一个完整的大作业：淘宝口红销量top10的销量和评价

Posted 2020-10-12 09方俊晖

网站：淘宝口红搜索页

https://s.taobao.com/search?q=%E5%8F%A3%E7%BA%A2&sort=sale-desc
先爬取该页面前十的口红的商品名、销售量、价格、评分以及评论数，发现该网页使用了json的方式，使用正则表达式匹配字段，抓取我们
所需要的信息。启用用户代理爬取数据，预防该网站的反爬手段，并把结果存入到csv文件中，效果如下。

成功爬取到淘宝口红top10的基本信息后，发现评论并不在同一页面上，并且该页面存在着进入评论页的关键字，爬取下来后放入一个列表中，然后用循环整个列表和页数，使用

正则表达式，匹配评论的关键字，成功爬取淘宝top10口红的评论进十万条，如下图所示。

 

完整的源代码如下：

from urllib import request
import re
import csv
import time
itemId=[]
sellerId=[]
links=[]
titles=[]
# ,\'商品评分\',\'评论总数\'
def get_product_info():
    fileName = \'商品.csv\'
    comment_file = open(fileName, \'w\', newline=\'\')
    write = csv.writer(comment_file)
    write.writerow([\'商品名\', \'连接\', \'销售量\', \'价格\', \'地址\',\'商品评分\',\'评论总数\'])
    comment_file.close()

    fileName2 = \'评价.csv\'
    productfile = open(fileName2, \'w\', newline=\'\')
    product_write = csv.writer(productfile)
    product_write.writerow([\'商品id\',\'商品名\',\'时间\', \'颜色分类\', \'评价\'])
    productfile.close()


def get_product():
    global itemId
    global sellerId
    global titles
    url = \'https://s.taobao.com/search?q=%E5%8F%A3%E7%BA%A2&sort=sale-desc\'
    head = {}
    # 写入User Agent信息
    head[
        \'User-Agent\'] = \'Mozilla/5.0 (Linux; android 4.1.1; Nexus 7 Build/JRO03D) AppleWebKit/535.19 (Khtml, like Gecko) Chrome/18.0.1025.166  Safari/535.19\'
    # 创建Request对象
    req = request.Request(url, headers=head)
    # 传入创建好的Request对象
    response = request.urlopen(req, timeout=30)
    # 读取响应信息并解码
    html = response.read().decode(\'utf-8\')
    # 打印信息
    pattam_id = \'"nid":"(.*?)"\'
    raw_title = \'"raw_title":"(.*?)"\'
    view_price = \'"view_price":"(.*?)"\'
    view_sales = \'"view_sales":"(.*?)"\'
    item_loc = \'"item_loc":"(.*?)"\'
    user_id = \'"user_id":"(.*?)"\'
    all_id = re.compile(pattam_id).findall(html)
    all_title = re.compile(raw_title).findall(html)
    all_price = re.compile(view_price).findall(html)
    all_sales = re.compile(view_sales).findall(html)
    all_loc = re.compile(item_loc).findall(html)
    all_userid = re.compile(user_id).findall(html)
    print("开始收集信息")
    try:
        for i in range(10):
            this_id = all_id[i]
            this_title = all_title[i]
            this_price = all_price[i]
            this_sales = all_sales[i]
            this_loc = all_loc[i]
            this_userid = all_userid[i]
            id = str(this_id)
            title = str(this_title)
            price = str(this_price)
            sales = str(this_sales)
            loc = str(this_loc)
            uid = str(this_userid)
            link = \'https://item.taobao.com/item.htm?id=\' + str(id)
            shoplink = \'https://dsr-rate.tmall.com/list_dsr_info.htm?itemId=\' +str(id)
            head = {}
            # 写入User Agent信息
            head[
                \'User-Agent\'] = \'Mozilla/5.0 (Linux; Android 4.1.1; Nexus 7 Build/JRO03D) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.166  Safari/535.19\'
            # 创建Request对象
            req2 = request.Request(shoplink, headers=head)
            # 传入创建好的Request对象
            response2 = request.urlopen(req2, timeout=30)
            # 读取响应信息并解码
            html2 = response2.read().decode(\'utf-8\')
            gradeAvg = \'"gradeAvg":(.*?,)"\'
            rateTotal = \'"rateTotal":(.*?,)"\'
            all_gradeAvg = re.compile(gradeAvg).findall(html2)
            all_rateTotal = re.compile(rateTotal).findall(html2)
            this_gradeAvg = all_gradeAvg
            this_rateTotal = all_rateTotal
            gradeAvg = str(this_gradeAvg)[2:-3]
            rateTotal = str(this_rateTotal)[2:-3]
            # print("平均分:" + gradeAvg)
            # print("评论总数：" + rateTotal)
            # print("商品名：" + title)
            # print("连接：" + link)
            # print("销售量:" + sales)
            # print("价格：" + price)
            # print("地址:" + loc)
            itemId.append(id)
            sellerId.append(uid)
            titles.append(title)
            comment_file = open(\'商品.csv\', \'a\', newline=\'\')
            write = csv.writer(comment_file)
            write.writerow([title, link, sales, price, loc,gradeAvg,rateTotal])
            comment_file.close()
    except (req.ConnectionError, IndexError, UnicodeEncodeError, TimeoutError) as e:
        print(e.args)
    except response.URLError as e:
        print(e.reason)
    except IOError as e:
        print(e)
    # HTTPError
    except response.HTTPError as e:
        print(e.code)
    print("商品基本信息收集完毕")







