学了一下python正则表达式，写一个百度图片爬虫玩玩。
当技术遇上心术不正的人，就成我这样的2B青年了。
python3.6开发。程序已经打包好，下载地址: http://pan.baidu.com/s/1bpalugf 密码：kfk4
#!/usr/local/env python
from tkinter import *
import re,os,requests,hashlib,threading
from PIL import Image
class Application(Frame):
def __init__(self, master=None):
Frame.__init__(self, master)
self.school=threading.local()
self.pack()
self.createWidgets()
def createWidgets(self):
self.nameLabel=Label(self,text='请输入关键词:')
self.nameLabel.grid(row=0,sticky=W)
self.nameInput = Entry(self)
self.nameInput.grid(row=0,column=1)
self.picys=IntVar()
self.Checkbutton = Checkbutton(self,text='图片压缩',variable=self.picys)
self.Checkbutton.grid(row=1,column=0,columnspan=2,sticky=W)
self.alertButton = Button(self, text='下载',command=self.gorun)
self.alertButton.grid(row=1,column=1,sticky=E)
def cddir(self):
keyword=self.nameInput.get()
os.chdir('C:\\Users\\Administrator\\Desktop\\')
if os.path.exists(keyword) ==False:
os.mkdir(keyword)
os.chdir(keyword)
def gorun(self):
self.cddir()
word=self.nameInput.get()
x=0
for i in range(5):
t=threading.Thread(target=self.xiazai,args=(x,word,))
t.start()
x+=20
if i == 4:
t.join()
self.delfile()
if self.picys.get() == 1:
self.suoxiao()
def xiazai(self,page,word):
baidupn=self.school.student=page
num=1
for i in range(50):
url='https://image.baidu.com/search/flip?tn=baiduimage&ie=utf-8'
payload={'word':word,'pn':baidupn}
html = requests.get(url,params=payload).text
regular='"objURL":"(.*?)",'
pic=re.findall(regular,html)
baidupn+=100
for tu in pic:
try:
dl=requests.get(tu,timeout=60)
pic_name=os.path.basename(tu)
if pic_name in os.walk(os.getcwd()):
continue
else:
if ('?' in pic_name) or ('&' in pic_name) or ('.' not in pic_name):
pic_name='%s%s' %(num,'.jpg')
with open(pic_name,"wb") as code:
code.write(dl.content)
requests.session().keep_alive = False
dl.close()
num+=1
except requests.exceptions.ConnectionError:
#print('这张图片下载失败了,图片地址',tu)
continue
def suoxiao(self):
self.cddir()
filedir=os.walk(os.getcwd())
for i in filedir:
for tplb in i[2]:
if ('jpg' in tplb) or ('jpeg' in tplb):
try:
im=Image.open(tplb)
w,h=im.size
if w > 500:
im.thumbnail((w//2,h//2))
im.save(tplb,'jpeg')
im.close()
except OSError:
print('跳过此文件')
def md5sum(self,filename):
f=open(filename, 'rb')
md5=hashlib.md5()
while True:
fb = f.read(8096)
if not fb:
break
md5.update(fb)
f.close()
return (md5.hexdigest())
def delfile(self):
all_md5={}
self.cddir()
filedir=os.walk(os.getcwd())
for i in filedir:
for tlie in i[2]:
if self.md5sum(tlie) in all_md5.values():
os.remove(tlie)
else:
all_md5[tlie]=self.md5sum(tlie)
app=Application()
app.master.title('图片下载器')
app.mainloop()
无耻的求一下赞助
