如何在Python中移除网页内的accordion/fusion-panel相关CSS代码

我用Python结合BeautifulSoup编写了如下代码，提取指定class的article标签下的div内容，处理后可得到所需文本，但偶尔会混入accordion/fusion-panel相关的CSS样式代码，请问如何删除这些内容？

原代码：

toggle=[]
for i in soup.find_all('article', {'class': 'post post-113572 type-post status-publish format-standard has-post-thumbnail hentry category-kurzprogramm'}):
    per_toggle=i.find('div')
    print(per_toggle)
    toggle.append(per_toggle)


    New_toggle=[]
    for each in toggle:
        new_each = str(each).replace('<br/', '')
        pattern = re.compile('<.*?>')
        result = re.sub(pattern, '', new_each)
        print (result)
        New_toggle.append(result)
        df=pd.DataFrame(New_toggle)

混入的CSS代码示例：

#accordion-113572-1 .fusion-panel:hover{ border-color: #d4d4d4 } #accordion-113572-1 .fusion-panel { border-color:#d4d4d4; }.fusion-accordian  #accordion-113572-1 .panel-title a .fa-fusion-box{ color: #ffffff;}.fusion-accordian  #accordion-113572-1 .panel-title a .fa-fusion-box:before{ font-size: 13px; width: 13px;}.fusion-accordian  #accordion-113572-1 .panel-title a{font-size:19px;color:#191919;font-family:""Barlow Condensed"";font-weight:500;}.fusion-accordian  #accordion-113572-1 .toggle-content{font-size:20px;color:#000000;font-family:""Barlow"";font-weight:regular;}.fusion-accordian  #accordion-113572-1 .fa-fusion-box { background-color: #009ee0;border-color: #009ee0;}.fusion-accordian  #accordion-113572-1 .panel-title a:not(.active):hover, #accordion-113572-1 .fusion-toggle-boxed-mode:hover .panel-title a { color: #009ee0;}.fusion-accordian  #accordion-113572-1 .panel-title .active .fa-fusion-box,.fusion-accordian  #accordion-113572-1 .panel-title a:not(.active):hover .fa-fusion-box { background-color: #009ee0!important;border-color: #009ee0!important;}.fusion-accordian  #accordion-113572-1 .panel-f0b74f27ef9a21ca8 .panel-title a{}.fusion-accordian  #accordion-113572-1 .panel-f0b74f27ef9a21ca8 .toggle-content{}

方法1：提前移除含CSS的标签（推荐） #

这些CSS通常存放在<style>标签，或是带有特定标识的容器里，在解析阶段直接删除这些元素，从根源避免冗余内容混入：

import re
import pandas as pd
from bs4 import BeautifulSoup

# 先移除页面所有style标签
for style_tag in soup.find_all('style'):
    style_tag.decompose()

# 移除所有包含fusion-panel或accordion关键词的容器
for panel in soup.find_all(class_=re.compile(r'fusion-panel|accordion')):
    panel.decompose()

# 再提取目标内容
toggle = []
for article in soup.find_all('article', {'class': 'post post-113572 type-post status-publish format-standard has-post-thumbnail hentry category-kurzprogramm'}):
    per_toggle = article.find('div')
    if per_toggle:
        toggle.append(per_toggle)

New_toggle = []
for each in toggle:
    # 用get_text直接提取纯文本，自动去除HTML标签
    pure_text = each.get_text(strip=True, separator=' ')
    New_toggle.append(pure_text)

df = pd.DataFrame(New_toggle, columns=['提取内容'])

方法2：正则匹配清除已混入的CSS文本 #

如果CSS已经混入到提取的文本中，可以用正则表达式匹配并删除符合特征的CSS代码：

# 在原有文本处理步骤后添加以下代码
css_pattern = re.compile(r'#accordion-\d+-\d+ .*?\{\s*.*?\s*\}|.[a-z0-9-]+ .*?\{\s*.*?\s*\}')
clean_result = css_pattern.sub('', result)
New_toggle.append(clean_result)

方法3：优化原有文本提取逻辑 #

原代码将标签转成字符串再处理容易引入冗余，改用BeautifulSoup原生的get_text()方法提取纯文本，能大幅减少垃圾内容：

New_toggle = []
for each in toggle:
    # 提取纯文本，去除首尾空白并统一分隔符
    pure_text = each.get_text(strip=True, separator=' ')
    # 清理可能残留的CSS
    css_pattern = re.compile(r'#accordion-\d+-\d+ .*?\{\s*.*?\s*\}')
    clean_text = css_pattern.sub('', pure_text)
    New_toggle.append(clean_text)

df = pd.DataFrame(New_toggle)

内容的提问来源于stack exchange，提问作者AnGeL