Sorry that the title wasn't very clear, basically I have a list with a whole series of url's, with the intention of downloading the ones that are pictures. Is there anyway to check if the webpage is an image, so that I can just skip over the ones that arent?
Thanks in advance
You can use requests module. Make a head request and check the content type. Head request will not download the response body.
import requests
response = requests.head(url)
print response.headers.get('content-type')
There is no reliable way. But you could find a solution that might be "good enough" in your case.
You could look at the file extension if it is present in the url e.g., .png, .jpg could indicate an image:
>>> import os
>>> name = url2filename('http://example.com/a.png?q=1')
>>> os.path.splitext(name)[1]
'.png'
>>> import mimetypes
>>> mimetypes.guess_type(name)[0]
'image/png'
where url2filename() function is defined here.
You could inspect Content-Type http header:
>>> import urllib.request
>>> r = urllib.request.urlopen(url) # make HTTP GET request, read headers
>>> r.headers.get_content_type()
'image/png'
>>> r.headers.get_content_maintype()
'image'
>>> r.headers.get_content_subtype()
'png'
You could check the very beginning of the http body for magic numbers indicating image files e.g., jpeg may start with b'\xff\xd8\xff\xe0' or:
>>> prefix = r.read(8)
>>> prefix # .png image
b'\x89PNG\r\n\x1a\n'
As @pafcu suggested in the answer to the related question, you could use imghdr.what() function:
>>> import imghdr
>>> imghdr.what(None, b'\x89PNG\r\n\x1a\n')
'png'
You can use mimetypes https://docs.python.org/3.0/library/mimetypes.html
import urllib
from mimetypes import guess_extension
url="http://example.com/image.png"
source = urllib.urlopen(url)
extension = guess_extension(source.info()['Content-Type'])
print extension
this will return "png"
