Bug 73128 - Clean up code in WK2/ChromeClientClient related to viewport handling
Summary: Clean up code in WK2/ChromeClientClient related to viewport handling
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebKit2 (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
Depends on:
Reported: 2011-11-25 05:21 PST by Kenneth Rohde Christiansen
Modified: 2011-11-25 05:53 PST (History)
2 users (show)

See Also:

Patch (6.46 KB, patch)
2011-11-25 05:22 PST, Kenneth Rohde Christiansen
hausmann: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Kenneth Rohde Christiansen 2011-11-25 05:21:25 PST
1) Make sure that we only call resizeToContentsIfNeeded when we are using the tiled backing store and fixed layout as well.

2) Guard the scrollbar code so that it is only called when scroll delegation is not used. Do similarily for the frameset code, which makes no sense with frame flattening.

Comment 1 Kenneth Rohde Christiansen 2011-11-25 05:22:34 PST
Created attachment 116603 [details]
Comment 2 zalan 2011-11-25 05:43:06 PST
a small observation.
+            This method is only called for the main frame, so the main frame
+            check has been removed.
It looks to me a null check rather than a mainframe check. However, m_page->useFixedLayout() does eliminate the need for the null check, so the code is correct.
Comment 3 Simon Hausmann 2011-11-25 05:47:06 PST
Comment on attachment 116603 [details]

View in context: https://bugs.webkit.org/attachment.cgi?id=116603&action=review

r=me but I'll let you decide on cq+ based on the #ifdef comment

> Source/WebKit2/WebProcess/WebCoreSupport/WebChromeClient.cpp:448
> +    if (m_page->useFixedLayout())
> +        m_page->resizeToContentsIfNeeded();
> +#endif

For WK2 it does seem a bit strange to have USE(TILED_BACKING_STORE) for this. Do we actually need the #ifdef?
Comment 4 Kenneth Rohde Christiansen 2011-11-25 05:53:24 PST
Comment on attachment 116603 [details]

landed in r101167