電影天堂爬蟲

# 電影天堂爬蟲

**今日目標**

爬取電影天堂前30頁最新電影以及下載地址
```python
# 1. 打印程序執行時間
# 2. 數據爬下來后做處理(字符串),定義成字典
# 3. 一條龍: 獲取 -> 調用解析 -> 數據處理

from urllib import request
import time
import re
import pymysql
import random

class MaoyanSpider(object):
    def __init__(self):
        self.url = 'https://maoyan.com/board/4?offset={}'
        self.headers={'User-Agent':'Mozilla/5.0'}
        # 用於記錄頁數
        self.page = 1
        # 創建數據庫連接對象和游標對象
        self.db = pymysql.connect(
            'localhost','root','123456','maoyandb',charset='utf8'
        )
        self.cursor = self.db.cursor()


    # 獲取網頁
    def get_page(self,url):
        req = request.Request(
            url=url,
            headers=self.headers
        )
        res = request.urlopen(req)
        html = res.read().decode('utf-8')
        # 直接調用解析函數
        self.parse_page(html)

    # 解析網頁
    def parse_page(self,html):
        pattren = re.compile('<div class="movie-item-info">.*?title="(.*?)".*?class="star">(.*?)</p>.*?releasetime">(.*?)</p>',re.S)
        r_list = pattren.findall(html)
        print(r_list)
        self.write_page(r_list)

    # 存入mysql數據庫(executemany([ [],[],[] ]))
    def write_page(self,r_list):
        # 定義空列表
        film_list = []
        ins = 'insert into filmtab values(%s,%s,%s)'
        # 處理數據,放到大列表film_list中
        for rt in r_list:
            one_film = [
                rt[0],rt[1].strip(),rt[2].strip()[5:15]
            ]
            # 添加到大列表中
            film_list.append(one_film)
        # 一次數據庫IO把1頁數據存入
        self.cursor.executemany(ins,film_list)
        # 提交到數據庫執行
        self.db.commit()


    def main(self):
        for offset in range(0,31,10):
            url = self.url.format(offset)
            self.get_page(url)
            time.sleep(random.randint(1,3))
            print('第%d頁爬取完成' % self.page)
            self.page += 1

        # 斷開數據庫(所有頁爬完之后)
        self.cursor.close()
        self.db.close()


if __name__ == '__main__':
    start = time.time()
    spider = MaoyanSpider()
    spider.main()
    end = time.time()
    print('執行時間: %.2f' % (end-start))
```