Great question! Your dictionary comprehension gets the job done, but there are much more readable, Pythonic approaches that align with the "stack and map" pattern you're asking about—where you iterate through each item (map) and accumulate ("stack") them into the appropriate groups. Let's break down a few options:

1. Using collections.defaultdict (Most Intuitive "Stack" Approach) #

This is my go-to for reverse mappings. It handles list initialization automatically, so you can focus on stacking each file into its owner's group:

from collections import defaultdict

files = {'Code.py': 'Stan', 'Output.txt': 'Randy', 'Input.txt': 'Randy'}
mapped = defaultdict(list)

# Iterate through each file-owner pair (map step)
for file_name, owner in files.items():
    # Stack the file into the owner's list
    mapped[owner].append(file_name)

# Optional: Convert back to a regular dict if needed
mapped = dict(mapped)

This gives you exactly the result you want: {'Stan': ['Code.py'], 'Randy': ['Output.txt', 'Input.txt']}. The logic is linear and easy to debug—perfect for when you need to modify the mapping logic later.

2. Manual Dictionary Accumulation (No External Dependencies) #

If you don't want to import defaultdict, you can handle list creation manually. This follows the same stack-and-map pattern, just with explicit checks:

files = {'Code.py': 'Stan', 'Output.txt': 'Randy', 'Input.txt': 'Randy'}
mapped = {}

for file_name, owner in files.items():
    # If the owner isn't in the result yet, create an empty list for them
    if owner not in mapped:
        mapped[owner] = []
    # Stack the file onto their list
    mapped[owner].append(file_name)

This is a bit more verbose, but it's self-contained and great for understanding the underlying mechanics of reverse mapping.

How This Compares to Your Original Approach #

Your nested dictionary comprehension works, but it's harder to parse at a glance—especially for someone reading your code later. The stack-and-map style loops above are more maintainable, and they make it easier to add extra logic (like filtering certain files or transforming owner names) without turning the code into a nested mess.

内容的提问来源于stack exchange，提问作者ThisGuyCantEven