This project can be used as a starter for learning python api service development, the api service is developed using Sanic web framework, SQLAlchemy sql toolkit, etc. There is also an article Python Sanic 高并发服务开发指南 for learning this project.
To students of course: 最新的实战项目目录结构已按照 干净架构 进行了整理，另外还使用了 IoC 容器 来简化业务对象的管理。代码结构调整较大，如想获取之前版本，可查看提交记录。
- Sanic Asynchronous web framework and server
- SQLAlchemy SQL toolkit and orm framework
- AIOMySQL Asynchronous access mysql
- AIORedis Asynchronous access redis
- MarshMallow Object serialization
- Fire CLI application framework
- APScheduler Run interval jobs
- Dependency Injector Dependency injection microframework for Python
| Path | Method | Description |
|---|---|---|
| /user/register | POST | Register |
| /user/login | POST | Login |
| /user/logout | GET | Logout |
| /user/logged | GET | Get logged user |
| /user/modify | POST | Modify logged user |
| /user/info | GET | Get user info |
| /user/follow | POST | Follow user |
| /user/unfollow | POST | Unfollow user |
| /user/following | GET | Following users of someone |
| /user/follower | GET | Fans of some user |
| /user/sendMobileVerifyCode | POST | Send mobile verify code |
| /post/publish | POST | Publish post |
| /post/delete | POST | Delete post |
| /post/info | GET | Get post info |
| /post/published | GET | Get published posts of some user |
| /post/like | POST | Like post |
| /post/unlike | POST | Unlike post |
| /post/liked | GET | Liked posts of some user |
| /post/following | GET | Posts of following users of someone |
| /file/upload | POST | Upload file |
| /file/info | GET | Get file meta info |
| /message/ws | Websocket | Create websocket connection for message |
This project need python v3.7+.
git clone https://github.com/jaggerwang/sanic-in-practice.git && cd sanic-in-practice
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txtYou can quit the virtual environment by execute command deactivate.
Install mysql and redis server, and start them. After mysql started, create a database for this project, and a user to access this database.
CREATE DATABASE `sip`;
CREATE USER 'sip'@'%' IDENTIFIED BY '123456';
GRANT ALL PRIVILEGES ON `sip`.* TO 'sip'@'%';Change configs in weiguan/config/base.py as your need, especially mysql, redis and path related configs. You can also change configs by environment variables, you need add WG_ prefix to each config you want to change.
python -u -m weiguan.cli.app model create-tablespython -u -m weiguan.api.apppython -u -m weiguan.cli.app schedule runThe api service's endpoint is http://localhost:8000/.
You need install Docker and Docker Compose at first.
Change the content of docker-compose.yml as your need, especially the host path of mounted volumes.
docker-compose up -dIt will start server, scheduler, mysql and redis services. If you need to stop and remove all services, you can execute command docker-compose down. The container port 8000 is mapping to the same port on local host, so the endpoint of api service is same as the local way.
When first start mysql, it will auto create a database sip and a user sip with password 123446 to access this database. The password of root user is also 123456.
Login to server container and create tables as previous.
docker container exec -it sanic-in-practice_server_1 bash
python -u -m weiguan.cli.app model create-tablesThis project support developing in VSCode's remote container, you can create your own development environment in just one click. You need install vscode and it's extension "Remote - Containers" at first.
Configure docker-compose.dev.yml and .devcontainer/devcontainer.json as your need, especially the host path of mount volumes. The file docker-compose.dev.yml will override some configs of the base file docker-compose.yml for better developing experience. Such as auto build application image, mount local project folder into container, etc. It also mount host's ~/.ssh folder into container for executing git command in container.
Click remote development button at bottom-left corner of vscode, or open command palette, then execute command Remote-Containers: Reopen Folder in Container. It'll build application image, and start a server container, including it's dependency containers, as this project's development environment.
The workspace folder of the remote project is at /workspace, which mounted the local project's folder. You can now run and restart the remote project as nomal local project. If you need execute some command, you can login to the server container or use docker container exec.
The left steps is the same as run application by docker compose, such as create database and tables, access apis, etc.