pyppeteer从基本使用到启发式爬虫

> pyppeteer 没有api手册,各种api的使用参考puppeteer手册 https://zhaoqize.github.io/puppeteer-api-zh_CN/

基本使用

尝试开始

  以简单的方式开始

 1 import asyncio
 2 from pyppeteer import launch
 3 
 4 async def main(url):
 5     browser = await launch()
 6     page = await  browser.newPage()
 7     res = await page.goto(url, options={'timeout': 30000})
 8 
 9     data = await page.content()
10     title = await page.title()
11     resp_cookies = await page.cookies()  # cookie
12     resp_headers = res.headers  # 响应头
13     resp_status = res.status  # 响应状态
14 
15     print(data)
16     print(title)
17     print(resp_headers)
18     print(resp_status)
19 
20 
21 if __name__ == '__main__':
22     url = "http://demo.aisec.cn/demo/aisec/"
23     asyncio.get_event_loop().run_until_complete(main(url))

launch选项

  启动 Chromium 实例的方法 launch

browser = await pyppeteer.launch({
    'headless': False,            # 关闭无头模式
    'devtools': True,             # 控制界面的显示,用来调试
    'executablePath': '你下载的Chromium.app/Contents/MacOS/Chromiu',
    'args': [ 
        '--disable-extensions',
        '--hide-scrollbars',
        '--disable-bundled-ppapi-flash',
        '--mute-audio',
        '--no-sandbox',           # --no-sandbox 在 docker 里使用时需要加入的参数,不然会报错
        '--disable-setuid-sandbox',
        '--disable-gpu',
        '--disable-xss-auditor',
    ],
    'dumpio': True,               # 解决浏览器多开卡死
})

截屏

1 async def main(url):
2     browser = await launch()
3     page = await  browser.newPage()
4     await page.setViewport(viewport={'width': 1280, 'height': 1800})
5     res = await page.goto(url, waitUntil=["networkidle0", "load", "domcontentloaded"],options={'timeout': 0})
6     await page.screenshot({'path': 'example.png'})
7 
8 url = "https://zhaoqize.github.io/puppeteer-api-zh_CN/#?product=Puppeteer&version=v2.0.0&show=api-puppeteerlaunchoptions"
9 asyncio.get_event_loop().run_until_complete(main(url))

基本使用

  pyppeteer的一些基本使用,包括网络请求,返回cookie、title、html、响应头、执行js、获取html元素等用法。

 1 async def main(url):
 2     browser = await launch()
 3     page = await  browser.newPage()
 4 
 5     await page.setUserAgent('Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.67 Safari/537.36')
 6     
 7     # 是否启用JS,enabled设为False,则无渲染效果
 8     await page.setJavaScriptEnabled(enabled=True)
 9 
10     # 注入 js 文件
11     CURDIR = Path(__file__).parent
12     JS_AJAX_HOOK_LIB = str(CURDIR / 'static' / 'ajaxhook.min.js')
13     await page.addScriptTag(path=JS_AJAX_HOOK_LIB)
14 
15     res = await page.goto(url, waitUntil=["networkidle0", "load", "domcontentloaded"],options={'timeout': 0})
16 
17     # cookie
18     cookies = await page.cookies()
19 
20     # 网站 title
21     title = await page.title()
22 
23     # html 内容
24     contents = await page.content()
25 
26     # 响应头
27     res_headers = res.headers
28 
29     # 响应状态
30     res_status = res.status
31 
32     # 等待
33     await asyncio.sleep(2)
34 
35     # 滚动到页面底部
36     await page.evaluate('window.scrollBy(0, document.body.scrollHeight)')
37 
38     # 在网页上执行js 脚本
39     dimensions = await page.evaluate(pageFunction='''() => {
40             return {
41                  document.documentElement.clientWidth,  // 页面宽度
42                 height: document.documentElement.clientHeight,  // 页面高度
43                 deviceScaleFactor: window.devicePixelRatio,  // 像素比 1.0000000149011612
44             }
45         }''', force_expr=False)  # force_expr=False  执行的是函数  True 则执行表达式
46 
47     """
48     抓取内容
49 
50     Page.querySelector()  # 选择器
51     Page.querySelectorAll()
52     Page.xpath()          # xpath 表达式
53 
54     Page.J(), Page.JJ(), and Page.Jx()   # 简写
55     """
56     element = await page.querySelector(".feed-infinite-wrapper > ul>li")  # 只抓取一个
57 
58     # 获取所有文本内容  执行 js
59     content = await page.evaluate('(element) => element.textContent', element)
60 
61     elements = await page.xpath('//div[@class="title-box"]/a')
62 
63     elements = await page.querySelectorAll(".title-box a")
64     for item in elements:
65         print(await item.getProperty('textContent'))
66         # <pyppeteer.execution_context.JSHandle object at 0x000002220E7FE518>
67 
68         # 获取文本
69         title_str = await (await item.getProperty('textContent')).jsonValue()
70 
71         # 获取链接
72         title_link = await (await item.getProperty('href')).jsonValue()
73 
74 
75     # 键盘输入
76     await page.type('#identifierId', username)
77 
78     # 点击下一步
79     await page.click('#identifierNext > content')
80     page.mouse  # 模拟真实点击
81 
82     
83     await browser.close()

