Skip to content

feat(editorjs): add batteries-included bundle package and make core headless - #183

Merged
gohabereg merged 15 commits into
mainfrom
feat/editorjs-bundle-package
Aug 27, 2026
Merged

gohabereg merged 15 commits into
mainfrom
feat/editorjs-bundle-package

Conversation

@gohabereg

@gohabereg gohabereg commented Jul 22, 2026 •

Copy link
Copy Markdown
Member

Why

Default tools, plugins, infra (DOM adapter, collaboration), and UI were hardcoded into Core's constructor, so @editorjs/core couldn't be consumed as a pure engine and there was no batteries-included package a v2 user could drop in.

What Changes

  • New package @editorjs/editorjs (packages/editorjs): a batteries-included entry point whose default export EditorJS composes Core + DOMAdapters + CollaborationManager + default tools (paragraph, bold, italic, link) + plugins (clipboard, shortcuts) + UI. It auto-initializes in the constructor and exposes an isReady promise, and honors a v2-style config.tools map merged over the defaults with override-by-name.
  • Core is now headless (BREAKING): the constructor no longer registers any default tools/plugins/infra/UI. initialize() fails loudly — it throws when no rendering adapter is registered or the defaultBlock tool is missing, and re-throws init failures instead of swallowing them with console.error.
  • ShortcutsPlugin is consumed from @editorjs/shortcuts-plugin — extracted from core by feat(core): extract ShortcutsPlugin into @editorjs/shortcuts-plugin #180, which landed on main while this PR was open and is merged in here. The bundle registers it as a default plugin; core neither ships nor registers it.
  • packages/ui/index.html fixed: the demo page constructed Core with no adapter and no block tool, relying on the defaults core used to register itself, so it threw under the new fail-loud validation. It now registers DOMAdapters and Paragraph explicitly.
  • Dead deps dropped from @editorjs/ui: @editorjs/dom-adapters, @editorjs/core and editorjs-v2 were production dependencies that packages/ui/src never imported (dom-adapters has been unused since the package was bootstrapped in chore(ui): ui package bootstrapped #96). Core and dom-adapters move to devDependencies for the demo page; editorjs-v2 is dropped. Matching stale entries removed from the vite externals list and both tsconfigs' project references.
  • Dependency alias: the published @editorjs/editorjs v2 dep is aliased as editorjs-v2 (npm:@editorjs/editorjs@…) in core, sdk, tools/paragraph to free the package name. Non-behavioral; import specifiers only.
  • Playground now consumes @editorjs/editorjs instead of wiring Core + UI by hand.

Merged main (#184, #166, #185, #186, #188)

main is merged in as of 3b61926. Five conflicts, three of them substantive:

  • packages/core/src/index.ts — main had re-asserted the batteries-included constructor (this.use(Paragraph), ClipboardPlugin, DOMAdapters, …) and wraps initialize() in a try/catch that logs. Kept this PR's headless + fail-loud version, and took from main only what was genuinely new: the PluginRegistry wiring, and moving UndoRedoManager resolution to after plugin/tool init. That reorder is main's deliberate fix — the manager relies on defaultPrevented being set by plugins — and since merge-base had it before, this branch only looked like it disagreed.
  • sdk/entities/{BaseTool,InlineTool}.ts + tools/paragraph — main stopped importing ToolConfig from the v2 package and now defines it in BaseTool.ts; this branch had renamed that dep to editorjs-v2. Took both: ToolConfig from the SDK, remaining v2 type imports via the alias. This is what sdk spec now requires anyway ("ToolConfig SHALL be an SDK-owned contract, not a re-exported passthrough from @editorjs/editorjs"). Confirmed no ToolConfig import resolves to the v2 package anywhere.
  • packages/ui/package.json — union of both devDependency sets.

Verification

  • Full workspace build, lint, and tests pass post-merge. Core: 179/179. Bundle: 8/8. openspec validate --all --strict: 15/15.
  • New packages/core/src/index.spec.ts covers the fail-loud contract, which previously had no tests at all — core had no index.spec.ts. It drives a real Core (real IoC containers and model) with stub adapter/block tool/plugin registered via use(), asserting both throwing paths (missing adapter, missing/mismatched defaultBlock) and the succeeding path. No DOM environment needed: with an empty document and stub registrations the boot sequence never reaches the holder.
  • packages/ui/index.html verified in-browser: renders all three paragraphs, no console errors. The ui build output is byte-identical after the dependency cleanup (85.72 kB / 67.57 kB, 28 modules), confirming the removed externals were no-ops.
  • Playground verified end-to-end in-browser, through the bundle: renders the configured paragraphs, typing updates the model, the inline toolbar appears on selection and its B button applies <b>.

Keyboard shortcuts: now working (was blocked on #184)

CMD+B works, as of the main merge. It could not work on this branch before: ShortcutsPlugin subscribes to ui:key-down (KeydownUIEvent), but nothing dispatched that event — packages/ui/src/Blocks/Blocks.ts handled CMD+Z / CMD+SHIFT+Z locally and returned early for every other key. Registering the plugin was necessary but not sufficient. The producer side is the BlocksUI keydown dispatch from #184, now merged in at Blocks.ts:125.

