因为工作原因,需要爬取相关网站的数据做统计。包括中基协网站和天眼查部分数据。
一、中基协网站
爬取思路:
1.查看目标页:http://gs.amac.org.cn/amac-infodisc/api/pof/manager?rand=0.9775162173180119&page=%s&size=50
发现有随机数字串(刷新反爬措施),以及页码和每页信息条数,可以用来拼接爬取url
用一个循环爬取所有展示页面,用到requests库请求访问页面以及random函数生成随机数
返回的是json数据,直接用request的json函数解析数据。
2.save函数用来保存目标页面的详细数据,可根据需要爬取。
1 import requests 2 import random 3 import json 4 5 def save(school_datas): 6 for data1 in school_datas: 7 # print(data) 8 id = data1[\'id\'] 9 managerName = data1[\'managerName\'] 10 artificialPersonName = data1[\'artificialPersonName\'] 11 regAdrAgg = data1[\'regAdrAgg\'] 12 registerNo = data1[\'registerNo\'] 13 print(id, managerName, artificialPersonName, regAdrAgg,registerNo) 14 15 for i in range(0, 427): 16 print("第%s页====================="%str(i)) 17 header={ 18 \'Accept\':\'application/json, text/javascript, */*; q=0.01\', 19 \'Accept-Encoding\':\'gzip, deflate\', 20 \'Connection\':\'keep-alive\', 21 \'Host\':\'gs.amac.org.cn\', 22 \'Origin\':\'http://gs.amac.org.cn\', 23 \'Referer\':\'http://gs.amac.org.cn/amac-infodisc/res/pof/manager/managerList.html\', 24 \'User-Agent\':\'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/61.0.3163.100 Safari/537.36\' 25 } 26 r=random.random() 27 print(str(r)) 28 30 # json={"rand":\'0.0045470034372876444\',"page":str(i),"size":"50"} 31 # http://gs.amac.org.cn/amac-infodisc/api/pof/manager?rand=0.9775162173180119&page=1&size=50 32 # data= requests.post("http://gs.amac.org.cn/amac-infodisc/api/pof/manager",json={\'rand\':str(r),\'page\':str(i),\'size\':\'50\'},headers=header)#.json() 33 url="http://gs.amac.org.cn/amac-infodisc/api/pof/manager?rand=0.9775162173180119&page=%s&size=50" 34 data= requests.post(url%i,json={\'rand\':str(r),\'page\':str(i),\'size\':\'50\'}).json() 35 40 41 # print (type(r)) 42 # print (r.status_code) 43 45 # print (r.cookies) 46 # print(r.text,"\n") 47 # print(r.json()) 48 55 56 print("每一页信息条数——>", len(data[\'content\'])) 57 print("全部信息条数——>", data["totalElements"]) 58 print("每页有——>", data["size"]) 59 print("总页数-->>", data["totalPages"]) 60 61 school_datas = data["content"] 62 save(school_datas)
二、天眼查
爬取思路:
1,首先考虑将协会提供的已经登记的广州所有基金或投资公司名单一个个读入到天眼查首页搜索框内(大概7000个公司):(page=requests.get(url1))
2,这样会得到一个返回的搜索结果页面(tree=html.fromstring(page.text)),此页面必定是按名称匹配最精确的排序,因此可以直接考虑第一条信息(名单上的公司名称)
3,通过解析第一条信息(href1=tree.xpath("//*[@id=\'web-content\']/div/div/div/div[1]/div[3]/div/div[2]/div[1]/a/@href")),就可得到它转向的下级静态页面地址,爬取此页面数据即可得到需要的每一个具体公司详细资料!(采用的是xpath读取数据)
4,将得到数据保存到excel表格里面。(如果搜索首页爬取被禁止,可以考虑移动端!)
注意:天眼查反扒措施做的很严格,频繁爬取就会触发反爬取规则,需要用到time函数降低爬取速度,但如果考虑时间成本,可以采用代理池方法爬取此网站的数据!
