用单进程、多线程并发、多线程分别实现爬一个或多个网站的所有链接,用浏览器打开所有链接并保存截图 python

#coding=utf-8
import requests
import re,os,time,ConfigParser
from selenium import webdriver
from multiprocessing.dummy import Pool

######单进程#####

#创建保存截图的目录

def createImagesPath():
    dirname=os.path.dirname(os.path.abspath(__file__)) 
    #print dirname
    imges_path=os.path.join(dirname,time.strftime("%Y%m%d")) 
    #print imges_path
    try:
        if not os.path.exists(imges_path):
            os.mkdir(imges_path)   
        print u"截图保存的目录:",imges_path
        return imges_path 
    except Exception,e:
        print e

#从文件中获取要爬的网站地址
def getWebUrls(web_urls_file_path):
    web_urls=[]
    try:
        with open(web_urls_file_path) as fp:  
            lines=fp.readlines() 
        for line in lines: 
            if line.strip():
                web_urls.append(line.strip()) 
        return web_urls 
    except Exception,e:
        print e

#获取单个网站的所有有效链接
def getLinks(web_url):
    try:
        response=requests.get(web_url) 
        #print response
        html=response.text
        links=re.findall(r'href="(.*?)"',html)
        valid_links=[]
        invalid_links=[]     
        for link in links:  
            if link.strip().startswith("//"):   
                valid_links.append("http:"+link.strip())
            elif link.strip()=="" or link.strip()=="#" or link.strip()=="/" or link.strip().count("javascript")>=1 or link.strip().count("mailto:") >= 1:   
                invalid_links.append(link)
            elif re.search(r"(\.jpg)|(\.jpeg)|(\.gif)|(\.ico)|(\.png)|(\.js)|(\.css)$",link.strip()) or re.match(r'/[^/].*',link.strip()):
                invalid_links.append(link)
            else:
                valid_links.append(link.strip())
        valid_links=list(set(valid_links))  
        return valid_links     
    except Exception,e:
        print e

#保存有效的链接到.txt文件
def saveLinks(links):
    dirname=os.path.dirname(os.path.abspath(__file__)) 
    #print dirname
    links_path=os.path.join(dirname,time.strftime("%Y%m%d")) 
    #print links_path
    try:
        if not os.path.exists(links_path):
            os.mkdir(links_path)   
        links_file_path=os.path.join(links_path,"links.txt")
        print u"链接保存的路径:",links_file_path   
        with open(links_file_path,"w") as fp: 
            fp.writelines([link+"\n" for link in links])   
    except Exception,e:
        print e

#模拟浏览器打开链接并保存截图
class OpenLinkAndSaveImg(object):
    def __init__(self,browser_type):
        try:
            configFilePath=os.path.dirname(os.path.abspath(__file__))+"\\browserAndDriver.ini"
            print u"浏览器驱动配置文件路径:",configFilePath  
            cf=ConfigParser.ConfigParser()
            cf.read(configFilePath)
            browser_type=browser_type.strip().lower()
            driver_path=cf.get("browser_driver",browser_type.strip()).strip()
            print u"浏览器:%s ,驱动位置:%s"%(browser_type,driver_path)
            if browser_type=="ie":
                self.driver=webdriver.Ie(executable_path=eval(driver_path))
            elif browser_type=="chrome":
                self.driver=webdriver.Chrome(executable_path=eval(driver_path))
            elif browser_type=="firefox":
                self.driver=webdriver.Firefox(executable_path=eval(driver_path))
            else:
                print "invalid browser!"
        except Exception,e:
            print e
   
    #打开链接并保存截图
    def openLinkAndSaveImg(self,link_index_imgspath): 
        try:
            link,index,imgspath=link_index_imgspath   
            self.driver.get(link)   
            self.driver.maximize_window()   
            self.driver.get_screenshot_as_file(os.path.join(imgspath,str(index+1)+".png"))
        except Exception,e:
            print e
 
    def end(self):
        self.driver.quit()

if __name__=="__main__":

#单进程
    imgs_path=createImagesPath()
    #weburls=getWebUrls(r"e:\\urls.txt")
    weburls=getWebUrls(os.path.dirname(os.path.abspath(__file__))+"\\urls.txt")
    links=[]    
    start_time=time.time()
    for weburl in weburls:
        links+=getLinks(weburl)
    print u"链接数:%s ;获取所有链接耗时:%s"%(len(links),time.time()-start_time)
    saveLinks(links)
    start_time1=time.time()
    open_link_and_save_img=OpenLinkAndSaveImg("ie")
    for i in range(len(links)):
        open_link_and_save_img.openLinkAndSaveImg((links[i],i,imgs_path))
    open_link_and_save_img.end()
    print u"单进程打开所有链接并截图耗时耗时:",time.time()-start_time1

######多线程(线程池并发执行)######

#coding=utf-8
import requests
import re,os,time,ConfigParser
from selenium import webdriver
from multiprocessing.dummy import Pool

