加速解析器:HTML到数据库

我需要插入所有的html标签和属性到数据库

el.driver.get(url_page)
txthtml = el.driver.page_source
soup = BeautifulSoup(txthtml, "html.parser")
body = soup.find('html')
html_parse(body, el, url_page_id, 0, 0, 0,url_page)

def html_parse(html, el, url_page_id, level, i, parent_id, url_page):
    txt = ""
    if len(html.text) > 0:
       txt = html.text.replace("\n","").replace("\t","").replace("\r","")
    ta = tag_list()
    ta.p_id = el.id
    ta.page_id = url_page_id
    ta.level = level
    ta.number = i
    ta.txt = txt
    ta.name = html.name
    ta.parent_id = parent_id
    ta.html = str(html)
    ta.save()
    insert_attr(html, el.id, url_page_id, ta.id, url_page)
    children = list(html.children)
    j = 0
    for child in children:
        if child.name is None:
            continue
        j = j + 1
        html_parse(child, el, url_page_id, level + 1, j, ta.id, url_page)

当我有递归函数html_parse时

• el-驾驶员等级
• level-DOM中的level
• parent_id—父项的id
• url_页面-当前url
• 标签列表-插入当前标签

每个html_parse函数都运行得很快,但是完整的html解析在每个大html页面上运行大约4-5分钟。

如何加快代码的速度?