Home > Enterprise >  Avoiding code duplication in Flask RESTful endpoints
Avoiding code duplication in Flask RESTful endpoints

Time:09-17

I am writing APIs in Flask RESTful framework. For every end point, I need to do a check which are same in structure but process inside are different. Here is the snippet.

def get_all_users():
    users = [{"name":"John","dob":"12-10-1990"}, {"name":"David","dob":"25-03-1995"}, {"name":"Maria","dob":"30-02-1998"}]
    return users

Endpoint 1:

class ClassOne(Resource):
    def get(self):
        # Get required info from body
        res_body = request.json
        name = res_body.get('name', None)
        item_list = []
        if name is None:
            users = get_all_users()
            for user in users:
                item_list.append(user['name'])
            return {"nameList":item_list}
        else:
            user = get_user_data(user_info=name)
            if user:
                name = user['name']
                return {"Message":"User {} exists!".format(name)}
            else:
                return {"Error":"User does not exist"}

        return "output based on the above process"

Endpoint 2:

class ClassTwo(Resource):
    def get(self):
        # Get required info from body
        res_body = request.json
        name = res_body.get('name', None)
        item_list = []
        if name is None:
            users = get_all_users()
            for user in users:
                item_list.append(user['dob'])
            return {"nameList":item_list}
        else:
            user = get_user_data(user_info=name)
            if user:
                dob = user['dob']
                return {"Message":"User dob is {}".format(dob)}
            else:
                return {"Error":"User does not exist"}
        
        return "output based on the above process"

What is the best way to write these codes by avoiding code duplication.

CodePudding user response:

Assuming Resource doesn't have a .get(), your endpoint classes could inherit from a common Process class or whatever you want to call it, in addition to Resource:

class Process:
    def get(self, user_key):
        res_body = request.json
        name = res_body.get("name")
        if name is None:
            item_list = []
            users = get_all_users()
            for user in users:
                item_list.append(user.get(user_key))
            return {"nameList": item_list}
        else:
            user = get_user_data(user_info=name)
            if user:
                info = user.get(user_key)
                return {"Message": f"User {user_key} is {info}"}
        return {"Error": "User does not exist"}


class ClassOne(Resource, Process):
    def __init__(self):
        super().__init__()
   
    def get(self):
        return super().get(user_key="name")


class ClassTwo(Resource, Process):
    def __init__(self):
        super().__init__()
   
    def get(self):
        return super().get(user_key="dob")

Here's a quick demo of this inheritance structure:

In [1]: class Grandparent:  # Resource
   ...:     pass
   ...:

In [2]: class Parent:  # Process
   ...:     def get(self, x, y):
   ...:         print(x, y)
   ...:

In [3]: class ChildA(Grandparent, Parent):  # Endpoint 1
   ...:     def get(self):
   ...:         super().get(1, 2)
   ...:

In [4]: class ChildB(Grandparent, Parent):  # Endpoint 2
   ...:     def get(self):
   ...:         super().get("hello", "world")
   ...:

In [5]: a = ChildA()

In [6]: b = ChildB()

In [7]: a.get()
1 2

In [8]: b.get()
hello world

Alternatively, you could simply have Process inherit from Resource and then have your endpoints inherit from Process.

  • Related