Bug 34239 – Implement CSSOM Range.getClientRects for collapsed selections

• Component: DOM
• Status: RESOLVED FIXED

[Julie Parent]

The spec used to say to return null when the endpoints were the same, but now has been updated to handle this case.  See http://www.w3.org/TR/2009/WD-cssom-view-20090804/#extensions-to-the-range-interface for full details.

Example: http://www.plexode.com/cgi-bin/eval3.py#ht=%3Cdiv%20contentEditable%3E%3Cdiv%20id%3D'bar'%3Efooo%3C%2Fdiv%3E%3C%2Fdiv%3E&ohh=1&ohj=1&jt=var%20bar%20%3D%20gebi('bar')%3B%0Avar%20range%20%3D%20document.createRange()%3B%0Arange.setStart(bar.firstChild%2C%200)%3B%0Arange.setEnd(bar.firstChild%2C%200)%3B%0A%0Avar%20rects%20%3D%20range.getClientRects()%3B%0Avar%20rect%20%3D%20rects.item(0)%3B%0A%0Aalert('top%3A%20'%20%2B%20rect.top%20%2B%20'%20bottom%3A%20'%20%2Brect.bottom%20%2B%20'%20left%3A%20'%20%2B%20rect.left%20%2B%20'%20right%3A%20'%20%2B%20rect.right%20%2B%20'%20width%3A%20'%20%2Brect.width%20%2B%20'%20height%3A%20'%20%2B%20rect.height)%3B&ojh=1&ojj=0&ms=100&oth=0&otj=0&cex=1

WebKit returns null, Firefox 3.7a returns a rect with width = 0 and height = line height (and appropriate top, left, right, bottom values as well).

[Ojan Vafai]

Created attachment 47682 [details]
Patch

[Ojan Vafai]

Committed r54117: <http://trac.webkit.org/changeset/54117>

[Simon Fraser (smfr)]

Why did this alter the behavior of absoluteQuadsForRange() but not absoluteRectsForRange()? Now one uses !isEmpty(), and the other uses if (r.height()), which is inconsistent.

[Simon Fraser (smfr)]

Bug 62478 exists for that.