#创建保存截图的目录
def createImagesPath():
    dirname=os.path.dirname(os.path.abspath(__file__)) 
    #print dirname
    imges_path=os.path.join(dirname,time.strftime("%Y%m%d")) 
    #print imges_path
    try:
        if not os.path.exists(imges_path): 
            os.mkdir(imges_path)   
        print u"所有截图保存的目录:",imges_path
        return imges_path 
    except Exception,e:
        print e

#从文件中获取要爬的网站地址
def getWebUrls(web_urls_file_path):
    web_urls=[]
    try:
        with open(web_urls_file_path) as fp:
            lines=fp.readlines() 
        for line in lines: 
            if line.strip():
                web_urls.append(line.strip())  
        return web_urls 
    except Exception,e:
        print e

#获取单个网站的所有有效链接
def getLinks(web_url):
    try:
        response=requests.get(web_url) 
        #print response
        html=response.text
        links=re.findall(r'href="(.*?)"',html)
        valid_links=[]
        invalid_links=[]       
        for link in links:    
            if link.strip().startswith("//"):   
                valid_links.append("http:"+link.strip())
            elif link.strip()=="" or link.strip()=="#" or link.strip()=="/" or link.strip().count("javascript")>=1 or link.strip().count("mailto:") >= 1:   
                invalid_links.append(link)
            elif re.search(r"(\.jpg)|(\.jpeg)|(\.gif)|(\.ico)|(\.png)|(\.js)|(\.css)$",link.strip()) or re.match(r'/[^/].*',link.strip()):
                invalid_links.append(link)
            else:
                valid_links.append(link.strip())
        valid_links=list(set(valid_links))  
        return valid_links     
    except Exception,e:
        print e

#保存有效的链接到.txt文件
def saveLinks(links):
    dirname=os.path.dirname(os.path.abspath(__file__)) 
    #print dirname
    links_path=os.path.join(dirname,time.strftime("%Y%m%d")) 
    #print links_path
    try:
        if not os.path.exists(links_path):
            os.mkdir(links_path)  
        links_file_path=os.path.join(links_path,"links.txt")
        print u"所有链接保存的路径:",links_file_path  
        with open(links_file_path,"w") as fp:   
            fp.writelines([link+"\n" for link in links])   
    except Exception,e:
        print e

#获取浏览器和驱动
def getBrowserAndDriver(browser_type):
    try:
        configFilePath=os.path.dirname(os.path.abspath(__file__))+"\\browserAndDriver.ini"
        print u"浏览器驱动配置文件路径:",configFilePath    
        cf=ConfigParser.ConfigParser()
        cf.read(configFilePath)
        browser_type=browser_type.strip().lower()
        driver_path=cf.get("browser_driver",browser_type.strip()).strip()
        print u"浏览器:%s ,驱动位置:%s"%(browser_type,driver_path)
        return browser_type,driver_path
    except Exception,e:
        print e

#打开链接并保存截图
def openLinkAndSaveImg(browser_driver_link_index_imgspath):  
        try:
            browser,driverpath,link,index,imgspath=browser_driver_link_index_imgspath   
            command="webdriver."+browser.capitalize()+"(executable_path="+driverpath+")"
            driver=eval(command)
            driver.get(link)      
            driver.maximize_window()   
            driver.get_screenshot_as_file(os.path.join(imgspath,str(index+1)+".png"))  
            driver.quit()
        except Exception,e:
            print e

if __name__=="__main__":
    imgs_path=createImagesPath()
    #weburls=getWebUrls(r"e:\\urls.txt")
    weburls=getWebUrls(os.path.dirname(os.path.abspath(__file__))+"\\urls.txt")
    p=Pool(5)
    start_time1=time.time()
    links_list=p.map(getLinks,weburls)
    end_time1=time.time()
    links=[]
    for link_list in links_list:
        links+=link_list
    saveLinks(links)
    print u"链接数:%s ;获取所有链接耗时:%s"%(len(links),end_time1-start_time1)
    browser,driver=getBrowserAndDriver("ie")
    browser_driver_link_index_imgspath=zip([browser]*len(links),[driver]*len(links),links,range(len(links)),[imgs_path]*len(links))
    start_time2=time.time()
    p.map(openLinkAndSaveImg,browser_driver_link_index_imgspath)
    p.close()
    p.join()
    print u"多线程打开所有链接并截图耗时:",time.time()-start_time2

######多线程######

#coding=utf-8
import requests
import re,os,time,ConfigParser
from selenium import webdriver
from multiprocessing.dummy import Pool
import Queue
import threading

#创建保存截图的目录
def createImagesPath():
    dirname=os.path.dirname(os.path.abspath(__file__)) 
    #print dirname
    imges_path=os.path.join(dirname,time.strftime("%Y%m%d")) 
    #print imges_path
    try:
        if not os.path.exists(imges_path):
            os.mkdir(imges_path)   
        print u"所有截图保存的目录:",imges_path
        return imges_path 
    except Exception,e:
        print e

