Summary: | [iOS][WK2] Set up the resize events | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Benjamin Poulain <benjamin> | ||||||
Component: | New Bugs | Assignee: | Benjamin Poulain <benjamin> | ||||||
Status: | RESOLVED FIXED | ||||||||
Severity: | Normal | CC: | simon.fraser | ||||||
Priority: | P2 | ||||||||
Version: | 528+ (Nightly build) | ||||||||
Hardware: | Unspecified | ||||||||
OS: | Unspecified | ||||||||
Attachments: |
|
Description
Benjamin Poulain
2014-05-08 21:51:38 PDT
Created attachment 231131 [details]
Patch
Comment on attachment 231131 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=231131&action=review This is so tricky that it really demands test coverage; we need to figure out how to test this in an automated way. I’m really concerned that we will break this in the future. > Source/WebCore/WebCore.exp.in:2451 > +__ZN7WebCore9FrameView27setCustomSizeForResizeEventENS_7IntSizeE Should run the sort-export-file script. I’m pretty sure that this will sort lower in the file, because it’s a literal ASCII sort, so 9F will come after 10. > Source/WebKit2/Shared/VisibleContentRectUpdateInfo.h:54 > + , m_isChangingObscuredInsetsInteractively(isChangingObscuredInsetsInteractively) We want to initialize this to false above in the default constructor too. > Source/WebKit2/WebProcess/WebPage/ios/WebPageIOS.mm:2039 > + frameView.setCustomSizeForResizeEvent(roundedIntSize(unobscuredContentRectSizeInContentCoordinates)); I worry about the use of rounded in all the various cases like this, but I suppose it’s not new. Is rounding correct as opposed to ceiling or floor? Created attachment 231189 [details]
Patch
Comment on attachment 231189 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=231189&action=review > Source/WebKit2/ChangeLog:8 > + Wire the UI Process updates to frameview to send the resize events appropriately. FrameView > Source/WebKit2/ChangeLog:29 > + Since JavaScript now gets two chances to change the content, we query the actual scroll offset Can you explain the "now gets two chances" here in the change log? > Source/WebCore/page/FrameView.h:701 > + bool m_useCustomSizeForResizeEvent; You forgot to initialize this in the constructor. Committed r168556: <http://trac.webkit.org/changeset/168556> |