请求多个url

 1 import asyncio
 2 import pyppeteer
 3 from collections import namedtuple
 4 
 5 Response = namedtuple("rs", "title url html cookies headers history status")
 6 
 7 
 8 async def get_html(url):
 9     browser = await pyppeteer.launch(headless=True, args=['--no-sandbox'])
10     page = await  browser.newPage()
11     res = await page.goto(url, options={'timeout': 30000})
12     data = await page.content()
13     title = await page.title()
14     resp_cookies = await page.cookies()  # cookie
15     resp_headers = res.headers  # 响应头
16     resp_status = res.status  # 响应状态
17     print(data)
18     print(title)
19     print(resp_headers)
20     print(resp_status)
21     return title
22 
23 
24 if __name__ == '__main__':
25     url_list = ["https://www.toutiao.com/",
26                 "http://jandan.net/ooxx/page-8#comments",
27                 "https://www.12306.cn/index/"
28                ]
29     task = [get_html(url) for url in url_list]
30 
31     loop = asyncio.get_event_loop()
32     results = loop.run_until_complete(asyncio.gather(*task))
33     for res in results:
34         print(res)

模拟输入

1 # 模拟输入 账号密码  {'delay': rand_int()} 为输入时间
2     await page.type('#TPL_username_1', "sadfasdfasdf")
3     await page.type('#TPL_password_1', "123456789", )
4     
5     await page.waitFor(1000)
6     await page.click("#J_SubmitStatic")

爬取京东商城

  一个使用案例,出自别人的博客。

 1 import requests
 2 from bs4 import BeautifulSoup
 3 from pyppeteer import launch
 4 import asyncio
 5 
 6 
 7 def screen_size():
 8     """使用tkinter获取屏幕大小"""
 9     import tkinter
10     tk = tkinter.Tk()
11     width = tk.winfo_screenwidth()
12     height = tk.winfo_screenheight()
13     tk.quit()
14     return width, height
15 
16 async def main(url):
17     browser = await launch({'args': ['--no-sandbox'], })
18     page = await browser.newPage()
19     width, height = screen_size()
20     await page.setViewport(viewport={"width": width, "height": height})
21     await page.setJavaScriptEnabled(enabled=True)
22     await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 Edge/16.16299')
23     await page.goto(url)
24     await page.evaluate('window.scrollBy(0, document.body.scrollHeight)')
25 
26     await asyncio.sleep(1)
27 
28     # content = await page.content()
29     li_list = await page.xpath('//*[@id="J_goodsList"]/ul/li')
30 
31     # print(li_list)
32     item_list = []
33     for li in li_list:
34         a = await li.xpath('.//div[@class="p-img"]/a')
35         detail_url = await (await a[0].getProperty("href")).jsonValue()
36         promo_words = await (await a[0].getProperty("title")).jsonValue()
37         a_ = await li.xpath('.//div[@class="p-commit"]/strong/a')
38         p_commit = await (await a_[0].getProperty("textContent")).jsonValue()
39         i = await li.xpath('./div/div[3]/strong/i')
40         price = await (await i[0].getProperty("textContent")).jsonValue()
41         em = await li.xpath('./div/div[4]/a/em')
42         title = await (await em[0].getProperty("textContent")).jsonValue()
43         item = {
44             "title": title,
45             "detail_url": detail_url,
46             "promo_words": promo_words,
47             'p_commit': p_commit,
48             'price': price
49         }
50         item_list.append(item)
51         # print(item)
52         # break
53     # print(content)
54 
55     await page_close(browser)
56     return item_list
57 
58 
59 async def page_close(browser):
60     for _page in await browser.pages():
61         await _page.close()
62     await browser.close()
63 
64 
65 msg = "手机"
66 url = "https://search.jd.com/Search?keyword={}&enc=utf-8&qrst=1&rt=1&stop=1&vt=2&wq={}&cid2=653&cid3=655&page={}"
67 
68 task_list = []
69 for i in range(1, 6):
70     page = i * 2 - 1
71     url = url.format(msg, msg, page)
72     task_list.append(main(url))
73 
74 loop = asyncio.get_event_loop()
75 results = loop.run_until_complete(asyncio.gather(*task_list))
76 # print(results, len(results))
77 for i in results:
78     print(i, len(i))
79 
80 # soup = BeautifulSoup(content, 'lxml')
81 # div = soup.find('div', id='J_goodsList')
82 # for i, li in enumerate(div.find_all('li', class_='gl-item')):
83 #     if li.select('.p-img a'):
84 #         print(li.select('.p-img a')[0]['href'], i)
85 #         print(li.select('.p-price i')[0].get_text(), i)
86 #         print(li.select('.p-name em')[0].text, i)
87 #     else:
88 #         print("#" * 200)
89 #         print(li)

