfigDict(json_schema_extra={"description": "Created item"})
    id: int
    name: str

# Default Values
# * Path: /create-item from LAMINA_USE_OBJECT_NAME=function
# * Method: POST
# * Response Status Codes: 200, 400, 500
# * Authentication: API Key in `Authorization` header
@lamina(schema_in=CreateItemIn, schema_out=CreateItemOut)
def create_item(request: Request) -> Dict[str, Any]:
    return {"id": 1, "name": request.data.name}


# Later (e.g., in a CLI or during startup)
spec = get_openapi_spec(title="My API", version="1.0.0", host="api.example.com", base_path="/v1")

# Dump as JSON
import json
print(json.dumps(spec, indent=2))
```

Example with custom settings:

```python
import os
from pydantic import BaseModel

class ParamsIn(BaseModel):
    paginate: bool
    next_token: str | None

class ErrorOut(BaseModel):
    detail: str

production_only = os.getenv("ENVIROMENT") == "production"

@lamina(
    path="/items/{id}",  # View Path
    params_in=ParamsIn,
    schema_in=CreateItemIn,
    schema_out=CreateItemOut,
    responses={503: {"schema": ErrorOut}},  # Extra responses
    methods=["GET", "POST"],  # Methods
    tags=["Item"],  # Tags
    add_to_spec=production_only  # Add to OpenApi spec?
)
def get_item(request: Request) -> Dict[str, Any]:
    """This is the Summary of the View in Swagger.

    This is the _description_ of the view in Swagger.

    * You can use GitHub Flavored Markdown (GFM) here, including tables and strikethrough.
    * Mermaid diagrams are supported in the docs, but are omitted from OpenAPI descriptions.
    * Everything below Args/Returns is ignored.

    Args:
        request (Request): Lamina Request Object.
    Returns:
        Dict[str, Any]: A dictionary containing the item details.
    """
        return {"id": 1, "name": request.data.name}

# Custom bearer auth
spec = get_openapi_spec(
    title="My API",
    version="1.0.0",
    security_schemes={"BearerAuth": {"type": "http", "scheme": "bearer"}},
    security=[{"BearerAuth": []}],
)
```

## Contributing

Contributions are welcome! Here's how you can help:

1. **Fork the repository** and clone it locally
2. **Create a new branch** for your feature or bugfix
3. **Make your changes** and add tests if applicable
4. **Run the tests** to ensure they pass: `make test`
5. **Submit a pull request** with a clear description of your changes

Please make sure your code follows the project's style guidelines by running:
```shell
poetry run make lint
```

### Development Setup

1. Clone the repository
2. Install dependencies with Poetry:
   ```shell
   poetry install
   ```
3. Install pre-commit hooks:
   ```shell
   poetry run pre-commit install
   ```

## License

This project is licensed under the terms of the MIT license.

