http://code.google.com/p/chromium/issues/detail?id=118181 Chrome Version : 18.0.1025.56 beta URLs (if applicable) : http://www.thesokolows.com/colorresearch/pedigrees/AARON_OF_QUIETUDE.xhtml Other browsers tested: Add OK or FAIL after other browsers where you have tested this issue: Safari 5: FAIL Firefox 4.x: OK IE 7/8/9: NA What steps will reproduce the problem? 1. From the example URL, click on the little eye symbol What is the expected result? Navigate to a picture of the horse (follow the link in the <a> element). What happens instead? The link in the containing element is followed. To aid in producing a reduction, the example URL contains this, where the "#male" bbox contains the "#eye" bbox: <use xlink:href="#couple" x="11" y="14" /> <a xlink:href="ADRIAN_ASHMORE.xhtml"> <use xlink:href= "#male" class="maleStyle nonflaxenStyle " x="11" y="14" /></a> <a xlink:href="ADRIAN_ASHMORE.xhtml"> <text font-size=".7" font-family="monospace" x="11.5" y="14.8" >ADRIAN ASHMORE</text></a> <text font-size=".6" font-family="monospace" x="15" y="15.6" >12877 -</text> <text font-size=".7" font-family="monospace" x="11.3" y="15.6" >CH</text> <a xlink:href="http://www.morganhorse.com/upload/photos/3922AdrianAshmoreNew.jpg" > <use xlink:href="#eye" x="13" y="14.9" /></a>
I think the reduction is confusing. Don't you mean <a xlink:href="outer.svg">Click <a xlink:href="inner.svg">Here</a></a> that we always navigate to outer.svg even when clicking on "Here"?
Yeah, my reduction is not really a reduction. Note that your anchor inside anchor is in fact illegal according to the spec. This might work, but I haven't tried yet. Still working through over 100 WebKit SVG bugs in crbug figuring out which ones are not yet in Bugzilla. <svg ...> <defs> <rect name="inner" x="10" y="10" width="20" height="20"/> </defs> <a href="outerHref"><rect x="5" y="5" width="30" height="30"/></a> <a href="innerHref"><use xlink:href="#inner"/></a> </svg> When you click at 15,15 you will apparently go to outerHref. Regardless, there is some bug that the referenced site reveals.
Is this still a problem in trunk?
This does indeed seem to have been fixed. Both the original bug report test case and a layout test I constructed both work as expected in Chrome 19.0.1084.24 beta. I think I'll upload the test anyway, just for future regression testing.
Created attachment 137747 [details] Patch
Comment on attachment 137747 [details] Patch Looks good.
Comment on attachment 137747 [details] Patch Clearing flags on attachment: 137747 Committed r114750: <http://trac.webkit.org/changeset/114750>
All reviewed patches have been landed. Closing bug.