一、搭建Python環境

1. 本帖使用的是Python 2.7
2. 涉及到的模組：spynner, scrapy, bs4, pymmssql

二、要獲取的天貓資料

三、資料抓取流程

四、原始碼

#coding:utf-8
import spynner
from scrapy.selector import Selector
from bs4 import BeautifulSoup
import random
import pymssql


#------------------------接資料庫-----------------------------#
server="localhost"
 

user="sa"
password = "123456"
conn=pymssql.connect(server,user,password,"TmallData")
if conn:
    print "DataBase connecting successfully!"
else:
    print "DataBase connecting error!"
cursor=conn.cursor()
#----------------------定義網頁操作函式--------------------------#
def py_click_element(browser,pos):
 

    #點選網頁中的元素
    #pos example:'a[href="#description"]'
    browser.click(pos)
    browser.wait(random.randint(3,10))
    return browser

def py_click_xpath(browser,xpath):
    xpath=xpath+'/@href'
    inner_href=Selector(text=browser.html).xpath(xpath).extract()
    pos='a[href="'+str(inner_href[0])+'"]'
 

    browser=py_click_element(browser, pos)
    return browser

def py_webpage_load(browser,url):
    browser.load(url,load_timeout=60)
    browser.wait(10)
    return browser

def py_check_element(browser,xpath):
    #按照xpath查詢元素，如果存在則返回True，否則返回False
    if Selector(text=browser.html).xpath(xpath).extract()!=[]:
        return True
    else:
        return False

def py_extract_xpath(browser,xpath):
    if py_check_element(browser, xpath):
        return Selector(text=browser.html).xpath(xpath).extract()[0]
    else:
        return "none"

def py_extract_xpaths(browser,xpaths):
    #批量提取網頁內容
    length=len(xpaths)
    results=[0]*length
    for i in range(length):
        results[i]=py_extract_xpath(browser, xpaths[i])
    return results

#-----------------------------資料庫操作函式---------------------------#


#-----------------------------資料提取函式----------------------------#
def py_getDealReord(doc):
    soup=BeautifulSoup(doc,'lxml')
    tr=soup.find_all('tr')
    total_dealRecord=[([0]*5)for i in range(len(tr))] 
    i=-1
    for this_tr in tr:
        i=i+1
        td_user=this_tr.find_all('td',attrs={'class':"cell-align-l buyer"})
        for this_td in td_user:
            total_dealRecord[i][0]=this_td.getText().strip(' ')
            #print username
        td_style=this_tr.find_all('td',attrs={'class':"cell-align-l style"})
        for this_td in td_style:
            total_dealRecord[i][1]=this_td.getText(',').strip(' ')
            #print style
        td_quantity=this_tr.find_all('td',attrs={'class':"quantity"})
        for this_td in td_quantity:
            total_dealRecord[i][2]=this_td.getText().strip(' ')
            #print quantity
        td_dealtime=this_tr.find_all('td',attrs={'class':"dealtime"})
        for this_td in td_dealtime:
            total_dealRecord[i][3]=this_td.find('p',attrs={'class':"date"}).getText()
            total_dealRecord[i][4]=this_td.find('p',attrs={'class':"time"}).getText()
    return total_dealRecord
#--------------------獲取要抓取的所有商品連結-----------------------#
cursor.execute("""
select * from ProductURLs where BrandName='NB'
""")


file=open("H:\\Eclipse\\TmallCrawling\\HTMLParse\\errLog.txt")
InProductInfo=cursor.fetchall()
browser=spynner.Browser()
for temp_InProductInfo in InProductInfo:

    url='https:'+temp_InProductInfo[2]

