Python - 从多个 JSON 文件写入 CSV

Question

我整天都在尝试将多个 JSON 文件从 nys.gov 网站上提取的有关 COVID 数据的多个 JSON 文件写入一个（或多个）CSV 文件。

我可以成功地连接 JSON 文件，但无法以我可以用来制作图表的格式将它们连接在一起。我知道问题出在我的 for 循环中，但经过多次尝试，我还没有找到一种成功的方法，可以将数据附加到一个 csv 或创建多个 csv 文件，然后我可以在 pandas 中使用。这是我的代码，目前它似乎正在遍历循环并将最终的 json 转储到我的 csv 中......

import csv
import datetime 
import json
import pandas as pd
import urllib.request as request

one_day = datetime.datetime.today() - datetime.timedelta(days=1)
two_days = datetime.datetime.today() - datetime.timedelta(days=2)
thr_days = datetime.datetime.today() - datetime.timedelta(days=3)
for_days = datetime.datetime.today() - datetime.timedelta(days=4)
fiv_days = datetime.datetime.today() - datetime.timedelta(days=5)
six_days = datetime.datetime.today() - datetime.timedelta(days=6)
sev_days = datetime.datetime.today() - datetime.timedelta(days=7)
egt_days = datetime.datetime.today() - datetime.timedelta(days=8)

one_day_str = one_day.strftime("%Y-%m-%dT00:00:00.000")
two_day_str = two_days.strftime("%Y-%m-%dT00:00:00.000")
thr_day_str = thr_days.strftime("%Y-%m-%dT00:00:00.000")
for_day_str = for_days.strftime("%Y-%m-%dT00:00:00.000")
fiv_day_str = fiv_days.strftime("%Y-%m-%dT00:00:00.000")
six_day_str = six_days.strftime("%Y-%m-%dT00:00:00.000")
sev_day_str = sev_days.strftime("%Y-%m-%dT00:00:00.000")
egt_day_str = egt_days.strftime("%Y-%m-%dT00:00:00.000")

url_one = 'https://health.data.ny.gov/resource/xdss-u53e.json?test_date=' + one_day_str
url_two = 'https://health.data.ny.gov/resource/xdss-u53e.json?test_date=' + two_day_str
url_thr = 'https://health.data.ny.gov/resource/xdss-u53e.json?test_date=' + thr_day_str
url_for = 'https://health.data.ny.gov/resource/xdss-u53e.json?test_date=' + for_day_str
url_fiv = 'https://health.data.ny.gov/resource/xdss-u53e.json?test_date=' + fiv_day_str
url_six = 'https://health.data.ny.gov/resource/xdss-u53e.json?test_date=' + six_day_str
url_sev = 'https://health.data.ny.gov/resource/xdss-u53e.json?test_date=' + sev_day_str
url_egt = 'https://health.data.ny.gov/resource/xdss-u53e.json?test_date=' + egt_day_str

url_lst = [url_one,url_two,url_thr, url_for, url_fiv, url_six, url_sev, url_egt]

d = []

def write_json(data, filename='data.json'):
    with open(filename,'w') as f: 
        json.dump(data, f, indent=4)

for url in url_lst:
    with request.urlopen(url) as response:
        #   print(url)
        source = response.read()
        data = json.loads(source)

    if len(data) == 0:
        continue

    with open ("covid.json", 'w') as outfile:
        json.dump(data, outfile)

    with open('covid.json') as json_data:
        j = json.load(json_data)
        d.append(j)
        write_json(d)  
    
    filename = "County Stats.csv"
    fields = ["Date","County", "New Positives", "All Positives", "New Tests", "All Tests"]
    with open(filename, 'w') as fw:
        cf = csv.writer(fw, lineterminator='\n')
        # write the header
        cf.writerow(fields)
        
        for counties in data:
            date = counties['test_date']
            cnty = counties['county']
            new_pos = counties['new_positives']
            cum_pos = counties['cumulative_number_of_positives']
            new_tests = counties['total_number_of_tests']
            cum_tests = counties['cumulative_number_of_tests']
            cf.writerow([date,cnty, new_pos, cum_pos, new_tests, cum_tests])

我可能介于 python 初学者和中级之间，所以请原谅任何糟糕的编码习惯。提前致谢。

• 康纳

Answer 1

好的，在我的前任教授和知己的帮助下，我想通了！

有几点：

1. 我没有为 for 循环编制索引，在循环和索引增量 (index+=1) 之前添加索引变量为 csv 文件名创建了一个可迭代对象。
2. 然后使用 fstring 每次循环运行时更新 csv 文件名，为提取的每个 JSON 资源创建一个单独的文件。

d = []
def write_json(data, filename='data.json'):
    with open(filename,'w') as f: 
        json.dump(data, f, indent=4)
 
index = 0
for url in url_lst:
    index +=1
    with request.urlopen(url) as response:
        source = response.read()
        data = json.loads(source)
 
        if len(data) == 0:
            continue

        with open ("covid.json", 'w') as outfile:
            json.dump(data, outfile)

        with open('covid.json') as json_data:
            j = json.load(json_data)
            d.append(j)
            write_json(d)  
        
        print(index)    
        fields = ["Date","County", "New Positives", "All Positives", "New Tests", "All Tests"]
        filename = f"County Stats {index}.csv"
        with open(filename, 'w') as fw:
            cf = csv.writer(fw, lineterminator='\n')
            # write the header
            cf.writerow(fields)
            for counties in data:   
                date = counties['test_date']
                cnty = counties['county']
                new_pos = counties['new_positives']
                cum_pos = counties['cumulative_number_of_positives']

                new_tests = counties['total_number_of_tests']
                cum_tests = counties['cumulative_number_of_tests']
                cf.writerow([date,cnty, new_pos, cum_pos, new_tests, cum_tests])