refactor: derive INTERACTIVE_ROLES from aria-query taxonomy (+ widen menu-pattern exception) by johanrd · Pull Request #27 · johanrd/eslint-plugin-ember

johanrd · 2026-04-21T06:57:07Z

Note

This is part of a series where Claude has audited eslint-plugin-ember against jsx-a11y, vuejs-accessibility, angular-eslint, lit-a11y and html-validate, ember-template-lint, and the HTML and WCAG specs.

Consolidates two hand-maintained INTERACTIVE_ROLES sets (in template-no-invalid-interactive and template-no-nested-interactive) into a single derivation sourced from aria-query's role taxonomy, and widens the composite-widget nesting exception so canonical WAI-ARIA APG patterns are not flagged.

Derivation

New lib/utils/interactive-roles.js — one shared set used by both rules:

const { roles } = require('aria-query');

function buildInteractiveRoleSet() {
  const result = new Set(['toolbar']);
  for (const [role, def] of roles) {
    if (def.abstract) continue;
    const descendsFromWidget = (def.superClass || []).some((chain) => chain.includes('widget'));
    if (descendsFromWidget) result.add(role);
  }
  return result;
}

All non-abstract roles with widget in any superClass chain per aria-query's taxonomy.
Plus explicit toolbar — it supports aria-activedescendant and is widget-like in practice but does not descend from widget in aria-query's data. Matches jsx-a11y's same manual addition.
tooltip is not included. Per WAI-ARIA 1.2 §5.3.3 "Document Structure Roles", tooltip is a document-structure role, not a widget; the spec says "The following roles describe structures that organize content in a page. Document structures are not usually interactive." aria-query confirms: tooltip.superClass is [['roletype', 'structure', 'section']] — no widget.

Matches jsx-a11y's isInteractiveRole derivation verbatim. (Note: lit-a11y uses the same derivation but explicitly excludes progressbar because its value is effectively readonly.)

Final set: 35 roles.

Composite-widget nesting exception

Previously template-no-nested-interactive hard-coded a single exception: menuitem-inside-menuitem. This PR widens it using a derived COMPOSITE_WIDGET_CHILDREN map driven by aria-query's requiredOwnedElements (a documented ARIA concept — WAI-ARIA 1.2 §5.2.6 Required Owned Elements), with superClass inheritance applied so treegrid inherits from both grid and tree.

Note: the superClass-based inheritance step is a composition of two spec facts (the superClass relation + requiredOwnedElements on ancestors), not a verbatim spec rule — it's a reasonable inference that matches APG patterns, but strictly speaking derived rather than directly cited.

Plus an explicit MENUITEM_ROLES → menu rule for the submenu direction (documented in APG's Menu pattern; aria-query does not encode this via requiredOwnedElements).

Relation to surrounding branches

lib/utils/is-native-element.js (+83 lines) has no consumer on this branch. It's included as a byte-identical copy of the canonical version owned by refactor: extract isNativeElement util (fix component-vs-HTML-tag misclassification) #50 (fix/is-component-invocation-util) so that whichever of refactor: derive INTERACTIVE_ROLES from aria-query taxonomy (+ widen menu-pattern exception) #27 / refactor: extract isNativeElement util (fix component-vs-HTML-tag misclassification) #50 / sibling branches merges first, the rest don't pick up a conflict on this file. Happy to drop it from this PR and defer to refactor: extract isNativeElement util (fix component-vs-HTML-tag misclassification) #50 if review prefers.
<details> handling comes from master via rebase. The rule now uses isAllowedDetailsChild (introduced by BUGFIX: false positive: interactive flow content inside <details> ember-cli/eslint-plugin-ember#2713 on master — summary-first-child carve-out plus flow content in the disclosed panel). No change to that helper in this PR; called out because a reviewer looking at the pre-rebase diff will see the older isSummaryFirstChildOfDetails disappear.

github-actions · 2026-04-21T06:57:49Z

🏎️ Benchmark Comparison

	Benchmark	Control (p50)	Experiment (p50)	Δ
⚪	js small	13.28 ms	13.38 ms	+0.7%
🟢	js medium	6.75 ms	6.45 ms	-4.5%
⚪	js large	2.60 ms	2.58 ms	-0.8%
⚪	gjs small	1.11 ms	1.11 ms	+0.3%
⚪	gjs medium	554.44 µs	553.00 µs	-0.3%
⚪	gjs large	220.43 µs	219.14 µs	-0.6%
⚪	gts small	1.10 ms	1.10 ms	+0.1%
⚪	gts medium	552.62 µs	552.22 µs	-0.1%
⚪	gts large	218.21 µs	220.24 µs	+0.9%

