几个月前。。。省略一堆剧情。。。直接请求json文件爬取企业信息未成功,在知乎提问后,得到解决,有大佬说带上全部headers和cookie是可以的,我就又去试了下,果然可以(之前自己试的时候不行,没搞清楚为什么突然可以了、、、)。但是即使sleep几秒后,爬虫还是会死掉,会浏览器再次访问,会发现需要验证才可以继续浏览。咨询了一些大佬,又查了查资料,看来是要用代理了,没有深入去研究。听说天眼查本事就是搞爬虫的。。。这里就贴下未解决验证码的半成品吧,代码写的很烂,也没有清洗数据,凑活着看看吧先。
当中尝试换UA避开验证,不过失败了。大佬们有新的办法欢迎告知。有错误欢迎指出,共同学习。
import requests
import MySQLdb
import time
import random #打开数据库链接
db = MySQLdb.connect(host="localhost", user="root", passwd="你的密码",db="test",use_unicode=True,charset="utf8")
cursor = db.cursor()
# 如果数据已经存在,使用excute()方法删除表
cursor.execute("DROP TABLE IF EXISTS tianyancha") #创建数据表 SQL语句
sql = """CREATE TABLE tianyancha(
industry VARCHAR(20),
base VARCHAR(10),
id VARCHAR(20),
name VARCHAR(50) NOT NULL,
legalPersonName VARCHAR(50),
regStatus VARCHAR(10),
score VARCHAR(10)
)
""" cursor.execute(sql) urls=["http://www.tianyancha.com/search/%E7%BB%9F%E8%AE%A1.json?&pn="+str(i) for i in range(38,51)]
UA = ["Mozilla/5.0 (Windows NT 10.0; WOW64; rv:49.0) Gecko/20100101 Firefox/49.0","Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/53.0.2785.143 Safari/537.36"]
headers={
"Accept":"application/json,text/plain, */*",
"Accept-Encoding":"gzip, deflate",
"Accept-Language":"zh-CN,zh;q=0.8,en-US;q=0.5,en;q=0.3",
"CheckError":"check",
"Cookie":"TYCID=79ecda1ebc7243bb8e0e61001fa62e45; tnet=219.217.246.3; Hm_lvt_e92c8d65d92d534b0fc290df538b4758=1478185016,1478185105; Hm_lpvt_e92c8d65d92d534b0fc290df538b4758=1478185127; RTYCID=f6052f4746504a92a9449adf8c1aad4d; aliyungf_tc=AQAAAIi4rlRU9QIAA/bZ26bXAnGDUsL8; _pk_id.1.4c4c=ff85a162bc61332e.1478185118.1.1478185128.1478185118.; _pk_ref.1.4c4c=%5B%22%22%2C%22%22%2C1478185118%2C%22http%3A%2F%2Fwww.tianyancha.com%2F%22%5D; _pk_ses.1.4c4c=*; token=70f49be8c16c4cb290fa7d05c8a60638; _utm=-24s2tr4st24-9n8d32849t38sk97hh8",
"Referer":"http://bj.tianyancha.com/search",
"Tyc-From":"normal",
"User-Agent":random.choice(UA),
"loop":"null"
} def get_data(url):
data = requests.get(url=url,headers=headers).json()
data = data["data"]
# print(data)
for i in range(len(data)):
score = data[i]["score"]
base = data[i]["base"]
regStatus = data[i]["regStatus"]
industry = data[i]["industry"]
legalPersonName = data[i]["legalPersonName"]
name = data[i]["name"]
id = data[i]["id"]
sql_save = """INSERT INTO tianyancha\
SET industry=%s,base=%s, id=%s, name=%s, legalPersonName=%s,regStatus=%s,score=%s"""
cursor.execute(sql_save,(industry,base, id, name, legalPersonName, regStatus, score))
db.commit()
print(name) for url in urls:
time.sleep(3)
try:
get_data(url)
except:
print(url)
pass #关闭数据库连接
db.close()
输出: