import urllib
import urllib.request
from bs4 import BeautifulSoup
import os
from string import ascii_lowercase
def make_soup(url):
thepage = urllib.request.urlopen(url)
soupdata = BeautifulSoup(thepage, "html.parser")
return soupdata
def take_image(n):
i=1
proxy = urllib.request.ProxyHandler({'http': '127.0.0.1'})
opener = urllib.request.build_opener(proxy)
urllib.request.install_opener(opener)
urllib.request.urlopen(n)
soup1 = make_soup(n)
for img in soup1.select('a[href] img'):
temp = img.find_parent('a', href=True)['href']
try:
if temp[:1] == "/":
image = "https://,,,.hk" + temp
else:
image = temp
print(image)
filename = str(i)
i = i + 1
imagefile = open(filename + ".png", 'wb')
imagefile.write(urllib.request.urlopen(image).read())
imagefile.close()
except:
continue
with open('E:\Test.txt', 'r') as file:
for line in file:
n = line
take_image(n)
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