def get_product_comment():
# 具体商品获取评论
# 前十销量商品
    global title
    for i in range(10):
        print("正在收集第{}件商品评论".format(str(i + 1)))
        for j in range(1,551):
            # 商品评论的url
            detaillinks="https://rate.tmall.com/list_detail_rate.htm?itemId="+itemId[i]+"&sellerId="+sellerId[i]+"&currentPage="+str(j)
            head = {}
            # 写入User Agent信息
            head[\'User-Agent\'] = \'Mozilla/5.0 (Linux; Android 4.1.1; Nexus 7 Build/JRO03D) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.166  Safari/535.19\'
            req1 = request.Request(detaillinks, headers=head)
            # 传入创建好的Request对象
            response1 = request.urlopen(req1,timeout=30)
            # 读取响应信息并解码
            html1 = response1.read().decode(\'gbk\')
            # 打印信息
            # 评论
            rateContent = \'"rateContent":"(.*?)"\'
            # 时间
            rateDate = \'"rateDate":"(.*?)"\'
            # 颜色
            auctionSku = \'"auctionSku":"(.*?)"\'
            all_date = re.compile(rateDate).findall(html1)
            all_content = re.compile(rateContent).findall(html1)
            all_sku = re.compile(auctionSku).findall(html1)
            # 获取全部评论
            try:

                for k in range(0, len(all_content)):
                    this_date = all_date[k]
                    this_content = all_content[k]
                    this_sku = all_sku[k]
                    date = str(this_date)
                    content = str(this_content)
                    sku = str(this_sku)
                    # print("时间:" + date)
                    # print(sku)
                    # print("评价:" + content)
                    productfile = open(\'评价.csv\', \'a\', newline=\'\')
                    product_write = csv.writer(productfile)
                    product_write.writerow([itemId[i] + "\\t", titles[i], date, sku, content])
                    productfile.close()
            except (req1.ConnectionError, IndexError, UnicodeEncodeError, TimeoutError) as e:
                print(e.args)
            # URLError产生的原因：网络无连接，即本机无法上网；连接不到特定的服务器；服务器不存在
            except response1.URLError as e:
                print(e.reason)
            # HTTPError
            except response1.HTTPError as e:
                print(e.code)
            except IOError as e:
                print(e)
        print("第{}件商品评论收集完成".format(str(i+1)))


if __name__ == "__main__":
    start=time.time()
    get_product_info()
    get_product()
    # get_product_comment()
    end=time.time()
    total=end-start
    print(\'本次爬行用时:{:.2f}s!\'.format(total))
from urllib import request
import re
import csv
import time
itemId=[]
sellerId=[]
links=[]
titles=[]
# ,\'商品评分\',\'评论总数\'
def get_product_info():
    fileName = \'商品.csv\'
    comment_file = open(fileName, \'w\', newline=\'\')
    write = csv.writer(comment_file)
    write.writerow([\'商品名\', \'连接\', \'销售量\', \'价格\', \'地址\',\'商品评分\',\'评论总数\'])
    comment_file.close()

    fileName2 = \'评价.csv\'
    productfile = open(fileName2, \'w\', newline=\'\')
    product_write = csv.writer(productfile)
    product_write.writerow([\'商品id\',\'商品名\',\'时间\', \'颜色分类\', \'评价\'])
    productfile.close()