Re-verified in the browser through the bundle, on a selection inside a paragraph:

Action Result
CMD+B selection bolds, inline toolbar's B goes active
CMD+B again reverts to regular weight, B inactive
CMD+I selection italicizes, i goes active

The toolbar's active state tracking the text confirms it round-tripped through the model rather than being a contenteditable side effect — isActive is reading back the fragment the shortcut wrote — and exercises both the Format and Unformat branches of getFormattingOptions. The handler table is populated from the namespaced static options.plugins.shortcuts.shortcut that Bold and Italic declare, which is also #184's shape.

Task 4.6 in the archived change is closed accordingly; that change now has no open tasks.

(An earlier revision of this description claimed CMD+B was verified end-to-end, before it could possibly work. That was wrong — thanks @Reversean for pushing on it. The claim above is a fresh in-browser check made after the merge, not a restatement of the old one.)

Specs

Implemented via OpenSpec change add-editorjs-bundle-package (archived in this PR). Because the shortcuts extraction landed mid-review, the archived change was updated in place rather than superseded: proposal, design, tasks and the editorjs-bundle delta now reflect shortcuts being registered from its own package. Main specs updated: core (headless + fail-loud scenarios, now with a validation pointer to the new spec file) and the new editorjs-bundle capability.

Auditing all 15 specs against the merged tree turned up further drift, fixed in two commits kept deliberately separate:

  • 8139d2f (this branch's own gap) — architecture still asserted "Default adapter is DOMAdapters", i.e. that Core's constructor calls this.use(DOMAdapters). Accurate at merge-base and still accurate on main, but false here, and it directly contradicted core's "No adapter registered fails loudly". The bundle change wrote deltas for core and editorjs-bundle and missed that the same fact was asserted in architecture too.
  • 43b5513 (pre-existing on main, not introduced here) — tools described Bold/Italic as declaring flat shortcut: 'CMD+B' / 'CMD+I' properties; feat(core): expose plugin public APIs and tool-directed plugin options #184 moved them to options.plugins.shortcuts.shortcut without updating the spec, leaving it contradicting shortcuts-plugin's own "Legacy flat shortcut key is ignored" scenario. plugin-public-api and tool-plugin-options also still carried the generated TBD - created by archiving change add-plugin-public-api placeholder purposes.

The second commit is unrelated to this PR's subject and is a clean standalone cherry-pick onto main if you'd rather it not ride along — happy to drop it.

Follow-ups (not in this PR)

The same audit found four spec/code gaps that all trace to #184 and need code, not wording. Flagging rather than fixing, since they're outside this PR's scope:

  1. Plugin teardown is specified but unimplemented. core ("Registry is cleared on teardown") and sdk ("Destroying a plugin") both describe destroy() running on teardown. Core has no teardown path, nothing calls plugin destroy(), and PluginRegistry.unregister() has no production caller.
  2. Duplicate/empty plugin names are only rejected when a plugin exposes a publicApi. Core#initializePlugin guards the registry.register() call on instance.publicApi !== undefined, so two same-named plugins without one register silently — while core and plugin-public-api state the rejection unconditionally.
  3. The integrator-facing api.plugins surface doesn't exist. plugin-public-api requires it "on the API surface available to the editor's integrator", but neither Core (only use()/initialize()) nor the bundle's EditorJS (only isReady) exposes one.
  4. CollaborationManager logs Unknown event type DataNodeAddedEvent three times per page load — no case for that event, falls through to default: console.error. Pre-existing at merge-base and on main; already noted as out of scope in task 4.4.

@github-actions

github-actions Bot commented Jul 22, 2026 •

Copy link
Copy Markdown
Contributor

Unit Tests

Package Coverage Delta
@editorjs/core 73.14% -17.52% 🔴
@editorjs/clipboard-plugin 66.66% 0% ⚪️
@editorjs/collaboration-manager 85.81% 0% ⚪️
@editorjs/dom-adapters 86.95% 0% ⚪️
@editorjs/ot-server 20% 0% ⚪️
@editorjs/model 98.24% 0% ⚪️
@editorjs/editorjs 100% N/A
@editorjs/shortcuts-plugin 100% 0% ⚪️
@editorjs/model-types 59.82% 0% ⚪️

Mutation Tests

Package Mutation score Dashboard URL
@editorjs/core 51.59% 🔴 Dashboard
@editorjs/model No files to mutate found.
@editorjs/clipboard-plugin No files to mutate found.
@editorjs/collaboration-manager No files to mutate found.
@editorjs/shortcuts-plugin No files to mutate found.
@editorjs/dom-adapters No files to mutate found.

…eadless

Introduce `@editorjs/editorjs`, a batteries-included entry point that composes
`@editorjs/core` with the default DOM adapter, collaboration, tools, plugins,
and UI, auto-initializes in its constructor, and exposes an `isReady` promise.
It honors a v2-style `config.tools` map merged over the defaults with
override-by-name.

To support it, `Core` becomes a headless engine:
- Remove the hardcoded default tool/plugin/infra `use()` calls from the
  constructor; drop the corresponding dependencies.
- `initialize()` now fails loudly: it throws when no rendering adapter is
  registered or the `defaultBlock` tool is missing, and re-throws init
  failures instead of swallowing them.

`ShortcutsPlugin` stays in `@editorjs/core` (no longer auto-registered); its
relocation is handled by a separate change.

