<爬虫实战>糗事百科

1.糗事百科段子.py

# 目标：爬取糗事百科段子信息（文字）
# 信息包括：作者头像，作者名字，作者等级，段子内容，好笑数目，评论数目
# 解析用学过的几种方法都实验一下①正则表达式.②BeautifulSoup③xpath
import requests
import re  # 正则表达式
import json
from bs4 import BeautifulSoup  # BS
from lxml import etree  # xpath


def get_one_page(url):
	response = requests.get(url)
	if response.status_code == 200:
		return response.text
	return None


def zhengze_parse(html):
	pattern = re.compile(
		'<img src="//(.*?)".*?alt="(.*?)".*?<a.*?<div class=".*?">(.*?)</div>'
		+ '.*?<div class=.*?<span>(.*?)</span>.*?<span class=".*?".*?<i.*?>(.*?)<'
		+ '.*?<i.*?>(.*?)<',
		re.S)
	items = re.findall(pattern, html)
	for item in items:
		content = item[3].replace('<br/>', '').strip()
		content = content.replace('x01', '')
		if item[5] == '京公网安备11010502031601号':
			break
		yield {
			'image': "http://" + item[0],
			'name': item[1],
			'grade': item[2],
			'content': content,
			'fun_Num': item[4],
			'com_Num': item[5]
		}


def soup_parse(html):
	soup = BeautifulSoup(html, 'lxml')
	for data in soup.find_all('div', class_='article'):
		image = "http:" + data.img['src']
		name = data.img['alt']
		# 匿名用户没有等级
		if name=="匿名用户":
			grade = "匿名用户"
		else:
			grade = data.find('div', class_='articleGender').text
		content = data.find('div', class_='content').span.text.strip()
		fun_Num = data.find('i', class_='number').text
		com_Num = data.find('a', class_='qiushi_comments').i.text

		yield {
			'image': image,
			'name': name,
			'grade': grade,
			'content': content,
			'fun_Num': fun_Num,
			'com_Num': com_Num,
		}


def xpath_parse(html):
	html = etree.HTML(html)
	for data in html.xpath('//div[@class="col1"]/div'):
		image = "http:"+ str(data.xpath('.//img/@src')[0])
		name = data.xpath('.//img/@alt')[0]
		if name == '匿名用户':
			grade = '匿名用户'
		else:
			grade = data.xpath('./div[1]/div/text()')[0]
		content = data.xpath('./a/div/span/text()')[0:]
		content = str(content).strip().replace('\n','')
		fun_Num = data.xpath('./div[2]/span[1]/i/text()')[0]
		com_Num = data.xpath('.//div[2]/span[2]/a/i/text()')[0]
		# print(image, name, grade, content, fun_Num, com_Num)
		yield {
			'image': image,
			'name': name,
			'grade': grade,
			'content': content,
			'fun_Num': fun_Num,
			'com_Num': com_Num,
		}


def write_to_file(content, flag):
	with open('糗百段子(' + str(flag) + ').txt', 'a', encoding='utf-8')as f:
		f.write(json.dumps(content, ensure_ascii=False) + '
')


def search(Num):
	url = 'https://www.qiushibaike.com/text/page/' + str(Num) + '/'
	html = get_one_page(url)
	# 正则匹配不到匿名用户的等级，不会匹配匿名用户的段子，所以少一些数据
	# 稍微加个判断逻辑就行了，懒得弄了
	for item in zhengze_parse(html):
		write_to_file(item, '正则表达式')

	for item in soup_parse(html):
		write_to_file(item, 'BS4')

	for item in xpath_parse(html):
		write_to_file(item, 'xpath')
	page = str(Num)
	print("正在爬取第" + page + '页')


def main():
	# 提供页码
	for Num in range(1, 14):
		search(Num)
	print("爬取完成")


if __name__ == '__main__':
	# 入口
	main()

　　

2.打包

pyinstaller -F 糗事百科段子.py

　　

3.运行效果

网页上匿名用户段子的显示情况