如何将 json 内容写入 csv 文件？

Question

我正在使用 api 调用来请求数据，它返回 JSON。我需要将此响应保存到 CSV 文件中。

我能够提取 JSON 响应，但在写入 CSV 时遇到问题。

# import statements
import json         # package to work with json
import requests     # package to make http requests
import csv

# set global variables
group_id = 0000
api_token = 'yyyy'
api_url = 'https://api.samsara.com/v1'

endpoint_url = api_url + '/fleet/drivers'

# query params for the request
my_params = {"access_token": api_token}

# body data to send with the request
my_data = {"groupId": group_id}

# send POST request to endpoint
resp = requests.post(url = endpoint_url, params = my_params, json = my_data)

# pull out the json
array = resp.json()
text = json.dumps(array)

csvFile = open('/tmp/output.csv','w')
csvwriter = csv.writer(csvFile)
count = 0
for line in text:
    if count == 0:
        header = line.keys()
        csvwriter.writerow(header)
        count += 1
    csvwriter.writerow(line.values())
csvFile.close()

当我运行上面的代码时，我得到以下错误：

Traceback（最近一次调用最后一次）：文件“get_driver_test4.py”，行 34，在 header = line.keys() AttributeError: 'str' object has no attribute 'keys'

Answer 1

将您的问题分解为不同的功能，如下所示：

import json
import requests
import csv

def get_data():
    group_id = 0000
    api_token = 'yyyy'
    api_url = 'https://api.samsara.com/v1'
    endpoint_url = api_url + '/fleet/drivers'

    my_params = {"access_token": api_token}
    my_data = {"groupId": group_id}
    resp = requests.post(url = endpoint_url, params = my_params, json = my_data)
    array = resp.json()
    text = json.dumps(array)

    return text


def write_file(filename, text):
    dataset = json.loads(text)
    drivers = dataset['drivers']

    csvFile = open(filename,'w')
    csvwriter = csv.writer(csvFile)

    # write header
    if len(drivers) > 0:
        keys = drivers[0].keys()
        csvwriter.writerow(keys)

    # write data
    for line in drivers:
       csvwriter.writerow(line.values())

    csvFile.close()

text = get_data()
write_file('output.csv', text)

这样，您可以将 write_file 函数与 get_data 函数分开测试。

需要注意的是0000 的GroupID。如果您输入了这样的组 id，您可能会收到 JSON 解析错误。如果您真的想要 4 个字符的数字，请找到一种引用它们的方法，例如 "0000"。如果您只对数字感兴趣，请将0000 替换为0。

为了测试你的理论，在 python 文件的底部使用如下内容：

text = '{"drivers": [{"id": 134763, "groupId": "0000", "vehicleId": 212014918234742, "currentVehicleId": 212014918234742, "username": "ablah", "name": "Art Blah"}, {"id": 134764, "groupId": "0000", "vehicleId": 212014918234709, "currentVehicleId": 212014918234709, "username": "btodd", "name": "Barry Todd"}]}'
write_file('output.csv', text)

结果文件

id,groupId,vehicleId,currentVehicleId,username,name
134763,0000,212014918234742,212014918234742,ablah,Art Blah
134764,0000,212014918234709,212014918234709,btodd,Barry Todd

Answer 2

好吧，这个怎么样：

# import statements
import json         # package to work with json
import requests     # package to make http requests
import csv
resp = """
{
    "drivers":[
        {
            "id":134763,
            "groupId":1234,
            "vehicleId":212014918234742,
            "currentVehicleId":212014918234742,
            "username":"ablah",
            "name":"Art Blah"
        },
        {
            "id":134764,
            "groupId":1234,
            "vehicleId":212014918234709,
            "currentVehicleId":212014918234709,
            "username":"btodd",
            "name":"Barry Todd"
        }
    ]
}
"""
text = json.loads(resp)
csvFile = open('output.csv','w')
csvwriter = csv.writer(csvFile, delimiter=',')
line = text["drivers"]
csvwriter.writerow(["id","groupId","vehicleId","currentVehicleId","username","name"])
for l in line:
    csvwriter.writerow([l["id"],l["groupId"],l["vehicleId"],l["currentVehicleId"],l["username"],l["name"]])
    
    
csvFile.close()

您可以根据需要调整代码。

输出.csv

id,groupId,vehicleId,currentVehicleId,username,name

134763,1234,212014918234742,212014918234742,ablah,Art Blah

134764,1234,212014918234709,212014918234709,btodd,Barry Todd

顺便说一句……

在您提供的示例 json 中，groupId 0000 在 python 中引发错误，因为它不是有效数字。考虑将 0000 设为字符串“0000”

更多参考资料...

https://docs.python.org/3/library/csv.html