Automatic generate a flask server project by command.(easy-flask)

Start

Install

1

pip3 install easy-flask-restful

Make a flask server project at current dir:

1
2
3
4
5
6
7
8
9

easy-flask -n demo_server
# or
easy-flask --new=demo_server

# start server in debug mode
cd demo_server
python3 app.py
# or
sh control.sh start|restart|stop

Introduce

By this tool you don`t need care how to make a flask server project.
What you need to do is write api-file in easy format inside dir api.
This project will automatically load uri by api-file, and pack json response with same format.

Project structure

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

.
├── api
│   ├── __init__.py
│   └── demo
│       ├── __init__.py
│       └── demo_api.py
├── conf
│   ├── __init__.py
│   ├── config.ini
├── internal
│   ├── __init__.py
│   ├── error.py
│   └── utils.py
├── app.py
├── start.py
├── control.sh
└── logs
    └── app.log

api - write api files here.

internal.error - define error info here.

app.py - start app by this file, you can rename it.

conf - config for application

control.sh - run app with command by gunicorn

Write Api File

For example:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

from flask import g

from internal import Resource


class DemoApi(Resource):

    uri = ['/api/demo']

    def get(self):
        req_p = {
            'num': {'type': int, 'min': 10, 'max': 100},
            'print': {'type': str, 'required': True},
            'default': {'type': str, 'default': 'demo'}
        }

        data, err = self.parse_request_data(req_pattern=req_p)
        if err:
            g.logger.warning(err)
            return err
        g.logger.info('success')
        return {'result': data}

uri - required by all api file

g.logger is loaded before each request for making different log_id.

self.parse_request_data can help you to parse params, it will return ParamsError when param invalid.

Normal response:

1
2
3
4
5
6
7
8
9
10
11
12

{
    "errno": 0,
    "data": {
        "result": {
            "num": 100,
            "print": "aaaa",
            "default": "demo"
        }
    },
    "msg": "success",
    "log_id": "f40c889d1b5744c7a87a9045aea8595c"
}

Error response:

return errno and msg according to error.py.

1
2
3
4
5

{
    "errno": 2,
    "msg": "params error:num check max invalid",
    "log_id": "7c53531b93a4406e9d3050bae5b99968"
}

Suggest

A real project must have complex logic and interactions with DB or other service.
By this tool, you just can get a simple framework. So I give some suggestions here for writing better code.

• logic dir (or other name) for writing complex logic, keep simple in api file.
• if you need db operation, use flask-sqlalchemy, and mkdir dto.
• client dir for interact with other service, and make BaseClient class(inherited by other client), because you never want to write same code for many times.