xss验证

 1 import asyncio 
 2 from pyppeteer import launch 
 3 payload = '<script>xianzhi(3.1415926535)</script>' 
 4 url = "http://192.168.81.132/vuls/sqli/?id=1'>{payload}".format(payload=payload) 
 5 def xss_auditor(url, message): 
 6     if message == 3.1415926535:         
 7         print('xss_auditor_found:', payload)         
 8         print(url) 
 9         
10 async def main(): 
11     browser = await launch(headless=False, args=['--disable-xss-auditor']) 
12     page = await browser.newPage() 
13     await page.exposeFunction( 
14         'xianzhi', lambda message: xss_auditor(url, message) 
15     ) 
16     await page.goto(url) 
17     await page.close() 
18     
19 asyncio.get_event_loop().run_until_complete(main())

启发式爬虫

  启发式爬虫要实现捕获ajax请求、遍历表单、触发事件。

捕获 ajax 链接

  启用请求拦截器await page.setRequestInterception(True),会激活 request.abort,request.continue 和 request.respond 方法,每个请求都将停止,这样可以拦截ajax请求。
  page.on('request', get_ajax)遇到request请求时触发。

 1 import asyncio
 2 from pyppeteer import launch
 3 
 4 async def get_ajax(req):
 5 
 6     res = {"method":req.method,"url":req.url,"data": "" if req.postData == None else req.postData}
 7     print(res)
 8     await req.continue_()
 9 
10 async def main(url):
11     browser = await launch()
12     page = await  browser.newPage()
13 
14     await page.setRequestInterception(True)
15     page.on('request', get_ajax)
16 
17     res = await page.goto(url)

绕过webdriver检测

 1 async def page_evaluate(self, page):
 2     '''window.navigator.webdriver=false'''
 3 
 4     await page.evaluate('''() =>{ Object.defineProperties(navigator,{ webdriver:{ get: () => undefined } }) }''')  # 以下为插入中间js,将淘宝会为了检测浏览器而调用的js修改其结果。
 5     await page.evaluate('''() =>{ window.navigator.chrome = { runtime: {},  }; }''')
 6     await page.evaluate('''() =>{ Object.defineProperty(navigator, 'languages', { get: () => ['en-US', 'en'] }); }''')
 7     await page.evaluate('''() =>{ Object.defineProperty(navigator, 'plugins', { get: () => [1, 2, 3, 4, 5,6], }); }''')
 8 
 9 async def main(self):
10     browser = await launch()
11     page = await browser.newPage()
12 
13     await self.page_evaluate(page)

处理弹框

  更多处理方法可以在puppeteer手册中查找

1 async def dialog_control(self, dialog):
2     await dialog.accept()
3 
4 async def main(self):
5         browser = await launch()
6         page = await browser.newPage()
7         page.on('dialog', self.dialog_control)

  用匿名函数表达

1 page.on('dialog', lambda dialog: dialog.accept())

