requests和lxml实现爬虫的方法

如下所示:

# requests模块来请求页面
# lxml模块的html构建selector选择器(格式化响应response)
# from lxml import html
# import requests

# response = requests.get(url).content

# selector = html.formatstring(response)

# hrefs = selector.xpath('/html/body//div[@class='feed-item _j_feed_item']/a/@href')

# 以url = 'https://www.mafengwo.cn/gonglve/ziyouxing/2033.html'为例子

# python 2.7
import requests
from lxml import html
import os
# 获取首页中子页的url链接
def get_page_urls(url):
  response = requests.get(url).content
  # 通过lxml的html来构建选择器
  selector = html.fromstring(response)
  urls = []
  for i in selector.xpath("/html/body//div[@class='feed-item _j_feed_item']/a/@href"):
    urls.append(i)
  return urls
# get title from a child's html(div[@class='title'])
def get_page_a_title(url):
  '''url is ziyouxing's a@href'''
  response = requests.get(url).content
  selector = html.fromstring(response)
  # get xpath by chrome's tool --> /html/body//div[@class='title']/text()
  a_title = selector.xpath("/html/body//div[@class='title']/text()")
  return a_title
# 获取页面选择器(通过lxml的html构建)
def get_selector(url):
  response = requests.get(url).content
  selector = html.fromstring(response)
  return selector
# 通过chrome的开发者工具分析html页面结构后发现,我们需要获取的文本内容主要显示在div[@class='l-topic']和div[@class='p-section']中
# 获取所需的文本内容
 def get_page_content(selector):
   # /html/body/div[2]/div[2]/div[1]/div[@class='l-topic']/p/text()
   page_title = selector.xpath("//div[@class='l-topic']/p/text()")
   # /html/body/div[2]/div[2]/div[1]/div[2]/div[15]/div[@class='p-section']/text()
   page_content = selector.xpath("//div[@class='p-section']/text()")
   return page_title,page_content
# 获取页面中的图片url地址
def get_image_urls(selector):
  imagesrcs = selector.xpath("//img[@class='_j_lazyload']/@src")
  return imagesrcs
# 获取图片的标题

def get_image_title(selector, num)
  # num 是从2开始的
  url = "/html/body/div[2]/div[2]/div[1]/div[2]/div["+num+"]/span[@class='img-an']/text()"
  if selector.xpath(url) is not None:
    image_title = selector.xpath(url)
  else:
    image_title = "map"+str(num) # 没有就起一个
  return image_title
# 下载图片

def downloadimages(selector,number):
  '''number是用来计数的'''
  urls = get_image_urls()
  num = 2
  amount = len(urls)
  for url in urls:
    image_title = get_image_title(selector, num)
    filename = "/home/WorkSpace/tour/words/result"+number+"/+"image_title+".jpg"
    if not os.path.exists(filename):
      os.makedirs(filename)
    print('downloading %s image %s' %(number, image_title))
    with open(filename, 'wb') as f:
      f.write(requests.get(url).content)
    num += 1
  print "已经下载了%s张图" %num
# 入口,启动并把获取的数据存入文件中
if __name__ =='__main__':
  url = 'https://www.mafengwo.cn/gonglve/ziyouxing/2033.html'
  urls = get_page_urls(url)
  # turn to get response from html
  number = 1
  for i in urls:
    selector = get_selector(i)
    # download images
    downloadimages(selector,number)
    # get text and write into a file
    page_title, page_content = get_page_content(selector)
    result = page_title+'\n'+page_content+'\n\n'
    path = "/home/WorkSpace/tour/words/result"+num+"/"
    if not os.path.exists(filename):
      os.makedirs(filename)
    filename = path + "num"+".txt"
    with open(filename,'wb') as f:
      f.write(result)
    print result

到此就结束了该爬虫,爬取页面前一定要认真分析html结构,有些页面是由js生成,该页面比较简单,没涉及到js的处理,日后的随笔中会有相关分享

相关推荐