🟢 faster · 🔴 slower · 🟠 slightly slower · ⚪ within 2%

Full mitata output

clk: ~2.76 GHz
cpu: AMD EPYC 9V74 80-Core Processor
runtime: node 24.15.0 (x64-linux)

benchmark                   avg (min … max) p75 / p99    (min … top 1%)
------------------------------------------- -------------------------------
js small (control)            16.53 ms/iter  18.57 ms █                    
                      (11.33 ms … 33.36 ms)  29.02 ms █ █                  
                    (  5.70 mb …  10.13 mb)   7.25 mb ███▃█▁▁█▃▃▃▃▃▁▁▁▁▃▆▁▆

js small (experiment)         13.97 ms/iter  14.52 ms    █▅                
                      (11.78 ms … 20.06 ms)  20.05 ms █▃███▃ ▃             
                    (  6.19 mb …   7.86 mb)   6.83 mb ████████▄▄▄██▁▄▁▁▁▁▁▄

                             ┌                                            ┐
                             ╷┌───────────┬────┐                          ╷
          js small (control) ├┤           │    ├──────────────────────────┤
                             ╵└───────────┴────┘                          ╵
                              ╷ ┌───┬┐             ╷
       js small (experiment)  ├─┤   │├─────────────┤
                              ╵ └───┴┘             ╵
                             └                                            ┘
                             11.33 ms           20.17 ms           29.02 ms

summary
  js small (experiment)
   1.18x faster than js small (control)

------------------------------------------- -------------------------------
js medium (control)            7.48 ms/iter   7.66 ms  █                   
                       (6.19 ms … 16.75 ms)  14.83 ms ▆█▄                  
                    (  2.66 mb …   4.61 mb)   3.54 mb ███▄▄▅▃▃▄▁▁▁▂▁▁▁▂▁▁▁▃

js medium (experiment)         7.25 ms/iter   7.76 ms  █                   
                       (6.05 ms … 14.02 ms)  12.57 ms  █                   
                    (  3.01 mb …   4.18 mb)   3.53 mb ▇█▆▃▂▅▂▃▄▂▁▂▁▁▂▃▂▁▁▁▂

                             ┌                                            ┐
                              ╷┌────┬┐                                    ╷
         js medium (control)  ├┤    │├────────────────────────────────────┤
                              ╵└────┴┘                                    ╵
                             ╷┌────┬──┐                       ╷
      js medium (experiment) ├┤    │  ├───────────────────────┤
                             ╵└────┴──┘                       ╵
                             └                                            ┘
                             6.05 ms           10.44 ms            14.83 ms

summary
  js medium (experiment)
   1.03x faster than js medium (control)

------------------------------------------- -------------------------------
js large (control)             3.10 ms/iter   2.92 ms  █                   
                       (2.17 ms … 11.47 ms)   8.49 ms ██▅                  
                    (241.26 kb …   3.11 mb)   1.44 mb ███▄▄▃▂▁▂▂▂▂▂▂▁▁▁▁▁▂▁

js large (experiment)          2.85 ms/iter   2.74 ms  █                   
                        (2.35 ms … 7.86 ms)   5.96 ms ▅█▃                  
                    (250.08 kb …   2.62 mb)   1.43 mb ███▂▂▂▃▂▂▂▁▁▁▁▁▁▁▁▁▁▂

                             ┌                                            ┐
                             ╷┌─────┬                                     ╷
          js large (control) ├┤     │─────────────────────────────────────┤
                             ╵└─────┴                                     ╵
                              ╷┌──┬                     ╷
       js large (experiment)  ├┤  │─────────────────────┤
                              ╵└──┴                     ╵
                             └                                            ┘
                             2.17 ms            5.33 ms             8.49 ms

summary
  js large (experiment)
   1.09x faster than js large (control)

