某车之家字体反爬升级

def repalce_source_code(self, html):
    #  转为 编码 比如: \uec8e
    html = html.encode("latin-1", "backslashreplace").decode("utf-8")
    for utf_code, word in self.new_unicode_map.items():
        html = html.replace("\\u%s" % utf_code[3:].lower(), word)
    # 再次将替换后的字符转为正常unicode
    html = html.encode("latin-1", "backslashreplace").decode("utf-8")
    # 转为中文
    html = html.encode("utf-8").decode("unicode_escape")
    return html

def get_subject_content(self):
    # 使用xpath 获取 主贴, 先获取主贴, 只替换主贴内容
    xp_html = etree.HTML(self.page_html)
    subject_text = ''.join(xp_html.xpath('//div[@class="tz-paragraph"]//text()'))
    return self.repalce_source_code(subject_text)