演示python单线程、多线程、多进程、协程
1 import requests,json,random 2 import re,threading,time 3 from lxml import etree 4 5 lock=threading.Lock() 6 semaphore=threading.Semaphore(100) ###每次限制只能100线程 7 8 user_agent_list = [ 9 "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/22.0.1207.1 Safari/537.1" , 10 "Mozilla/5.0 (X11; CrOS i686 2268.111.0) AppleWebKit/536.11 (KHTML, like Gecko) Chrome/20.0.1132.57 Safari/536.11", 11 "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/536.6 (KHTML, like Gecko) Chrome/20.0.1092.0 Safari/536.6", 12 "Mozilla/5.0 (Windows NT 6.2) AppleWebKit/536.6 (KHTML, like Gecko) Chrome/20.0.1090.0 Safari/536.6", 13 "Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/19.77.34.5 Safari/537.1", 14 "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/536.5 (KHTML, like Gecko) Chrome/19.0.1084.9 Safari/536.5", 15 "Mozilla/5.0 (Windows NT 6.0) AppleWebKit/536.5 (KHTML, like Gecko) Chrome/19.0.1084.36 Safari/536.5", 16 "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1063.0 Safari/536.3", 17 "Mozilla/5.0 (Windows NT 5.1) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1063.0 Safari/536.3", 18 "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_0) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1063.0 Safari/536.3", 19 "Mozilla/5.0 (Windows NT 6.2) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1062.0 Safari/536.3", 20 "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1062.0 Safari/536.3", 21 "Mozilla/5.0 (Windows NT 6.2) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1061.1 Safari/536.3", 22 "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1061.1 Safari/536.3", 23 "Mozilla/5.0 (Windows NT 6.1) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1061.1 Safari/536.3", 24 "Mozilla/5.0 (Windows NT 6.2) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1061.0 Safari/536.3", 25 "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.24 (KHTML, like Gecko) Chrome/19.0.1055.1 Safari/535.24", 26 "Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/535.24 (KHTML, like Gecko) Chrome/19.0.1055.1 Safari/535.24" 27 ] 28 count=0 29 30 def sina(page_url): ##列表页 31 if semaphore.acquire(): 32 header={} 33 34 header[‘User-Agent‘]=random.choice(user_agent_list) 35 header.update({ 36 "Host":"platform.sina.com.cn", 37 38 #"Cookie":"global_cookie=fb1g6d0w64d2cmu86sv4g9n3va0j137sk48; vh_newhouse=3_1491312022_2816%5B%3A%7C%40%7C%3A%5D833300ee3177d88529c7aa418942ece9; newhouse_user_guid=2F163DE7-8201-7FA9-2FB6-E507FE6F03B1; SoufunSessionID_Esf=3_1495389730_232; sf_source=; s=; showAdsh=1; hlist_xfadhq_SZ=0%7c2017%2f5%2f25+1%3a21%3a47%7c; city=sz; __utmt_t0=1; __utmt_t1=1; __utmt_t2=1; logGuid=a768dd46-b85b-47f4-a7a0-0a6596cab4cd; __utma=147393320.1111837171.1491290389.1495646208.1495650134.9; __utmb=147393320.12.10.1495650134; __utmc=147393320; __utmz=147393320.1495650134.9.4.utmcsr=esf.sz.fang.com|utmccn=(referral)|utmcmd=referral|utmcct=/; unique_cookie=U_cqyov4ut5vv1al8e2858qhzgt17j2z06mph*14" 39 }) 40 while(1): 41 content=‘‘ 42 try: 43 content=requests.get(page_url,headers=header,timeout=5).content 44 45 except Exception as e: 46 print e 47 if content!=‘‘: 48 break 49 50 51 52 53 jsona=re.findall(‘jQuery191012358189839869738_1495880348059\(([\s\S]*?"