Free the `@editorjs/editorjs` package name by aliasing the published v2
dependency as `editorjs-v2` (`npm:@editorjs/editorjs@…`) across core, sdk, ui,
and tools/paragraph.

The playground now consumes `@editorjs/editorjs` instead of wiring `Core` and
UI by hand.

Specs updated via OpenSpec change `add-editorjs-bundle-package` (archived).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
@gohabereg
gohabereg force-pushed the feat/editorjs-bundle-package branch from 433933c to 182b01d Compare July 22, 2026 21:29
gohabereg and others added 4 commits July 22, 2026 22:46
…gration test

Core no longer depends on `@editorjs/dom-adapters`, so its dist is not built
in CI when checking `@editorjs/core`. The leftover `jest.unstable_mockModule`
call still resolved the specifier, failing the suite (and the Stryker dry run)
with "Cannot find module '@editorjs/dom-adapters'".

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Runs lint, build and unit tests for the new bundle package on pull requests
and merge groups, mirroring the other package workflows. Mutation tests are
disabled since the package has no Stryker config yet.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
`tsc --build` only produces output for TypeScript project references, so
`@editorjs/ui` — which is bundled by vite — had no `dist/ui.js` after building
`@editorjs/editorjs` from a clean checkout. Jest then failed to resolve
`@editorjs/ui` in the bundle test.

Build dependencies first via `yarn workspaces foreach`, mirroring how the
playground package builds its own dependency graph.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR introduces a new batteries-included @editorjs/editorjs bundle package and refactors @editorjs/core into a truly headless engine by removing hardcoded defaults and adding fail-loud precondition checks. It also aliases the legacy v2 @editorjs/editorjs dependency as editorjs-v2 across the workspace to free the @editorjs/editorjs name, and updates the playground to consume the new bundle.

Changes:

  • Add packages/editorjs (@editorjs/editorjs) that composes Core + DOM adapter + collaboration + default tools/plugins + UI and auto-initializes with isReady.
  • Make @editorjs/core headless (no default use() registrations) and add synchronous initialization precondition validation.
  • Alias v2 @editorjs/editorjs as editorjs-v2 and update imports/dependencies; switch the playground to the new bundle.

Reviewed changes

Copilot reviewed 46 out of 47 changed files in this pull request and generated 6 comments.

Show a summary per file
File Description
yarn.lock Updates workspace lockfile for new package and editorjs-v2 aliasing.
packages/ui/package.json Replaces direct v2 @editorjs/editorjs dependency with editorjs-v2 alias.
packages/tools/paragraph/src/index.ts Updates v2 type imports to use editorjs-v2.
packages/tools/paragraph/package.json Swaps v2 @editorjs/editorjs dependency for editorjs-v2 alias.
packages/sdk/src/tools/facades/BlockTuneFacade.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/src/tools/facades/BlockToolFacade.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/src/tools/facades/BaseToolFacade.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/src/entities/InlineTool.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/src/entities/EventBus/events/core/BlockAddedCoreEvent.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/src/entities/Config.ts Updates EditorConfig type source to editorjs-v2.
packages/sdk/src/entities/BlockTune.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/src/entities/BlockTool.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/src/entities/BaseTool.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/src/api/BlocksAPI.ts Updates v2 type imports to use editorjs-v2.
packages/sdk/package.json Replaces v2 @editorjs/editorjs dependency with editorjs-v2 alias.
packages/playground/src/App.vue Switches from manual Core+UI wiring to new EditorJS() bundle usage.
packages/playground/package.json Simplifies deps by depending on @editorjs/editorjs bundle.
packages/editorjs/tsconfig.json Adds TS project config for new bundle package.
packages/editorjs/tsconfig.eslint.json Adds ESLint TS project config for the new package.
packages/editorjs/tsconfig.build.json Adds build-specific TS project config for the new package.
packages/editorjs/src/mergeTools.ts Adds override-by-name merge helper for default + user tool constructors.
packages/editorjs/src/mergeTools.spec.ts Adds unit tests for tool merging semantics.
packages/editorjs/src/index.ts Implements EditorJS bundle class composition + isReady.
packages/editorjs/src/index.spec.ts Adds bundle tests for composition and isReady resolve/reject behavior.
packages/editorjs/package.json Defines the new @editorjs/editorjs workspace package and scripts.
packages/editorjs/jest.config.ts Adds Jest config for new package tests (ESM + TS).
packages/editorjs/eslint.config.mjs Adds ESLint config for the new package.
packages/editorjs/.gitignore Adds package-local ignores for build/test artifacts.
packages/core/tsconfig.json Removes build references to defaults now moved out of core.
packages/core/tsconfig.build.json Removes build references to defaults now moved out of core.
packages/core/src/utils/composeDataFromVersion2.ts Updates v2 type imports to use editorjs-v2.
packages/core/src/tools/ToolsManager.ts Updates v2 type imports to use editorjs-v2.
packages/core/src/tools/ToolsFactory.ts Updates v2 type imports to use editorjs-v2.
packages/core/src/index.ts Removes default registrations and adds precondition validation + rethrow behavior.
packages/core/src/components/BlockRenderer.ts Switches v2 imports to editorjs-v2.
packages/core/src/components/BlockManager.ts Switches v2 imports to editorjs-v2.
packages/core/src/api/BlocksAPI.integration.spec.ts Adjusts test scaffolding/mocks in response to refactors.
packages/core/package.json Drops baked-in defaults; adds editorjs-v2 alias dependency.
openspec/specs/editorjs-bundle/spec.md Adds main spec for the new bundle capability.
openspec/specs/core/spec.md Updates core spec to reflect headless + fail-loud initialization behavior.
openspec/changes/archive/2026-07-22-add-editorjs-bundle-package/tasks.md Archives change tasks and implementation checklist.
openspec/changes/archive/2026-07-22-add-editorjs-bundle-package/specs/editorjs-bundle/spec.md Archives delta spec for new bundle capability.
openspec/changes/archive/2026-07-22-add-editorjs-bundle-package/specs/core/spec.md Archives delta spec for core headless behavior changes.
openspec/changes/archive/2026-07-22-add-editorjs-bundle-package/proposal.md Archives the change proposal for traceability.
openspec/changes/archive/2026-07-22-add-editorjs-bundle-package/design.md Archives the design decisions and trade-offs.
openspec/changes/archive/2026-07-22-add-editorjs-bundle-package/.openspec.yaml Adds archived change metadata for OpenSpec workflow.
.github/workflows/editorjs.yml Adds CI workflow to run package-check for @editorjs/editorjs.

