Closed Bug 573085 Opened 15 years ago Closed 14 years ago

Intermittent failure in a11y/accessible/test_nsIAccessible_focus.html | Focus wasn't received for element with ID aria-link.

Categories

(Core :: Disability Access APIs, defect)

Product:
Core
Component:
Disability Access APIs
Platform:
x86
macOS
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED WORKSFORME

People

(Reporter: ehsan.akhgari, Assigned: davidb)

References

Details

(Keywords: intermittent-failure)

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1276882231.1276883593.21429.gz Rev3 Fedora 12 mozilla-central opt test mochitest-other on 2010/06/18 10:30:31 11306 INFO TEST-START | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html 11307 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | before wait for focus -- loaded: loading active window: ([object ChromeWindow]) chrome://browser/content/browser.xul focused window: ([object XPCNativeWrapper [object Window]]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html desired window: ([object Window]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html child window: ([object XPCNativeWrapper [object Window]]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html docshell visible: true 11308 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | must wait for load 11309 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | already focused 11310 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | maybe run tests <load:false, focus:true> -- loaded: loading active window: ([object ChromeWindow]) chrome://browser/content/browser.xul focused window: ([object XPCNativeWrapper [object Window]]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html desired window: ([object Window]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html child window: ([object XPCNativeWrapper [object Window]]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html docshell visible: true 11311 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | waitForEvent called <type:load, target[object HTMLDocument]> -- loaded: complete active window: ([object ChromeWindow]) chrome://browser/content/browser.xul focused window: ([object XPCNativeWrapper [object Window]]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html desired window: ([object Window]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html child window: ([object XPCNativeWrapper [object Window]]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html docshell visible: true 11312 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | maybe run tests <load:true, focus:true> -- loaded: complete active window: ([object ChromeWindow]) chrome://browser/content/browser.xul focused window: ([object XPCNativeWrapper [object Window]]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html desired window: ([object Window]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html child window: ([object XPCNativeWrapper [object Window]]) chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html docshell visible: true 11313 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | Focus wasn't received for element with ID aria-link. 11314 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | No focused state for element with ID aria-link. 11315 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | Focus wasn't received for element with ID aria-link2. 11316 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | No focused state for element with ID aria-link2. 11317 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | Focus wasn't received for element with ID link. 11318 INFO TEST-PASS | chrome://mochikit/content/a11y/accessible/test_nsIAccessible_focus.html | No focused state for element with ID link. This was first seen on mozilla-central on this changeset <http://hg.mozilla.org/mozilla-central/rev/d4156799e66a> but bz has seen this on the try server without his patches.
So this is sort of a regression from bug 563327 in that it became more likely to fail. The structure of the test is this: 28 gFocusManager.listenElement(linkAcc, ID, doTest2); 29 linkAcc.takeFocus(); where listenElement does: 68 registerA11yEventListener(nsIAccessibleEvent.EVENT_FOCUS, this); ... 76 window.setTimeout( 77 function(aFocusMgr) 78 { 79 aFocusMgr.checkWasFocusHandled(); 80 }, 0, this); and checkWasFocusHandled checks that the EVENT_FOCUS fired. Now what this means is that we have a race between the async EVENT_FOCUS firing and the async timeout execution. The former presumably happens when we flush the event queue. The latter happens whenever it happens; it's actually rather complicated. But it's not likely to happen less than 10ms after the setTimeout call, whereas the queue flush _used_ to be likely to thus happen and is now less likely to do. Ideally, we'd run the checkWasFocusHandled after the a11y queue flush happens, but we don't really have a way to detect that, right? The other option is to do the check off an async timer from the notification I plan to add in bug 569520.
Blocks: 563327
I just pushed http://hg.mozilla.org/mozilla-central/rev/0bb70936ae1c to bump the timeout for now, in hopes of reducing incidence of orange.
(In reply to comment #7) Makes sense. > Ideally, we'd run the checkWasFocusHandled after the a11y queue flush happens, > but we don't really have a way to detect that, right? Not currently. > The other option is to do the check off an async timer from the notification I > plan to add in bug 569520. Ok not sure what notification you will add there. It might make sense. (In reply to comment #8) > I just pushed http://hg.mozilla.org/mozilla-central/rev/0bb70936ae1c to bump > the timeout for now, in hopes of reducing incidence of orange. Thanks.
(In reply to comment #10) > > The other option is to do the check off an async timer from the notification I > > plan to add in bug 569520. > > Ok not sure what notification you will add there. One that will tell script when a refresh is in process (and the refresh will make a sync call to the queue flush in a11y).
(In reply to comment #11) > (In reply to comment #10) > > > The other option is to do the check off an async timer from the notification I > > > plan to add in bug 569520. > > > > Ok not sure what notification you will add there. > > One that will tell script when a refresh is in process (and the refresh will > make a sync call to the queue flush in a11y). OK great!
Depends on: 569520
I'll own this for now, we should be able to check off of bz's async timer :)
Assignee: nobody → bolterbugz
no reports during half of year, close as worksforme
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → WORKSFORME
Whiteboard: [orange]
You need to log in before you can comment on or make changes to this bug.