}]}})‘,content)[0] 54 #print jsona 55 dict= json.loads(jsona) 56 #print type(dict) 57 #print dict 58 #print dict[‘result‘][‘data‘] 59 for l in dict[‘result‘][‘data‘]: 60 title= l[‘title‘] 61 url= l[‘url‘] 62 biaoqian=get_biaoqian(url) 63 64 lock.acquire() 65 global count 66 count+=1 67 print time.strftime(‘%H:%M:%S‘,time.localtime(time.time())),‘ ‘,count 68 print ‘列表页:‘ 69 70 print ‘ title: %s\n url: %s‘%(title,url) 71 72 print ‘详情页:‘ 73 print ‘ biaoqian: %s \n‘%(biaoqian) 74 print ‘**************************************************************‘ 75 lock.release() 76 77 semaphore.release() 78 79 80 81 def get_biaoqian(url): ###新闻页,爬取标签 82 83 header={‘User-Agent‘:random.choice(user_agent_list)} 84 header.update({"Host":"mil.news.sina.com.cn"}) 85 86 while(1): 87 content=‘‘ 88 try: 89 content=requests.get(url,headers=header,timeout=10).content 90 except Exception as e: 91 #print e 92 pass 93 if content!=‘‘: 94 break 95 96 97 se=etree.HTML(content) 98 #print etree.tounicode(se) 99 biaoqian=se.xpath(‘//p[@class="art_keywords"]/a/text()‘) 100 return ‘ ‘.join(biaoqian) 101 102 103 104 105 def singe_req(): 106 for i in range(1,301): 107 page_url=‘http://platform.sina.com.cn/news/news_list?app_key=2872801998&channel=mil&cat_1=lishi&show_all=0&show_cat=1&show_ext=1&tag=1&format=json&page=%s&show_num=10&callback=jQuery191012358189839869738_1495880348059&_=1495880348069‘%i 108 sina(page_url) 109 print ‘over‘ 110 111 def threading_red(): 112 threads=[] 113 for i in range(1,301): 114 t=threading.Thread(target=sina,args=(‘http://platform.sina.com.cn/news/news_list?app_key=2872801998&channel=mil&cat_1=lishi&show_all=0&show_cat=1&show_ext=1&tag=1&format=json&page=%s&show_num=10&callback=jQuery191012358189839869738_1495880348059&_=1495880348069‘%i,)) 115 threads.append(t) 116 t.start() 117 for t in threads: 118 t.join() 119 print ‘over‘ 120 121 def muiltiprocessing_req(): 122 import multiprocessing 123 pool = multiprocessing.Pool(100) 124 #pool = multiprocessing.Pool(multiprocessing.cpu_count()) 125 126 pool.map(sina, [‘http://platform.sina.com.cn/news/news_list?app_key=2872801998&channel=mil&cat_1=lishi&show_all=0&show_cat=1&show_ext=1&tag=1&format=json&page=%s&show_num=10&callback=jQuery191012358189839869738_1495880348059&_=1495880348069‘%i for i in range(1,301)]) 127 pool.close() 128 pool.join() 129 print ‘over‘ 130 131 def gevent_req(): 132 ######################利用pool###################### 133 from gevent import monkey 134 from gevent.pool import Pool 135 136 monkey.patch_all() 137 pool = Pool(100) 138 data= pool.map(sina, [‘http://platform.sina.com.cn/news/news_list?app_key=2872801998&channel=mil&cat_1=lishi&show_all=0&show_cat=1&show_ext=1&tag=1&format=json&page=%s&show_num=10&callback=jQuery191012358189839869738_1495880348059&_=1495880348069‘%i for i in range(1,301)]) 139 print ‘over‘ 140 141 if __name__==‘__main__‘: 142 pass 143 singe_req() ##单线程 144 #threading_red() ###多线程 145 #muiltiprocessing_req() ####多进程146 #gevent_req() ##协程
这篇主要是用四种方法来实现爬虫。无论是100线程还是100进程或者100协程,网速都撑满了,爬取速度很快,单线程对网速利用很不充分,当然就爬取缓慢。
特别是我之前在面试房极客时候,那主管告诉我,他说他看了网上说python多线程是假的,所以他从来就没使用过多线程,只用多进程,他认为多线程不能加快爬虫速度。
关于这一点我是非常确定python多线程能加快爬取速度的,因为我使用多线程的时间很长,那主管应该只看了一半,python对cpu密集型速度提升不了多少,但对于io密集型的速度提升是立竿见影的,特别是对timeout比较大的网站,多线程爬取优势非常明显,因为爬虫是打开页面,请求服务器后端,服务器后端操作数据库查询数据,数据库返回给后端返回给前段,这种属于io密集型,多线程在爬虫和性能测试都是可以的。而多进程实在是开销太大了,开100进程,任务管理器可以看到100个python.exe,每个占用20M内存,多进程启动时候占用cpu极高。爬虫是非常适合多线程的,或者利用协程也可以。
发下运行结果:
时间: 2024-10-22 02:10:46