标签:
一是用微软提供的扩展库win32com来操作IE:
win32com可以获得类似js里面的document对象,但貌似是只读的(文档都没找到)。
二是用selenium的webdriver:
selenium则提供了Chrome,IE,FireFox等的支持,每种浏览器都有execute_script和find_element_by_xx方法,可以方便的执行js脚本(包括修改元素)和读取html里面的元素。不足是selenium只提供对python2.6和2.7的支持。
三是用python自带的HTMLParser解析:
HTMLParser则是需要自己写个类继承基类,重写解析元素的方法。
1.win32com
1 #将滚动条滑到底,最多滑动20000像素 2 #模拟键盘右键,查看多张图片 3 import sys 4 import win32com.client,win32api 5 import urllib.request 6 import time 7 import os 8 def main(): 9 #获取参数 10 url=sys.argv[1] 11 #操作IE 12 ie=win32com.client.Dispatch("InternetExplorer.Application") 13 ie.Navigate(url) 14 ie.Visible=True 15 last_url=‘‘ 16 dir_name=‘‘ 17 while last_url!=url: 18 print(‘\nThe URL is:‘,url,‘\n‘) 19 while ie.ReadyState != 4: 20 time.sleep(1) 21 while ie.Document.readyState != "complete": 22 time.sleep(1) 23 #滑动滚动条 24 win=ie.Document.parentWindow 25 lastY=-1; 26 for i in range(40): 27 win.scrollTo(0,500*i) 28 nowY=win.pageYOffset 29 if(nowY==lastY): 30 break 31 lastY=nowY 32 time.sleep(0.4) 33 print(‘Document load state:‘,ie.Document.readyState) 34 doc=ie.Document 35 #第一次需要创建目录 36 if(dir_name==‘‘): 37 root_dir=‘E:\\img‘ 38 dir_name=root_dir+‘\\‘+doc.title 39 dir_name=dir_name.replace(‘|‘,‘-‘) 40 if(os.path.exists(root_dir)!=True): 41 os.mkdir(root_dir) 42 if(os.path.exists(dir_name)!=True): 43 os.mkdir(dir_name) 44 all_image=doc.images 45 print(‘共有‘,all_image.length,‘张图片‘) 46 count=0; 47 for img in all_image: 48 if(img.id==‘b_img‘): 49 count=count+1 50 print(count,img.src) 51 time.sleep(1) 52 img_file=urllib.request.urlopen(img.src) 53 byte=img_file.read() 54 print(count,‘donwload complete!‘,‘-‘*10,‘size:‘,‘{:.3}‘.format(byte.__len__()/1024),‘KB‘) 55 if(byte.__len__()>7000): 56 file_name=img.src.replace(‘/‘,‘_‘) 57 file_name=file_name.replace(‘:‘,‘_‘) 58 end=file_name.__len__() 59 if(file_name.rfind(‘!‘)!=-1): 60 end=file_name.rfind(‘!‘) 61 if(file_name.rfind(‘?‘)!=-1): 62 end=file_name.rfind(‘?‘) 63 file_name=file_name[:end] 64 write_file=open(dir_name+‘\\‘+file_name,‘wb‘) 65 write_file.write(byte) 66 write_file.close() 67 print(count,file_name,‘complete!‘) 68 #下一张 69 last_url=url 70 win32api.keybd_event(39,0) 71 time.sleep(1) 72 url=ie.Document.url 73 print(last_url,url) 74 #ie.Quit() 75 if __name__ == ‘__main__‘: 76 main()
2.selenium
1 # -*- coding: cp936 -*- 2 import sys 3 import urllib 4 import time 5 import os 6 from selenium import webdriver 7 def main(): 8 #获取参数 9 url=sys.argv[1] 10 #操作IE 11 driver=webdriver.Chrome() 12 driver.get(url) 13 driver.execute_script("window.scrollTo(0, document.body.scrollHeight);") 14 #创建目录 15 dir_name=driver.find_element_by_tag_name(‘title‘).text 16 print dir_name 17 root_dir=‘E:\\img‘ 18 dir_name=root_dir+‘\\‘+dir_name 19 dir_name=dir_name.replace(‘|‘,‘-‘) 20 if(os.path.exists(root_dir)!=True): 21 os.mkdir(root_dir) 22 if(os.path.exists(dir_name)!=True): 23 os.mkdir(dir_name) 24 images=driver.find_elements_by_tag_name(‘img‘) 25 count=0 26 for image in images: 27 count=count+1 28 image_url=str(image.get_attribute(‘src‘)) 29 img_file=urllib.urlopen(image_url) 30 byte=img_file.read() 31 print count,‘donwload complete!‘,‘-‘*10,‘size:‘,byte.__len__()/1024,‘KB‘ 32 if(byte.__len__()>7000): 33 file_name=image_url.replace(‘/‘,‘_‘) 34 file_name=file_name.replace(‘:‘,‘_‘) 35 end=file_name.__len__() 36 if(file_name.rfind(‘!‘)!=-1): 37 end=file_name.rfind(‘!‘) 38 if(file_name.rfind(‘?‘)!=-1): 39 end=file_name.rfind(‘?‘) 40 file_name=file_name[:end] 41 write_file=open(dir_name+‘\\‘+file_name,‘wb‘) 42 write_file.write(byte) 43 write_file.close() 44 print count,file_name,‘complete!‘ 45 driver.quit() 46 if __name__ == ‘__main__‘: 47 main()
3.HTMLParser:
1 # import modules used here -- sys is a very standard one 2 import sys 3 import urllib.request 4 # Gather our code in a main() function 5 from html.parser import HTMLParser 6 class MyHTMLParser(HTMLParser): 7 def handle_starttag(self,tag,attrs): 8 if(tag==‘img‘): 9 for attr in attrs: 10 if(attr[0]==‘src‘): 11 img_file=urllib.request.urlopen(attr[1]) 12 byte=img_file.read() 13 #文件大于1000b则生成文件,添加计数,下载多少图片,显示html代码 14 if(byte.__len__()>1000): 15 file_name=attr[1].replace(‘/‘,‘_‘) 16 file_name=file_name.replace(‘:‘,‘_‘) 17 end=file_name.__len__() 18 if(file_name.rfind(‘!‘)!=-1): 19 end=file_name.rfind(‘!‘) 20 if(file_name.rfind(‘?‘)!=-1): 21 end=file_name.rfind(‘?‘) 22 file_name=file_name[:end] 23 ## print(file_name) 24 write_file=open(‘E:\\img\\‘+file_name,‘wb‘) 25 write_file.write(byte) 26 write_file.close() 27 def main(): 28 #获取参数 29 url=sys.argv[1] 30 print(‘\nThe URL is:‘,url,‘\n‘) 31 #读取url所指向的资源 32 html_file=urllib.request.urlopen(url) 33 byte_content=html_file.read() 34 #将html网页保存起来 35 url_file=open(‘E:\\img\\html\\result.htm‘,‘wb‘) 36 url_file.write(byte_content) 37 url_file.close() 38 #从字节转换为字符串 39 s=str(byte_content, encoding = "utf-8") 40 #print(s) 41 #bytes.decode(html_file.read()) 42 parser=MyHTMLParser(strict=False) 43 parser.feed(s) 44 # Standard boilerplate to call the main() function to begin 45 # the program. 46 if __name__ == ‘__main__‘: 47 main()
标签:
原文地址:http://www.cnblogs.com/liu-ke/p/5092391.html