from bs4 import BeautifulSoup

import re

import urlparse

class HtmlParser(object):   

    def _get_new_urls(self, page_url, soup):

        new_urls=set()

        links=soup.find_all('a',href=re.compile(r"/view/\d+\.html"))

        for link in links:

            new_url=link['href']

            new_full_url=urlparse.urljoin(page_url,new_url)

            new_urls.add(new_full_url)

        return new_urls

    # class="lemmaWgt-lemmaTitle-title"

    def _get_new_data(self, page_url, soup):

        res_data={}

        res_data['url']=page_url

        title_node=soup.find('dd',class_="lemmaWgt-lemmaTitle-title").find("h1")

        res_data['title']=title_node.get_text()

        summary_node=soup.find('div',class_='lemma-summary')

        res_data['summary']=summary_node.get_text()

        return res_data

    def parse(self,page_url,html_cont):

        if page_url is None or html_cont is None:

            return

        soup= BeautifulSoup(html_cont,'html.parser',from_encoding= 'utf-8')

        new_urls=self._get_new_urls(page_url,soup)

        new_data=self._get_new_data(page_url,soup)

        return new_urls,new_data 

百度的url 不一样，python页是这个:https://baike.baidu.com/item/Python/407313

里面的链接也有变化