Bug 16933

Summary: Popup on opentable.com is off by ~(10,10)
Product: WebKit Reporter: Cameo Wood <cameowood>
Component: EvangelismAssignee: Nobody <webkit-unassigned>
Status: RESOLVED FIXED    
Severity: Normal CC: aroben, mitz, mrowe, vicki, webkit
Priority: P2 Keywords: InRadar
Version: 528+ (Nightly build)   
Hardware: PC   
OS: All   
URL: http://www.opentable.com/

Description Cameo Wood 2008-01-18 23:26:21 PST
When on www.opentable.com, click on "san Francisco" then in the search field, type in "Ruth's Chris"  As you type, a popup should appear with autocomplete options.  In FF/IE this popup is flush directly underneath the search box.  in Safari, it has a 10,10 offset to the bottom right.

olliej said  "the div of interest is PopupDiv, containing a table with id smartInputFloater"
Comment 1 mitz 2008-01-18 23:31:56 PST
Spoofing as Firefox corrects the problem.
Comment 2 mitz 2008-01-18 23:40:13 PST
(In reply to comment #1)
> Spoofing as Firefox corrects the problem.

So does spoofing as WinIE. The site might employ a workaround for Safari 2 offest{Top,Left,Parent} quirks that is no longer necessary.
Comment 3 Mark Rowe (bdash) 2008-01-19 09:21:28 PST
In http://www.opentable.com/jscripts/wick.js:

		//hack: browser-specific adjustments.
		if (!siw.isGecko && !siw.isWinIE) x += 8;
		if (!siw.isGecko && !siw.isWinIE) y += 10;

That looks like it could easily be the cause.  wick.js comes from http://wick.sourceforge.net/.  This incorrect code is part of the latest version, 0.1, which was released in 2005.  Should this be moved to "Evangelism"?
Comment 4 mitz 2008-01-19 11:02:48 PST
(In reply to comment #3)
> Should this be moved to "Evangelism"?

I say "yes".
Comment 5 Adam Roben (:aroben) 2008-01-21 08:28:54 PST
<rdar://problem/5697482>
Comment 6 Robert Blaut 2008-07-17 03:28:04 PDT
As I tested today the issue is no longer reproducible in WebKit r35203. Any confirmations?
Comment 7 Robert Blaut 2008-12-23 05:28:43 PST
(In reply to comment #6)
> As I tested today the issue is no longer reproducible in WebKit r35203. Any
> confirmations?
> 

I think the bug report should be closed now.