[Webkit-unassigned] [Bug 54594] Expose isValidProtocol() in KURL.h.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=54594





--- Comment #9 from James Kozianski <koz at chromium.org>  2011-02-17 13:14:06 PST ---
(In reply to comment #8)
> (From update of attachment 82714 [details])
> This seems tricky.  How flexible should this be?  Should this allow leading/trailing spaces?  Why is this only implemented for the Google side?  is "a" a valid scheme?  Seems this says it is.  Can we test this?

There is already an implementation in KURL.cpp of which this is just a copy (see my above comment). And I think the most intuitive behaviour for this function would be to validate exactly what the spec says at http://tools.ietf.org/html/rfc3986#section-3.1.

As for testing, this function isn't exposed to the browser directly (it is used internally by KURL.cpp), but in my other patch (bug 54438) which causes navigator.registerProtocolHandler() to use this function to verify protocols passed to it, I add test cases that exercise this function directly.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.