RESOLVED FIXED Bug 202272
Support accessibility for <figure> element on iOS.
https://bugs.webkit.org/show_bug.cgi?id=202272
Summary Support accessibility for <figure> element on iOS.
Andres Gonzalez
Reported 2019-09-26 09:21:12 PDT
Support accessibility for <figure> element on iOS.
Attachments
Patch (8.93 KB, patch)
2019-09-26 09:38 PDT, Andres Gonzalez
no flags
Patch (11.83 KB, patch)
2019-09-27 13:27 PDT, Andres Gonzalez
no flags
Patch (12.27 KB, patch)
2019-09-27 14:38 PDT, Andres Gonzalez
no flags
Patch (1.27 KB, patch)
2019-10-01 09:02 PDT, Andres Gonzalez
no flags
Andres Gonzalez
Comment 1 2019-09-26 09:38:18 PDT
chris fleizach
Comment 2 2019-09-26 10:08:49 PDT
Comment on attachment 379648 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=379648&action=review > Source/WebCore/accessibility/AccessibilityObject.cpp:2618 > + return AXFigureText(); It looks like we also have this line in the macOS wrapper case AccessibilityRole::Figure: return AXFigureText(); I think we can remove that since we're adding this into the lowest level of role description behavior however. I think we still want people to be able to override figure text with their own aria_roledescription maybe this would be better to put in the base accessibility wrapper and remove it from the Mac side...
Andres Gonzalez
Comment 3 2019-09-27 13:27:41 PDT
Andres Gonzalez
Comment 4 2019-09-27 13:33:55 PDT
Fixed the aria-roledescription issue that you pointed out, updated the Mac wrapper and updated the roles-exposed test that was testing the figure element.
chris fleizach
Comment 5 2019-09-27 14:02:21 PDT
Comment on attachment 379758 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=379758&action=review > LayoutTests/accessibility/ios-simulator/figure-element.html:33 > +<figure id="figure5" title="Figure 5."> can you add one more case for a figure that overrides with aria-roledescription ?
Andres Gonzalez
Comment 6 2019-09-27 14:38:29 PDT
Andres Gonzalez
Comment 7 2019-09-27 14:45:54 PDT
figure 4 has aria-roledescription attribute.
WebKit Commit Bot
Comment 8 2019-09-27 15:15:28 PDT
Comment on attachment 379768 [details] Patch Clearing flags on attachment: 379768 Committed r250462: <https://trac.webkit.org/changeset/250462>
WebKit Commit Bot
Comment 9 2019-09-27 15:15:30 PDT
All reviewed patches have been landed. Closing bug.
Radar WebKit Bug Importer
Comment 10 2019-09-27 15:16:20 PDT
Truitt Savell
Comment 11 2019-09-30 15:36:04 PDT
It looks like this change https://trac.webkit.org/changeset/250462/webkit broke accessibility/roles-exposed.html History: https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=accessibility%2Froles-exposed.html Diff: --- /Volumes/Data/slave/highsierra-release-tests-wk1/build/layout-test-results/accessibility/roles-exposed-expected.txt +++ /Volumes/Data/slave/highsierra-release-tests-wk1/build/layout-test-results/accessibility/roles-exposed-actual.txt @@ -129,6 +129,11 @@ AXSubrole: AXRoleDescription: +figure + AXRole: AXGroup + AXSubrole: + AXRoleDescription: figure + footer AXRole: AXGroup AXSubrole: AXLandmarkContentInfo I was able to reproduce this locally. Is this expected behavior and need a rebasline?
chris fleizach
Comment 12 2019-09-30 15:39:28 PDT
(In reply to Truitt Savell from comment #11) > It looks like this change https://trac.webkit.org/changeset/250462/webkit > > broke accessibility/roles-exposed.html > > History: > https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard. > html#showAllRuns=true&tests=accessibility%2Froles-exposed.html > > Diff: > --- > /Volumes/Data/slave/highsierra-release-tests-wk1/build/layout-test-results/ > accessibility/roles-exposed-expected.txt > +++ > /Volumes/Data/slave/highsierra-release-tests-wk1/build/layout-test-results/ > accessibility/roles-exposed-actual.txt > @@ -129,6 +129,11 @@ > AXSubrole: > AXRoleDescription: > > +figure > + AXRole: AXGroup > + AXSubrole: > + AXRoleDescription: figure > + > footer > AXRole: AXGroup > AXSubrole: AXLandmarkContentInfo > > I was able to reproduce this locally. Is this expected behavior and need a > rebasline? Yes, just need a WK1 rebase line
Aakash Jain
Comment 13 2019-10-01 06:58:45 PDT
> Yes, just need a WK1 rebase line Can we rebaseline quickly. This is slowing down EWS.
Andres Gonzalez
Comment 14 2019-10-01 09:02:11 PDT
Reopening to attach new patch.
Andres Gonzalez
Comment 15 2019-10-01 09:02:13 PDT
Andres Gonzalez
Comment 16 2019-10-01 09:05:22 PDT
Latest patch fixes WK1 test. Sorry about this oversight.
Aakash Jain
Comment 17 2019-10-01 09:12:14 PDT
Comment on attachment 379907 [details] Patch r=me if it passes EWS.
Alexey Proskuryakov
Comment 18 2019-10-01 09:29:55 PDT
*** Bug 202404 has been marked as a duplicate of this bug. ***
WebKit Commit Bot
Comment 19 2019-10-01 10:34:49 PDT
Comment on attachment 379907 [details] Patch Clearing flags on attachment: 379907 Committed r250564: <https://trac.webkit.org/changeset/250564>
WebKit Commit Bot
Comment 20 2019-10-01 10:34:51 PDT
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.