Python爬虫存谷歌搜索图片

之前做box相关的learning项目，需要好多box的图片，就写了这个爬虫。
功能相当于在google image中搜索相关关键字，批量保存前几页图片。

from selenium import webdriver  
import time  
import urllib  
from selenium.webdriver.common.keys import Keys

# 查询图片的关键字
keys = ['cupboard','glass box']

# chrome driver绝对路径
driver = webdriver.Chrome('C:\Program Files (x86)\Google\Chrome\Application\chromedriver.exe')  


for key in keys:
    driver.get("https://www.google.co.jp/imghp?hl=zh-CN&tab=wi&ei=qJgtWINugqDSBJTzqqgI&ved=0EKouCBMoAQ")
    elem = driver.find_element_by_name("q")
    elem.send_keys(key)
    elem.send_keys(Keys.RETURN)
    #driver.fine_element_by_name("btnG").click()

    xpath = '//div[@id="rg_s"]/div/a'  

    driver.maximize_window()  
    #driver.set_window_size(1000,30000)
    #time.sleep(5)
    img_url_dic = {}  
    pos = 0  
    m = 0 
    #urlfile = open(key+".txt",'w')
    urlfile = open("url.txt",'a')
    for i in range(5): 
        print i
        pos += i*500 
        #js = "document.documentElement.scrollTop=%d" % pos  
        js = "window.scrollBy(0,%d)" % pos
        driver.execute_script(js)    
        time.sleep(3)
        for element in driver.find_elements_by_xpath(xpath): 
            img_url = element.get_attribute('href')   
            if img_url != None and not img_url_dic.has_key(img_url):  
                img_url_dic[img_url] = ''  
                img_url = img_url[img_url.index("=")+1:img_url.index("&")] #first image url - smallest
                img_url = img_url.replace("%3A",":") 
                img_url = img_url.replace("%2F","/")
                
                #####save picture
                #data = urllib.urlopen(img_url).read()  
                #filename = img_url[img_url.rindex("/")+1:]
                #f = open("pictures\\"+filename, 'wb')  
                #f.write(data)  
                #f.close()  
                
                #####save img url
                #print img_url
                urlfile.write(img_url)
                urlfile.write('\n')
    urlfile.close()
driver.close()