过滤掉img、css等请求

 1 async def goto(page, url):
 2     '''请求加载是否完成,无网都需要处理
 3     '''
 4     while True:
 5         try:
 6             await page.goto(url, {
 7                 'timeout': 0,
 8                 'waitUntil': 'networkidle0'
 9             })
10             break
11         except (pyppeteer.errors.NetworkError,
12                 pyppeteer.errors.PageError) as ex:
13             # 无网络 'net::ERR_INTERNET_DISCONNECTED','net::ERR_TUNNEL_CONNECTION_FAILED'
14             if 'net::' in str(ex):
15                 await asyncio.sleep(10)
16             else:
17                 raise
18 
19 async def request_check(self, req):
20         '''filter requests
21         Disable image loading
22         '''
23         if req.resourceType in ["image", "media", "eventsource", "websocket", "stylesheet", "font"]:
24             await req.abort()
25         elif "logout" in req.url or "delete" in req.url or "signout" in req.url:
26             await req.abort()
27         else:
28             await req.continue_()
29 
30 async def main(self):
31         browser = await launch()
32         page = await browser.newPage()
33         await page.setRequestInterception(True)
34         page.on('request', self.request_check)

获取表单

  通过运行js获取表单,先在浏览器中测试通过后再转入脚本运行。

 1 async def get_urls(page):
 2     res = await page.evaluate('''() => {
 3         var urls = [];
 4         //get form
 5         function GetForm() {
 6             var f = document.forms;
 7             for (var i = 0; i < f.length; i++) {
 8                 url = f[i].action;
 9                 //input 
10                 var inputs = f[i].getElementsByTagName('*');
11                 var requestdata = "";
12                 var len = inputs.length;
13                 
14                 for (var j = 0; j < len; j++) {
15                     if(inputs[j].hasAttributes("*")== true){    
16                         if (j < len - 1) {
17                             if(inputs[j].hasAttributes("name") && inputs[j].name !=undefined && inputs[j].name !=""){
18                                 requestdata = requestdata + inputs[j].name 
19                             }  
20                             else{
21                                 continue
22                             }
23                             if(inputs[j].hasAttributes("value") && inputs[j].value !="" && inputs[j].value !=undefined){
24                                 requestdata = requestdata + "=" + inputs[j].value + "&";
25                             }
26                             else{
27                                 requestdata = requestdata + "=123123&";
28                             } 
29                         }
30                         if (j == len - 1) {
31                             if(inputs[j].hasAttributes("name") && inputs[j].name !=undefined && inputs[j].name !=""){
32                                    requestdata = requestdata + inputs[j].name 
33                             }
34                             else{
35                                 continue
36                             }
37                             if(inputs[j].hasAttributes("value") && inputs[j].value !="" && inputs[j].value !=undefined){
38                                 requestdata = requestdata + "=" + inputs[j].value ;
39                             }
40                             else{
41                                 requestdata = requestdata + "=123123";
42                             }
43                         }
44                     }
45                 }  
46                 res = {"method": "POST", "url": url, "data": requestdata};
47                 if  (urls.indexOf(res) < 0) {
48                     urls.push(res)
49                 }
50             }
51         }
52         GetForm()
53         return urls;
54     }''') 
55     return res

获取 href

  还是现在浏览器中测试成功后再加入脚本

 1 () => {
 2         var urls = [];
 3         //get href
 4          function getallurl() {
 5             tag_dict = {'a': 'href','link': 'href','area': 'href','img': 'src','embed': 'src','video': 'src','audio': 'src'}
 6             for(var tag in tag_dict){
 7                 var src = tag_dict[tag];
 8                 var elements = document.getElementsByTagName(tag);
 9                 for (var i = 0; i < elements.length; i++) {
10                     res = {"method":"GET", "url": elements[i][src], "data": ""};
11                     if (urls.indexOf(res) < 0 && elements[i][src].indexOf("javascript:") < 0 && elements[i][src].indexOf("mailto:") < 0) {
12                         urls.push(res);
13                     }
14                 }
15             }
16         }
17 
18         getallurl();
19         return urls;
20     }

