[rfc][icedtea-web] Fix regression in recent PR1204 patch

[Omair Majid]

On 10/02/2013 05:18 PM, Andrew Azores wrote:
> Well, I ended up running into another corner case (as I mentioned on
> IRC) and decided it was easier to ditch the URI class and go back to
> manually constructing URLs :( . This is sort of ugly IMO but it
> completely sidesteps the percent-encoding issues involved with using
> URI, which was quickly making that solution uglier and uglier anyway.
> It's still enhanced over the pre-PR1204 fix version though, as that one
> was missing some of the optional parts in a URL. It had no user info and
> no query string, notably. Now it does, and tests for them too!

> I didn't write tests for that one.

That should be fine; it's a helper method that is exercised by other
tests already, right?

Patch looks good to me.

Cheers,
Omair

-- 
PGP Key: 66484681 (http://pgp.mit.edu/)
Fingerprint = F072 555B 0A17 3957 4E95  0056 F286 F14F 6648 4681