[uf-discuss] [citation] url field

[Scott Reynen]

On Dec 1, 2006, at 10:44 PM, Ross Singer wrote:

> But how many 'citable' URLs contain ISBNs?

Apparently every book on Amazon.com, BarnesAndNoble.com and ISBN.nu  
has a URL containing its ISBN.

> DOIs are likely to be in a
> URL, but they are, sadly, impossible to discern (don't let the 10.foo
> fool you -- a DOI can contain any character and as many characters
> until whitespace - there is no way to know if the DOI ends before the
> URL does).

I don't know of any tools for extracting DOIs from URLs, so  
skepticism of that may well be warranted.  But it's clearly possible  
to extract ISBNs from URLs because people are doing it.

> Also, where do get that Jon Udell's Library Lookup 'is reliably
> extracting ISBNs from URLs'?  LibraryLookup finds ISBNs in /HTML/ and
> writes a URL to a library catalog based on it.

Sorry, that's just false.  Here's the description:

"Let's say you're on a book-related site (Amazon, BN, isbn.nu) and  
your current page's URL includes an ISBN."

Note the word "URL."  And here's an example LibraryLookup bookmarklet:

javascript:var%20re=/([\/-]|is[bs]n=)(\d{7,9}[\dX])/i;if(re.test 
(location.href)==true){var%20isbn=RegExp.$2;void(win=window.open 
('http://ksclib.keene.edu'+'/search/ 
i='+isbn,'LibraryLookup','scrollbars=1,resizable=1,location=1,width=575, 
height=500'))}

Note the "location.href" -- that's the URL.  There's no parsing of  
the HTML in this JavaScript.

Peace,
Scott