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

Comment thread packages/core/src/index.ts Outdated
Comment on lines +178 to +182
} catch (error) {
console.error('Editor.js initialization failed', error);

throw error;
}
Comment on lines 11 to 14
import { TOKENS } from '../tokens.js';
import ToolsManager from '../tools/ToolsManager.js';
import { BlockAPI } from '@editorjs/editorjs';
import { BlockAPI } from 'editorjs-v2';
import {
Comment on lines 16 to 20
import { TOKENS } from '../tokens.js';
import ToolsManager from '../tools/ToolsManager.js';
import { BlockToolData } from '@editorjs/editorjs';
import { BlockToolData } from 'editorjs-v2';
import { CoreConfigValidated, EventBus } from '@editorjs/sdk';

Comment on lines 62 to 65
});

core
.use(EditorjsUI)
.use(BlocksUI)
.use(InlineToolbarUI)
.use(ToolbarUI)
.use(ToolboxUI)
.initialize();
void editor.isReady;
});
Comment on lines +22 to +26
if (userTools !== undefined) {
for (const [name, tool] of Object.entries(userTools)) {
merged.set(name, tool);
}
}
Comment on lines +3 to +5
## Purpose
TBD - created by archiving change add-editorjs-bundle-package. Update Purpose after archive.
## Requirements
- Core.initialize() no longer logs to console.error before re-throwing
  (dead try/catch removed)
- BlockRenderer/BlockManager import editorjs-v2 types with `import type`
  instead of pulling it in as a runtime dependency
- Playground attaches a .catch() to editor.isReady instead of discarding
  the promise, avoiding an unhandled rejection
- mergeTools() now throws when a config.tools key doesn't match the
  tool's static name, since Core looks up tools by name, not map key
- editorjs-bundle spec: replaced TBD Purpose and documented the new
  fail-loud mismatch scenario

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Comment thread packages/editorjs/src/index.ts Resolved
Comment thread packages/core/src/index.ts Resolved
Comment thread packages/core/src/index.ts Resolved
gohabereg and others added 5 commits August 4, 2026 19:09
Resolve conflicts in favor of the headless-core direction: `@editorjs/core`
keeps no default tool/plugin/infra registrations, dependencies, or project
references, so main's `@editorjs/shortcuts-plugin` wiring in core is dropped.

Register `ShortcutsPlugin` from the newly extracted `@editorjs/shortcuts-plugin`
package in the `@editorjs/editorjs` bundle instead, which is where default
plugins now live. The core spec's keyboard-shortcuts requirement is removed
(it moved to `openspec/specs/shortcuts-plugin/spec.md` on main), and the bundle
spec's default-composition scenario now lists shortcuts alongside clipboard.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
…ange

The `@editorjs/shortcuts-plugin` extraction landed on main while this PR was
open, so the archived `add-editorjs-bundle-package` change is updated in place
rather than superseded: the bundle now registers `ShortcutsPlugin` from that
package instead of deferring it.

- delta + main `editorjs-bundle` spec: default composition lists clipboard and
  shortcuts
- proposal/design: shortcuts is consumed from `@editorjs/shortcuts-plugin`, not
  left behind in core
- tasks 2.3/3.2/3.3/3.4 reflect what shipped; 4.6 added to re-verify CMD+B
  end-to-end through the bundle

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
`packages/ui/index.html` constructed `Core` with no rendering adapter and no
block tool, relying on the defaults core used to register itself. Now that
`initialize()` validates its preconditions, that page threw immediately. Register
`DOMAdapters` and `Paragraph` explicitly.

While wiring that up: `@editorjs/dom-adapters`, `@editorjs/core` and `editorjs-v2`
were declared as production dependencies of `@editorjs/ui` but never imported by
`packages/ui/src` — dom-adapters has been unused since the package was
bootstrapped in #96. Only the demo page needs core and dom-adapters, so they move
to devDependencies alongside the new `@editorjs/paragraph`; `editorjs-v2` is
dropped outright. The matching stale entries are removed from the vite externals
list and from both tsconfigs' project references, which were forcing needless
build-order edges from ui to core and dom-adapters.

