[Webkit-unassigned] [Bug 136291] New: Style's "altText" field (webkit-alt property) is not properly managed.

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Aug 27 10:02:34 PDT 2014


https://bugs.webkit.org/show_bug.cgi?id=136291

           Summary: Style's "altText" field (webkit-alt property) is not
                    properly managed.
           Product: WebKit
           Version: 528+ (Nightly build)
          Platform: Unspecified
        OS/Version: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Accessibility
        AssignedTo: webkit-unassigned at lists.webkit.org
        ReportedBy: jfernandez at igalia.com
                CC: webkit-bug-importer at group.apple.com


The recent support of -webkit-alt property for dealing with alternative text in IMG html tags is not properly managed by the RenderStyle (StyleNonInheritedData) and ContentData (ImageContentData) classes. 

As the rest of the fields defined in the RenderStyle class, via the XXXData classes, the constructor using an already instantiated object should copy the field value, in this case the altText and ContentData.altText fields. The bug is not easily reproduced, but its root cause comes from the fact that the StyleNonInheritedData (and the other XXData classes) is defined using the DataRef template, so it uses the access() method to handle its fields. The access() method creates a new instance the first time is called, and a copy reference otherwise. 

So, the bug consists of the altText String pointer being lost when accessing the StyleNonInheritedData fields during the CSS cascade, for instance. This seems to happen, at least during the testing I've done so far, only when setting the webkit-alt string directly; I have not been able to reproduce the issue using the html attribute mechanisms.

This issue seems to affect only the LayoutTests/accessibility/webkit-alt-for-css-content.html tests, which on the other hand is the only one using/testing this new property. 

Finally, I wonder why the mentioned tests is mac specific; I think the test and the property are general enough to be cross platform. So, I'd like to use this bug to define a new cross-platform test.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.



More information about the webkit-unassigned mailing list