Spider -- MySQL数据库 之 增量爬取

把电影天堂数据存入MySQL数据库 - 增量爬取

# 思路
# 1、MySQL中新建表 urltab,存储所有爬取过的链接的指纹
# 2、在爬取之前,先判断该指纹是否爬取过,如果爬取过,则不再继续爬取

1、建库建表

# 建库建表
create database filmskydb charset utf8;

use filmskydb;

create table request_finger(
finger char(32)
)charset=utf8;
create table filmtab(
name varchar(200),
download varchar(500)
)charset=utf8;

 

2、完整代码

from urllib import request
import re
from useragents import ua_list　　# 自己个人写的模块，提供随机User-Agent
import time
import random
import pymysql
from hashlib import md5
​
​
class FilmSkySpider(object):
    def __init__(self):
        # 一级页面url地址
        self.url = 'https://www.dytt8.net/html/gndy/dyzz/list_23_{}.html'
        self.db = pymysql.connect('localhost', 'root', '123456', 'filmskydb', charset='utf8')
        self.cursor = self.db.cursor()
​
    # 获取html功能函数
    def get_html(self, url):
        headers = {
            'User-Agent': random.choice(ua_list)
        }
        req = request.Request(url=url, headers=headers)
        res = request.urlopen(req)
        # 通过网站查看网页源码,查看网站charset='gb2312'
        # 如果遇到解码错误,识别不了一些字符,则 ignore 忽略掉
        html = res.read().decode('gb2312', 'ignore')
​
        return html
​
    # 正则解析功能函数
    def re_func(self, re_bds, html):
        pattern = re.compile(re_bds, re.S)
        r_list = pattern.findall(html)
​
        return r_list
​
    # 获取数据函数
    def parse_page(self, one_url):
        html = self.get_html(one_url)
        re_bds = r'<table width="100%".*?<td width="5%".*?<a href="(.*?)".*?ulink">.*?</table>'
        # one_page_list: ['/html/xxx','/html/xxx','/html/xxx']
        one_page_list = self.re_func(re_bds, html)
​
        for href in one_page_list:
            two_url = 'https://www.dytt8.net' + href
            # 生成指纹 - md5加密
            s = md5()
            s.update(two_url.encode())
            two_url_md5 = s.hexdigest()
            # 判断链接是否需要抓取
            if self.is_go_on(two_url_md5):
                self.parse_two_page(two_url)
                # 爬取完成此链接后将指纹放到数据库表中
                ins = 'insert into request_finger values(%s)'
                self.cursor.execute(ins, [two_url_md5])
                self.db.commit()
                # uniform: 浮点数,爬取1个电影信息后sleep
                time.sleep(random.uniform(1, 3))
​
​
    def is_go_on(self, two_url_md5):
        # 爬取之前先到数据库中查询比对
        sel = 'select finger from request_finger where finger=%s'
        # 开始抓取之前,先来判断该链接之前是否抓取过
        result = self.cursor.execute(sel, [two_url_md5])
        if not result:
            return True
​
​
    # 解析二级页面数据
    def parse_two_page(self, two_url):
        item = {}
        html = self.get_html(two_url)
        re_bds = r'<div class="title_all"><h1><font color=#07519a>(.*?)</font></h1></div>.*?<td style="WORD-WRAP.*?>.*?>(.*?)</a>'
        # two_page_list: [('名称1','ftp://xxxx.mkv')]
        two_page_list = self.re_func(re_bds, html)
​
        item['name'] = two_page_list[0][0].strip()
        item['download'] = two_page_list[0][1].strip()
​
        ins = 'insert into filmtab values(%s,%s)'
        film_list = [
            item['name'], item['download']
        ]
        self.cursor.execute(ins, film_list)
        self.db.commit()
        print(film_list)
​
​
    def main(self):
        for page in range(1, 201):
            one_url = self.url.format(page)
            self.parse_page(one_url)
            # uniform: 浮点数
            time.sleep(random.uniform(1, 3))
​
​
if __name__ == '__main__':
    spider = FilmSkySpider()
    spider.main()