Comments on: Untiny that url!

By: Sam Johnston

Sam Johnston — Wed, 22 Apr 2009 09:10:40 +0000

G'day,

Did you know that "rel" is a space separated list, so "alternate shorter" is interpreted as "alternate" and "shorter". "alternate" refers to the content itself, not the link (its used for e.g. atom, pdf, text versions of the same) and for "shorter", what's shorter? the content?

If you want to avoid any confusion (including the rel=short[_- ]?ur[il] mess) then rel=shortlink is the answer you're looking for.

Sam

By: Saleh

Saleh — Fri, 17 Apr 2009 22:15:45 +0000

You can try http://untiny.com http://untiny.me .. It has been there for a while and supports over 75 tiny services It has about 10 addons as well (based on API http://untiny.com/api ) in Firefox, Mac, Linux … http://untiny.com/extra

It’s in Arabic in time being, but it will be available in English hopefully after few days.

You can check the English version while I’m translating it :=) at http://alzaid.ws/labs/untiny-en

Thanks

By: pib

pib — Tue, 14 Apr 2009 23:23:22 +0000

@Bill: The idea is that people should use these functions in their Twitter clients or apps that pull in tweets for one reason or another. I didn’t really intend that people would run a script in bash just to find out what URL something goes to.

By: Bill

Bill — Tue, 14 Apr 2009 21:46:21 +0000

Perhaps I'm missing something but how do you propose someone would utilize one of these tools to "untiny" a url in a non-intrusive way; wouldn't most people continue to just click on the URL even if they risk being Rick Rolled?

Would going to the command line really be any less a pain in the ass than just clicking on the link and finding out where it goes?

It seems like a more inline approach is what is needed (a greasemonkey script or something similar) that will show the expanded URL within the browser.

By: pib

pib — Tue, 14 Apr 2009 06:49:42 +0000

@Alexander: That’s a good point. I didn’t even notice that function existed. It seems it just does a GET request, sadly, but it’s still better than having all the code in my version above.

By: Alexander Nyquist

Alexander Nyquist — Tue, 14 Apr 2009 05:26:57 +0000

For the PHP version, you have the get_headers()-function.

By: Nathan de Vries

Nathan de Vries — Mon, 13 Apr 2009 14:21:06 +0000

You can perform a HEAD request using Ruby’s Net::HTTP like this http://gist.github.com/94463

By: skeptomai

skeptomai — Sun, 12 Apr 2009 22:49:27 +0000

And a Common Lisp version, using Edi Weitz’s Drakma http client: (require ‘drakma) (lambda (url) (multiple-value-bind (body return headers) (drakma:http-request url :method :head :redirect nil) (or (cdr (assoc :LOCATION headers)) url)))

By: pib

pib — Sun, 12 Apr 2009 21:22:51 +0000

@Dan Farina

I know it's not idiomatic Python, and I even came up with that same shorter solution (with a try: except: to catch invalid URLS), but the problem with that version is that it actually does two HTTP requests. I couldn't find a way to make it only do the first request.

If somebody knows how to make urllib or urllib2 not follow redirects, then I will change it. Until then, this is the shortest version I could come up with that didn't involve doing some crazy stuff with subclassing urllib2.HTTPRedirectHandler

By: Dan Farina

Dan Farina — Sun, 12 Apr 2009 20:29:30 +0000

Not idiomatic Python. The first line in httplib docs says:

"This module defines classes which implement the client side of the HTTP and HTTPS protocols. It is normally not used directly — the module urllib uses it to handle URLs that use HTTP and HTTPS."

With urllib, the following will suffice:

import urllib

def untinyurl(tinyurl):
    return urllib.urlopen(tinyurl).geturl()

A sample interaction is as follows:

>>> untinyurl('http://tinyurl.com/c8f5bz')
'http://blog.paulbonser.com/2009/04/11/untiny-that-url/'