第五次作业

作业1：

使用Selenium框架爬取京东商城某类商品信息及图片。

代码如下：

from selenium import webdriver
from selenium.webdriver.chrome.options import Options
import urllib.request
import threading
import sqlite3
import os
import datetime
from selenium.webdriver.common.keys import Keys
import time
import pymysql

class MySpider:
    headers = {
        "User-Agent": "Mozilla/5.0 (Windows; U; Windows NT 6.0 x64; en-US; rv:1.9pre) Gecko/2008072421 Minefield/3.0.2pre"}

    imagePath = "D:/data/download"
    #No=0
    def startUp(self, url, key):
        # # Initializing Chrome browser
        #chrome_options = Options()
        #chrome_options.add_argument('--headless')
        #chrome_options.add_argument('--disable-gpu')
        self.driver = webdriver.Chrome()#chrome_options=chrome_options)

        # Initializing variables1

        self.threads = []
        self.No = 0
        self.imgNo = 0
        # Initializing database
        try:
            self.con = sqlite3.connect("phones.db")
            self.cursor = self.con.cursor()
            try:
                # 如果有表就删除
                self.cursor.execute("drop table phones")
            except:
                pass

            try:
                #  建立新的表
                sql = "create  table  phones  (mNo  varchar(32) primary key, mMark varchar(256),mPrice varchar(32),mNote varchar(1024),mFile varchar(256))"
                self.cursor.execute(sql)
            except:
                pass

        except Exception as err:
            print(err)
        try:
            if not os.path.exists(MySpider.imagePath):
                os.mkdir(MySpider.imagePath)
            images = os.listdir(MySpider.imagePath)
            # 将爬取到的图片信息存入指定文件夹
            for img in images:
                s = os.path.join(MySpider.imagePath, img)
                os.remove(s)
        except Exception as err:
            print(err)

        self.driver.get(url)
        keyInput = self.driver.find_element_by_id("key")
        keyInput.send_keys(key)
        #在键盘输入enter
        keyInput.send_keys(Keys.ENTER)

    def closeUp(self):
        try:

            self.con.commit()
            self.con.close()
            self.driver.close()

        except Exception as err:
            print(err);


    def insertDB(self, mNo, mMark, mPrice, mNote, mFile):
        try:
            self.cursor.execute(
                "insert into computer (mNo,mMark,mPrice,mNote,mFile) values (%s,%s,%s,%s,%s)",
                (mNo, mMark, mPrice, mNote, mFile))
        except Exception as err:
            print(err)


    def showDB(self):
        try:
            con = sqlite3.connect("phones.db")
            cursor =con.cursor()
            print("%-8s%-16s%-8s%-16s%s"%("No", "Mark", "Price", "Image", "Note"))
            cursor.execute("select mNo,mMark,mPrice,mFile,mNote from phones  order by mNo")

            rows = cursor.fetchall()
            for row in rows:
                print("%-8s %-16s %-8s %-16s %s" % (row[0], row[1], row[2], row[3],row[4]))

            con.close()
        except Exception as err:
            print(err)



    def download(self, src1, src2, mFile):
        data = None
        if src1:
            try:
                req = urllib.request.Request(src1, headers=MySpider.headers)
                resp = urllib.request.urlopen(req, timeout=10)
                data = resp.read()
            except:
                pass
        if not data and src2:
            try:
                req = urllib.request.Request(src2, headers=MySpider.headers)
                resp = urllib.request.urlopen(req, timeout=10)
                data = resp.read()
            except:
                pass
        if data:
            print("download begin", mFile)
            fobj = open(MySpider.imagePath + "\" + mFile, "wb")
            fobj.write(data)
            fobj.close()
            print("download finish", mFile)


    def processSpider(self):
        try:
            time.sleep(1)
            print(self.driver.current_url)
            lis =self.driver.find_elements_by_xpath("//div[@id='J_goodsList']//li[@class='gl-item']")
            for li in lis:
                # We find that the image is either in src or in data-lazy-img attribute
                try:
                    src1 = li.find_element_by_xpath(".//div[@class='p-img']//a//img").get_attribute("src")
                except:
                    src1 = ""

