AttributeError: 'NoneType' 对象没有属性 'assign' |使用 Pandas 的数据框 Python

Question

我正在尝试使用网络抓取来访问数据，并使用 pandas 将其制成数据框。使用以下代码，我已经能够获取数据框。我想将所有数据帧与 append 组合成一个大数据帧。

import requests
import re
import pandas as pd
from urllib.parse import unquote
from json import loads
from bs4 import BeautifulSoup

# Download URL
url = "https://riwayat-file-covid-19-dki-jakarta-jakartagis.hub.arcgis.com/"
req = requests.get(url)

# Get encoded JSON from HTML source
encoded_data = re.search("window\.__SITE=\"(.*)\"", req.text).groups()[0]

# Decode and load as dictionary
json_data = loads(unquote(encoded_data))

# Get the HTML source code for the links
html_src = json_data["site"]["data"]["values"]["layout"]["sections"][1]["rows"][0]["cards"][0]["component"]["settings"]["markdown"]

# Parse it using BeautifulSoup
soup = BeautifulSoup(html_src, 'html.parser')

# Get links
links = soup.find_all('a')

# For each link...
link_list = []
id_list = []
date_list = []
dataframe_csv = []

for link in links:
    if "2021" in link.text:
       link_list.append(link.text+" - "+link.attrs['href'])

link_list.remove("31 Januari 2021 Pukul 10.00 - https://drive.google.com/file/d/1vd1tToQbx3A420KMDA63aKviLjgGPJMd/view?usp=sharing")

for i in link_list:
    id_list.append(i.split("/")[5])
    date_list.append(i.split("/")[0][:-21])
    
for ID in id_list:
    dataframe_csv.append("https://docs.google.com/spreadsheets/d/"+ID+"/export?format=csv")

我想使用循环组合我拥有的所有数据帧。对于每个循环，我想删除索引 0 行并添加一个新列，即日期。代码如下：

date_num = 0
df_total = pd.DataFrame()

for i in dataframe_csv:
    df = pd.read_csv(i)
    df = df.drop(index=df.index[0], axis=0, inplace=True)
    df = df.assign(Date = date_list[date_num])
    
    date_num += 1
    
    df_total.append(df,ignore_index=True)

问题是，我收到这样的错误：

AttributeError                            Traceback (most recent call last)
<ipython-input-11-ef67f0a87a8e> in <module>
      5     df = pd.read_csv(i)
      6     df = df.drop(index=df.index[0], axis=0, inplace=True)
----> 7     df = df.assign(Date = date_list[date_num])
      8 
      9     date_num += 1

AttributeError: 'NoneType' object has no attribute 'assign'

Answer 1

inplace=True 直接修改数据框，所以要么删除它：

date_num = 0
df_total = pd.DataFrame()

for i in dataframe_csv:
    df = pd.read_csv(i)
    df = df.drop(index=df.index[0], axis=0)
    df = df.assign(Date = date_list[date_num])
    
    date_num += 1
    
    df_total.append(df,ignore_index=True)

或者不分配回去：

date_num = 0
df_total = pd.DataFrame()

for i in dataframe_csv:
    df = pd.read_csv(i)
    df.drop(index=df.index[0], axis=0, inplace=True)
    df = df.assign(Date = date_list[date_num])
    
    date_num += 1
    
    df_total.append(df,ignore_index=True)

如drop 的documentation 中所述：

就地：布尔值，默认为 False
如果为 False，则返回一个副本。否则，就地操作并返回None。