一个简单的python爬虫,爬取知乎
- 主要实现 爬取一个收藏夹 里 所有问题答案下的 图片
- 文字信息暂未收录，可自行实现，比图片更简单
- 具体代码里有详细注释，请自行阅读
项目源码：
1 # -*- coding:utf-8 -*- 2 3 from spider import SpiderHTML 4 from multiprocessing import Pool 5 import sys,urllib,http,os,random,re,time 6 __author__ = 'waiting' 7 ''' 8 使用了第三方的类库 BeautifulSoup4，请自行安装 9 需要目录下的spider.py文件 10 运行环境：python3.4,windows7 11 ''' 12 13 #收藏夹的地址 14 url = 'https://www.zhihu.com/collection/30822111' #page参数改为代码添加 15 16 #本地存放的路径,不存在会自动创建 17 store_path = 'E:\zhihu收藏夹\会员才知道的世界' 18 19 class zhihuCollectionSpider(SpiderHTML): 20 def __init__(self,pageStart, pageEnd, url): 21 self._url = url 22 self._pageStart = int(pageStart) 23 self._pageEnd = int(pageEnd)+1 24 self.downLimit = 0 #低于此赞同的答案不收录 25 26 def start(self): 27 for page in range(self._pageStart,self._pageEnd): #收藏夹的页数 28 url = self._url + '?page='+str(page) 29 content = self.getUrl(url) 30 questionList = content.find_all('div',class_='zm-item') 31 for question in questionList: #收藏夹的每个问题 32 Qtitle = question.find('h2',class_='zm-item-title') 33 if Qtitle is None: #被和谐了 34 continue 35 36 questionStr = Qtitle.a.string 37 Qurl = 'https://www.zhihu.com'+Qtitle.a['href'] #问题题目 38 Qtitle = re.sub(r'[\/:*?"<>]','#',Qtitle.a.string) #windows文件/目录名不支持的特殊符号 39 try: 40 print('-----正在获取问题:'+Qtitle+'-----') #获取到问题的链接和标题，进入抓取 41 except UnicodeEncodeError: 42 print(r'---问题含有特殊字符无法显示---') 43 try: 44 Qcontent = self.getUrl(Qurl) 45 except: 46 print('!!!!获取出错!!!!!') 47 pass 48 answerList = Qcontent.find_all('div',class_='zm-item-answer zm-item-expanded') 49 self._processAnswer(answerList,Qtitle) #处理问题的答案 50 time.sleep(5) 51 52 53 def _processAnswer(self,answerList,Qtitle): 54 j = 0 55 for answer in answerList: 56 j = j + 1 57 58 upvoted = int(answer.find('span',class_='count').string.replace('K','000')) #获得此答案赞同数 59 if upvoted < self.downLimit: 60 continue 61 authorInfo = answer.find('div',class_='zm-item-answer-author-info') #获取作者信息 62 author = {'introduction':'','link':''} 63 try: 64 author['name'] = authorInfo.find('a',class_='author-link').string #获得作者的名字 65 author['introduction'] = str(authorInfo.find('span',class_='bio')['title']) #获得作者的简介 66 author['link'] = authorInfo.find('a',class_='author-link')['href'] 67 except AttributeError: 68 author['name'] = '匿名用户'+str(j) 69 except TypeError: #简介为空的情况 70 pass #匿名用户没有链接 71 72 file_name = os.path.join(store_path,Qtitle,'info',author['name']+'_info.txt') 73 if os.path.exists(file_name): #已经抓取过 74 continue 75 76 self.saveText(file_name,'{introduction} {link}'.format(**author)) #保存作者的信息 77 print('正在获取用户`{name}`的答案'.format(**author)) 78 answerContent = answer.find('div',class_='zm-editable-content clearfix') 79 if answerContent is None: #被举报的用户没有答案内容 80 continue 81 82 imgs = answerContent.find_all('img') 83 if len(imgs) == 0: #答案没有上图 84 pass 85 else: 86 self._getImgFromAnswer(imgs,Qtitle,**author) 87 88 #收录图片 89 def _getImgFromAnswer(self,imgs,Qtitle,**author): 90 i = 0 91 for img in imgs: 92 if 'inline-image' in img['class']: #不抓取知乎的小图 93 continue 94 i = i + 1 95 imgUrl = img['src'] 96 extension = os.path.splitext(imgUrl)[1] 97 path_name = os.path.join(store_path,Qtitle,author['name']+'_'+str(i)+extension) 98 try: 99 self.saveImg(imgUrl,path_name) #捕获各种图片异常，流程不中断 100 except: 101 pass 102 103 #收录文字 104 def _getTextFromAnswer(self): 105 pass 106 107 #命令行下运行，例：zhihu.py 1 5 获取1到5页的数据 108 if __name__ == '__main__': 109 page, limit, paramsNum= 1, 0, len(sys.argv) 110 if paramsNum>=3: 111 page, pageEnd = sys.argv[1], sys.argv[2] 112 elif paramsNum == 2: 113 page = sys.argv[1] 114 pageEnd = page 115 else: 116 page,pageEnd = 1,1 117 118 spider = zhihuCollectionSpider(page,pageEnd,url) 119 spider.start()
很多初学者，对Python的概念都是模糊不清的，C语言、Python能做什么，学的时候，该按照什么线路去学习，学完往哪方面发展，想深入了解，详情可以点击有道云笔记链接了解：http://note.youdao.com/noteshare?id=e4fa02e7b56d7909a27674cdb3da08aa