  更多js调用参考https://github.com/wilson9x1/crawler_phantomjs

捕获onevent

 1 function getonevents() {
 2     // eval all on event
 3     var nodes = document.all;
 4     for(var i=0;i<nodes.length;i++){
 5         var attrs = nodes[i].attributes;
 6         for(var j=0;j<attrs.length;j++){
 7             attr_name = attrs[j].nodeName;
 8             attr_value = attrs[j].nodeValue.replace(/return.*;/g,'');
 9             if(attr_name.substr(0,2) == "on"){
10                 if (onevents.indexOf(attr_value) < 0) {
11                     onevents.push(attr_value);
12                 }
13             }
14             if(attr_name == "href"){
15                 javascript_code = attr_value.match("javascript:(.*)")
16                if (javascript_code) {
17                     if (onevents.indexOf(attr_value) < 0) {
18                         onevents.push(attr_value);
19                     }
20                }
21             }                           
22         }
23     }
24 }

  最终运行效果:

  1 #!/usr/bin/env python3
  2 import asyncio
  3 import hashlib
  4 from pyppeteer import launch
  5 from collections import OrderedDict
  6 
  7 '''
  8 pyppeteer/connection.py
  9 self._ws = websockets.client.connect(
 10     # self._url, max_size=None, loop=self._loop)
 11     self._url, max_size=None, loop=self._loop, ping_interval=None, ping_timeout=None)
 12 '''
 13 
 14 class DynamicCrawler(object):
 15     ''' Dynamic Crawler 
 16     crawler ajax onclick tags
 17     '''
 18     
 19     def __init__(self, target):
 20         super(DynamicCrawler, self).__init__()
 21         self.urls = []
 22         self.target = target
 23         self.timeout = 50000
 24         self.browser_setting = {
 25             'headless': False,
 26             'devtools': True,             # Console
 27             # 'executablePath': 'Chromium.app/Contents/MacOS/Chromiu',
 28             'ignoreHTTPSErrors': True,
 29             # 'slowMo':100,
 30             'args': [ 
 31                 '--disable-extensions',
 32                 '--hide-scrollbars',
 33                 '--disable-bundled-ppapi-flash',
 34                 '--mute-audio',
 35                 '--no-sandbox',
 36                 '--disable-setuid-sandbox',
 37                 '--disable-gpu',
 38                 '--disable-xss-auditor',
 39             ],
 40             'dumpio': True,
 41         }
 42         
 43         
 44     async def request_check(self, req):
 45         '''filter requests
 46         Disable image loading
 47         '''
 48 
 49         res = {"method":req.method,"url":req.url,"data": "" if req.postData == None else req.postData}
 50         res.update({"hash":self.get_hash(str(res))})
 51         self.urls.append(res)
 52         if req.resourceType in ["image", "media", "eventsource", "websocket", "stylesheet", "font"]:
 53         # if req.resourceType in ["image", "media", "websocket"]:
 54             await req.abort()
 55         elif "logout" in req.url or "delete" in req.url or "signout" in req.url:
 56             await req.abort()
 57         else:
 58             await req.continue_()
 59 
 60 
 61     async def frame_nav(self, frameTo):
 62         res = {"method":"Frame","url":frameTo.url}
 63         res.update({"hash":self.get_hash(str(res))})
 64         self.urls.append(res)
 65 
 66 
 67     async def page_evaluate(self, page):
 68         '''window.navigator.webdriver=false'''
 69 
 70         await page.evaluate('''() =>{ Object.defineProperties(navigator,{ webdriver:{ get: () => undefined } }) }''')  # 以下为插入中间js,将淘宝会为了检测浏览器而调用的js修改其结果。
 71         await page.evaluate('''() =>{ window.navigator.chrome = { runtime: {},  }; }''')
 72         await page.evaluate('''() =>{ Object.defineProperty(navigator, 'languages', { get: () => ['en-US', 'en'] }); }''')
 73         await page.evaluate('''() =>{ Object.defineProperty(navigator, 'plugins', { get: () => [1, 2, 3, 4, 5,6], }); }''')
 74 
 75 
 76     async def get_urls(self, page):
 77         res = await page.evaluate('''() => {
 78             var urls = [];
 79             var onevents = [];
 80             var onclickstrs = [];
 81             //get form
 82             function GetForm() {
 83                 var f = document.forms;
 84                 for (var i = 0; i < f.length; i++) {
 85                     url = f[i].action;
 86                     //input 
 87                     var inputs = f[i].getElementsByTagName('*');
 88                     var requestdata = "";
 89                     var len = inputs.length;
 90                     
 91                     for (var j = 0; j < len; j++) {
 92                         if(inputs[j].hasAttributes("*")== true){    
 93                             if (j < len - 1) {
 94                                 if(inputs[j].hasAttributes("name") && inputs[j].name !=undefined && inputs[j].name !=""){
 95                                     requestdata = requestdata + inputs[j].name 
 96                                 }  
 97                                 else{
 98                                     continue
 99                                 }
100                                 if(inputs[j].hasAttributes("value") && inputs[j].value !="" && inputs[j].value !=undefined){
101                                     requestdata = requestdata + "=" + inputs[j].value + "&";
102                                 }
103                                 else{
104                                     requestdata = requestdata + "=ATBvea182&";
105                                 } 
106                             }
107                             if (j == len - 1) {
108                                 if(inputs[j].hasAttributes("name") && inputs[j].name !=undefined && inputs[j].name !=""){
109                                        requestdata = requestdata + inputs[j].name 
110                                 }
111                                 else{
112                                     continue
113                                 }
114                                 if(inputs[j].hasAttributes("value") && inputs[j].value !="" && inputs[j].value !=undefined){
