Well, actually, this is a Python library for accessing xkcd.

It does contain five scripts- one of which does download all xkcds onto your computer. (I haven't actually downloaded all ~850 yet though). There are two others, for opening up random/last xkcds in your web browser, and two others for downloading the random/latest xkcd to your computer.

Plus, you can use the functions available in the library for doing other things.

You need Python 2.x (2.7 is the latest version) to install: you can get it here.

As of version 1.1, you no longer need the feedparser module. I've modified it to use xkcd's JSON interface, which is good for two reasons: the RSS feed sometimes has things in it that aren't comic) which break the old implemention, and secondly: it removes the only third-party dependency.

The download of xkcd.zip includes seven files (minus the readme, which has all this text):

-dowallxkcd.py: script to download all xkcds
-dowlastxkcd.py: script to download latest xkcd
-dowrandxkcd.py: script to download random xkcd
-lastxkcd.py: script to open a random xkcd in browser
-randxkcd.py: script to open the latest xkcd in browser
-xkcd.py: the library itself
-xkcd-1.0.zip: a zipped Python package - if you know how to install Python packages just install this

To install manually on Windows, you'd put xkcd.py in C:\Python27\lib\site-packages (wherever Python is installed, then \lib\site-packages). The scripts can go anywhere.

To install manually on Linux... it goes to the same lib/site-packages folder, but I'm not sure where this is- either /lib/python/ or /usr/lib/python, probably.

Nice, I hope if this is popular that this doesn't overload their servers, though. O.o (eg: if someone decided to post the script on 4chan or another ultra-busy community site)

Ah cool!

Does this handle this?

Lol SirCmpwn

It would be funny if the script actually found a comic hidden there.

Nope, none hidden there.  But there is a comic hidden deep within xkcd.

graphmastur, you will now share all knowledge you have on the subject, or face my wrath.

Oh, I've seen the unixkcd source before, but I'm not able to see it from my phone.  Could you do me the favor of putting the beautified code in a spoiler, seeing as it may be quite a while before I can see it?

To answer the original question(s).

It does handle 404, because all Internet access runs in a try-except block, and it simply skips over 404 if it fails.

It can't be changed to use other websites, because it depends on the way xkcd works- the page encoding saying "direct link to comic", the RSS feed,  and the page numbers for comics: "xkcd.com/843".