                try:
                    src2 = li.find_element_by_xpath(".//div[@class='p-img']//a//img").get_attribute("data-lazy-img")
                except:
                    src2 = ""


                try:
                    price = li.find_element_by_xpath(".//div[@class='p-price']//i").text
                except:
                    price = "0"

                try:
                    note = li.find_element_by_xpath(".//div[@class='p-name p-name-type-2']//em").text
                    mark = note.split(" ")[0]
                    mark = mark.replace("爱心东东
", "")
                    mark = mark.replace(",", "")
                    note = note.replace("爱心东东
", "")
                    note = note.replace(",", "")

                except:
                    note = ""
                    mark = ""

                self.No = self.No + 1
                no = str(self.No)
                while len(no) < 6:
                    no = "0" + no
                print(no, mark, price)
                if src1:
                    src1 = urllib.request.urljoin(self.driver.current_url, src1)
                    p = src1.rfind(".")
                    mFile = no + src1[p:]
                elif src2:
                    src2 = urllib.request.urljoin(self.driver.current_url, src2)
                    p = src2.rfind(".")
                    mFile = no + src2[p:]
                if src1 or src2:
                    T = threading.Thread(target=self.download, args=(src1, src2, mFile))
                    T.setDaemon(False)
                    T.start()
                    self.threads.append(T)
                else:
                    mFile = ""
                self.insertDB(no, mark, price, note, mFile)
         # 取下一页的数据，直到最后一页
            try:
                self.driver.find_element_by_xpath("//span[@class='p-num']//a[@class='pn-next disabled']")
            except:
                nextPage = self.driver.find_element_by_xpath("//span[@class='p-num']//a[@class='pn-next']")
                time.sleep(10)
                #if(self.No<50):
                nextPage.click()
                self.processSpider()
        except Exception as err:
            print(err)

    def executeSpider(self, url, key):
        starttime = datetime.datetime.now()
        print("Spider starting......")
        self.startUp(url, key)
        print("Spider processing......")
        self.processSpider()
        print("Spider closing......")
        self.closeUp()
        for t in self.threads:
            t.join()
        print("Spider completed......")
        endtime = datetime.datetime.now()
        elapsed = (endtime - starttime).seconds
        print("Total ", elapsed, " seconds elapsed")


url = "http://www.jd.com"
spider = MySpider()
while True:
    print("1.爬取")
    print("2.显示")
    print("3.退出")
    s = input("请选择(1,2,3):")
    if s == "1":
        spider.executeSpider(url, "诺基亚老人机")
        continue
    elif s == "2":
        spider.showDB()
        continue
    elif s == "3":
        break

运行结果

心得体会

1.根据老师的代码了解了selenium框架爬取网页的步骤
2.selenium框架应该不适合大规模数据爬取吧，这速度真不敢恭维。。

作业2：

使用Selenium框架+ MySQL数据库存储技术路线爬取“沪深A股”、“上证A股”、“深证A股”3个板块的股票数据信息。

代码如下：

from selenium import webdriver
from selenium.webdriver.chrome.options import Options
import time
import pymysql


def spider():
    global n
    time.sleep(1)
    lis = driver.find_elements_by_xpath('//*[@id="table_wrapper-table"]/tbody//tr')
    for li in lis:
        array = li.text.split(" ")
        count = array[0]
        id = array[1]
        name = array[2]
        new_price = array[6]
        extent = array[7]
        change_price = array[8]
        number = array[9]
        money = array[10]
        promote = array[11]
        highest = array[12]
        lowest = array[13]
        today_begin = array[14]
        yesterday_over = array[15]
        cursor.execute("insert into stock" + str(n) + "(count,id,sName,new_price,extent,change_price,sNumber,money,"
                                                      "promote,highest,lowest,today_begin,yesterday_over) "
                       "values( %s, % s, % s, % s, % s, % s, % s, % s, % s, % s, % s, % s, % s)",
                       (count, id, name, new_price, extent, change_price, number, money, promote, highest,
                        lowest, today_begin, yesterday_over))