The built bundle is byte-identical (85.72 kB / 67.57 kB, 28 modules), confirming
the removed externals were no-ops. Demo page verified in a browser: renders all
three paragraphs, no console errors.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The headless refactor made `initialize()` validate that a rendering adapter and
the `defaultBlock` tool are registered, but core had no `index.spec.ts` at all, so
the new contract — and the two error messages callers are meant to act on — was
untested while `openspec/specs/core/spec.md` already carried its scenarios.

Drives a real `Core` with real IoC containers and model, registering stub
adapter/block tool/plugin through `use()`, per review feedback on PR #183:

- throws naming the rendering adapter when none is registered
- throws naming `defaultBlock` when it is missing, and when a differently named
  block tool is registered instead
- resolves when adapter, default block tool and plugin are all registered
- constructs registered plugins during boot

No DOM environment is needed: with an empty document and stub registrations the
boot sequence never reaches into the holder, so this runs under core's existing
`node` test environment with no new dependencies. Core: 166/166 pass.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
…ducer

Re-verified in the playground: registering `ShortcutsPlugin` in the bundle is not
enough to make CMD+B work. The plugin subscribes to `ui:key-down`, but no code on
this branch dispatches `KeydownUIEvent` — `Blocks.ts` handles CMD+Z/CMD+SHIFT+Z
inline and returns early for every other key. The dispatch arrives with PR #184.

Isolated the failure: the inline toolbar's B button applies `<b>` to the same
selection, so formatting works and only the shortcut path is dead.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@gohabereg
gohabereg requested a review from Reversean August 4, 2026 18:45
gohabereg and others added 4 commits August 27, 2026 01:01
Conflicts resolved:
- packages/core/src/index.ts: kept the headless Core from this branch (no default
  tool/plugin/adapter registration, fail-loud `#validatePreconditions`, no
  swallowing try/catch) while taking main's `PluginRegistry` wiring and its
  reordering of `UndoRedoManager` to after plugin/tool init.
- packages/sdk/src/entities/BaseTool.ts, InlineTool.ts and
  packages/tools/paragraph/src/index.ts: `ToolConfig` now comes from the SDK's own
  definition (main) instead of the v2 package; remaining v2 type imports keep this
  branch's `editorjs-v2` alias, since `@editorjs/editorjs` is now the bundle.
- packages/ui/package.json: union of both devDependency sets.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
… headless Core

Two loose ends this branch owned:

`architecture` still asserted "Default adapter is DOMAdapters" — that Core's
constructor calls `this.use(DOMAdapters)`. True at merge-base and still true on
main, but this branch made Core headless and dropped that call, so the scenario
was false here and contradicted core/spec.md's "No adapter registered fails
loudly". The bundle change wrote deltas for `core` and `editorjs-bundle` and
missed that the same fact was asserted in `architecture` too. Replaced with "No
adapter is bound by default", and tightened the parent requirement to say Core
never binds an adapter itself.

Task 4.6 was blocked on nothing dispatching `KeydownUIEvent`. That producer
landed with #184 (Blocks.ts:125), now merged in. Re-verified in the playground
through the bundle: CMD+B bolds and toggles off, CMD+I italicizes, toolbar
active states track the model.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Not this branch's debt — these are already wrong on main today, and arrived here
via the merge. Fixing them while the context is fresh rather than leaving them.

`tools` described Bold/Italic as declaring flat `shortcut: 'CMD+B'` / `'CMD+I'`
properties. #184 moved both to `static options.plugins.shortcuts.shortcut`
without updating the spec, which left it contradicting shortcuts-plugin's own
"Legacy flat shortcut key is ignored" scenario.

`plugin-public-api` and `tool-plugin-options` still carried the generated
"TBD - created by archiving change add-plugin-public-api" placeholder. Written
up as mirror images of each other — API a plugin exposes to consumers vs.
config a tool addresses to a plugin — cross-linked both ways.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@gohabereg
gohabereg added this pull request to the merge queue Aug 27, 2026
Merged via the queue into main with commit 253ca57 Aug 27, 2026
70 checks passed
@gohabereg
gohabereg deleted the feat/editorjs-bundle-package branch August 27, 2026 19:34
gohabereg added a commit that referenced this pull request Aug 27, 2026
main landed the squashed bundle package (#183) plus work this branch predates,
so the overlap had to be reconciled by hand:

- ShortcutsPlugin: kept main's extraction into @editorjs/shortcuts-plugin — the
  bundle depends on it, registers it as a default plugin, and core no longer
  ships it. Added to editorjs's package.json, tsconfigs and index.ts alongside
  ClipboardPlugin.
- ToolConfig now comes from @editorjs/sdk (#188) instead of editorjs-v2; dropped
  the alias import in sdk entities and paragraph while keeping this branch's
  ARIA changes in paragraph.
- Core.initialize resolves UndoRedoManager after plugins/tools (main's ordering)
  and registers PluginRegistry.
- packages/ui: took main's chokidar dev script and dependency reshuffle, kept
  this branch's @editorjs/ui-kit ^2.0.0 pin.
- openspec: took main's archived bundle change and updated core/editorjs-bundle
  specs; this branch's add-editor-aria-semantics change is untouched.
- Playwright/e2e tooling, .gitignore, eslint config and the e2e CI job are kept
  from this branch.

