Smart Inline Completions: Use Cases, Examples, and Exercises
34%
Ask Your Code Questions
You can use Copilot to understand a code snippet by asking questions about it. Remember that the models powering Copilot are LLMs, so they understand human language and can answer questions based on a given context. The context is the code snippet you provide, and the question is the prompt you task Copilot with.
Let's say you have the following code snippet:
class Book:
def __init__(self, title, author):
self.title = title
self.author = author
self.is_borrowed = False
class Patron:
def __init__(self, name):
self.name = name
self.books = []
def borrow(self, book):
if not book.is_borrowed:
self.books.append(book)
book.is_borrowed = True
return f"{book.title} has been borrowed."
return "Book is currently unavailable."
class Library:
def __init__(self):
self.inventory = []
def add_book(self, title, author):
self.inventory.append(Book(title, author))
def get_book_info(self, title):
for book in self.inventory:
if book.title == title:
return f"{title} by {book.author}, {'Available' if not book.is_borrowed else 'Borrowed'}"
return "Book not found."
# Usage
lib = Library()
lib.add_book("Animal Farm", "George Orwell")
lib.add_book("Brave New World", "Aldous Huxley")
patron = Patron("Alice")
print(patron.borrow(lib.inventory[0])) # Attempt to borrow Animal Farm
print(lib.get_book_info("Animal Farm"))
At the end of the code, you can add your question and wait for Copilot to generate the answer. For example, you can ask the following:
# What does the Book class represent and what are its attributes?
# [Wait for the suggestion]
A better approach would be to add an indication that you're asking a question. This can be done by adding Q: before the question and A: before the answer.
Building with GitHub Copilot
From Autocomplete to Autonomous AgentsEnroll now to unlock all content and receive all future updates for free.