Advent of Code - Day 3

< Prev Day Next Day >

My Code

import re

with open("day3.txt") as f:
    instructions = f.read()
    multipliers = re.findall("mul\\(\\d\\d?\\d?,\\d\\d?\\d?\\)", instructions)
    
    sum = 0
    for item in multipliers:
        nums = list(map(int,item.strip("mul()").split(",")))
        sum += nums[0] * nums[1]

print(sum)

ChatGPT’s Improved Code

import re

# Open the file using a context manager and read its content
with open("day3.txt", "r") as file:
    instructions = file.read()

# Use a more descriptive variable name for the matches
pattern = r"mul\((\d{1,3}),(\d{1,3})\)"
matches = re.findall(pattern, instructions)

# Calculate the sum in a more Pythonic way using a generator expression
total_sum = sum(int(a) * int(b) for a, b in matches)

# Print the result
print(total_sum)

What I Learned

Regex review

• * means the character before can be matched 0 or >=1 number of times, not necessarily just once
• Can group characters together with []. [abx]* means match 0 or more characters in the group a,b,x
• + matches the character before it 1 or more times (NOT 0)
• ? matches the character before it 1 or 0 times (NOT multiple)
• {m,n} matches the character before it at least m times and at most n times
• {m} with no comma matches the character before it exactly m times

Capturing Groups in regex

• We can avoid the need to “strip” the ‘mul’ portion of the matched strings using a capturing group (“()”) in the regex.
• Wrapping a portion of the regex in parantheses means that the whole regex will be matched, but only the items inside the group will be returned
• Doing this makes the sum code much simpler, because we can use a generator expression inside of sum() instead of having to write out the full loop

Better Syntax

• It is cleaner to distinctly label the pattern for the regex, and pass this to findall.
• It is also easier to read when we store the matches variable.

Callista Krebs, 2024