    BrandName=temp_InProductInfo[0]
    ProductType=temp_InProductInfo[1]
    print BrandName,'\t',ProductType,'\t',url
    #url= 'https://detail.tmall.com/item.htm?id=524425656711&rn=77636d6db8dea5e30060976fdaf9768d&abbucket=19' 

    try:
        browser=py_webpage_load(browser, url)
    except:
        print "Loading webpage failed."
        file.write(url)
        file.write('\n')
        continue

    xpaths=['//*[@id="J_PromoPrice"]/dd/div/span/text()',\
        '//*[@id="J_StrPriceModBox"]/dd/span/text()',\
        '//*[@id="J_DetailMeta"]/div[1]/div[1]/div/div[1]/h1/text()',\
        '//*[@id="J_PostageToggleCont"]/p/span/text()',\
        '//*[@id="J_EmStock"]/text()',\
        '//*[@id="J_CollectCount"]/text()',\
        '//*[@id="J_ItemRates"]/div/span[2]/text()',\
        '//*[@id="J_DetailMeta"]/div[1]/div[1]/div/ul/li[1]/div/span[2]/text()']
    out_ProductInfo=py_extract_xpaths(browser,xpaths)
    browser=py_click_element(browser,'a[href="#description"]')
    ProductProperty=py_extract_xpath(browser, '//*[@id="J_AttrUL"]')
    soup=BeautifulSoup(ProductProperty,'lxml')
    li=soup.find_all('li')
    prop=''
    for this_li in li:
        prop=prop+this_li.getText()+'\\'
    prop=prop[0:len(prop)-1]
    out_ProductProperty=prop
    print out_ProductProperty
    cursor.execute("""
    Insert into py_ProductInfo values(%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)
    """,(BrandName,ProductType,url,\
         out_ProductInfo[2],out_ProductInfo[1],\
         out_ProductInfo[0],out_ProductInfo[7],\
         out_ProductInfo[1],out_ProductInfo[3],\
         out_ProductInfo[4],out_ProductInfo[5],\
         out_ProductProperty))
    conn.commit()
    Deal_PageCount=0
    browser=py_click_element(browser, 'a[href="#J_DealRecord"]')
    #browser.browse(True)
    DealRecord=py_extract_xpath(browser, '//*[@id="J_showBuyerList"]/table/tbody')
    out_DealRecord=py_getDealReord(DealRecord)
    for temp_DealRecord in out_DealRecord:
        if str(temp_DealRecord[4])=='0':
            continue
        cursor.execute("""
        Insert into DealRecord values(%s,%s,%s,%s,%s,%s)
        """,(url,temp_DealRecord[0],temp_DealRecord[1],\
             temp_DealRecord[2],temp_DealRecord[3],\
             temp_DealRecord[4]))
        conn.commit()
    Deal_PageCount=Deal_PageCount+1
    print "Page ",Deal_PageCount
    for i in range(6):
        if (i==0) or (i==2):
            continue
        xpath='//*[@id="J_showBuyerList"]/div/div/a['+str(i)+']'
        if py_check_element(browser,xpath):
            browser=py_click_xpath(browser, xpath)
            DealRecord=py_extract_xpath(browser, '//*[@id="J_showBuyerList"]/table/tbody')
            out_DealRecord=py_getDealReord(DealRecord)
            for temp_DealRecord in out_DealRecord:
                if str(temp_DealRecord[4])=='0':
                    continue
                cursor.execute("""
                Insert into DealRecord values(%s,%s,%s,%s,%s,%s)
                """,(url,temp_DealRecord[0],temp_DealRecord[1],\
                     temp_DealRecord[2],temp_DealRecord[3],\
                     temp_DealRecord[4]))
                conn.commit()
            Deal_PageCount=Deal_PageCount+1
            print "Page ",Deal_PageCount
    while py_check_element(browser, '//*[@id="J_showBuyerList"]/div/div/a[6]'):
        browser=py_click_xpath(browser, '//*[@id="J_showBuyerList"]/div/div/a[6]')
        DealRecord=py_extract_xpath(browser, '//*[@id="J_showBuyerList"]/table/tbody')
        out_DealRecord=py_getDealReord(DealRecord)
        for temp_DealRecord in out_DealRecord:
            if str(temp_DealRecord[4])=='0':
                continue
            cursor.execute("""
            Insert into DealRecord values(%s,%s,%s,%s,%s,%s)
            """,(url,temp_DealRecord[0],temp_DealRecord[1],\
                 temp_DealRecord[2],temp_DealRecord[3],\
                 temp_DealRecord[4]))
            conn.commit()
        Deal_PageCount=Deal_PageCount+1
        print "Page ",Deal_PageCount