仅将第一行写入CSV-不明白为什么

Question

I need to convert multiple JSON file to a CSV file. I have a data.zip folder that contains 2000 .JSON file. I am able to load the JSON file and print the data inside each JSON file with print normalized. But when I write it to CSV it only shows 1 line of data. How do I write all the .JSONS data tot he CSV file.

import zipfile
from pandas.io.json import json_normalize

def get_text():
    with zipfile.ZipFile("data.zip", "r") as z:
         for filename in z.namelist():
             with z.open(filename) as f:
                  data = json.load(f)
                  normalized = json_normalize(data)
                  print normalized
                  normalized.to_csv('data.csv', encoding="utf-8-sig")

Answer 1

Default for .to_csv is to write a new file. You need to append - mode='a'. See:

Panda's Write CSV - Append vs. Write

Answer 2

As some people explained you're overwriting your CSV file instead of appending it. Try appending and only writing headersThe first time around.

Here's an example:

def get_text():
    add_csv_header = True
    with zipfile.ZipFile("data.zip", "r") as z:
         for filename in z.namelist():
             with z.open(filename) as f:
                  data = json.load(f)
                  normalized = json_normalize(data)
                  print normalized
                  normalized.to_csv('data.csv', encoding="utf-8-sig", mode='a', header=add_csv_header)
                  add_csv_header = False