115                                     requestdata = requestdata + "=" + inputs[j].value ;
116                                 }
117                                 else{
118                                     requestdata = requestdata + "=ATBvea182";
119                                 }
120                             }
121                         }
122                     }  
123                     res = {"method": "POST", "url": url, "data": requestdata};
124                     if  (urls.indexOf(res) < 0) {
125                         urls.push(res)
126                     }
127                 }
128             }
129             
130             // get all href
131             function getallurl() {
132                 GetForm();
133                 tag_dict = {'a': 'href','link': 'href','area': 'href','img': 'src','embed': 'src','video': 'src','audio': 'src'}
134                 for(var tag in tag_dict){
135                     var src = tag_dict[tag];
136                     var elements = document.getElementsByTagName(tag);
137                     for (var i = 0; i < elements.length; i++) {
138                         res = {"method":"GET", "url": elements[i][src], "data": ""};
139                         if (urls.indexOf(res) < 0 && elements[i][src].indexOf("javascript:") < 0 && elements[i][src].indexOf("mailto:") < 0) {
140                             urls.push(res);
141                         }
142                     }
143                 }
144             }
145 
146             //get onevent
147             function getonevents() {
148                 // eval all on event
149                 var nodes = document.all;
150                 for(var i=0;i<nodes.length;i++){
151                     var attrs = nodes[i].attributes;
152                     for(var j=0;j<attrs.length;j++){
153                         attr_name = attrs[j].nodeName;
154                         attr_value = attrs[j].nodeValue.replace(/return.*/g,'').replace(/return.*;/g,'');
155                         if(attr_name.substr(0,2) == "on"){
156                             if (onevents.indexOf(attr_value) < 0) {
157                                 onevents.push(attr_value);
158                             }
159                         }
160                         if(attr_name == "href"){
161                             javascript_code = attr_value.match("javascript:(.*)")
162                            if (javascript_code) {
163                                 if (onevents.indexOf(attr_value) < 0) {
164                                     onevents.push(attr_value);
165                                 }
166                            }
167                         }                           
168                     }
169                 }
170             }
171 
172             function doloop(i) {
173                 getallurl();
174                 getonevents();
175                 if (onevents.length ==0) {
176                     return;
177                 }
178                 if (i == (onevents.length - 1)) {
179                     try {
180                         eval(onevents[i]);
181                     }catch(err) {
182                         return;
183                     }
184                     getallurl();
185                 }
186                 else {
187                     try {
188                         eval(onevents[i]);
189                         i = i + 1; //1
190                         doloop(i);
191                     }catch(err) {
192                         i = i + 1; //1
193                         doloop(i);
194                     } 
195                 }
196             }
197 
198             function main() {
199                 doloop(0);
200             }
201             main();
202             return urls;
203         }''') 
204         return res
205 
206 
207     def duplicate(self, datas):
208         ''' Data deduplication
209         '''
210         b = OrderedDict()
211         for item in datas:
212             b.setdefault(item['hash'],{**item,})
213         b = list(b.values())
214         return b
215 
216 
217     def get_hash(self, content):
218         md = hashlib.md5() 
219         md.update(content.encode('utf-8')) 
220         return md.hexdigest()[:5]
221 
222 
223     async def main(self):
224         browser = await launch(self.browser_setting)
225         # browser = await launch()
226         page = await browser.newPage()
227 
228         await self.page_evaluate(page)
229 
230         await page.setUserAgent('Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.67 Safari/537.36')
231 
232         await page.setJavaScriptEnabled(True)
233         # 禁止加载图片接口
234         await page.setRequestInterception(True)
235         page.on('request', self.request_check)
236 
237         page.on('framenavigated', self.frame_nav)
238         # 处理弹框
239         page.on('dialog', lambda dialog: dialog.accept())
240 
241         req = await page.goto(self.target, waitUntil=["networkidle0", "load", "domcontentloaded"], options={'timeout': self.timeout})
242         cookies = await page.cookies()
243 
244         title = await page.title()
245         # html 内容
246         contents = await page.content()
247 
248         # 响应头
249         res_headers = req.headers
250 
251         # 响应状态
252         res_status = req.status
253 
254         getUrls = await self.get_urls(page)
255         for res in getUrls:
256             res.update({"hash":self.get_hash(str(res))})
257             self.urls.append(res)
258         
259         print (title)
260         print (cookies)
261         # print (res_headers)
262         
263         for i in self.duplicate(self.urls):
264             print(i)
265 
266         await page.close()
267         await browser.close()
268 
269 
270     def run(self):
271         asyncio.get_event_loop().run_until_complete(self.main())
272 
273 
274 
275 # a = DynamicCrawler("http://demo.aisec.cn/demo/aisec/")
276 a = DynamicCrawler("http://192.168.81.132/vuls/config/setup.php")
277 
278 a.run()

