Summary: | AX: HTML5 landmark (and related) elements should not be ignored. | ||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Takeshi Kurosawa <taken.spc> | ||||||||||||
Component: | Accessibility | Assignee: | Nobody <webkit-unassigned> | ||||||||||||
Status: | RESOLVED FIXED | ||||||||||||||
Severity: | Normal | CC: | aboxhall, apinheiro, cfleizach, commit-queue, dmazzoni, faulkner.steve, jcraig, jdiggs, mario, samuel_white, webkit-bug-importer | ||||||||||||
Priority: | P2 | Keywords: | InRadar | ||||||||||||
Version: | 528+ (Nightly build) | ||||||||||||||
Hardware: | All | ||||||||||||||
OS: | All | ||||||||||||||
Bug Depends on: | |||||||||||||||
Bug Blocks: | 109013 | ||||||||||||||
Attachments: |
|
Created attachment 235219 [details]
Patch
Patch.
Attachment 235219 [details] did not pass style-queue:
ERROR: Source/WebCore/accessibility/AccessibilityRenderObject.cpp:1257: A case label should not be indented, but line up with its switch statement. [whitespace/indent] [4]
Total errors found: 1 in 8 files
If any of these errors are false positives, please file a bug against check-webkit-style.
Comment on attachment 235219 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=235219&action=review patch looks ok besides style issue. thanks > Source/WebCore/accessibility/AccessibilityRenderObject.cpp:1257 > + case AudioRole: don't indent here Created attachment 235248 [details]
Patch (indetation fix)
Fixed indentation.
Comment on attachment 235248 [details] Patch (indetation fix) View in context: https://bugs.webkit.org/attachment.cgi?id=235248&action=review > Source/WebCore/accessibility/AccessibilityObject.cpp:320 > + || role == FooterRole We definitely don't want this. <footer> can be used hundreds of times per page. For example, each post in a Facebook feed can have a footer. There's another open bug about Footer versus ContentInfo. This would make all of them landmarks. You only want footer to map to content info when it's not a descendant of an article or section: http://www.w3.org/html/wg/drafts/html/CR/dom.html#sec-strong-native-semantics Just pull this out and it'd be okay to address the rest in a separate bug. related to bug 109013 (In reply to comment #6) > You only want footer to map to content info when it's not a… Should have read: > You only want footer to *be a landmark* when it's not a… Created attachment 235286 [details]
Patch (drop footer element)
Drop footer element as per review comment
Created attachment 235287 [details]
Patch (drop footer element)
Oops! The previous patch contains wrong ChangeLogs.
Comment on attachment 235287 [details]
Patch (drop footer element)
thanks
Thanks Takeshi! Let us know if you need a CQ+, too. Comment on attachment 235287 [details]
Patch (drop footer element)
Adding CQ+
Comment on attachment 235287 [details] Patch (drop footer element) Clearing flags on attachment: 235287 Committed r173428: <http://trac.webkit.org/changeset/173428> All reviewed patches have been landed. Closing bug. (In reply to comment #6) > (From update of attachment 235248 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=235248&action=review > > > Source/WebCore/accessibility/AccessibilityObject.cpp:320 > > + || role == FooterRole > > We definitely don't want this. <footer> can be used hundreds of times per page. For example, each post in a Facebook feed can have a footer. There's another open bug about Footer versus ContentInfo. This would make all of them landmarks. > > You only want footer to map to content info when it's not a descendant of an article or section: > http://www.w3.org/html/wg/drafts/html/CR/dom.html#sec-strong-native-semantics > > Just pull this out and it'd be okay to address the rest in a separate bug. - header (LandmarkBannerRole) is same "header element that is not a descendant of an article or section element." http://www.w3.org/html/wg/drafts/html/CR/dom.html#sec-strong-native-semantics the original implementation of these requirements (for header/footer) were in webkit, Firefox and Chrome followed suit. Steve, if that is not working as expected, we'll need a second bug for that work. |
Created attachment 235216 [details] Testcase WebKit ignores and doesn't expose following HTML5 landmark (and related) elements when they have block render object children: - header (LandmarkBannerRole) - aside (LandmarkComplementaryRole) - footer (FooterRole) - address (LandmarkContentInfoRole) - main (LandmarkMainRole) - nav (LandmarkNavigationRole) - article (DocumentArticleRole) - section (DocumentRegionRole) - dt (DescriptionListTermRole) - dd (DescriptionListDetailRole) We should not ignore these elements and should expose them. # Although dt and dd are not related to landmark, we should expose.