def get_product():
    global itemId
    global sellerId
    global titles
    url = \'https://s.taobao.com/search?q=%E5%8F%A3%E7%BA%A2&sort=sale-desc\'
    head = {}
    # 写入User Agent信息
    head[
        \'User-Agent\'] = \'Mozilla/5.0 (Linux; Android 4.1.1; Nexus 7 Build/JRO03D) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.166  Safari/535.19\'
    # 创建Request对象
    req = request.Request(url, headers=head)
    # 传入创建好的Request对象
    response = request.urlopen(req, timeout=30)
    # 读取响应信息并解码
    html = response.read().decode(\'utf-8\')
    # 打印信息
    pattam_id = \'"nid":"(.*?)"\'
    raw_title = \'"raw_title":"(.*?)"\'
    view_price = \'"view_price":"(.*?)"\'
    view_sales = \'"view_sales":"(.*?)"\'
    item_loc = \'"item_loc":"(.*?)"\'
    user_id = \'"user_id":"(.*?)"\'
    all_id = re.compile(pattam_id).findall(html)
    all_title = re.compile(raw_title).findall(html)
    all_price = re.compile(view_price).findall(html)
    all_sales = re.compile(view_sales).findall(html)
    all_loc = re.compile(item_loc).findall(html)
    all_userid = re.compile(user_id).findall(html)
    print("开始收集信息")
    try:
        for i in range(10):
            this_id = all_id[i]
            this_title = all_title[i]
            this_price = all_price[i]
            this_sales = all_sales[i]
            this_loc = all_loc[i]
            this_userid = all_userid[i]
            id = str(this_id)
            title = str(this_title)
            price = str(this_price)
            sales = str(this_sales)
            loc = str(this_loc)
            uid = str(this_userid)
            link = \'https://item.taobao.com/item.htm?id=\' + str(id)
            shoplink = \'https://dsr-rate.tmall.com/list_dsr_info.htm?itemId=\' +str(id)
            head = {}
            # 写入User Agent信息
            head[
                \'User-Agent\'] = \'Mozilla/5.0 (Linux; Android 4.1.1; Nexus 7 Build/JRO03D) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.166  Safari/535.19\'
            # 创建Request对象
            req2 = request.Request(shoplink, headers=head)
            # 传入创建好的Request对象
            response2 = request.urlopen(req2, timeout=30)
            # 读取响应信息并解码
            html2 = response2.read().decode(\'utf-8\')
            gradeAvg = \'"gradeAvg":(.*?,)"\'
            rateTotal = \'"rateTotal":(.*?,)"\'
            all_gradeAvg = re.compile(gradeAvg).findall(html2)
            all_rateTotal = re.compile(rateTotal).findall(html2)
            this_gradeAvg = all_gradeAvg
            this_rateTotal = all_rateTotal
            gradeAvg = str(this_gradeAvg)[2:-3]
            rateTotal = str(this_rateTotal)[2:-3]
            # print("平均分:" + gradeAvg)
            # print("评论总数：" + rateTotal)
            # print("商品名：" + title)
            # print("连接：" + link)
            # print("销售量:" + sales)
            # print("价格：" + price)
            # print("地址:" + loc)
            itemId.append(id)
            sellerId.append(uid)
            titles.append(title)
            comment_file = open(\'商品.csv\', \'a\', newline=\'\')
            write = csv.writer(comment_file)
            write.writerow([title, link, sales, price, loc,gradeAvg,rateTotal])
            comment_file.close()
    except (req.ConnectionError, IndexError, UnicodeEncodeError, TimeoutError) as e:
        print(e.args)
    except response.URLError as e:
        print(e.reason)
    except IOError as e:
        print(e)
    # HTTPError
    except response.HTTPError as e:
        print(e.code)
    print("商品基本信息收集完毕")







def get_product_comment():
# 具体商品获取评论
# 前十销量商品
    global title
    for i in range(10):
        print("正在收集第{}件商品评论".format(str(i + 1)))
        for j in range(1,551):
            # 商品评论的url
            detaillinks="https://rate.tmall.com/list_detail_rate.htm?itemId="+itemId[i]+"&sellerId="+sellerId[i]+"&currentPage="+str(j)
            head = {}
            # 写入User Agent信息
            head[\'User-Agent\'] = \'Mozilla/5.0 (Linux; Android 4.1.1; Nexus 7 Build/JRO03D) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.166  Safari/535.19\'
            req1 = request.Request(detaillinks, headers=head)
            # 传入创建好的Request对象
            response1 = request.urlopen(req1,timeout=30)
            # 读取响应信息并解码
            html1 = response1.read().decode(\'gbk\')
            # 打印信息
            # 评论
            rateContent = \'"rateContent":"(.*?)"\'
            # 时间
            rateDate = \'"rateDate":"(.*?)"\'
            # 颜色
            auctionSku = \'"auctionSku":"(.*?)"\'
            all_date = re.compile(rateDate).findall(html1)
            all_content = re.compile(rateContent).findall(html1)
            all_sku = re.compile(auctionSku).findall(html1)
            # 获取全部评论
            try:

                for k in range(0, len(all_content)):
                    this_date = all_date[k]
                    this_content = all_content[k]
                    this_sku = all_sku[k]
                    date = str(this_date)
                    content = str(this_content)
                    sku = str(this_sku)
                    # print("时间:" + date)
                    # print(sku)
                    # print("评价:" + content)
                    productfile = open(\'评价.csv\', \'a\', newline=\'\')
                    product_write = csv.writer(productfile)
                    product_write.writerow([itemId[i] + "\\t", titles[i], date, sku, content])
                    productfile.close()
            except (req1.ConnectionError, IndexError, UnicodeEncodeError, TimeoutError) as e:
                print(e.args)
            # URLError产生的原因：网络无连接，即本机无法上网；连接不到特定的服务器；服务器不存在
            except response1.URLError as e:
                print(e.reason)
            # HTTPError
            except response1.HTTPError as e:
                print(e.code)
            except IOError as e:
                print(e)
        print("第{}件商品评论收集完成".format(str(i+1)))


if __name__ == "__main__":
    start=time.time()
    get_product_info()
    get_product()
    # get_product_comment()
    end=time.time()
    total=end-start
    print(\'本次爬行用时:{:.2f}s!\'.format(total))