yarn.lock was regenerated from main's and still has no @editorjs/ui-kit entry:
2.0.0 is not published yet, so `yarn install` fails exactly as it did on this
branch before the merge. Verified with the workspace's tsc/jest directly —
core (179), editorjs (8) pass; core, paragraph, shortcuts-plugin and editorjs
typecheck; packages/ui's three errors are the unpublished ui-kit 2.0.0 APIs.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
gohabereg added a commit to SAY-5/document-model that referenced this pull request Oct 16, 2026
…-js#187)

* feat(editorjs): add batteries-included bundle package and make core headless

Introduce `@editorjs/editorjs`, a batteries-included entry point that composes
`@editorjs/core` with the default DOM adapter, collaboration, tools, plugins,
and UI, auto-initializes in its constructor, and exposes an `isReady` promise.
It honors a v2-style `config.tools` map merged over the defaults with
override-by-name.

To support it, `Core` becomes a headless engine:
- Remove the hardcoded default tool/plugin/infra `use()` calls from the
  constructor; drop the corresponding dependencies.
- `initialize()` now fails loudly: it throws when no rendering adapter is
  registered or the `defaultBlock` tool is missing, and re-throws init
  failures instead of swallowing them.

`ShortcutsPlugin` stays in `@editorjs/core` (no longer auto-registered); its
relocation is handled by a separate change.

Free the `@editorjs/editorjs` package name by aliasing the published v2
dependency as `editorjs-v2` (`npm:@editorjs/editorjs@…`) across core, sdk, ui,
and tools/paragraph.

The playground now consumes `@editorjs/editorjs` instead of wiring `Core` and
UI by hand.

Specs updated via OpenSpec change `add-editorjs-bundle-package` (archived).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(core): drop stale @editorjs/dom-adapters mock from BlocksAPI integration test

Core no longer depends on `@editorjs/dom-adapters`, so its dist is not built
in CI when checking `@editorjs/core`. The leftover `jest.unstable_mockModule`
call still resolved the specifier, failing the suite (and the Stryker dry run)
with "Cannot find module '@editorjs/dom-adapters'".

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* ci: add package check workflow for @editorjs/editorjs

Runs lint, build and unit tests for the new bundle package on pull requests
and merge groups, mirroring the other package workflows. Mutation tests are
disabled since the package has no Stryker config yet.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(editorjs): build workspace dependencies before the bundle

`tsc --build` only produces output for TypeScript project references, so
`@editorjs/ui` — which is bundled by vite — had no `dist/ui.js` after building
`@editorjs/editorjs` from a clean checkout. Jest then failed to resolve
`@editorjs/ui` in the bundle test.

Build dependencies first via `yarn workspaces foreach`, mirroring how the
playground package builds its own dependency graph.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(editorjs): address Copilot review comments on PR editor-js#183

- Core.initialize() no longer logs to console.error before re-throwing
  (dead try/catch removed)
- BlockRenderer/BlockManager import editorjs-v2 types with `import type`
  instead of pulling it in as a runtime dependency
- Playground attaches a .catch() to editor.isReady instead of discarding
  the promise, avoiding an unhandled rejection
- mergeTools() now throws when a config.tools key doesn't match the
  tool's static name, since Core looks up tools by name, not map key
- editorjs-bundle spec: replaced TBD Purpose and documented the new
  fail-loud mismatch scenario

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* test(editorjs): add Playwright e2e suite and propose ARIA semantics change

Adds a Playwright e2e setup for @editorjs/editorjs (Vite-served fixture
mounting the real bundle from src/, no build step) covering paragraph
rendering, typing, and inline-toolbar bold. Documents a gotcha where
locator.press()/selectText() desync Editor.js's native caret tracking.

Writing these tests surfaced that @editorjs/ui and the paragraph tool
emit no ARIA roles/names anywhere, forcing CSS-class selectors instead
of getByRole/getByLabel. Proposes openspec change
add-editor-aria-semantics to close that gap.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(openspec/aria): move role=textbox from holder to per-block; add webkit e2e target

Replace role="textbox" on the blocks holder with role="group" to avoid
nesting a textbox inside another textbox — contenteditable elements
implicitly carry role="textbox" via HTML-AAM regardless of explicit role,
so putting it on the holder too would be an ARIA anti-pattern.

Instead, assign role="textbox" + aria-multiline="true" + aria-label to
each block's own contenteditable element (starting with Paragraph). Update
proposal, design decisions, specs (ui + tools), and task checklist to
reflect the revised approach.

Also adds webkit (Desktop Safari) as a second Playwright project alongside
chromium in packages/editorjs/playwright.config.ts.

* feat(ui,paragraph): expose ARIA semantics and cover them with e2e

The editor emitted zero roles or accessible names, which made it unusable
with a screen reader and forced the e2e suite onto CSS-class selectors.

