Bug 74014

Created attachment 118254 [details] Patch

Comment on attachment 118254 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=118254&action=review > Source/WebCore/testing/Internals.idl:88 > + float getDeviceScaleFactor(in Document document) raises(DOMException); Drop the 'get'

Created attachment 118259 [details] Patch

(In reply to comment #2) > (From update of attachment 118254 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=118254&action=review > > > Source/WebCore/testing/Internals.idl:88 > > + float getDeviceScaleFactor(in Document document) raises(DOMException); > > Drop the 'get' Done. I uploaded a new patch to make sure bots go green first.

Attachment 118259 [details] did not pass style-queue: Failed to run "['Tools/Scripts/update-webkit']" exit_code: 9 Updating OpenSource Index mismatch: 26f0902d4a952ed5da5ec33d1949597fa79abdda != e9bcf9b2e57e236c20b02d45c6e274893b01a1f7 rereading f1b18930bbf5e73c83f2b0018b501e8f64330752 M Source/JavaScriptCore/wtf/HashTraits.h M Source/JavaScriptCore/ChangeLog M Source/WebCore/dom/ChildListMutationScope.cpp M Source/WebCore/ChangeLog 102267 = 67d4f2de4c2f4a2b6f1b3adfd20b50620fad3051 already exists! Why are we refetching it? at /usr/lib/git-core/git-svn line 5210 Died at Tools/Scripts/update-webkit line 158. If any of these errors are false positives, please file a bug against check-webkit-style.

Comment on attachment 118259 [details] Patch Attachment 118259 [details] did not pass gtk-ews (gtk): Output: http://queues.webkit.org/results/10750291

Created attachment 121288 [details] Patch

I have refreshed the patch. Hopefully the EWS bots are happy.

Comment on attachment 121288 [details] Patch Attachment 121288 [details] did not pass gtk-ews (gtk): Output: http://queues.webkit.org/results/11120421

Created attachment 121289 [details] Patch

Using EWS bots to fix build issues on other platforms...

Comment on attachment 121289 [details] Patch Attachment 121289 [details] did not pass gtk-ews (gtk): Output: http://queues.webkit.org/results/11144011

Comment on attachment 121289 [details] Patch Please use LayoutTestController instead of Internals because setting/getting the device scale factor can be done by WebKit public API. The Internals object is for inspecting hidden, internal state of WebCore. It's OK to take care of Chromium for now, skip for other ports if you file bugs for them.

(In reply to comment #13) > (From update of attachment 121289 [details]) > Please use LayoutTestController instead of Internals because setting/getting the device scale factor can be done by WebKit public API. > The Internals object is for inspecting hidden, internal state of WebCore. > > It's OK to take care of Chromium for now, skip for other ports if you file bugs for them. windows.internals is cross platform. There's already page scaling code in windows.internals, and this code is related. There's no reason why this parameter has to behave differently between platforms.

I guess not all platforms will make a public API for setting this but instead hardcode it per platform, so I guess internals is totally ok for this.

(In reply to comment #13) > (From update of attachment 121289 [details]) > Please use LayoutTestController instead of Internals because setting/getting the device scale factor can be done by WebKit public API. > The Internals object is for inspecting hidden, internal state of WebCore. > > It's OK to take care of Chromium for now, skip for other ports if you file bugs for them. Morita, This is a fairly cross-platform setting and introducing a layout test suite for scaling and fixed layout seems to be beneficial for many ports of WebKit. Do you think it makes sense to put this into window.internals? There was a thread on webkit-dev about this a few minutes ago and the consensus seemed to be moving code into window.internals is a good idea overall. Thanks.