That would definitely work.  That could even be tied into a more general 
URL API (I think Mozilla might have something similar?).  Something like:

var urlObject = chrome.url.parse("cnn.com/story"); // Could throw an 
exception if the URL is not valid
var aDomain = urlObject.domain; // "cnn.com"
var aPath = urlObject.path; // "/story"
var bookmarkableURL = urlObject.fullURL; // "http://cnn.com/story";

Colin

Nick Baum wrote:
> Alternatively, could we expose the urlFixerUpper as an extension API?
>
> Then Colin could do 
> chrome.bookmarks.create(...chrome.urlFixerUpper(myUrl)...);
>
> Nick
>
> On Wed, Sep 16, 2009 at 5:18 PM, Colin Bleckner <[email protected] 
> <mailto:[email protected]>> wrote:
>
>     Right now I've hacked together a function that checks to see if
>     the URL I'm about to add starts with known protocols and, if it
>     doesn't, automatically prepends "http://"; to it.  This (kind of)
>     works, but I'm still discovering that it doesn't handle certain
>     protocols like "chrome://" or <awesome new protocol that you guys
>     add in Chrome 5.0 that I didn't know existed> and I'm fiddling
>     with it more often than I'd like.  My options are to either sit
>     down and write (or find) a complete URI parser in Javascript or
>     use whatever you guys have already built (and are going to keep up
>     to date for me!).  You can imagine which option I prefer. :)
>
>     Colin
>
>
>     Aaron Boodman wrote:
>>     On Wed, Sep 16, 2009 at 1:06 PM, Erik Kay <[email protected]> 
>> <mailto:[email protected]> wrote:
>>       
>>>     This is a really good idea.  Apparently, I'm not running the URL
>>>     through our URLFixerUpper class (I'm not making that name up), which
>>>     does a bunch of stuff to fix up URLs that someone may have typed in or
>>>     copied and pasted in (not just the http:// example you gave).
>>>
>>>     I filed http://code.google.com/p/chromium/issues/detail?id=22038 for
>>>     this.  In the future, feel free to just go ahead and file the bug
>>>     directly.
>>>         
>>     I would have leaned the other way, that our API should be pedantic and
>>     require a correct URL. In some cases, passing an incomplete URL is a
>>     bug on the developer's part and they would want to know at development
>>     time, rather than storing bad data.
>>
>>     - a
>>       
>
>     >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Chromium-extensions" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/chromium-extensions?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to