爬虫库
使用简单的requests库,这是一个阻塞的库,速度比较慢。
解析使用XPATH表达式
总体采用类的形式
多线程
使用concurrent.future并发模块,建立线程池,把future对象扔进去执行即可实现并发爬取效果
数据存储
使用Python ORM sqlalchemy保存到数据库,也可以使用自带的csv模块存在CSV中。
API接口
因为API接口存在数据保护情况,一个电影的每一个分类只能抓取前25页,全部评论、好评、中评、差评所有分类能爬100页,每页有20个数据,即最多为两千条数据。
因为时效性原因,不保证代码能爬到数据,只是给大家一个参考思路,上代码:
1 from datetime import datetime 2 import random 3 import csv 4 from concurrent.futures import ThreadPoolExecutor, as_completed 5 6 from lxml import etree 7 import pymysql 8 import requests 9 10 from models import create_session, Comments 11 12 #随机UA 13 USERAGENT = [ 14 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.163 Safari/535.1', 15 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36', 16 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0) Gecko/20100101 Firefox/6.0', 17 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/534.50 (KHTML, like Gecko) Version/5.1 Safari/534.50', 18 'Opera/9.80 (Windows NT 6.1; U; zh-cn) Presto/2.9.168 Version/11.50', 19 'Mozilla/5.0 (Windows; U; Windows NT 6.1; ) AppleWebKit/534.12 (KHTML, like Gecko) Maxthon/3.0 Safari/534.12' 20 ] 21 22 23 class CommentFetcher: 24 headers = {'User-Agent': ''} 25 cookie = '' 26 cookies = {'cookie': cookie} 27 # cookie为登录后的cookie,需要自行复制 28 base_node = '//div[@class="comment-item"]' 29 30 31 def __init__(self, movie_id, start, type=''): 32 ''' 33 :type: 全部评论:'', 好评:h 中评:m 差评:l 34 :movie_id: 影片的ID号 35 :start: 开始的记录数,0-480 36 ''' 37 self.movie_id = movie_id 38 self.start = start 39 self.type = type 40 self.url = 'https://movie.douban.com/subject/{id}/comments?start={start}&limit=20&sort=new_score\&status=P&percent_type={type}&comments_only=1'.format( 41 id=str(self.movie_id), 42 start=str(self.start), 43 type=self.type 44 ) 45 #创建数据库连接 46 self.session = create_session() 47 48 #随机useragent 49 def _random_UA(self): 50 self.headers['User-Agent'] = random.choice(USERAGENT) 51 52 53 #获取api接口,使用get方法,返回的数据为json数据,需要提取里面的HTML 54 def _get(self): 55 self._random_UA() 56 res = '' 57 try: 58 res = requests.get(self.url, cookies=self.cookies, headers=self.headers) 59 res = res.json()['html'] 60 except Exception as e: 61 print('IP被封,请使用代理IP') 62 print('正在获取{} 开始的记录'.format(self.start)) 63 return res 64 65 def _parse(self): 66 res = self._get() 67 dom = etree.HTML(res) 68 69 #id号 70 self.id = dom.xpath(self.base_node + '/@data-cid') 71 #用户名 72 self.username = dom.xpath(self.base_node + '/div[@class="avatar"]/a/@title') 73 #用户连接 74 self.user_center = dom.xpath(self.base_node + '/div[@class="avatar"]/a/@href') 75 #点赞数 76 self.vote = dom.xpath(self.base_node + '//span[@class="votes"]/text()') 77 #星级 78 self.star = dom.xpath(self.base_node + '//span[contains(@class,"rating")]/@title') 79 #发表时间 80 self.time = dom.xpath(self.base_node + '//span[@class="comment-time "]/@title') 81 #评论内容 所有span标签class名为short的节点文本 82 self.content = dom.xpath(self.base_node + '//span[@class="short"]/text()') 83 84 #保存到数据库 85 def save_to_database(self): 86 self._parse() 87 for i in range(len(self.id)): 88 try: 89 comment = Comments( 90 id=int(self.id[i]), 91 username=self.username[i], 92 user_center=self.user_center[i], 93 vote=int(self.vote[i]), 94 star=self.star[i], 95 time=datetime.strptime(self.time[i], '%Y-%m-%d %H:%M:%S'), 96 content=self.content[i] 97 ) 98 99 self.session.add(comment) 100 self.session.commit() 101 return 'finish' 102 103 104 except pymysql.err.IntegrityError as e: 105 print('数据重复,不做任何处理') 106 107 except Exception as e: 108 #数据添加错误,回滚 109 self.session.rollback() 110 111 finally: 112 #关闭数据库连接 113 self.session.close() 114 115 #保存到csv 116 def save_to_csv(self): 117 self._parse() 118 f = open('comment.csv', 'w', encoding='utf-8') 119 csv_in = csv.writer(f, dialect='excel') 120 for i in range(len(self.id)): 121 csv_in.writerow([ 122 int(self.id[i]), 123 self.username[i], 124 self.user_center[i], 125 int(self.vote[i]), 126 self.time[i], 127 self.content[i] 128 ]) 129 f.close() 130 131 132 if __name__ == '__main__': 133 with ThreadPoolExecutor(max_workers=4) as executor: 134 futures = [] 135 for i in ['', 'h', 'm', 'l']: 136 for j in range(25): 137 fetcher = CommentFetcher(movie_id=26266893, start=j * 20, type=i) 138 futures.append(executor.submit(fetcher.save_to_csv)) 139 140 for f in as_completed(futures): 141 try: 142 res = f.done() 143 if res: 144 ret_data = f.result() 145 if ret_data == 'finish': 146 print('{} 成功保存数据'.format(str(f))) 147 except Exception as e: 148 f.cancel()