Reviewed by Beth, approved in principle by Adele, who couldn't be troubled
authorggaren <ggaren@268f45cc-cd09-0410-ab3c-d52691b4dbfc>
Fri, 20 Oct 2006 19:14:17 +0000 (19:14 +0000)
committerggaren <ggaren@268f45cc-cd09-0410-ab3c-d52691b4dbfc>
Fri, 20 Oct 2006 19:14:17 +0000 (19:14 +0000)
commit8a51b2075ea79c2e43927f1132288d2872f3b23e
tree669e6191783210c146eab24950ad5afa9cf57111
parentb210dd4b0cc30a8c6b6dcf112b8ab1b72cb2c01d
    Reviewed by Beth, approved in principle by Adele, who couldn't be troubled
        for an actual review.

        Fixed http://bugs.webkit.org/show_bug.cgi?id=11345
        1 fewer shouldChangeSelection delegate call for fast/forms/focus-control-to-page

        I updated this test to use <input type="search"> because that's the only
        remaining AppKit control, and the test's intention is verify tabbing
        between AppKit and non-AppKit controls.

        I updated the results because the behavior change is good.

        Here's why the test regressed:

        Our editing code is cra-zay. When AppKit calls nextKeyView, we not only
        calculate the next key element in the page, we also focus it as a side-effect.
        As a result, we used to fire a focus delegate twice, because AppKit happened
        to call nextKeyView twice.

        This test first regressed when Adele changed listboxes to be non-AppKit
        controls (r16718). That change took AppKit out of test.

        The regression redoubled itself when Adam moved TAB handling into the DOM
        (r16741), taking AppKit out of TABing completely.

        * fast/forms/focus-control-to-page-expected.txt:
        * fast/forms/focus-control-to-page.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@17166 268f45cc-cd09-0410-ab3c-d52691b4dbfc
LayoutTests/ChangeLog
LayoutTests/fast/forms/focus-control-to-page-expected.txt
LayoutTests/fast/forms/focus-control-to-page.html