采集小工具，目前采集主要针对知乎文章与评论,今天刚开始弄，会不断更新完善
目前效果（测试站点 ：科技；测试连接：http://zhihu.sogou.com/include/pc/pc/topic/topic2_0.html）
1.输入框输入站点与连接
2.点击提交链接进行采集（会判断链接是否有效，文本框显示输入的站点与连接）
3.每次输入的站点与连接存到同目录下的txt文件中
4.点击采集所有采集（将txt中所有链接进行采集）
数据库显示
tkinter 代码
import tkinter
import tkinter.filedialog
import tkinter.messagebox
from tkinter.scrolledtext import ScrolledText
from threading import Thread
from yanshi.hot import *
class Cmpfile:
def __init__(self) :
self.list1=[]
self.list2=[]
self.item={}
root=tkinter.Tk()
self.root=root
self.root.title(‘知乎‘)
self.root.minsize(400,350)
self.Menu1()
self.Label1()
self.root.mainloop()
def thread_up(self,func):
t = Thread(target=func) # 此时线程是新建状态
t.setDaemon(True)
t.start() # 启动线程
def up(self):
entry1=self.entry1.get()
entry2=self.entry2.get()
self.text.insert(‘insert‘,entry1+‘-‘+entry2+‘\n‘)
tkinter.messagebox.showinfo(‘温馨提示‘, ‘开始采集‘)
self.item[entry1]=entry2
print(self.item)
self.list1.append(self.item)
a=zhihu(self.list1)
b=a.starts()
tkinter.messagebox.showinfo(‘温馨提示‘,b)
if b==‘采集结束‘:
f = open(‘./1.txt‘, ‘a‘)
f.write(entry1 + ‘:‘ + entry2 + ‘\n‘)
f.close
self.entry1.delete(0,‘end‘)
self.entry2.delete(0,‘end‘)
self.item={}
self.list1=[]
def load(self):
self.clear()
f=open(‘./1.txt‘,‘r‘)
one_list=f.readlines()
for one in one_list:
self.text.insert(‘insert‘,one+‘\n‘)
f.close()
def clear(self):
self.text.delete(0.0,‘end‘)
def all(self):
f=open(‘./1.txt‘,‘r‘)
one_list=f.readlines()
for one in one_list:
self.text.insert(‘insert‘, one)
item = {}
pattern=‘(.*?)-(.*?)\n‘
re_list=re.findall(pattern,one)
if re_list==[]:
pass
else:
item[re_list[0][0]]=re_list[0][1]
self.list2.append(item)
self.list2=[]
f.close()
#用place基础布局
def Menu1(self):
# 添加菜单
menu = tkinter.Menu(self.root)
# 添加查看子menu
lookmenu = tkinter.Menu(menu, tearoff=0, bg=‘purple‘, fg=‘white‘)
# 添加编辑子menu
menu.add_cascade(menu=lookmenu, label=‘日志‘)
# 添加帮助子menu
menu.add_cascade(menu=lookmenu, label=‘帮助‘)
#添加登录子menu
menu.add_cascade(menu=lookmenu, label=‘登录‘)
# 添加查看子menu
menu.add_cascade(menu=lookmenu, label=‘查看‘)
self.root.config(menu=menu)
def Label1(self):
label1=tkinter.Label(self.root,text = ‘sitename‘,height=1,width=1,pady=3,bd=3).place(relx=0.02,rely=0.05,relwidth=0.2)
label2=tkinter.Label(self.root,text = ‘link‘,height=1,width=1,pady=3,bd=3).place(relx=0.02,rely=0.15,relwidth=0.2)
self.entry1 = tkinter.Entry(self.root, width=40, bg=‘white‘, bd=5)
self.entry1.place(relx=0.25, rely=0.05, relwidth=0.7)
self.entry2 = tkinter.Entry(self.root, width=40, bg=‘white‘, bd=5)
self.entry2.place(relx=0.25, rely=0.15, relwidth=0.7)
button1 = tkinter.Button(self.root, text=‘提交链接‘, height=1, width=8, pady=5, bd=1,command=lambda :self.thread_up(self.up)).place(x=105, y=100)
