100074 2012-10-22 22:12:08 -0700 [chromium] svg/as-image/image-respects-pageScaleFactor-change.html should use dumpAsText(true) 2013-04-09 17:07:28 -0700 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) Unspecified Unspecified RESOLVED WONTFIX P2 Normal --- 1 noel.gordon schenney dpranke schenney oldest_to_newest 748273 0 noel.gordon 2012-10-22 22:12:08 -0700 Test currently uses dumpAsText(1) but that's not supported by chromium DRT so NRWT won't compare the expected pixel results :/ 748559 1 schenney 2012-10-23 06:16:03 -0700 So dumpAsText(1) means dump as text and compare as image, but do not dump the render tree? How 'bout that. I can try to get it implemented, or maybe Dirk has time. Still, the test seems to me like it should be a ref-test, so I think I will make it so. 748572 2 noel.gordon 2012-10-23 06:33:12 -0700 Change it to dumpAsText(true), that'll work everywhere. Then submit new pixel results for chromium. 748822 3 dpranke 2012-10-23 11:27:52 -0700 As far as I know, dumpAsText(True) means do not dump a render tree and do not produce a pixel result, on all ports. I don't think there's a difference between dumpAsText(true) and dumpAsText(1), nor a way to get non-render-tree text output but still get a pixel result. 749260 4 noel.gordon 2012-10-23 19:53:23 -0700 (In reply to comment #3) > As far as I know, dumpAsText(True) means do not dump a render tree and do not produce a pixel result, on all ports. It means dump text & pixels and do not dump render tree and yes, we admit it's confusing http://trac.webkit.org/wiki/Writing%20Layout%20Tests%20for%20DumpRenderTree#Howtodisablerendertreedumpsbutkeeppixelsandtext > I don't think there's a difference between dumpAsText(true) and dumpAsText(1) Agree; there should not be a difference. However the chromium DRT only accepts a boolean and that creates a difference on the Chromium port :) I can fix that later. As suggested, just change this test to use dumpAsText(true), and check in new chromium pixel results. I've fixed the other cases of dumpAsText(1) on bug 100076. 872516 5 schenney 2013-04-09 17:07:28 -0700 Marked LayoutTest bugs, bugs with Chromium IDs, and some others as WontFix. Test failure bugs still are trackable via TestExpectations or disabled unit tests.