AI, Engineering Assistance - Code Generation

Summary

Artificial intelligence techniques, such as machine learning and natural language processing, can be used to automatically generate code snippets, functions, or even entire programs. It can also help avoid developers having to type repetitive or boilerplate code.

BRAINSTORMING IDEAS

Tip

Use AI to suggest innovative ideas and solutions based on a problem statement or requirements.

Tools: Employ Natural Language Processing (NLP) models or Generative AI to brainstorm and propose diverse concepts.

Example Prompt: Generate some ideas for an arcade game based on pickle ball

AI

Response:

1. Pickleball Pro Smash:

• Gameplay: This game involves a fast-paced, arcade-style setting where players control a pickleball player aiming to hit the ball against a series of targets or obstacles.

• Objective: Players must time their shots to hit targets for points, break through barriers, or avoid obstacles while maintaining a rally to achieve a high score within a limited time.

• Special Features: Introduce different court settings (indoor, outdoor, beach) with varying obstacles and challenges. Power-ups like precision aiming or ball control might add depth to the gameplay.

…

Example Follow-Up Prompt(s):

• Give me more detail about the gameplay for the first idea
• Give me a suggestion for a game engine to use to create the game

---

CODE SKELETON CREATION

Use AI to generate basic code structures or skeletons based on given input or specifications.

Tools: Explore code generation models or templates within integrated development environments (IDEs) that assist in initial code setup.

Example Prompt: Generate secure code to retrieve a username a password from the command line in python

AI

Response:

import getpass
 
def get_credentials():  
 username = input(“Enter username: “)  
 password = getpass.getpass(“Enter password: “)
 return username, password
 
if __name__ == “__main__”:  
 retrieved_username, retrieved_password = get_credentials()  
 # Perform operations with retrieved_username and retrieved_password print(“Username:”, retrieved_username)  
 # Avoid printing passwords in real applications  
 # For demonstration purposes only  
 print(“Password:”, retrieved_password)

This script utilizes the getpass.getpass() function to securely read the password without displaying it on the console. Upon execution, it will prompt the user to enter a username and password in the terminal.

Remember, it’s essential to handle passwords securely. Avoid storing them in plain text or displaying them unnecessarily. Instead, consider hashing or encrypting sensitive information before storage and transmission.

---

TEST DATA GENERATION

Leverage AI algorithms to create diverse and comprehensive test data sets.

Tools: Apply AI-powered data generation tools that use machine learning to generate test data relevant to the codebase.

Example Prompt: Generate a set of test data of people with the following attributes: first name, last name, age, occupation, region, favorite color, favorite sport, and no. countries visited.

AI

Response:

test_data = [
 {  
 	“first_name”: “John”,
 	“last_name”: “Doe”,  
 	“age”: 30,  
 	“occupation”: “Engineer”,
 	“region”: “California, USA”,
 	“favorite_color”: “Blue”,
 	“favorite_sport”: “Basketball”,
 	“countries_visited”: 5
 },
 {
 	"first_name": "Emma",  
 	"last_name": "Johnson",  
 	"age": 32,  
 	"occupation": "Software Engineer",
 	"region": "London, UK",
 	"favorite_color": "Blue",
 	"favorite_sport": "Tennis",
 	"countries_visited": 8
  },
 # ...
]
 
# Accessing individual data  
print("First Person's Last Name:", people_data[0]["last_name"])
print("Second Person's Age:", people_data[1]["age"])

This code creates a list of dictionaries, where each dictionary holds the attributes of an individual such as first name, last name, age, occupation, region, favorite color, favorite sport, and number of countries visited. You can access and manipulate individual data by indexing the list and using dictionary keys.

Example Follow-Up Prompt: Create a text fixture that loads this data from a file.

AI

Response:

1. First, create a JSON file named test_data.json.

2. Then, create a pytest test script, let’s call it test_data_fixture.py, which will load this JSON file as a fixture:

import json import pytest
 
@pytest.fixture  
def loaded_test_data():
 with open(‘test_data.json’, ‘r’) as file:
 	test_data = json.load(file)
 	return test_data
 
def test_loaded_test_data(loaded_test_data):  
 assert len(loaded_test_data) == 3 # Assuming there are three entries in the test data  
 # Add more assertions as needed to test the loaded data ...