Bug 100733

Summary:

3d rotation with [0, 0, 0] direction vector should not be applied

Product:

WebKit

Reporter:

Zoltan Nyul <zoltan.nyul>

Component:

Layout and Rendering

Assignee:

Zoltan Nyul <zoltan.nyul>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

cdumez, gyuyoung.kim, jamesr, kenneth, laszlo.gombos, leviw, simon.fraser, tonikitoo, webkit.review.bot

Priority:

Version:

528+ (Nightly build)

Hardware:

Unspecified

OS:

Unspecified

URL:

http://dev.w3.org/csswg/css3-3d-transforms/#transform-functions

Attachments:

Description	Flags
patch	none
patch	none
Changelog fixed	none
patch	none

Zoltan Nyul

Reported 2012-10-30 01:09:37 PDT

A direction vector that cannot be normalized, such as [0, 0, 0], will cause the rotation to not be applied, but webkit applies it with [1, 0, 0] direction vector.

Attachments
patch (4.88 KB, patch) 2012-10-30 01:19 PDT, Zoltan Nyul	no flags	Details Formatted Diff Diff
patch (4.91 KB, patch) 2012-10-30 01:35 PDT, Zoltan Nyul	no flags	Details Formatted Diff Diff
Changelog fixed (4.90 KB, patch) 2012-10-30 01:40 PDT, Zoltan Nyul	no flags	Details Formatted Diff Diff
patch (5.01 KB, patch) 2012-10-30 01:55 PDT, Zoltan Nyul	no flags	Details Formatted Diff Diff
Show Obsolete (3) View All Add attachment proposed patch, testcase, etc.

Zoltan Nyul

Comment 1 2012-10-30 01:19:52 PDT

Created attachment 171391 [details] patch

Chris Dumez

Comment 2 2012-10-30 01:25:26 PDT

Comment on attachment 171391 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=171391&action=review > LayoutTests/ChangeLog:8 > + Addig test for 3d rotation with [0, 0, 0] direction vector. "Addig" -> "Add" > Source/WebCore/ChangeLog:8 > + A direction vector that cannot be normalized, such as [0, 0, 0], will cause the rotation to not be applied, Is there a clear mention of this in the spec? If so, we should probably point to it to justify this change. > Source/WebCore/ChangeLog:11 > + No new tests (OOPS!). Please state that it is covered by transforms/3d/general/3dtransform-values.html test case.

Zoltan Nyul

Comment 3 2012-10-30 01:28:59 PDT

(In reply to comment #2) > (From update of attachment 171391 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=171391&action=review > > > LayoutTests/ChangeLog:8 > > + Addig test for 3d rotation with [0, 0, 0] direction vector. > > "Addig" -> "Add" > > > Source/WebCore/ChangeLog:8 > > + A direction vector that cannot be normalized, such as [0, 0, 0], will cause the rotation to not be applied, > > Is there a clear mention of this in the spec? If so, we should probably point to it to justify this change. Yes, here, http://dev.w3.org/csswg/css3-3d-transforms/ in the "8. The Transformation Functions". > > > Source/WebCore/ChangeLog:11 > > + No new tests (OOPS!). > > Please state that it is covered by transforms/3d/general/3dtransform-values.html test case. Ok

Zoltan Nyul

Comment 4 2012-10-30 01:35:05 PDT

Created attachment 171396 [details] patch

Chris Dumez

Comment 5 2012-10-30 01:37:04 PDT

Comment on attachment 171396 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=171396&action=review > LayoutTests/ChangeLog:8 > + Addig test for 3d rotation with [0, 0, 0] direction vector. "Addig" -> "Add" > Source/WebCore/ChangeLog:8 > + A direction vector that cannot be normalized, such as [0, 0, 0], will cause the rotation to not be applied, Is this stated clearly in the spec, if so, you should point to it.

Zoltan Nyul

Comment 6 2012-10-30 01:40:12 PDT

Created attachment 171398 [details] Changelog fixed

Zoltan Nyul

Comment 7 2012-10-30 01:55:51 PDT

Created attachment 171400 [details] patch

Chris Dumez

Comment 8 2012-10-30 02:11:25 PDT

Comment on attachment 171400 [details] patch LGTM.

Levi Weintraub

Comment 9 2012-10-30 02:39:03 PDT

Comment on attachment 171400 [details] patch LGTM.