button2 = tkinter.Button(self.root, text=‘载入文本‘, height=1, width=8, pady=5, bd=1,command=self.load).place(x=190, y=100)
button3 = tkinter.Button(self.root, text=‘采集所有‘, height=1, width=8, pady=5, bd=1,command=self.all).place(x=275, y=100)
self.text=ScrolledText(self.root,height=8,width=37,bg=‘white‘,pady=3,bd=3)
self.text.place(x=100, y=150)
button4 = tkinter.Button(self.root, text=‘清空‘, height=1, width=8, pady=5, bd=1,command=self.clear).place(x=105 ,y=280)
#实例化对象
one=Cmpfile()
采集代码另外的py文件中，运行时引用
import re
import pymysql
import time
import datetime
import requests
from lxml import etree
from multiprocessing.dummy import Pool as ThreadPool
class zhihu(object):
def __init__(self,urls):
self.url_list=urls
self.headers={
‘User-Agent‘:‘Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.108 Safari/537.36‘,
‘Connection‘:‘keep-alive‘,
‘Host‘:‘zhihu.sogou.com‘,
‘Referer‘:‘http://zhihu.sogou.com/‘,}
self.headers1={
‘User-Agent‘: ‘Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.108 Safari/537.36‘,
‘Referer‘: ‘https://www.zhihu.com/‘,
}
def ToResponse(self,res):
res.encoding=res.apparent_encoding
args=etree.HTML(res.text)
return args
def request_url(self,url):
response=requests.get(url,headers=self.headers)
response=self.ToResponse(response)
link_list = response.xpath(‘//li/p[@class="tit"]/a/@href‘)
return link_list
def connectdb(self):
print(‘连接到mysql服务器...‘)
# 打开数据库连接
db = pymysql.connect("服务器", "root", "123456", "zhihu",charset=‘utf8‘)
print(‘连接上了!‘)
return db
def get_link(self,item):
for key,value in item.items():
url=value
groupname=key
link_list=self.request_url(url)
db = self.connectdb()
cursor = db.cursor()
for link in link_list:
print(link)
response = requests.get(url=link, headers=self.headers1)
response = self.ToResponse(response)
IR_GROUPNAME = ‘问答社区‘
IR_SITENAME=‘搜狗知乎‘
IR_CHANNEL=groupname
IR_URLNAME=link
BBSNUM=‘0‘
IR_LASTTIME = datetime.datetime.now().strftime(‘%Y-%m-%d %H:%M:%S‘)
try:
IR_URLTITLE = response.xpath(‘//h1[@class="QuestionHeader-title"]/text()‘)[0]
IR_QUESTION = ‘‘.join(response.xpath("//span[@class=‘RichText‘]//text()")) if response.xpath(
"//span[@class=‘RichText‘]//text()") else ‘‘
IR_RETURN = response.xpath("//div[@class=‘List-header‘]//h4/span/text()")[0]
IR_FOLLOW = response.xpath("//div[@class=‘QuestionFollowStatus‘]//div[@class=‘NumberBoard-item‘][1]//strong/text()")[0]
IR_VIEW = response.xpath("//div[@class=‘QuestionFollowStatus‘]//div[@class=‘NumberBoard-item‘][2]//strong/text()")[
0]
print(IR_GROUPNAME, IR_URLTITLE, IR_QUESTION, IR_RETURN, IR_FOLLOW, IR_VIEW)
