- From: Rick Byers <rbyers@google.com>
- Date: Thu, 14 Nov 2013 15:16:55 -0500
- To: "public-pointer-events@w3.org" <public-pointer-events@w3.org>
- Message-ID: <CAFUtAY_ajBwdicYdSsb20kz5TOiCTPgxWkSwj+imW0K+hX+XrQ@mail.gmail.com>
Hi, The touch-action processing model ( http://www.w3.org/TR/pointerevents/#the-touch-action-css-property) says that the touch-action property is restricted to block-level elements - which simplifies implementations a little. But it indicates that hit testing still needs to be done against inline elements since they may flow outside their containing block. In particular "then examine the touch-action property of each element between the hit tested element and the element with the default touch behavior". So for example, in this test http://jsbin.com/AROLEjEH/1/edit, touching the blue text outside the box should not scroll (since the hit-tested element is the span, and it has a block-level ancestor with touch-action: none). But my understanding of the whole point of the block-level restriction was to simplify the hit testing. So we technically aren't getting that simplification the way the spec is worded today. Not surprisingly, IE does NOT conform to the spec here. It appears to only hit test against the block-level elements (at least in the simple scenarios I tried). This distinction makes a huge difference for implementations (and impacts future compatibility). Unfortunately I think we need to make a normative change to the spec here to say explicitly that hit testing occurs only on block-level elements (and just passes through any inline elements, as if they had 'pointer-events: none'). Rick
Received on Thursday, 14 November 2013 20:17:46 UTC