Closed Bug 1844058 Opened 8 months ago Closed 3 months ago

"Filter rules containing this property" control is not accessible with keyboard alone


(DevTools :: Inspector: Rules, defect, P2)



(Accessibility Severity:s2, firefox122 fixed)

122 Branch
Accessibility Severity s2
Tracking Status
firefox122 --- fixed


(Reporter: nchevobbe, Assigned: nchevobbe)


(Blocks 1 open bug)


(Keywords: access)


(2 files)

“Filter rules containing this property” control is not accessible with keyboard alone and a screen reader can only focus it in the browse/read all mode, the role is missing as well. NVDA cannot focus the element even in the browse mode and it splits the announcement in two strings: “clickable Filter rules containing” then (press Down Arrow) “this property”

Accessibility Severity: --- → s2
Severity: S2 → S3
Priority: -- → P2

The severity field for this bug is set to S3. However, the accessibility severity is higher, .
:jdescottes, could you consider increasing the severity?

For more information, please visit BugBot documentation.

Flags: needinfo?(jdescottes)
Flags: needinfo?(jdescottes)
Depends on: 1861674
Assignee: nobody → nchevobbe
Pushed by
[devtools] Use <button> for Rules view filter property icon. r=devtools-reviewers,accessibility-frontend-reviewers,ayeddi,ochameau.
Closed: 3 months ago
Resolution: --- → FIXED
Target Milestone: --- → 122 Branch

== Change summary for alert #40403 (as of Sun, 26 Nov 2023 17:08:12 GMT) ==


Ratio Test Platform Options Absolute values (old vs new)
6% damp custom.inspector.manyrules.deselectnode windows10-64-shippable-qr e10s fission stylo webrender 197.65 -> 210.21
6% damp custom.inspector.manyrules.deselectnode windows10-64-shippable-qr e10s fission stylo webrender-sw 200.53 -> 211.96
5% damp custom.inspector.manyrules.deselectnode linux1804-64-shippable-qr e10s fission stylo webrender-sw 234.82 -> 247.60

For up to date results, see:

"good" profile:
"bad" profile:

(both on OSX though, and the regression did not seem to occur on osx)

Not obvious what's taking more time here, but this is a bit surprising that replacing a div with a button has such an impact
Emilio, would you know from the top of your head what could explain such regression?

Flags: needinfo?(emilio)

Not off-hand, no. There's some overflow event there that happens in one profile but not other, maybe some styles on windows unexpectedly causing overflow and more JS to run?

Flags: needinfo?(emilio)
You need to log in before you can comment on or make changes to this bug.