The m_contentPosition variable is scaled position which is scaled with content scale factor. It can be confused with real position (which is known in the webcore), so it should be used carefully. I fix the wrong usage of m_contentPosition in this bug.
Created attachment 206426 [details] Patch
Comment on attachment 206426 [details] Patch Attachment 206426 [details] did not pass mac-ews (mac): Output: http://webkit-queues.appspot.com/results/1048524 New failing tests: fullscreen/full-screen-iframe-with-max-width-height.html
Created attachment 206430 [details] Archive of layout-test-results from webkit-ews-05 for mac-mountainlion The attached test failures were seen while running run-webkit-tests on the mac-ews. Bot: webkit-ews-05 Port: mac-mountainlion Platform: Mac OS X 10.8.3
Comment on attachment 206426 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=206426&action=review Noam, could you take a look this patch ? > Source/WebKit2/ChangeLog:3 > + [EFL][WK2] Correct wrong usage of m_contentPosition variable in the WebView. It looks you need to change [EFL] with [CoordnatiedGraphics].
Comment on attachment 206426 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=206426&action=review > Source/WebKit2/ChangeLog:10 > + The m_contentPosition variable is scaled position which is scaled with > + content scale factor. It can be confused with real position which is > + known in the webcore, so it should be used carefully. I don't get what this says, or what this patch does differently. > Source/WebKit2/UIProcess/CoordinatedGraphics/WebView.cpp:248 > + FloatPoint position = contentPosition(); > + position.scale(1 / m_contentScaleFactor, 1 / m_contentScaleFactor); > + FloatRect visibleContentsRect(position, visibleContentsSize()); This does not explain itself, nor it is explained in the Changelog.
Comment on attachment 206426 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=206426&action=review >> Source/WebKit2/ChangeLog:3 >> + [EFL][WK2] Correct wrong usage of m_contentPosition variable in the WebView. > > It looks you need to change [EFL] with [CoordnatiedGraphics]. Yes, I will change that. >> Source/WebKit2/ChangeLog:10 >> + known in the webcore, so it should be used carefully. > > I don't get what this says, or what this patch does differently. OK, I will re-write description. >> Source/WebKit2/UIProcess/CoordinatedGraphics/WebView.cpp:248 >> + FloatRect visibleContentsRect(position, visibleContentsSize()); > > This does not explain itself, nor it is explained in the Changelog. the contentPosition() returns the scaled position, so we have to divide that with content scale factor before setting to the visibleContentrect. I will add description about this to the changelog.
Created attachment 206965 [details] Patch Re-write the changelog.
Comment on attachment 206965 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=206965&action=review > Source/WebKit2/ChangeLog:16 > + > + The CoordinatedGraphics has its own scaling logic which is called as > + contents scaling, so the content position of WebCore should be used > + after scaling with contents scale factor in the CoordinatedGraphics. > + The scaled content position is maintained in the WebView as a > + m_contentPosition, and we have to recognize that is scaled value. > + > + The m_contentPosition is regarded as a non-scaled value in the > + transformToScene() and updateViewportSize(), so I fix them correctly. > + Correct English please... I hate being grammar police (I'm not a native English speaker myself) but we have to maintain readability standard in the Changelog.
Created attachment 206980 [details] Patch Re-write changelog.
Comment on attachment 206980 [details] Patch Attachment 206980 [details] did not pass mac-ews (mac): Output: http://webkit-queues.appspot.com/results/1090892 New failing tests: media/video-zoom.html
Created attachment 206988 [details] Archive of layout-test-results from webkit-ews-05 for mac-mountainlion The attached test failures were seen while running run-webkit-tests on the mac-ews. Bot: webkit-ews-05 Port: mac-mountainlion Platform: Mac OS X 10.8.3
Created attachment 207049 [details] Patch Rebase.
Created attachment 214554 [details] Patch re-write Changelog.
*** Bug 124396 has been marked as a duplicate of this bug. ***
(In reply to comment #13) > Created an attachment (id=214554) [details] > Patch > > re-write Changelog. I think that we need this. Are there still issues to block this?
Comment on attachment 214554 [details] Patch Clearing flags on attachment: 214554 Committed r159864: <http://trac.webkit.org/changeset/159864>
All reviewed patches have been landed. Closing bug.