1 import xlrd 2 import requests 3 from lxml import html 4 import xlwt 5 import time 6 # # import os 7 # # print(os.getcwd()) 8 #打开源工作簿 9 data = xlrd.open_workbook(r\'C:\Users\lin\Desktop\wx\j.xlsx\')#字符串中\是被当作转义字符来使用,所以’d:\a.txt’会被转义成’d:\a.txt’这是正确路径,所以要双\\或者加r或者反/ 10 table = data.sheet_by_index(0) 11 #创建目标工作簿 12 book = xlwt.Workbook() 13 sheet1 = book.add_sheet(\'tyc_data\') 14 print(table.nrows) 15 # headers={ 16 # #\'Referer\':\'http://www.qichacha.com/search?key=%E5%B9%BF%E5%B7%9E%E5%B8%82%E9%BC%8E%E9%94%8B%E6%8E%A7%E8%82%A1%E6%9C%89%E9%99%90%E5%85%AC%E5%8F%B8\', 17 # \'User-Agent\':\'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/62.0.3202.94 Safari/537.36\', 18 # # \'Accept-Encoding\':\'gzip, deflate, br\', 19 # # \'Connection\':\'keep-alive\', 20 # # \'Cookie\':\'TYCID=b1367520cc2211e7ac3c4d1875dff195; undefined=b1367520cc2211e7ac3c4d1875dff195; ssuid=31903499; tyc-user-info=%257B%2522token%2522%253A%2522eyJhbGciOiJIUzUxMiJ9.eyJzdWIiOiIxNTMyMjMyNDc3OSIsImlhdCI6MTUxMDk4ODc4NSwiZXhwIjoxNTI2NTQwNzg1fQ.N7mPw74UB9NRY7rganfmUWM8DC_o85LgquR5qR-R8vV1YkS40wQLOsuHiZL8xOalpZnfIxTqb5wAYhAnpom8hw%2522%252C%2522integrity%2522%253A%25220%2525%2522%252C%2522state%2522%253A%25220%2522%252C%2522vnum%2522%253A%25220%2522%252C%2522onum%2522%253A%25220%2522%252C%2522mobile%2522%253A%252215322324779%2522%257D; auth_token=eyJhbGciOiJIUzUxMiJ9.eyJzdWIiOiIxNTMyMjMyNDc3OSIsImlhdCI6MTUxMDk4ODc4NSwiZXhwIjoxNTI2NTQwNzg1fQ.N7mPw74UB9NRY7rganfmUWM8DC_o85LgquR5qR-R8vV1YkS40wQLOsuHiZL8xOalpZnfIxTqb5wAYhAnpom8hw; RTYCID=a6e769163818420593243d0a51e56e0b; _csrf=RcxvwHNl4qydC+/W33R21g==; OA=vYm5oBQzSUVS6+DX2qis7FlZX+DsbDSKXygIqyxLjVH0n5HhrWY+oxVPIj9Pf6PZrP+HWwgSzGdxLkTHf9m+UJwWuc+PRPq+F6ASd2BjSMsLu772U8GZ7mcDQ+/byYHOa59bn0Qkib3TDKxviYpbPUvrKGpaeTgJqKznbPOVkosZd5j6Dr0dA8SIGjCWA6R6z4gchCC8oldSjK+QdZysAolmEJ+HyfaGBteSo65AWV0=; _csrf_bk=463462e0ebb3936aa444b2fec7cfdcf2; Hm_lvt_e92c8d65d92d534b0fc290df538b4758=1510983505,1510988772; Hm_lpvt_e92c8d65d92d534b0fc290df538b4758=1511107547\' 21 # } 22 for i in range(0,1264):#1224,1264 23 24 t1=table.cell(i, 1).value#第3行第2列 25 print("取单元格:",t1) 26 27 # c1=table.col_values(1) 28 # print(c1) 29 # url="http://www.qichacha.com/search?key=%s"%t1 30 # print(url) 31 # //*[@id="searchlist"]/table/tbody/tr/td[2]/a 32 # //*[@id="searchlist"]/table/tbody/tr/td[2]/a 33 34 35 36 37 url1=\'https://www.tianyancha.com/search?key=%s\'%t1 38 39 print("拼接后地址:",url1) 40 time.sleep(5) 41 page=requests.get(url1) 42 # print(headers) 43 print("拼接后访问:",page) 44 45 tree=html.fromstring(page.text) 46 47 print(tree) 48 #//*[@id="web-content"]/div/div/div/div[1]/div[3]/div/div[2]/div[1]/a #href地址 49 50 href1=tree.xpath("//*[@id=\'web-content\']/div/div/div/div[1]/div[3]/div/div[2]/div[1]/a/@href") 51 print(type(href1),len(href1)) 52 print(href1) 53 if len(href1)==0: 54 all_data=(1,1,1,1,1,1,1,1) 55 else: 56 # url2=\'https://m.tianyancha.com%s\'%href1[0] 57 url2=href1[0] 58 59 print("取搜索到的链接:",url2) 60 page2=requests.get(url2) 61 print(page2) 62 ####开始取数据 63 print("开始取目标数据") 64 # # zcd=tree.xpath("//table/tbody/tr/td[2]/a[@class=\'ma_h1\']") 65 # # # //*[@id="web-content"]/div/div/div/div[1]/div[3]/div[1]/div[2]/div[1]/a 66 tree2=html.fromstring(page2.text) 67 all_data=[] 68 69 70 all_data.append(tree2.xpath("//*[@id=\'company_web_top\']/div[2]/div[2]/div[1]/span[1]/text()")[0]) 71 all_data.append(tree2.xpath("//*[@id=\'_container_baseInfo\']/div/div[2]/table/tbody/tr[5]/td[2]/text()")[0]) 72 all_data.append(tree2.xpath("//*[@id=\'_container_baseInfo\']/div/div[1]/table/tbody/tr/td[2]/div[2]/div[2]/div/text()")[0]) 73 all_data.append(tree2.xpath("//*[@id=\'_container_baseInfo\']/div/div[1]/table/tbody/tr/td[2]/div[1]/div[2]/div/text()")[0]) 74 all_data.append(tree2.xpath("//*[@id=\'_container_baseInfo\']/div/div[2]/table/tbody/tr[4]/td[4]/text()")[0]) 75 all_data.append(tree2.xpath("//*[@id=\'_container_baseInfo\']/div/div[2]/table/tbody/tr[3]/td[4]/text()")[0]) 76 all_data.append(tree2.xpath("//*[@id=\'web-content\']/div/div/div[2]/div/div[2]/div/div[2]/div[3]/div/div[2]/div[1]/div[1]/span[1]/text()")) 77 all_data.append(tree2.xpath("//*[@id=\'web-content\']/div/div/div[2]/div/div[2]/div/div[2]/div[3]/div/div[2]/div[2]/div/span[1]/text()")) 78 79 print(all_data) 80 81 # # # 打印搜索页面所有链接地址 82 # djjg_page=tree.xpath("//*[@id=\'_container_baseInfo\']/div/div[2]/table/tbody/tr[5]/td[2]/text()") 83 # print(\'企业名称:\',all_data[0],\'登记机关:\',all_data[0],"\n注册时间:",all_data[1],"\n注册资本:",all_data[2],"\n核准日期:",all_data[3], \ 84 # "\n行业:",all_data[4],"\n自身风险:",all_data[5],"\周边风险:",all_data[6]) 85 86 # workbook = xlwt.Workbook(encoding = \'ascii\') 87 88 # for i2 in range(3): 89 # print(i2,j2,item,"\n") 90 91 # print(item,"\n") 92 for j2,item in zip(range(8),all_data): 93 print(i,j2,item) 94 sheet1.write(i,j2,item) 95 96 book.save(\'C:\\Users\\lin\\Desktop\\wx\\j2.xlsx\') 97 print(\'the excel save success:%s\'%i)
由于各种网站一直在变化,可能爬取策略一直需要变动 ,具体问题具体分析。