# 实现翻页功能
def nextpage():
    try:
        driver.find_element_by_xpath(
            "//div[@class='dataTables_paginate paging_input']//a[@class='next paginate_button disabled']")
        print("没有下一页了")
    except:
        nextPage = driver.find_element_by_xpath("//*[@id='main-table_paginate']/a[2]")
        time.sleep(3)
        nextPage.click()
        spider()


def changeStock1():
    try:
        global n
        n = 1
        # 初次运行创建stock1
        '''sql1 = "create table stock1(count int,id varchar(16),sName varchar(16),new_price varchar(16)," 
              "extent varchar(16),change_price varchar(16),sNumber varchar(16),money varchar(16),promote varchar(16)," 
              "highest varchar(16),lowest varchar(16),today_begin varchar(16),yesterday_over varchar(16));"
        cursor.execute(sql1)'''
        cursor.execute("delete from stock1")
        # 默认在沪深A股板块，无需点击
        '''hs = driver.find_element_by_xpath('//*[@id="nav_hs_a_board"]')
        time.sleep(3)
        hs.click()'''
        # 启动爬虫
        print("***开始爬取沪深A股的数据***")
        spider()
        # 因为数据过多，所以只进行一次翻页操作
        nextpage()
        print("***爬取结束，数据已存入数据库中***")
    except Exception as err:
        print(err)


def changeStock2():
    try:
        global n
        n = 2
        # 创建stock2
        '''sql2 = "create table stock2(count int,id varchar(16),sName varchar(16),new_price varchar(16),extent " 
               "varchar(16),change_price varchar(16),sNumber varchar(16),money varchar(16),promote varchar(16)," 
               "highest varchar(16),lowest varchar(16),today_begin varchar(16),yesterday_over varchar(16));"
        cursor.execute(sql2)'''
        cursor.execute("delete from stock2")
        # 进入上证A股板块
        sh = driver.find_element_by_xpath('//*[@id="nav_sh_a_board"]')
        time.sleep(3)
        sh.click()
        print("***开始爬取上证A股的数据***")
        spider()
        nextpage()
        print("***爬取结束，数据已存入数据库中***")
    except Exception as err:
        print(err)


def changeStock3():
    try:
        global n
        n = 3
        # 创建stock3
        '''sql3 = "create table stock3(count int,id varchar(16),sName varchar(16),new_price varchar(16)," 
               "extent varchar(16),change_price varchar(16),sNumber varchar(16),money varchar(16),promote varchar(16)," 
               "highest varchar(16),lowest varchar(16),today_begin varchar(16),yesterday_over varchar(16));"
        cursor.execute(sql3)'''
        cursor.execute("delete from stock3")
        # 进入深证A股板块
        sz = driver.find_element_by_xpath('//*[@id="nav_sz_a_board"]')
        time.sleep(3)
        sz.click()
        print("***开始爬取深证A股的数据***")
        spider()
        nextpage()
        print("***爬取结束，数据已存入数据库中***")
    except Exception as err:
        print(err)


# 进入网址后默认处在沪深A股板块
url = 'http://quote.eastmoney.com/center/gridlist.html#hs_a_board'
# 模拟时不打开浏览器
chrome_options = Options()
chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')
driver = webdriver.Chrome(chrome_options=chrome_options)
driver.get(url)
# 最大化浏览器窗口
driver.maximize_window()
# 连接数据库
con = pymysql.connect(host='127.0.0.1', port=3306, user='root', passwd='123456', db='mydb', charset='utf8')
cursor = con.cursor(pymysql.cursors.DictCursor)
# 选择板块进行数据爬取
print("股票信息如下：
"
      "1.沪深A股
"
      "2.上证A股
"
      "3.深证A股")
n = input("请选择你要爬取的板块：")
if n == "1":
    changeStock1()
if n == "2":
    changeStock2()
if n == "3":
    changeStock3()

# 关闭连接
con.commit()
con.close()

运行结果

心得体会

1.selenium爬取网页是模拟人在操作，所以当浏览器窗口太小时可能会出现“无效点击”的报错，我理解是因为所要点击的位置需要进行左右拖动才可以找到位置，可以通过最大化浏览器窗口
driver.maximize_window()来解决这个问题。
2.selenium爬取内容速度较慢，在程序运行过程中将数据存储于缓存，没有运行结束之前数据不会存储进数据库，只有关闭连接时才会将数据传输进数据库。
3.点击新的页面后需要sleep一小段时间，等待页面数据刷新，不然所爬取的数据会是原来页面的内容。

