Python实现嵌套JSON转CSV：新手遭遇输出格式不符问题求助

Hey there! Let's get your JSON-to-CSV conversion working exactly how you want it. The problem with your current code is that it's flattening all nested arrays into separate columns (like item_item_batters_batter_0_id), but we need to generate every possible combination of batter and topping paired with the base donut details—resulting in that m*n row count you mentioned.

Here's the Solution Code #

This script will iterate through each donut, then create a row for every batter-topping combination:

import json
import csv

def json_to_csv(json_data, csv_path):
    # Define the CSV headers to match your desired output
    headers = ["id", "type", "name", "batter", "topping"]
    rows = []
    
    # Loop through each top-level entry in your JSON array
    for entry in json_data:
        item = entry["item"]
        # Extract the core donut information that stays consistent across rows
        base_details = {
            "id": item["id"],
            "type": item["type"],
            "name": item["name"]
        }
        # Get the full list of batters and toppings for this donut
        all_batters = item["batters"]["batter"]
        all_toppings = item["topping"]
        
        # Generate every combination of batter and topping
        for batter in all_batters:
            batter_type = batter["type"]
            for topping in all_toppings:
                topping_type = topping["type"]
                # Combine base details with the current batter/topping pair
                row = {
                    **base_details,
                    "batter": batter_type,
                    "topping": topping_type
                }
                rows.append(row)
    
    # Write the collected rows to CSV
    with open(csv_path, "w", newline="", encoding="utf-8") as csv_file:
        writer = csv.DictWriter(csv_file, fieldnames=headers, quoting=csv.QUOTE_ALL)
        writer.writeheader()
        writer.writerows(rows)
    
    print(f"Done! Wrote {len(rows)} rows to {csv_path}")

if __name__ == "__main__":
    import sys
    # Simplified command line arguments (no need for the "node" parameter anymore)
    if len(sys.argv) != 3:
        print("\nUsage: python json_to_csv.py <json_in_file_path> <csv_out_file_path>\n")
        sys.exit(1)
    
    json_file = sys.argv[1]
    csv_output = sys.argv[2]
    
    # Load the JSON data from file
    with open(json_file, "r", encoding="utf-8") as f:
        raw_json = json.load(f)
    
    # Run the conversion
    json_to_csv(raw_json, csv_output)

Key Changes & Explanations #

• No more recursive flattening: Instead of turning arrays into columns, we use nested loops to create a row for every batter-topping pair (this is called a Cartesian product).
• Clean base details: We extract the fixed donut info (id, type, name) once per item, then reuse it for every combination.
• Simplified arguments: The original node parameter isn't needed here—we directly access the item key in each JSON entry.

How to Run It #

Use this command (adjust paths to match your files):

C:\Users\user1>python C:/Users/user1/PycharmProjects/pro1/venv/file3.py D:/jsonfile1.json D:/filecsv2.csv

What the Output Looks Like #

Your CSV will have exactly the format you want:

"id","type","name","batter","topping"
"0001","donut","Cake","Regular","None"
"0001","donut","Cake","Regular","Glazed"
"0001","donut","Cake","Regular","Sugar"
...
"0001","donut","Cake","Devil's Food","Maple"

This works even if you add more donut entries to your JSON—each will generate its own set of batter-topping rows automatically.

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