[挑战]爬byr论坛之1：主题帖列表

试一试自带的库python3.4 import requests from html.parser import HTMLParser p=''' title : %s articleid: %s author : %s lastreply: %s ''' class ByrParser(HTMLParser): def __init__(self): super().__init__() self.re_set() def re_set(self): self.enter_li=False self.is_top=None self.info=[] def handle_starttag(self, tag, attrs): if not self.enter_li and tag=='li': self.enter_li=True elif self.enter_li and self.is_top is None and tag=='a': if len(attrs)>1: self.is_top = True else: self.info.append(attrs[0][1].split('/')[-1]) def handle_endtag(self, tag): if self.enter_li and tag=='li': if self.info: self.output() self.re_set() def handle_data(self, data): if self.enter_li and not self.is_top: self.info.append(data) def output(self): articleid=self.info[0] title=self.info[1] author=self.info[4] lastreply=self.info[-1] print(p%(title,articleid,author,lastreply)) def tops(board): parser = ByrParser() parser.feed(requests.get("http://m.byr.cn/board/%s"%board).text)

nuan神继续这样的挑战啊， # encoding:utf-8 import requests, re from bs4 import BeautifulSoup def board(board_name): base_url = 'http://m.byr.cn/board/%s' href_rc = re.compile('/article/%s' % board_name) html_doc = requests.get(base_url % board_name).text bs = BeautifulSoup(html_doc) link_list = bs.find_all(name='a', attrs={'class': ''}, href=href_rc) for link_a in link_list: a_text = link_a.get_text() a_href = link_a.get('href') a_id = a_href[a_href.rfind('/') + 1:] user_list = link_a.parent.next_sibling.find_all('a')#父亲节点的兄弟节点 author, lastreply = user_list[0].get_text(), user_list[1].get_text() print("title: ", a_text) print("articleid: ", a_id) print("author: ", author) print("lastreply: ", lastreply) print() pass pass if __name__ == '__main__': board_name = 'Python' board(board_name) pass

其实只是相当于换了个库…… import urllib2 from lxml import etree def tops(board): html_text = urllib2.urlopen('http://m.byr.cn/board/%s' % board).read() html_tree = etree.HTML(html_text) lis = html_tree.xpath('/html/body/div[@id="wraper"]/div[@id="m_main"]/ul[@class="list sec"]/li') return [( li[0][0].text, li[0][0].attrib['href'].split('/')[-1], li[1][0].text, li[1][1].text, ) for li in lis if li[0][0].attrib.get('class') != 'top']

啊，在学校没网，占个楼

围观学习

好喜欢python，可惜不会用

学习了。

蛋疼一下…… #!/bin/bash [ -z "$1" ] && { exit 0; } url="http://m.byr.cn/board/$1" curl $url | sed -n 's/<\/li>/\n/gp' | grep -v 'class="top"' | grep '^<li'| \ gawk '{ match($0, /article\/.*\/([0-9]+)/, group) id = group[1] match($0, /[0-9]+">(.*)<\/a>\([0-9]+\)/, group) title = group[1] match($0, /user\/query\/([^"]+).+user\/query\/([^"]+)/, group) author = group[1] last = group[2] printf "title: %s\narticleid: %s\nauthor: %s\nlastreply: %s\n\n", title, id, author, last }' 【 在 nuanyangyang 的大作中提到: 】 : 要求：给定版名，爬北邮人论坛的该版，并返回它的第一页的非顶置贴的标题、帖子id、原作者和最后回复者。 : 限制： : 1. 语言不限，但考虑一下你能击败Python的优雅吗？ : ...................

进楼学习