autocrud 0.8.2

自動產生 CRUD API 的 Python 函式庫

AutoCRUD

AutoCRUD is a model-driven automated FastAPI: built-in versioning, permissions, and search, focused on getting business logic to production quickly.

✨ Features

• 🧠 Focus only on business and models: Developers only need to focus on business logic and domain model schema; metadata, indexing, events, permissions, and other foundational capabilities are automatically handled by the framework.
• ⚙️ Automated FastAPI: Apply a model with one line of code to automatically generate CRUD routes and OpenAPI/Swagger—zero boilerplate, zero manual binding.
• 🗂️ Versioning: Native support for full revision history, draft in-place editing, revision switching and restore—ideal for auditing, rollback, and draft workflows.
• 🔧 Highly customizable: Flexible route naming, indexed fields, event handlers, and permission checks.
• 🏎️ High performance: Built on FastAPI + msgspec for low latency and high throughput.

🧙 Starter Wizard

Use the interactive Starter Wizard to generate a ready-to-run AutoCRUD project with your models, storage, and permissions configured — no boilerplate needed.

👉 https://hychou0515.github.io/autocrud/wizard/

Feature Overview

Feature	Description
✅ Auto-generation (Schema → API/Storage)	Schema as Infrastructure: automatically generates routes, logic bindings, and storage mappings
✅ Versioning (Revision History)	Draft→Update / Stable→Append, complete parent revision chain
✅ Migration	Functional Converter, Lazy Upgrade on Read + Save
✅ Storage Architecture	Hybrid: Meta (SQL/Redis) + Payload (Object Store) + Blob
✅ Scalability	Object Storage with decoupled indexing for horizontal scaling
✅ Partial Update (PATCH)	Precise JSON Patch updates for speed and bandwidth efficiency
✅ Partial Read	Skip unnecessary fields at msgspec decode time for speed and bandwidth efficiency
✅ GraphQL Integration	Auto-generated Strawberry GraphQL Endpoint
✅ Blob Optimization	BlobStore deduplication and lazy loading
✅ Permissions	Three-tier RBAC (Global / Model / Resource) and custom checkers
✅ Event Hooks	Customizable Before / After / OnSuccess / OnError for every operation
✅ Route Templates	Standard CRUD plus plug-in custom endpoints
✅ Search & Index	Meta Store provides efficient filtering, sorting, pagination, and complex queries
✅ Audit / Logging	Post-event audit records and review workflows
✅ Message Queue	Built-in async job processing; manage Jobs as resources with versioning and states

Installation

pip install autocrud

Optional Dependencies

For S3 storage support:

pip install "autocrud[s3]"

For BlobStore automatic Content-Type detection:

pip install "autocrud[magic]"

autocrud[magic] depends on python-magic.

• Linux: Ensure libmagic is installed (e.g., on Ubuntu run sudo apt-get install libmagic1).
• Other OS: See the python-magic installation guide.

Documentation

https://hychou0515.github.io/autocrud/

Your First API

from datetime import datetime, timedelta
from fastapi import FastAPI
from fastapi.testclient import TestClient
from autocrud import AutoCRUD
from msgspec import Struct

class TodoItem(Struct):
    title: str
    completed: bool
    due: datetime

class TodoList(Struct):
    items: list[TodoItem]
    notes: str

# Create AutoCRUD
crud = AutoCRUD()
crud.add_model(TodoItem)
crud.add_model(TodoList)

app = FastAPI()
crud.apply(app)
crud.openapi(app)

uvicorn.run(app, host="0.0.0.0", port=8000, log_level="info")

Auto-generated CRUD Endpoints

• POST /todo-item - Create
• GET /todo-item/{id}/data - Read
• PATCH /todo-item/{id} - JSON Patch update
• DELETE /todo-item/{id} - Soft delete
• GET /todo-list/data - List with search support
• A dozen more auto endpoints

➡️ AutoCRUD User Guide

Operate Resources via ResourceManager

ResourceManager is the entry point for resource operations in AutoCRUD. It manages create, query, update, delete, and versioning of resources.

Its core is versioning: every create/update/patch generates a new revision_id (create new revision) and preserves full history; drafts (draft) can be repeatedly edited with modify (in-place, without creating a new revision), then switched to stable when confirmed. You can also list all revisions, read any revision, switch the current revision, or restore after a soft delete. Indexed queries support filtering, sorting, and pagination by metadata and data fields (indexed fields), making it ideal for auditing, rollback, and large-scale retrieval.

➡️ ResourceManager Guide

🚀 Quick Start

from datetime import datetime, timedelta
from fastapi import FastAPI
from fastapi.testclient import TestClient
from autocrud import AutoCRUD
from msgspec import Struct

class TodoItem(Struct):
    title: str
    completed: bool
    due: datetime

class TodoList(Struct):
    items: list[TodoItem]
    notes: str

# Create CRUD API
crud = AutoCRUD()
crud.add_model(TodoItem)
crud.add_model(TodoList)

app = FastAPI()
crud.apply(app)

# Test
client = TestClient(app)
resp = client.post("/todo-list", json={"items": [], "notes": "My todos"})
todo_id = resp.json()["resource_id"]

# Add an item using JSON Patch
client.patch(f"/todo-list/{todo_id}", json=[{
    "op": "add", 
    "path": "/items/-",
    "value": {
        "title": "Complete item",
        "completed": False,
        "due": (datetime.now() + timedelta(hours=1)).isoformat()
    }
}])

# Get the result
result = client.get(f"/todo-list/{todo_id}/data")
print(result.json())

Start the development server:

python -m fastapi dev main.py

Visit http://localhost:8000/docs to view the auto-generated API documentation.