------------------------------------------- -------------------------------
gjs small (control)            1.22 ms/iter   1.13 ms █                    
                        (1.08 ms … 5.90 ms)   5.61 ms █                    
                    (250.78 kb …   1.60 mb)   1.06 mb █▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

gjs small (experiment)         1.24 ms/iter   1.12 ms █                    
                        (1.08 ms … 6.43 ms)   5.64 ms █                    
                    (220.70 kb …   1.79 mb)   1.06 mb █▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

                             ┌                                            ┐
                             ┌┬                                           ╷
         gjs small (control) ││───────────────────────────────────────────┤
                             └┴                                           ╵
                             ┌─┬                                          ╷
      gjs small (experiment) │ │──────────────────────────────────────────┤
                             └─┴                                          ╵
                             └                                            ┘
                             1.08 ms            3.36 ms             5.64 ms

summary
  gjs small (control)
   1.01x faster than gjs small (experiment)

------------------------------------------- -------------------------------
gjs medium (control)         610.91 µs/iter 565.83 µs █                    
                      (534.77 µs … 5.29 ms)   2.96 ms █                    
                    ( 97.90 kb …   1.04 mb) 541.23 kb █▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

gjs medium (experiment)      599.82 µs/iter 559.37 µs █                    
                      (533.59 µs … 5.46 ms)   2.00 ms █                    
                    ( 74.82 kb …   1.24 mb) 540.72 kb █▃▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

                             ┌                                            ┐
                             ┌┬                                           ╷
        gjs medium (control) ││───────────────────────────────────────────┤
                             └┴                                           ╵
                             ┌┬                         ╷
     gjs medium (experiment) ││─────────────────────────┤
                             └┴                         ╵
                             └                                            ┘
                             533.59 µs           1.75 ms            2.96 ms

summary
  gjs medium (experiment)
   1.02x faster than gjs medium (control)

------------------------------------------- -------------------------------
gjs large (control)          257.97 µs/iter 229.98 µs █                    
                      (212.81 µs … 6.89 ms) 766.04 µs █                    
                    (  7.94 kb … 777.59 kb) 217.40 kb ██▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

gjs large (experiment)       239.17 µs/iter 226.55 µs  █                   
                      (212.61 µs … 4.87 ms) 304.07 µs  █▃                  
                    ( 16.88 kb … 738.41 kb) 216.38 kb ▅██▅▇▃▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁

                             ┌                                            ┐
                             ┌───┬                                        ╷
         gjs large (control) │   │────────────────────────────────────────┤
                             └───┴                                        ╵
                             ┌─┬    ╷
      gjs large (experiment) │ │────┤
                             └─┴    ╵
                             └                                            ┘
                             212.61 µs         489.32 µs          766.04 µs

summary
  gjs large (experiment)
   1.08x faster than gjs large (control)

------------------------------------------- -------------------------------
gts small (control)            1.20 ms/iter   1.11 ms █                    
                        (1.07 ms … 5.74 ms)   5.35 ms █                    
                    (648.78 kb …   1.57 mb)   1.06 mb █▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

gts small (experiment)         1.22 ms/iter   1.12 ms █                    
                        (1.07 ms … 6.28 ms)   5.38 ms █                    
                    (504.59 kb …   1.63 mb)   1.05 mb █▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

                             ┌                                            ┐
                             ┌┬                                           ╷
         gts small (control) ││───────────────────────────────────────────┤
                             └┴                                           ╵
                             ┌─┬                                          ╷
      gts small (experiment) │ │──────────────────────────────────────────┤
                             └─┴                                          ╵
                             └                                            ┘
                             1.07 ms            3.23 ms             5.38 ms

summary
  gts small (control)
   1.02x faster than gts small (experiment)

------------------------------------------- -------------------------------
gts medium (control)         598.95 µs/iter 559.63 µs █                    
                      (533.92 µs … 5.11 ms)   1.54 ms █                    
                    ( 59.05 kb …   1.26 mb) 541.47 kb ██▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

gts medium (experiment)      594.89 µs/iter 559.75 µs █                    
                      (532.64 µs … 5.46 ms)   1.83 ms █                    
                    (  1.42 kb …   1.07 mb) 540.45 kb █▄▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