作业3：

使用Selenium框架+MySQL爬取中国mooc网课程资源信息

代码如下：

from selenium import webdriver
from selenium.webdriver.chrome.options import Options
import time
import pymysql


def spider():
    time.sleep(1)
    global id
    # 找到所有课程以列表形式返回
    lis = driver.find_elements_by_xpath('//div[@class="_1aoKr"]//div[@class="_2mbYw"]')
    for li in lis:
        time.sleep(1)
        li.click()
        # 获取页面句柄，原网页为0，新网页为1
        window = driver.window_handles
        # 切换到新页面
        driver.switch_to.window(window[1])
        time.sleep(1)
        id += 1
        course = driver.find_element_by_xpath('//*[@id="g-body"]/div[1]/div/div[3]/div/div[1]/div[1]/span[1]').text
        teacher = driver.find_element_by_xpath('//*[@id="j-teacher"]//h3[@class="f-fc3"]').text
        collage = driver.find_element_by_xpath('//*[@id="j-teacher"]/div/a/img').get_attribute('alt')
        process = driver.find_element_by_xpath('//*[@id="course-enroll-info"]/div/div[1]/div[2]/div[1]').text
        count = driver.find_element_by_xpath('//*[@id="course-enroll-info"]/div/div[2]/div[1]/span').text
        brief = driver.find_element_by_xpath('//*[@id="j-rectxt2"]').text
        print(id)
        print(course)
        print(teacher)
        print(collage)
        print(process)
        print(count)
        print(brief)
        cursor.execute("insert into mooc(id, course, teacher, collage, process, count, brief) "
                       "values( % s, % s, % s, % s, % s, % s, % s)",
                       (id, course, teacher, collage, process, count, brief))
        # 关闭此窗口
        driver.close()
        # 切换回原网页，方便进行下一次操作
        driver.switch_to.window(window[0])
        print('***********************************')


def all_spider():
    global id
    spider()
    time.sleep(1)
    # id<21时仅爬取两页内容
    if (driver.find_elements_by_xpath('//a[@class="_3YiUU "]')[-1].text == '下一页') & (id < 21):
        next = driver.find_elements_by_xpath('//a[@class="_3YiUU "]')[-1]
        next.click()
        # 模拟鼠标向上滑动，回到页面顶部
        driver.execute_script("window.scrollTo(document.body.scrollHeight,0)")
        all_spider()


url = 'https://www.icourse163.org/channel/2001.htm'
# 模拟时不打开浏览器
chrome_options = Options()
'''chrome_options.add_argument('--headless')
chrome_options.add_argument('--disable-gpu')'''
driver = webdriver.Chrome(chrome_options=chrome_options)
driver.get(url)
# 最大化浏览器窗口
driver.maximize_window()
# 连接数据库
con = pymysql.connect(host='127.0.0.1', port=3306, user='root', passwd='123456', db='mydb', charset='utf8')
cursor = con.cursor(pymysql.cursors.DictCursor)
sql = 'create table mooc(id int,course varchar(32),teacher varchar(16),collage varchar(32),process varchar(64),' 
      'count varchar(64),brief text);'
'''cursor.execute(sql)'''
cursor.execute("delete from mooc")

id = 0
print("开始爬取")
all_spider()
print("爬取结束")
# 关闭连接
con.commit()
con.close()

运行结果

心得体会

1.通过点击事件跳转到一个新的页面，想要关闭新页面时要先获取句柄，通过句柄切换页面，不然driver.close()关闭的会是原窗口。
2.mooc的课程翻页部分好坑，上一页和下一页按钮的元素属性一模一样，如果没有通过文本信息进行判断，就会在最后一页和倒数第二页“反复横跳”，就是永远找得到try命令下的xpath目标，无法结束程序。
3.在爬取完当页课程信息，进行翻页操作后，程序无法找到下一课程信息，具体原因不太清楚，但可以通过js模拟鼠标向上滑动滚轮解决此问题。