- From: Jatinder Mann <jmann@microsoft.com>
- Date: Fri, 7 Oct 2011 17:21:02 +0000
- To: Anne van Kesteren <annevk@opera.com>, Arvind Jain <arvind@google.com>, Jonas Sicking <jonas@sicking.cc>
- CC: Boris Zbarsky <bzbarsky@mit.edu>, "public-web-perf@w3.org" <public-web-perf@w3.org>
That was my initial impression also. Further, if we begin including formatting information for sub-documents, why not include element visibility? E.g, my canvas element has been scrolled off page, so I should stop my animation. Scoping this API to the visibility of the top level document, as it is now, keeps it simple and still very useful in creating power- and CPU- efficient applications, particularly in the cases where the user choses to hide the page from view by interacting with the browser. Jatinder -----Original Message----- From: Anne van Kesteren [mailto:annevk@opera.com] Sent: Friday, October 07, 2011 3:15 AM To: Arvind Jain; Jonas Sicking Cc: Jatinder Mann; Boris Zbarsky; public-web-perf@w3.org Subject: Re: [PageVisibility] What should the visibility api return in display:none iframes? On Fri, 07 Oct 2011 02:28:18 +0200, Jonas Sicking <jonas@sicking.cc> wrote: > That means that if a page wants to only do certain actions when > displayed, it has to first check it's .hidden state, then walk up the > element.ownerDocument.defaultView.frameElement chain and at each > element use CSSOM to check if the iframe is hidden (which off the top > of my head I don't remember the API for). > > This seems severely more complex than simply checking .hidden. Does > anyone know of any webpages that we can check to see if they do this? > > What is the benefit of this approach? Are you also going to check for it being positioned off-screen, visibility:hidden, et cetera? It seems somewhat weird to have a display:none special case in this API. -- Anne van Kesteren http://annevankesteren.nl/
Received on Friday, 7 October 2011 17:21:33 UTC