一、介绍
本例子用Selenium +phantomjs爬取活动树(http://www.huodongshu.com/html/find_search.html?search_keyword=数字)的资讯信息,输入给定关键字抓取资讯信息。
给定关键字:数字;融合;电视
抓取信息内如下:
1、资讯标题
2、资讯链接
3、资讯时间
4、资讯来源
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[id="eventList"]').find('div[class="list"]')
2、抓取标题
抓取代码:title = element('h3').find('a').text().encode('utf8').strip()
3、抓取链接
抓取代码:url = 'http://www.huodongshu.com' + element('h2').find('a').attr('href')
4、抓取日期
抓取代码:element('div[class="info time"]').text().encode('utf8').strip()
5、抓取主办方
抓取代码:orgnizer = element('div[class="info title"]').find('a').text().encode('utf8').strip()
6、抓取地点
抓取代码:area = element('div[class="info site"]').text().encode('utf8').strip()
四、完整代码
# coding=utf-8 import os import re from selenium import webdriver import selenium.webdriver.support.ui as ui import time from datetime import datetime from selenium.webdriver.common.action_chains import ActionChains import IniFile # from threading import Thread from pyquery import PyQuery as pq import LogFile import mongoDB import urllib class huodongshuSpider(object): def __init__(self): logfile = os.path.join(os.path.dirname(os.getcwd()), time.strftime('%Y-%m-%d') + '.txt') self.log = LogFile.LogFile(logfile) configfile = os.path.join(os.path.dirname(os.getcwd()), 'setting.conf') cf = IniFile.ConfigFile(configfile) self.keyword_list = cf.GetValue("section", "meeting_keywords").split(';') websearch_url = cf.GetValue("huodongshu", "websearchurl") self.db = mongoDB.mongoDbBase() self.start_urls = [] for keyword in self.keyword_list: url = websearch_url + urllib.quote(keyword) self.start_urls.append(url) self.driver = webdriver.PhantomJS() self.wait = ui.WebDriverWait(self.driver, 2) self.driver.maximize_window() def scroll_foot(self): ''' 滚动条拉到底部 :return: ''' js = "" # 如何利用chrome驱动或phantomjs抓取 if self.driver.name == "chrome" or self.driver.name == 'phantomjs': js = "var q=document.body.scrollTop=10000" # 如何利用IE驱动抓取 elif self.driver.name == 'internet explorer': js = "var q=document.documentElement.scrollTop=10000" return self.driver.execute_script(js) def compareDate(self, dateLeft, dateRight): ''' 比较俩个日期的大小 :param dateLeft: 日期 格式2017-03-04 :param dateRight:日期 格式2017-03-04 :return: 1:左大于右,0:相等,-1:左小于右 ''' dls = dateLeft.split('-') drs = dateRight.split('-') if len(dls) > len(drs): return 1 if int(dls[0]) == int(drs[0]) and int(dls[1]) == int(drs[1]) and int(dls[2]) == int(drs[2]): return 0 if int(dls[0]) > int(drs[0]): return 1 elif int(dls[0]) == int(drs[0]) and int(dls[1]) > int(drs[1]): return 1 elif int(dls[0]) == int(drs[0]) and int(dls[1]) == int(drs[1]) and int(dls[2]) > int(drs[2]): return 1 return -1 def date_isValid(self, strDateText): ''' 判断日期时间字符串是否合法:如果给定时间大于当前时间是合法,或者说当前时间给定的范围内 :param strDateText: 三种格式 '2017.04.27 ~ 04.28'; '2017.04.20 08:30 ~ 12:30' ; '2015.12.29 ~ 2016.01.03' :return: True:合法;False:不合法 ''' datePattern = re.compile(r'd{4}-d{2}-d{2}') date = strDateText.replace('.', '-') strDate = re.findall(datePattern, date) currentDate = time.strftime('%Y-%m-%d') flag = False startdate = '' enddate = '' if len(strDate) == 2: # '2015.12.29 ~ 2016.01.03' if self.compareDate(strDate[1], currentDate) > 0: flag = True startdate = strDate[0] enddate = strDate[1] elif len(strDate) == 1: if date.find(':') > 0: # 2017.04.20 08:30 ~ 12:30 if self.compareDate(strDate[0], currentDate) >= 0: flag = True startdate = strDate[0] enddate = strDate[0] else: # 2017.07.13 ~ 07.15 startdate = strDate[0] enddate = date[0:5] + date[len(date) - 5:] if self.compareDate(enddate, currentDate) >= 0: flag = True return flag, startdate, enddate def log_print(self, msg): ''' # 日志函数 # :param msg: 日志信息 # :return: # ''' print '%s: %s' % (time.strftime('%Y-%m-%d %H-%M-%S'), msg) def scrapy_date(self): strsplit = '------------------------------------------------------------------------------------' for link in self.start_urls: # time.sleep(1) self.driver.get(link) selenium_html = self.driver.execute_script("return document.documentElement.outerHTML") doc = pq(selenium_html) infoList = [] self.log.WriteLog(strsplit) self.log_print(strsplit) Elements = doc('div[id="eventList"]').find('div[class="list"]') for element in Elements.items(): date = element('div[class="info time"]').text().encode('utf8').strip() flag, startdate, enddate = self.date_isValid(date) if flag: title = element('h2').find('a').text().encode('utf8').strip() for keyword in self.keyword_list: if title.find(keyword) > -1: url = 'http://www.huodongshu.com' + element('h2').find('a').attr('href') orgnizer = element('div[class="info title"]').find('a').text().encode('utf8').strip() area = element('div[class="info site"]').text().encode('utf8').strip() dictM = {'title': title, 'date': date, 'url': url, 'keyword': keyword, 'startdate': startdate, 'enddate': enddate, 'area': area, 'organizer': orgnizer} infoList.append(dictM) # self.log.WriteLog('title:%s'%title) # self.log.WriteLog('url:%s' % url) # self.log.WriteLog('orgnizer:%s' % orgnizer) # self.log.WriteLog('kword:%s' % keyword) # self.log.WriteLog('area:%s' % area) # self.log.WriteLog('-----------------------------------' ) self.log_print('title:%s' % title) self.log_print('url:%s' % url) self.log_print('date:%s' % date) self.log_print('orgnizer:%s' % orgnizer) self.log_print('kword:%s' % keyword) self.log_print('area:%s' % area) self.log_print('--------------------------' ) break if len(infoList)>0: self.db.SaveMeetings(infoList) self.driver.close() self.driver.quit() obj = huodongshuSpider() obj.scrapy_date()