我正在尝试修改整个值列,但是列表不断超出范围,我一直遇到问题。这是我的整个代码:

# Libraries
import json, requests
import pandas as pd
from pandas.io.json import json_normalize

# Set URL
url = 'https://api-v2.themuse.com/jobs'

# For loop to extract data
for i in range(100):
    data = json.loads(requests.get(
        url=url,
        params={'page': i}
    ).text)['results']

# JSON to PANDAS
data_norm = pd.read_json(json.dumps(data))

# Modify two columns' values
data_norm.locations = data_norm.locations.apply(lambda x: [{x[0]['name']}])
data_norm.publication_date = pd.to_datetime(data_norm.publication_date)


这里的问题是当我使用该功能时

data_norm.locations = data_norm.locations.apply(lambda x: [{x[0]['name']}])


我收到以下错误:

IndexError: list index out of range


理想情况下,我想从此更改location列:

0               [{'name': 'Seattle, WA'}]
1    [{'name': 'San Francisco Bay Area'}]
2             [{'name': 'Palo Alto, CA'}]
3                  [{'name': 'Reno, NV'}]
4                                      []
Name: locations, dtype: object


到这个:

0                     Seattle, WA
1          San Francisco Bay Area
2                   Palo Alto, CA
3                        Reno, NV
4
Name: locations, dtype: object

最佳答案

data_norm.locations = data_norm.locations.apply(lambda x:
                                                [{x[0].get('name', '')}]
                                                if len(x) > 0 else []
                                                )


请注意,这假设如果该条目包含至少一个元素,则第一个元素是字典。代码的问题在于,您试图访问空数组的第一个(索引0)元素。

编辑

要根据您的评论删除[{}]:

data_norm.locations = data_norm.locations.apply(lambda x:
                                                x[0].get('name', '')
                                                if len(x) > 0 else ''
                                                )

关于python - 应用功能导致列表索引超出范围,我们在Stack Overflow上找到一个类似的问题:https://stackoverflow.com/questions/41946703/

10-10 07:17