Bug 124836

Summary: AX: Add accessibility tests for MathML operators
Product: WebKit Reporter: Frédéric Wang (:fredw) <fred.wang>
Component: AccessibilityAssignee: Frédéric Wang (:fredw) <fred.wang>
Status: RESOLVED FIXED    
Severity: Normal CC: aboxhall, apinheiro, cfleizach, commit-queue, dmazzoni, jcraig, jdiggs, mario, samuel_white, webkit-bug-importer
Priority: P2 Keywords: InRadar
Version: 528+ (Nightly build)   
Hardware: All   
OS: All   
Bug Depends on: 99620, 139582    
Bug Blocks: 159208    
Attachments:
Description Flags
Patch
buildbot: commit-queue-
Archive of layout-test-results from ews126 for ios-simulator-wk2
none
Patch to get test expectations
none
Patch jdiggs: review+

Description Frédéric Wang (:fredw) 2013-11-25 04:22:53 PST
This is for when bug 99620 is fixed.

I'm not sure I see tests for

AccessibilityRenderObject::isMathOperator()                 (<mo> renderers or the anonymous created by <mfenced>)
AccessibilityRenderObject::isMathFenceOperator()            (fences created by <mfenced>)
AccessibilityRenderObject::isMathSeparatorOperator()        (separators created by <mfenced>)

After bugs 99620 is fixed, we will have the following attributes:

<mo fence="true"> => isMathFenceOperator
<mo separator="true"> => isMathSeparatorOperator
<mo separator="true" fence="true"> => isMathFenceOperator && isMathSeparatorOperator (not sure this really exists in real math)

Also, the values can be determined from the operator dictionary (http://www.w3.org/TR/MathML3/appendixc.html). For example the two markup should be equivalent for the AT:

<math><mo>(</mo><mi>x</mi><mo>,<mo><mi>y</mi><mo>;</mo><mo>]</mo></math>

and with explicit attributes

<math><mo fence="true">(</mo><mi>x</mi><mo separator="true">,<mo><mi>y</mi><mo separator="true">;</mo><mo fence="true">]</mo></math>

Similarly, we should have the equivalence between mfenced and its expanded form. See http://www.w3.org/TR/MathML3/chapter3.html#presm.mfenced
Comment 1 Frédéric Wang (:fredw) 2013-11-25 04:24:59 PST
There is also the separate questions of whether we want to expose other operator properties to the AT. Most of the new properties are purely visual but perhaps "LargeOp" and "Accent" have a semantic meaning that is worth exposing.
Comment 2 Radar WebKit Bug Importer 2014-02-20 00:24:02 PST
<rdar://problem/16119302>
Comment 3 Frédéric Wang (:fredw) 2014-02-20 00:29:55 PST
@Chris: now that bug 99620 is fixed, I let you ensure correct test coverage for the AT and decide whether it is worth exposing things like "LargeOp" or "Accent".
Comment 4 Frédéric Wang (:fredw) 2014-03-12 07:33:57 PDT
For the record: the accessibility test has been disabled in http://trac.webkit.org/changeset/165436 because the change broke the <mfenced> test. We need to be restore it and ensure it works with normal <mo> elements too.
Comment 5 Frédéric Wang (:fredw) 2016-06-28 03:08:59 PDT
Created attachment 282234 [details]
Patch
Comment 6 Build Bot 2016-06-28 04:09:25 PDT
Comment on attachment 282234 [details]
Patch

Attachment 282234 [details] did not pass ios-sim-ews (ios-simulator-wk2):
Output: http://webkit-queues.webkit.org/results/1585623

New failing tests:
animations/multiple-backgrounds.html
Comment 7 Build Bot 2016-06-28 04:09:29 PDT
Created attachment 282240 [details]
Archive of layout-test-results from ews126 for ios-simulator-wk2

The attached test failures were seen while running run-webkit-tests on the ios-sim-ews.
Bot: ews126  Port: ios-simulator-wk2  Platform: Mac OS X 10.11.4
Comment 8 Frédéric Wang (:fredw) 2016-06-28 04:58:13 PDT
Created attachment 282243 [details]
Patch to get test expectations
Comment 9 Frédéric Wang (:fredw) 2016-06-28 10:06:14 PDT
Created attachment 282252 [details]
Patch

Adding the mac expectation...
Comment 10 Joanmarie Diggs 2016-06-28 14:43:25 PDT
Comment on attachment 282252 [details]
Patch

Yay for more tests! :) Thanks!
Comment 11 Frédéric Wang (:fredw) 2016-06-28 21:32:08 PDT
Committed r202610: <http://trac.webkit.org/changeset/202610>