如何将CSV数据以字典形式读取并另存？附指定Commit数据详情

Alright, let's solve this problem step by step. You need to read a CSV file with sha and commit columns (where the commit column contains nested dictionary data), load it as a list of dictionaries, then write that data back to a new CSV file. Here's a practical Python solution using built-in modules:

Step 1: Import Required Modules #

We'll use Python's built-in csv module for handling CSV files, and ast to safely parse the string-formatted dictionary in the commit column:

import csv
import ast

Step 2: Read the Original CSV and Parse Data #

This code reads the CSV, converts each row to a dictionary, and parses the commit string into an actual Python dictionary:

# Initialize a list to store our parsed data
parsed_data = []

# Open the input CSV file
with open('input.csv', mode='r', newline='', encoding='utf-8') as infile:
    # Use DictReader to automatically map rows to dictionaries using headers
    reader = csv.DictReader(infile)
    
    for row in reader:
        # Parse the commit column's string into a dictionary
        # ast.literal_eval is safer than eval() for parsing literal structures
        row['commit'] = ast.literal_eval(row['commit'])
        parsed_data.append(row)

Step 3: Write the Parsed Data to a New CSV #

Now we'll write our list of dictionaries back to a new CSV. We need to convert the commit dictionary back to a string to store it in the CSV column:

# Open the output CSV file for writing
with open('output.csv', mode='w', newline='', encoding='utf-8') as outfile:
    # Define the column headers (matches the original CSV)
    fieldnames = ['sha', 'commit']
    writer = csv.DictWriter(outfile, fieldnames=fieldnames)
    
    # Write the header row first
    writer.writeheader()
    
    # Write each parsed row to the output CSV
    for item in parsed_data:
        writer.writerow({
            'sha': item['sha'],
            'commit': str(item['commit'])
        })

Key Notes #

• Safety with ast.literal_eval: We use this instead of eval() because it only evaluates safe literal Python structures (dictionaries, lists, strings, numbers), preventing execution of malicious code that could be hidden in the CSV.
• JSON Alternative: If your commit data uses double quotes (valid JSON format), you can replace ast.literal_eval(row['commit']) with json.loads(row['commit']) and str(item['commit']) with json.dumps(item['commit']) for strict JSON compliance.

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