                             ┌                                            ┐
                             ┌─┬                                ╷
        gts medium (control) │ │────────────────────────────────┤
                             └─┴                                ╵
                             ┌─┬                                          ╷
     gts medium (experiment) │ │──────────────────────────────────────────┤
                             └─┴                                          ╵
                             └                                            ┘
                             532.64 µs           1.18 ms            1.83 ms

summary
  gts medium (experiment)
   1.01x faster than gts medium (control)

------------------------------------------- -------------------------------
gts large (control)          240.44 µs/iter 225.77 µs ██                   
                      (211.20 µs … 4.71 ms) 439.40 µs ██                   
                    ( 17.26 kb … 802.22 kb) 216.89 kb ███▂▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁

gts large (experiment)       249.00 µs/iter 230.51 µs  █                   
                      (212.29 µs … 5.01 ms) 409.08 µs ▆█▂                  
                    (183.84 kb … 776.98 kb) 216.62 kb ███▄▂▂▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁

                             ┌                                            ┐
                             ╷┌────┬                                      ╷
         gts large (control) ├┤    │──────────────────────────────────────┤
                             ╵└────┴                                      ╵
                             ╷┌─────┬                               ╷
      gts large (experiment) ├┤     │───────────────────────────────┤
                             ╵└─────┴                               ╵
                             └                                            ┘
                             211.20 µs         325.30 µs          439.40 µs

summary
  gts large (control)
   1.04x faster than gts large (experiment)

Import INTERACTIVE_ROLES and COMPOSITE_WIDGET_CHILDREN from the shared lib/utils/interactive-roles.js util (introduced in #27 — byte-identical copy here so either PR can land first without conflict). Drop the hardcoded 19-role set previously duplicated inline in each rule. Behavior changes: - ARIA widget role set expands from 19 to 35 roles — picks up menubar, menu, listbox, tree, tablist, grid, treegrid, radiogroup, alertdialog, progressbar, and other widget-descended roles in aria-query's taxonomy that the hardcoded list missed. - tooltip is no longer treated as interactive. Per WAI-ARIA 1.2 §5.3.3, tooltip is a document-structure role, not a widget. #27's util reflects this (tooltip explicitly excluded). Old <div role="tooltip" onclick> test moves from valid to invalid — spec-correct. - Composite-widget nesting exception expanded via COMPOSITE_WIDGET_CHILDREN. Canonical APG patterns (<ul role="menubar"><li role="menuitem">..., <ul role="listbox"><li role="option">..., grid/row/gridcell, treegrid, radiogroup/radio) no longer flag as nested-interactive. Previously the rule only handled menuitem-in-menuitem explicitly. Pairs with #37's HTML-content-model authority split to complete the two-authority architecture: HTML §3.2.5.2.7 via html-interactive-content, ARIA widget taxonomy via interactive-roles. Each util cites one authority honestly; rules compose both.

Copilot

Pull request overview

Refactors accessibility role handling by centralizing INTERACTIVE_ROLES derivation from aria-query and expanding template-no-nested-interactive to allow canonical composite-widget role hierarchies (per requiredOwnedElements, plus an APG submenu exception).

Changes:

Introduce lib/utils/interactive-roles.js to derive INTERACTIVE_ROLES from aria-query (+ manual toolbar, exclude tooltip) and to compute COMPOSITE_WIDGET_CHILDREN.
Update template-no-invalid-interactive and template-no-nested-interactive to consume the shared role utilities.
Add/adjust tests to validate role derivation and composite-widget nesting exceptions (including tooltip now being treated as non-interactive).

Reviewed changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
lib/utils/interactive-roles.js	New shared derivation for interactive roles and composite-widget child/descendant allowances.
lib/rules/template-no-nested-interactive.js	Switch to derived role sets and broaden nesting exception to composite-widget patterns (+ submenu rule).
lib/rules/template-no-invalid-interactive.js	Switch to derived interactive role set; `tooltip` no longer treated as interactive.
tests/lib/utils/interactive-roles-test.js	New tests for `INTERACTIVE_ROLES` and `COMPOSITE_WIDGET_CHILDREN` derivation and drift detection.
tests/lib/rules/template-no-nested-interactive.js	Add valid cases for canonical composite-widget hierarchies.
tests/lib/rules/template-no-invalid-interactive.js	Remove `tooltip` from valid widget-role cases; add invalid case asserting `tooltip` is non-interactive.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Pull request overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated 1 comment.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

… pin gridcell/row/columnheader/rowheader)

