| 数据学习者官方网站(Datalearner)

```json
import re
import json
import pickle
from bs4 import BeautifulSoup

def extract_comments_by_html(html_data):

# 使用 BeautifulSoup 解析 HTML 数据
    soup = BeautifulSoup(html_data, 'html.parser')
    comments_ul = soup.find('ul', {'id': 'comments'})  # 定位评论列表
    comments = []  # 存储解析后的评论数据

# 遍历每个评论项
    for comment_item in comments_ul.find_all('li', class_='comment-item'):
        comment = {}
        comment["comment_id"] = comment_item["id"]  # 评论ID
        comment["author"] = comment_item.find(
            'a', class_='').text.strip()  # 发布者名字
        comment["author_profile"] = comment_item.find('a', class_='')[
            "href"]  # 发布者主页链接
        comment["pubtime"] = comment_item.find('span', class_='pubtime').text.strip()  # 发布时间
        try:
            content = comment_item.find('div', class_='reply-content').find('p').text.strip()  # 评论内容
        except AttributeError:
            content = ""
        comment["content"] = content

# 点赞数
        like_button = comment_item.find('a', class_='comment-vote')
        likes_text = like_button.text.strip().replace("赞", "").strip().replace(
            "(", "").replace(")", "") if like_button else "0"

likes = int(likes_text) if likes_text else 0
        comment["likes"] = likes

# 检测是否是回复
        reply_quote = comment_item.find('div', class_='reply-quote')
        if reply_quote:
            comment["is_reply"] = True
            reply_to_author = reply_quote.find('a').text.strip()  # 被回复的作者
            reply_to_content = reply_quote.find(
                'span', class_='all ref-content').text.strip()  # 被回复的评论内容
            comment["reply_to"] = {
                "author": reply_to_author,
                "content": reply_to_content
            }
        else:
            comment["is_reply"] = False
            comment["reply_to"] = None

comments.append(comment)

# 打印结果
    result = {"comments": comments}
    return result

def extract_comments(js_code):
    """从 JavaScript 代码中提取 comments 字段。"""

js_code = js_code.replace("var _COMMENTS_CONFIG = ", "").strip()
    js_code = js_code.rstrip(';')
    js_code = js_code.rstrip('[]')

pattern = r"('comments':.*?'total':\s*\d+)"
    match = re.search(pattern, js_code, re.DOTALL)

if match:
        comments_section = "{"+match.group(1).replace(
            "'comments':", "\"comments\":").replace("'total':", "\"total\":")+"}"
        print(comments_section)
        # 你可以进一步将其转换为字典对象
        comments_data = json.loads(comments_section)

else:
        print("没有找到匹配的内容进一步的内容")

return comments_data

comment_count = 0
for filename in os.listdir(comments_dir):
    if filename != "search_result_page.html" and filename.endswith("html"):

comments_file = os.path.join(comments_dir, filename)
        comments_json_file = comments_file + ".json"
        html_content = open(comments_file, encoding="utf-8").read()

print(comments_file)

match = re.search(
            r'var _COMMENTS_CONFIG = ({.*?});', html_content, re.DOTALL)

if match:
            comment_html = match.group(1)
            # print(comment_html)
            comments_data = extract_comments(comment_html)
            
        else:
            print("没有找到匹配的内容")
            comments_data = extract_comments_by_html(html_content)
        
        comment_count += len(comments_data["comments"])
        comments_data = json.dumps(comments_data)  # 将字典转换为 JSON 格式的字符串
        print(comments_data)
        
        print(comments_data)

print(comment_count)
```

欢迎大家关注DataLearner官方微信，接受最新的AI技术推送

相关博客

最热博客