- BlocksUI: role="group" on the holder, so the block textboxes are not
  nested inside a textbox the holder would otherwise imply
- Paragraph: role="textbox", aria-multiline, aria-label, and the configured
  placeholder as aria-placeholder
- ToolbarUI: role="toolbar" with a roving tabindex, named plus button,
  aria-haspopup/aria-expanded, and focus moved to the button on open so
  Safari restores it there rather than to <body>
- InlineToolbarUI: named toolbar, aria-activedescendant mirrored from
  ui-kit's highlight and made resolvable with aria-owns, plus a live region
  announcing the toolbar once per open
- Popover item roles and names come from @editorjs/ui-kit 2.0.0, fixed
  upstream rather than worked around from here

Covered by aria.spec.ts, axe.spec.ts and a VoiceOver suite driving a real
screen reader (voiceover.spec.ts, macOS-only, excluded from test:e2e).

Note: @editorjs/ui-kit 2.0.0 is not published yet, so the lockfile has no
entry for it and `yarn install` fails. The local `link:` resolution the work
was developed against has been removed deliberately — see tasks.md 5a.2/5a.3.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci: run the @editorjs/editorjs e2e suite on pull requests

The ARIA semantics the UI now depends on are enforced only by Playwright —
nothing in the type surface catches a dropped role or accessible name, so
until now a regression could land unnoticed.

Adds an `include-e2e` input to the reusable package-check workflow and an
`e2e-tests` composite action that installs chromium and webkit (cached by
lockfile hash) and runs `test:e2e`. Kept as its own job: it needs browsers
and takes minutes, and its failures read differently from a unit test's.
`voiceover.spec.ts` stays excluded — screen readers cannot run headless.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix(ui): announce the inline toolbar again after it is closed with Escape

Escape closes the popover inside ui-kit, without going through #hide(),
so #isVisible stayed true and every later appearance was taken for a
reselection while already open: the toolbar was never announced again.

Listen for PopoverEvent.Closed to reset it. The listener is removed
before the popover is destroyed, since destroy() emits Closed as well
and a rebuild would otherwise re-announce the toolbar on every
selection change.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci(editorjs): write the Playwright html report the e2e action uploads

With only the list reporter nothing was written to playwright-report/,
so the upload step had nothing to upload and the traces recorded on a
retry were lost with the runner.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci: key the Playwright browser cache by the Playwright version

Hashing yarn.lock dropped the ~400MB browser cache on any unrelated
dependency bump. Each Playwright release pins its own browser builds,
so the installed version is all the key needs.

Also bump upload-artifact to v7, in line with the other actions.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* chore(deps): bump @editorjs/dom to ^1.2.0 and dedupe it

@editorjs/ui-kit 2.0.0 depends on @editorjs/dom ^1.2.0, while the workspace
packages still asked for ^1.1.0. Yarn will not re-resolve an already-locked
entry, so the two ranges resolved separately and node_modules ended up with
two copies of @editorjs/dom - including a nested one under @editorjs/caret,
which pins ^1.1.0 transitively. Two copies split module state and break any
identity check across the boundary.

Widen the workspace ranges to ^1.2.0 and run `yarn dedupe @editorjs/dom` so
both ranges collapse onto a single 1.2.0 resolution. The lockfile also gains
the @editorjs/ui-kit 2.0.0 resolution it was missing.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* refactor(ui,editorjs): drop dead popover field and reuse the e2e helpers

`#popoverElement` was only ever written: assigned from `getElement()`, used as
the receiver for the id and the append, then nulled in the teardown. A local
covers all three, and dropping the field removes a second piece of state that
had to be kept in step with `#popover` for no one's benefit.

`editor.spec.ts` was the only spec reimplementing the mount-and-wait sequence
and the triple-click selection instead of importing `mountDocument` and
`selectParagraphText`. The readiness signal it waited on lives in the fixture,
so a change there would have fixed the three specs using the helpers and left
this one timing out.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* docs(openspec): record the ui-kit publish as done and fix a stale claim

5a.3 was still written as the blocking, not-yet-done publish. @editorjs/ui-kit
2.0.0 is published, the lockfile resolves it, and the suite was re-run against
the published package: build, 973 unit tests, 80 e2e on both engines, lint. The
task now also records the duplicate @editorjs/dom the regeneration surfaced and
the dedupe that resolved it, since @editorjs/caret still pins the old range.

The verification summary listed the same publish as an outstanding blocker and
its verdict turned on it; both now say otherwise.

Case 18's comment credited the CSS class with removing filtered items from the
accessibility tree. ui-kit sets `hidden` alongside the class and that is what
does it - as the spec and aria.spec.ts both already said.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* docs(openspec): drop the three deferred findings from tasks.md

10.5b, 10.7b and 10.8 were records of pre-existing bugs the change deliberately
did not fix, not work items. Every archived change here has zero unchecked
tasks, so as encoded they would have blocked archiving and reported 69/72.

The findings themselves are preserved outside the repo. What referenced them by
number - the proposal, the design note, 10.1c, 10.7a and the verification
summary - now describes the bug instead, so nothing points at a task that is
gone.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* chore(openspec): archive add-editor-aria-semantics

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
gohabereg added a commit to SAY-5/document-model that referenced this pull request Oct 16, 2026
* feat(editorjs): add batteries-included bundle package and make core headless

Introduce `@editorjs/editorjs`, a batteries-included entry point that composes
`@editorjs/core` with the default DOM adapter, collaboration, tools, plugins,
and UI, auto-initializes in its constructor, and exposes an `isReady` promise.
It honors a v2-style `config.tools` map merged over the defaults with
override-by-name.