#从文件中获取要爬的网站地址
def getWebUrls(web_urls_file_path):
    web_urls=[]
    try:
        with open(web_urls_file_path) as fp:
            lines=fp.readlines()  
        for line in lines: 
            if line.strip():
                web_urls.append(line.strip())
        return web_urls 
    except Exception,e:
        print e

#获取单个网站的所有有效链接
def getLinks(web_url):
    try:
        response=requests.get(web_url) 
        #print response
        html=response.text
        links=re.findall(r'href="(.*?)"',html)
        valid_links=[]
        invalid_links=[]       
        for link in links:      
            if link.strip().startswith("//"):   
                valid_links.append("http:"+link.strip()) 
            elif link.strip()=="" or link.strip()=="#" or link.strip()=="/" or link.strip().count("javascript")>=1 or link.strip().count("mailto:") >= 1:   
                invalid_links.append(link)
            elif re.search(r"(\.jpg)|(\.jpeg)|(\.gif)|(\.ico)|(\.png)|(\.js)|(\.css)$",link.strip()) or re.match(r'/[^/].*',link.strip()):
                invalid_links.append(link)
            else:
                valid_links.append(link.strip())
        valid_links=list(set(valid_links)) 
        return valid_links     
    except Exception,e:
        print e

#保存有效的链接到.txt文件(当前目录\年月日\links.txt)
def saveLinks(links):
    dirname=os.path.dirname(os.path.abspath(__file__)) 
    #print dirname
    links_path=os.path.join(dirname,time.strftime("%Y%m%d")) 
    #print links_path
    try:
        if not os.path.exists(links_path):  
            os.mkdir(links_path)   
        links_file_path=os.path.join(links_path,"links.txt") 
        print u"所有链接保存的路径:",links_file_path  
        with open(links_file_path,"w") as fp:   
            fp.writelines([link+"\n" for link in links])   
    except Exception,e:
        print e

#多线程
class MyThread(threading.Thread):
    def __init__(self,browser,queue):
        threading.Thread.__init__(self)
        self.queue=queue
        try:
            configFilePath=os.path.dirname(os.path.abspath(__file__))+"\\browserAndDriver.ini"
            #print u"浏览器驱动配置文件路径:",configFilePath
            cf=ConfigParser.ConfigParser()
            cf.read(configFilePath)
            browser_type=browser.strip().lower()
            driver_path=cf.get("browser_driver",browser_type.strip()).strip()
            #print u"浏览器:%s ,驱动位置:%s"%(browser_type,driver_path)
            if browser_type=="ie":
                self.driver=webdriver.Ie(executable_path=eval(driver_path))
            elif browser_type=="chrome":
                self.driver=webdriver.Chrome(executable_path=eval(driver_path))
            elif browser_type=="firefox":
                self.driver=webdriver.Firefox(executable_path=eval(driver_path))
            else:
                print "invalid browser!"

except Exception,e:
            print e

def run(self):
        print "Starting"+self.name
        openLinkAndSaveImg(self.driver,self.queue)
        self.driver.quit()

#打开链接并保存截图
def openLinkAndSaveImg(driver,queue):  
        while not queue.empty():
            queueLock.acquire()
            link_index_imgspath=queue.get()
            queueLock.release()
            try:             
                link,index,imgspath=link_index_imgspath   
                driver.get(link)     
                driver.maximize_window()  
                driver.get_screenshot_as_file(os.path.join(imgspath,str(index+1)+".png"))
            except Exception,e:
                print e

if __name__=="__main__":
    #多线程
    imgs_path=createImagesPath()
    #weburls=getWebUrls(r"e:\\urls.txt")
    weburls=getWebUrls(os.path.dirname(os.path.abspath(__file__))+"\\urls.txt")
    p=Pool(5)
    start_time1=time.time()
    links_list=p.map(getLinks,weburls)
    end_time1=time.time()
    links=[]
    for link_list in links_list:
        links+=link_list
    saveLinks(links)
    print u"链接数:%s ;获取所有链接耗时:%s"%(len(links),end_time1-start_time1)

link_index_imgspath=zip(links,range(len(links)),[imgs_path]*len(links))
    
    queueLock=threading.Lock()
    threads=[]
    link_index_imgspath_Queue=Queue.Queue(len(links))
    for element in link_index_imgspath:
        link_index_imgspath_Queue.put(element)

start_time2=time.time()

for i in range(5):
        thread=MyThread("ie",link_index_imgspath_Queue)
        thread.start()
        threads.append(thread)

for t in threads:
        t.join()

print u"多线程打开所有链接并截图耗时:",time.time()-start_time2

print "end!"

上一篇:javascript判断手机旋转横屏竖屏


下一篇:java中set的交集、差集、并集的简单实现