Python循环逻辑问题：如何将间接引用内容归类至根引用类别

Hey there! Let's tackle this problem—you want to group not just articles that directly quote A, but also those that quote articles which themselves quote A (like D and E quoting C, which quotes A). Here's how to adjust your code to handle these indirect references:

First, let's recap your existing setup so we're on the same page:

import pandas as pd

art1 = ["A",'B','C','D','E','F']
quote = ["",'A','A','C','C','A']
g2 = pd.DataFrame({"Article":art1,"Quote":quote})

This gives us a DataFrame where:

• A has no quotes
• B, C, F directly quote A
• D, E directly quote C (so they indirectly quote A)

The Core Idea: Trace the Reference Chain #

We need to build a "root reference" for each article—this is the final article that all its references lead back to. For D, that root is A (D → C → A); for E, it's also A.

Solution 1: Recursive Reference Traversal #

This uses a simple recursive function to climb up the reference chain until we hit a root (an article with no quote):

def find_root(article, quote_map):
    # Stop if the article has no quote, or quotes itself
    if not quote_map[article] or quote_map[article] == article:
        return article
    # Keep climbing up the reference chain
    return find_root(quote_map[article], quote_map)

# Create a dictionary mapping each article to its direct quote
quote_map = g2.set_index('Article')['Quote'].to_dict()

# Add a new column with the root reference for each article
g2['Root_Quote'] = g2['Article'].apply(lambda x: find_root(x, quote_map))

# Now get all articles whose root is A
articles_under_a = g2[g2['Root_Quote'] == 'A']['Article'].tolist()
print(articles_under_a)  # Output: ['A', 'B', 'C', 'D', 'E', 'F']

Solution 2: Iterative Traversal (For Long Reference Chains) #

If you're worried about hitting recursion limits with extra-long reference chains, use an iterative approach instead:

quote_map = g2.set_index('Article')['Quote'].to_dict()
root_reference = {}

for article in quote_map:
    current_article = article
    # Traverse up the chain until we find a root or a precomputed root
    while quote_map[current_article] and quote_map[current_article] != current_article and quote_map[current_article] not in root_reference:
        current_article = quote_map[current_article]
    # Assign the root reference for the original article
    root_reference[article] = root_reference.get(current_article, current_article)

# Map the root references back to your DataFrame
g2['Root_Quote'] = g2['Article'].map(root_reference)

# Get all articles linked to A
articles_under_a = g2[g2['Root_Quote'] == 'A']['Article'].tolist()
print(articles_under_a)  # Same output as above

How to Replace Your Existing For Loop #

If your original for loop only checked direct quotes (like q == 'A'), you can replace that logic with the Root_Quote column. Now you'll capture every article that traces back to A, whether directly or indirectly.

内容的提问来源于stack exchange，提问作者judy