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

… pin gridcell/row/columnheader/rowheader)

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

Copilot

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

Import INTERACTIVE_ROLES and COMPOSITE_WIDGET_CHILDREN from the shared lib/utils/interactive-roles.js util (introduced in #27 — byte-identical copy here so either PR can land first without conflict). Drop the hardcoded 19-role set previously duplicated inline in each rule. Behavior changes: - ARIA widget role set expands from 19 to 35 roles — picks up menubar, menu, listbox, tree, tablist, grid, treegrid, radiogroup, alertdialog, progressbar, and other widget-descended roles in aria-query's taxonomy that the hardcoded list missed. - tooltip is no longer treated as interactive. Per WAI-ARIA 1.2 §5.3.3, tooltip is a document-structure role, not a widget. #27's util reflects this (tooltip explicitly excluded). Old <div role="tooltip" onclick> test moves from valid to invalid — spec-correct. - Composite-widget nesting exception expanded via COMPOSITE_WIDGET_CHILDREN. Canonical APG patterns (<ul role="menubar"><li role="menuitem">..., <ul role="listbox"><li role="option">..., grid/row/gridcell, treegrid, radiogroup/radio) no longer flag as nested-interactive. Previously the rule only handled menuitem-in-menuitem explicitly. Pairs with #37's HTML-content-model authority split to complete the two-authority architecture: HTML §3.2.5.2.7 via html-interactive-content, ARIA widget taxonomy via interactive-roles. Each util cites one authority honestly; rules compose both.

Copilot

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

Copilot

Pull request overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated no new comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Consolidates two hand-maintained INTERACTIVE_ROLES sets (in template-no-invalid-interactive and template-no-nested-interactive) into a single derivation sourced from aria-query's role taxonomy. lib/utils/interactive-roles.js filters concrete roles whose superClass chain includes widget / command / composite / input / range, then adds tooltip and toolbar explicitly (same rationale as jsx-a11y: ARIA 1.2 categorizes tooltip as a widget but aria-query's superClass chain doesn't reflect that; toolbar supports aria-activedescendant and is widget-like in practice). Behavioral change: Adds 18 roles to the interactive set: columnheader, doc-backlink, doc-biblioref, doc-glossref, doc-noteref, grid, listbox, menu, menubar, meter, progressbar, radiogroup, row, rowheader, tablist, toolbar, tree, treegrid. These were silently treated as non-interactive — a composite widget (role="tree") could contain nested interactive elements without being flagged; an <a href> inside a role="menu" was accepted but nested-interactive per ARIA. No removals — tooltip was in the hand list and is preserved explicitly. Follow-up: template-no-nested-interactive's menu-pattern exception broadened from "menuitem inside menuitem" to "menu patterns per WAI-ARIA" — menuitem/menuitemcheckbox/menuitemradio inside menu/menubar/menuitem, and menu inside menuitem (for submenus), are the standard menu hierarchy.

Our earlier derivation was broader (widget|command|composite|input|range ancestors). jsx-a11y and lit-a11y both use the narrower 'widget ancestor' check. The only role our broader check added was 'meter' — spec-readonly and not interactive in peer-plugin terms. Narrow to match peer behavior exactly: - derivation: def.superClass.some(chain => chain.includes('widget')) - + 'toolbar' (doesn't descend from widget but supports activedescendant) - + 'tooltip' (ARIA 1.2 categorization is ambiguous; our existing tests treat it as interactive) No test changes — the new set is a subset of the previous except for meter (not covered by any existing test).

Per WAI-ARIA 1.2 §5.3.3 — Document Structure Roles: https://www.w3.org/TR/wai-aria-1.2/#tooltip Tooltip is explicitly listed under document-structure roles (alongside img, note, paragraph, etc.), not widget roles. The spec notes: 'Document structures are not usually interactive.' jsx-a11y and lit-a11y agree — neither adds tooltip to their interactive- role set (both add only toolbar). The one test case expecting tooltip to be treated as interactive moves accordingly. aria-query's superClass for tooltip is structure/section, which the narrow widget-ancestor filter correctly excludes.