sql = "INSERT INTO wenda (IR_GROUPNAME,IR_SITENAME,IR_CHANNEL,IR_URLTITLE,IR_QUESTION,IR_URLNAME,IR_LASTTIME,IR_VIEW,IR_FOLLOW,IR_RETURN,BBSNUM) VALUES (‘" + IR_GROUPNAME + "‘,‘" + IR_SITENAME + "‘,‘" + IR_CHANNEL + "‘,‘" + IR_URLTITLE + "‘,‘" + IR_QUESTION + "‘,‘" + IR_URLNAME + "‘,‘" + IR_LASTTIME + "‘,‘" + IR_VIEW+ "‘,‘" + IR_FOLLOW + "‘,‘" + IR_RETURN + "‘,‘" + BBSNUM + "‘)"
try:
# 执行sql语句
cursor.execute(sql)
# 提交到数据库执行
db.commit()
except Exception as e:
# Rollback in case there is any error
print(‘插入数据失败!‘)
print(e)
db.rollback()
#回帖
huifu = response.xpath("//div[@class=‘List‘]//div[@class=‘List-item‘]")
print(huifu)
num=1
for one in huifu:
IR_AUTHOR = one.xpath(".//div[@class=‘AuthorInfo-head‘]//a[@class=‘UserLink-link‘]/text()")[0] if one.xpath(
".//div[@class=‘AuthorInfo-head‘]//a[@class=‘UserLink-link‘]/text()") else ‘匿名用户‘
if IR_AUTHOR == ‘匿名用户‘:
IR_AUTHOR_LINK = ‘‘
else:
IR_AUTHOR_LINK = ‘https:‘ + one.xpath(".//div[@class=‘AuthorInfo-head‘]//a[@class=‘UserLink-link‘]/@href")[
0]
IR_RESPONSE = ‘‘.join(one.xpath(".//div[@class=‘RichContent-inner‘]//text()"))
IR_URLTIME = one.xpath(".//div[@class=‘ContentItem-time‘]//span/text()")[0]
pattern = r‘((\u53d1\u5e03\u4e8e|\u7f16\u8f91\u4e8e).*?(\d+-\d+-\d+|\d+:\d+))‘
gone = re.search(pattern, IR_URLTIME).group(1)
gtwo = re.search(pattern, IR_URLTIME).group(3)
if ‘昨天‘ not in gone and ‘-‘ not in gtwo:
IR_URLTIME = time.strftime("%Y/%m/%d") + ‘ ‘ + gtwo
elif ‘-‘ in gtwo:
IR_URLTIME = gtwo
else:
IR_URLTIME = str(datetime.date.today() - datetime.timedelta(days=1)) + ‘ ‘ + gtwo
IR_AGREE = one.xpath(".//button[@aria-label=‘赞同‘]/text()")[0]
if ‘K‘ in IR_AGREE:
IR_AGREE=str(int(float(IR_AGREE.replace(‘K‘,‘‘))*1000))
print(IR_AGREE)
pattern=‘\d+‘
IR_COMMENT = one.xpath(".//div[@class=‘ContentItem-actions RichContent-actions‘]/button[1]/text()")[0]
try:
IR_COMMENT = re.search(pattern, IR_COMMENT).group()
except:
IR_COMMENT = ‘0‘
BBSNUM=str(num)
sql = "INSERT INTO wenda (IR_GROUPNAME,IR_SITENAME,IR_CHANNEL,IR_URLTITLE,IR_URLNAME,IR_URLTIME,IR_LASTTIME,BBSNUM,IR_AUTHOR,IR_RESPONSE,IR_AGREE,IR_COMMENT) VALUES (‘" + IR_GROUPNAME + "‘,‘" + IR_SITENAME + "‘,‘" + IR_CHANNEL + "‘,‘" + IR_URLTITLE + "‘,‘" + IR_URLNAME + "‘,‘" + IR_URLTIME + "‘,‘" + IR_LASTTIME + "‘,‘" + BBSNUM + "‘,‘" + IR_AUTHOR + "‘,‘" + IR_RESPONSE + "‘,‘" + IR_AGREE + "‘,‘" + IR_COMMENT + "‘)"
try:
# 执行sql语句
cursor.execute(sql)
# 提交到数据库执行
db.commit()
except Exception as e:
# Rollback in case there is any error
print(‘插入数据失败!‘)
print(e)
db.rollback()
num+=1
time.sleep(1)
except Exception as e:
print(e)
def starts(self):
try:
pool = ThreadPool(5)
time3 = time.time()
pool.map(self.get_link, self.url_list)
pool.close()
pool.join()
time4 = time.time()
print (‘多线程耗时 : ‘ + str(time4 - time3) + ‘ s‘)
return ‘采集结束‘
except:
return ‘链接错误‘
