Error Handling in Python Requests Module When URL Does Not Exist

Question

I am trying to nail down the error handling for the requests module in python in order to be notified as and when a URL is unavailable, ie HTTPError, ConnectionError, Timeout etc...

The issue that I am having is that I seem to be getting status responses of 200 even on FAKE URLs

I have trawled through SO & various other web sources, tried many differing ways of seemingly trying to achieve the same goal but have so far come up empty.

I have boiled the code down to as basic as it gets to simplify things.

import requests

urls = ['http://fake-website.com', 
        'http://another-fake-website.com',
        'http://yet-another-fake-website.com',
        'http://google.com']

for url in urls:
    r = requests.get(url,timeout=1)
    try:
        r.raise_for_status()
    except:
        pass
    if r.status_code != 200:
        print ("Website Error: ", url, r)
    else:
        print ("Website Good: ", url, r)

I expected the first 3 URLs in the list to classed as 'Website Error:' as they are URLs that I have just made up. The final URL in the list is quite obviously real so should be the only one to be listed as 'Website Good:'

What is happening is the first URL produces a correct response to the code as it gives a response code of 503 but the next two URLs do not produce a status_code at all according to https://httpstatus.io/ but only display ERROR with Cannot find URI. another-fake-website.com another-fake-website.com:80 Cannot find URI. another-fake-website.com another-fake-website.com:80

So I expected all but the last URL in the list to be shown as 'Website Error:'

OUTPUT

when running script in Raspberry Pi

Python 2.7.9 (default, Sep 26 2018, 05:58:52) 
[GCC 4.9.2] on linux2
Type "copyright", "credits" or "license()" for more information.
>>> ================================ RESTART ================================
>>> 
('Website Error: ', 'http://fake-website.com', <Response [503]>)
('Website Good: ', 'http://another-fake-website.com', <Response [200]>)
('Website Good: ', 'http://yet-another-fake-website.com', <Response [200]>)
('Website Good: ', 'http://google.com', <Response [200]>)
>>>

If I enter all 4 URLs in to https://httpstatus.io/ I get this result:

It shows a 503, a 200 & two URLs that do not have a status code but rather just display Error

UPDATE

so I thought that I would check this in Windows using PowerShell & followed this example: https://stackoverflow.com/a/52762602/5251044

This is the output below

c:\Testing>powershell -executionpolicy bypass -File .\AnyName.ps1
0 - http://fake-website.com
200 - http://another-fake-website.com
200 - http://yet-another-fake-website.com
200 - http://google.com

as you can see, I am no further forward.

UPDATE 2

having had further discussions with Fozoro HERE & trying various options with no fix in sight I thought that I would try this code using urllib2 instead of requests

Here is the changed code

from urllib2 import urlopen
import socket

urls = ['http://another-fake-website.com',
        'http://fake-website.com',
        'http://yet-another-fake-website.com',
        'http://google.com',
        'dskjhkjdhskjh.com',
        'doioieowwros.com']

for url in urls:

    try:
        r  = urlopen(url, timeout = 5)
        r.getcode()
    except:
        pass
    if r.getcode() != 200:
        print ("Website Error: ", url, r.getcode())
    else:
        print ("Website Good: ", url, r.getcode())

Unfortunately the resulting output is still not correct but does differ slightly from the output of the previous code, see below:

Python 2.7.9 (default, Sep 26 2018, 05:58:52) 
[GCC 4.9.2] on linux2
Type "copyright", "credits" or "license()" for more information.
>>> ================================ RESTART ================================
>>> 
('Website Good: ', 'http://another-fake-website.com', 200)
('Website Good: ', 'http://fake-website.com', 200)
('Website Good: ', 'http://yet-another-fake-website.com', 200)
('Website Good: ', 'http://google.com', 200)
('Website Good: ', 'dskjhkjdhskjh.com', 200)
('Website Good: ', 'doioieowwros.com', 200)
>>> 

This time it is showing all 200 responses, very peculiar.

Answer 1

You should put r = requests.get(url,timeout=1) inside of the try: block. So your code needs to look like this:

import requests

urls = ['http://fake-website.com', 
        'http://another-fake-website.com',
        'http://yet-another-fake-website.com',
        'http://google.com']

for url in urls:
    try:
        r = requests.get(url,timeout=1)
        r.raise_for_status()
    except:
        pass
    if r.status_code != 200:
        print ("Website Error: ", url, r)
    else:
        print ("Website Good: ", url, r)

Output:

Website Error:  http://fake-website.com <Response [503]>
Website Error:  http://another-fake-website.com <Response [503]>
Website Error:  http://yet-another-fake-website.com <Response [503]>
Website Good:  http://google.com <Response [200]>

I hope this helps!

Answer 2

For me, the reason turned out to be a website served by my ISP about the URL being invalid - it's that website that returns a 200, not the fake one.

This can be verified by printing the content of the returned site with requests.get('http://fakesite').text