…erns Deriving INTERACTIVE_ROLES from aria-query's widget taxonomy added roles (`listbox`, `tablist`, `tree`, `treegrid`, `grid`, `radiogroup`, `row`, and their children) to `template-no-nested-interactive`'s interactive set. Without a corresponding exception, canonical WAI-ARIA APG composite widgets would trip the rule: <div role="listbox"><div role="option">A</div></div> <div role="tablist"><div role="tab">…</div></div> <div role="tree"><div role="treeitem">…</div></div> <div role="grid"><div role="row"><div role="gridcell">…</div></div></div> <div role="radiogroup"><div role="radio">…</div></div> Widen the existing menu-pattern exception into a general composite-widget exception driven by aria-query's `requiredOwnedElements` data. For each parent role, inherit `requiredOwnedElements` from ancestor roles in `superClass` (so `treegrid` picks up both `row` via `grid` and `treeitem` via `tree`) and close transitively over intermediate composite roles (so `grid → row → gridcell` is allowed at arbitrary depth in the hierarchy). The submenu pattern `menuitem → menu`, which aria-query does not express via `requiredOwnedElements`, is kept as an explicit special case. Also add an invalid-interactive test for `role="tooltip"` — now that tooltip is no longer in INTERACTIVE_ROLES (per WAI-ARIA 1.2 §5.3.3, it is a Document Structure role), a click handler on a tooltip is an invalid interactive handler on a non-interactive element.

Covers the full util surface: - Canonical widget roles (button, link, combobox, etc.) — in set - Composite-widget containers (listbox, grid, tablist, tree, treegrid) — in set - toolbar manual override — in set (documented exception) - tooltip exclusion — NOT in set (WAI-ARIA 1.2 §5.3.3) - progressbar included (documented divergence from lit-a11y) - Set-size pin (35 roles) — surfaces aria-query taxonomy drift as test failure - COMPOSITE_WIDGET_CHILDREN sanity (listbox/option, grid/gridcell transitivity, treegrid/row+treeitem superClass inheritance, radiogroup/radio, submenu) Also extend the util JSDoc to document progressbar inclusion vs lit-a11y's readonly-value-based exclusion — previously only noted in the PR body.

…ILDREN expansion (Copilot review)

…er rebase

…tails> body panel During the rebase that introduced isSummaryFirstChildOfDetails, the broader isAllowedDetailsChild check was lost. The disclosed panel of <details> is flow content — interactive children after <summary> are valid and must not be flagged. Restores isAllowedDetailsChild as a wrapper that delegates the <summary>-position check to isSummaryFirstChildOfDetails.

…HtmlInteractiveContent)

johanrd · 2026-04-27T14:50:53Z

Closing in favour of test/composite-widget-nesting. The INTERACTIVE_ROLES derivation, COMPOSITE_WIDGET_CHILDREN map, and the util test are already on master (landed via piecemeal commits). The only missing piece — composite-widget hierarchy valid cases in the rule test — will be added via a clean branch from current master. The remaining diff on this PR reverts ember-cli#2748 (deletes html-interactive-content.js) and drops 'use strict', both of which would be regressions.

johanrd requested a review from Copilot April 22, 2026 10:28

Copilot started reviewing on behalf of johanrd April 22, 2026 10:31 View session

Copilot AI reviewed Apr 22, 2026

View reviewed changes

Comment thread lib/utils/interactive-roles.js

Comment thread lib/rules/template-no-nested-interactive.js Outdated

johanrd requested a review from Copilot April 22, 2026 12:58

Copilot started reviewing on behalf of johanrd April 22, 2026 13:02 View session

Copilot AI reviewed Apr 22, 2026

View reviewed changes

Comment thread lib/rules/template-no-nested-interactive.js

johanrd added a commit that referenced this pull request Apr 22, 2026

sync: interactive-roles to canonical (PR #27) for byte-identical merges

1c04408

johanrd added a commit that referenced this pull request Apr 22, 2026

sync: interactive-roles to canonical (PR #27) for byte-identical merges

c710c9a

johanrd added a commit that referenced this pull request Apr 22, 2026

sync: interactive-roles to canonical (PR #27) for byte-identical merges

be19c2a

johanrd added a commit that referenced this pull request Apr 22, 2026

sync: interactive-roles to canonical (PR #27) for byte-identical merges

b7bae63

johanrd requested a review from Copilot April 24, 2026 08:36

Copilot started reviewing on behalf of johanrd April 24, 2026 08:36 View session