Section 1.1: Prioritizing Code Reusability

During my student days, I often neglected the idea of reusability:

def func1():

# step A

# step B

# step C

# step D

def func2():

# step A

# step B

# step C

# step E

def func3():

# step A

# step B

# step C

# step F

In the example above, steps A, B, and C are repeated across three functions. This redundancy not only makes the code less reusable but also complicates future updates.

A better approach in enterprise applications could involve creating a utility function:

def util_func():

# step A

# step B

# step C

def func1():

util_func()

# step D

def func2():

util_func()

# step E

def func3():

util_func()

# step F

By consolidating common steps into a single function, we enhance reusability and maintainability.

Section 1.2: Enhancing Code Readability

As a novice, I often used single-letter variable names like x, a, and b. However, after gaining experience, I've learned the value of descriptive naming, such as person_id and person_name.

I also overlooked type hinting, despite knowing it existed. Here’s an example without type hints:

def do_something(mylist, a):

return [i + a for i in mylist]

And with type hints:

from typing import List

def do_something(mylist: List[int], a: int) -> List[int]:

return [i + a for i in mylist]

Type hints improve code readability, making it easier for others to understand the expected types and return values.

Section 1.3: Implementing Authorization

In my early projects, I created APIs in FastAPI that were open to everyone. However, when developing enterprise applications, security is paramount. Each user should have defined roles that dictate their access to various endpoints.

Generally, role data is stored in a database, allowing for tailored access control based on user roles.

Section 1.4: Data Access Control

Beyond basic authorization, defining which data subsets users can access adds another layer of security. For example:

• Admins can access all data.
• Privileged users may access 80% of the data.
• Regular users can only interact with their own data.

This approach varies by application, emphasizing the necessity for tailored solutions.

Section 1.5: Utilizing Response Models in FastAPI

As a student, I often wrote minimal code to achieve functionality:

@app.get('/test')

def test(id):

return get_one(id)

Without a response model, understanding the output requires additional effort. To aid other developers, we can define a response model:

from pydantic import BaseModel

class Person(BaseModel):

name: str

age: int

gender: str

@app.get('/test', response_model=Person)

def test(id):

return get_one(id)

This approach clarifies what the API is expected to return, which can also be viewed in the Swagger UI.