- From: Philippe Le Hegaret <plh@w3.org>
- Date: Thu, 09 Jul 2015 19:21:12 -0400
- To: chaals@yandex-team.ru, Richard Schwerdtfeger <schwer@us.ibm.com>
- CC: HTML Accessibility Task Force <public-html-a11y@w3.org>, "public-canvas-api@w3.org" <public-canvas-api@w3.org>, mark.sadecki+w3c@gmail.com
We did manage to meet for one hour and looked into http://cptvitamin.github.io/wpt-canvas/2dcontext/hit-regions/addHitRegion.IDNotNullPreviousInList-manual.html Chrome exposed two separate issues: 1. the first region focusable area is misplaced on the screen, as mentioned by Rich. However, the second region is placed correctly. We think this needs to be reported (and Mark will follow up); 2. if there is a reflow after the second region is added, then the location reported will be incorrect. It seems to me that the Canvas spec is unclear on what gets reported to the user and the API. We'll need Rich to confirm that or not. The spec says: "inform the user of the location of the region representing the control given by the control member's value for this canvas element. The full location of the corresponding fallback element, pertaining to the control, as represented by the region is passed to the accessibility API, if supported." I might be misreading the text but I have several questions: For the first sentence: 1. If the region isn't focused (and it cannot be since we just created the region using addHitRegion!), why do we "inform the user" [1]? For example, why would we notify a magnifying tool if we're not focusing anyway? 2. if it doesn't matter, then what is the "location of the region"? Is it the hit region's bounding circumference subject to the clipping region? Is the implementation expected to deduce from that the region relative to the origin of the screen of the output device [2]? If yes, What is it if addHitRegion is invoked on a path that is not within the viewport but expected to scroll into view because it's not focused? For the second sentence: What is the "full location" of the element? Is it the location of the hit region's bounding circumference subject to the clipping region and relative to the initial containing block origin of the document [3]? As a side, the term "specified path" and "source path" are used in the algorithm but never defined either. Philippe [1] http://www.w3.org/TR/2dcontext/#inform [2] http://dev.w3.org/csswg/cssom-view/#screen [3] http://dev.w3.org/csswg/css-display-3/#initial-containing-block
Received on Thursday, 9 July 2015 23:21:20 UTC