在docker里使用

  在 window10 里开发很流程,部署到 windows server 上,可能由于配置比较差或其他原因,网站渲染很慢。
  可以放在容器里,效果明显。注意点是上面提到了的关闭沙盒模式,需要下一些浏览器的依赖,还有就是最好先把浏览器下好,做到镜像里,这样
就不会在容器里一个一个下了。

FROM python:slim

WORKDIR /usr/src/app

RUN apt-get update && apt-get install -y gconf-service libasound2 libatk1.0-0 libc6 libcairo2 libcups2 libdbus-1-3 libexpat1 libfontconfig1 libgcc1 libgconf-2-4 libgdk-pixbuf2.0-0 libglib2.0-0 libgtk-3-0 libnspr4 libpango-1.0-0 libpangocairo-1.0-0 libstdc++6 libx11-6 libx11-xcb1 libxcb1 libxcomposite1 libxcursor1 libxdamage1 libxext6 libxfixes3 libxi6 libxrandr2 libxrender1 libxss1 libxtst6 ca-certificates fonts-liberation libappindicator1 libnss3 lsb-release xdg-utils wget
RUN apt-get install -y vim

COPY requirements.txt ./
RUN pip install --no-cache-dir -r requirements.txt
RUN python -c "import pyppeteer;pyppeteer.chromium_downloader.download_chromium();"


COPY . .

VOLUME /data

References

[Puppeteer api手册] (https://zhaoqize.github.io/puppeteer-api-zh_CN/)
[Pyppeteer 使用笔记] (https://www.cnblogs.com/zhang-zi-yi/p/10820813.html)
[pyppeteer使用总结] (https://www.dust8.com/2018/06/03/pyppeteer使用总结/)
[pyppeteer最为核心类Page的接口方法(下)] (https://zhuanlan.zhihu.com/p/64170269)
[WEB2.0启发式爬虫实战_猪猪侠] (https://github.com/ring04h/papers/blob/master/WEB2.0启发式爬虫实战-猪猪侠-20180616.pdf)
[phantomjs爬虫] (https://github.com/wilson9x1/crawler_phantomjs)
[基于PhantomJS的动态爬虫引擎] (http://byd.dropsec.xyz/2019/05/21/基于PhantomJS的动态爬虫引擎/)
[漏扫动态爬虫实践] (https://0kee.360.cn/blog/漏扫动态爬虫实践/)
[启发式爬虫] (https://www.dazhuanlan.com/2019/11/15/5dcdb39c33fc8/)
[AISec漏洞扫描器测试平台] (http://demo.aisec.cn/demo/aisec/)

原文地址:https://www.cnblogs.com/trojan-z/p/12072211.html