[Python爬虫] 之二十九：Selenium +phantomjs 利用 pyquery抓取节目信息信息

　　

一、介绍

　　　　本例子用Selenium +phantomjs爬取节目（http://tv.cctv.com/epg/index.shtml?date=2018-03-25）的信息

　　　

　　二、网站信息

　　　　

　　　　

　　

　　三、数据抓取

　　　　针对上面的网站信息，来进行抓取

　　　　1、首先抓取信息列表

　　　　　　抓取代码：Elements = doc('div[class="epglist"]').find('ul')

　　　　2、节目名称，链接，时间

　　　　　　title = subEle('div[class="innerbox"]').find('h3').text().encode('utf8')

　　　　　　link = subEle('div[class="innerbox"]').find('p').find('a').attr('href')

　　　　　　strTime = subEle('div[class="innerbox"]').find('p').text().encode('utf8')

　　四，实现代码

# coding=utf-8
import os
import re
from selenium import webdriver
from datetime import datetime,timedelta
import selenium.webdriver.support.ui as ui
import time
from pyquery import PyQuery as pq
class cctvDriver:

    def __init__(self,startDate,endDate):
        #通过配置文件获取IEDriverServer.exe路径
        self.urls = self.getUrlsFromStartEndDate(startDate,endDate)
        IEDriverServer ='C:\Program Files\Internet Explorer\IEDriverServer.exe'
        self.driver = webdriver.Ie(IEDriverServer)
        self.driver.maximize_window()
        self.fileName = time.strftime('%Y-%m-%d')

    def compareDate(self, startDate, endDate):
        start_Date = time.strptime(startDate, "%Y-%m-%d")
        end_Date = time.strptime(endDate, "%Y-%m-%d")
        totalSeconds = (end_Date - start_Date).total_seconds()
        if totalSeconds >= 0:
            print endDate
            return True
        else:
            print startDate
            return False

    def compareTime(self, startTime, endTime):
        st = int(startTime.replace(':',""))
        et = int(endTime.replace(':',""))
        if st>et:
            return True
        else:
            return False

    def getUrlsFromStartEndDate(self,startDate,endDate):

        urls = []
        start_Date = datetime.strptime(startDate, "%Y-%m-%d")
        end_date = datetime.strptime(endDate, "%Y-%m-%d")
        ts = end_date-start_Date

        days = ts.days + 1
        index = 0
        for d in xrange(0,days):
            date = start_Date + timedelta(days=index)
            urls.append('http://tv.cctv.com/epg/index.shtml?date='+date.strftime("%Y-%m-%d"))
            index += 1
        return urls

    def WriteLog(self, message,date):
        fileName = os.path.join(os.getcwd(), 'cctvInfo/'+date + '.txt')
        with open(fileName, 'a') as f:
            f.write(message)

    def CatchData(self):
        className = "//div[@class='epglist']/ul"
        for url in self.urls:
            date = url.split('=')[1]
            start_Date = datetime.strptime(date, "%Y-%m-%d") + timedelta(days=-1)
            predate = start_Date.strftime("%Y-%m-%d")
            self.driver.get(url)
            time.sleep(5)
            selenium_html = self.driver.execute_script("return document.documentElement.outerHTML")
            doc = pq(selenium_html)
            Elements = doc('div[class="epglist"]').find('ul')
            message = ''
            recount = 0
            for element in Elements.items():
                channel = element.attr('id')
                subElements = element.find("li")

                for subEle in subElements.items():
                    strTime = subEle('div[class="innerbox"]').find('p').text().encode('utf8').strip().replace(
                        '回看', '').replace('直播','')
                    if strTime:
                        title = subEle('div[class="innerbox"]').find('h3').text().encode(
                            'utf8').strip().replace(
                            ',', '，')
                        link = subEle('div[class="innerbox"]').find('p').find('a').attr('href')
                        if self.compareTime(strTime.split('~')[0],strTime.split('~')[1]):
                            starttime = predate + " " + strTime.split('~')[0]
                        else:
                            starttime = date + " " + strTime.split('~')[0]
                        endtime = date + " " + strTime.split('~')[1]

                        mess = '\r\n{0},{1},{2},{3},{4}'.format(channel, title, starttime, endtime, link)
                        # print mess
                        message += mess
                        recount+=1
            if len(message)>10:
                self.WriteLog(message.strip(),date)
            print recount
        self.driver.close()
        self.driver.quit()

# #测试抓取微博数据
obj = cctvDriver('2018-01-01','2018-03-01')
obj.CatchData()