To support it, `Core` becomes a headless engine:
- Remove the hardcoded default tool/plugin/infra `use()` calls from the
  constructor; drop the corresponding dependencies.
- `initialize()` now fails loudly: it throws when no rendering adapter is
  registered or the `defaultBlock` tool is missing, and re-throws init
  failures instead of swallowing them.

`ShortcutsPlugin` stays in `@editorjs/core` (no longer auto-registered); its
relocation is handled by a separate change.

Free the `@editorjs/editorjs` package name by aliasing the published v2
dependency as `editorjs-v2` (`npm:@editorjs/editorjs@…`) across core, sdk, ui,
and tools/paragraph.

The playground now consumes `@editorjs/editorjs` instead of wiring `Core` and
UI by hand.

Specs updated via OpenSpec change `add-editorjs-bundle-package` (archived).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(core): drop stale @editorjs/dom-adapters mock from BlocksAPI integration test

Core no longer depends on `@editorjs/dom-adapters`, so its dist is not built
in CI when checking `@editorjs/core`. The leftover `jest.unstable_mockModule`
call still resolved the specifier, failing the suite (and the Stryker dry run)
with "Cannot find module '@editorjs/dom-adapters'".

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* ci: add package check workflow for @editorjs/editorjs

Runs lint, build and unit tests for the new bundle package on pull requests
and merge groups, mirroring the other package workflows. Mutation tests are
disabled since the package has no Stryker config yet.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(editorjs): build workspace dependencies before the bundle

`tsc --build` only produces output for TypeScript project references, so
`@editorjs/ui` — which is bundled by vite — had no `dist/ui.js` after building
`@editorjs/editorjs` from a clean checkout. Jest then failed to resolve
`@editorjs/ui` in the bundle test.

Build dependencies first via `yarn workspaces foreach`, mirroring how the
playground package builds its own dependency graph.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(editorjs): address Copilot review comments on PR editor-js#183

- Core.initialize() no longer logs to console.error before re-throwing
  (dead try/catch removed)
- BlockRenderer/BlockManager import editorjs-v2 types with `import type`
  instead of pulling it in as a runtime dependency
- Playground attaches a .catch() to editor.isReady instead of discarding
  the promise, avoiding an unhandled rejection
- mergeTools() now throws when a config.tools key doesn't match the
  tool's static name, since Core looks up tools by name, not map key
- editorjs-bundle spec: replaced TBD Purpose and documented the new
  fail-loud mismatch scenario

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* test(editorjs): add Playwright e2e suite and propose ARIA semantics change

Adds a Playwright e2e setup for @editorjs/editorjs (Vite-served fixture
mounting the real bundle from src/, no build step) covering paragraph
rendering, typing, and inline-toolbar bold. Documents a gotcha where
locator.press()/selectText() desync Editor.js's native caret tracking.

Writing these tests surfaced that @editorjs/ui and the paragraph tool
emit no ARIA roles/names anywhere, forcing CSS-class selectors instead
of getByRole/getByLabel. Proposes openspec change
add-editor-aria-semantics to close that gap.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(openspec/aria): move role=textbox from holder to per-block; add webkit e2e target

Replace role="textbox" on the blocks holder with role="group" to avoid
nesting a textbox inside another textbox — contenteditable elements
implicitly carry role="textbox" via HTML-AAM regardless of explicit role,
so putting it on the holder too would be an ARIA anti-pattern.

Instead, assign role="textbox" + aria-multiline="true" + aria-label to
each block's own contenteditable element (starting with Paragraph). Update
proposal, design decisions, specs (ui + tools), and task checklist to
reflect the revised approach.

Also adds webkit (Desktop Safari) as a second Playwright project alongside
chromium in packages/editorjs/playwright.config.ts.

* feat(ui,paragraph): expose ARIA semantics and cover them with e2e

The editor emitted zero roles or accessible names, which made it unusable
with a screen reader and forced the e2e suite onto CSS-class selectors.

- BlocksUI: role="group" on the holder, so the block textboxes are not
  nested inside a textbox the holder would otherwise imply
- Paragraph: role="textbox", aria-multiline, aria-label, and the configured
  placeholder as aria-placeholder
- ToolbarUI: role="toolbar" with a roving tabindex, named plus button,
  aria-haspopup/aria-expanded, and focus moved to the button on open so
  Safari restores it there rather than to <body>
- InlineToolbarUI: named toolbar, aria-activedescendant mirrored from
  ui-kit's highlight and made resolvable with aria-owns, plus a live region
  announcing the toolbar once per open
- Popover item roles and names come from @editorjs/ui-kit 2.0.0, fixed
  upstream rather than worked around from here

Covered by aria.spec.ts, axe.spec.ts and a VoiceOver suite driving a real
screen reader (voiceover.spec.ts, macOS-only, excluded from test:e2e).

Note: @editorjs/ui-kit 2.0.0 is not published yet, so the lockfile has no
entry for it and `yarn install` fails. The local `link:` resolution the work
was developed against has been removed deliberately — see tasks.md 5a.2/5a.3.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci: run the @editorjs/editorjs e2e suite on pull requests

The ARIA semantics the UI now depends on are enforced only by Playwright —
nothing in the type surface catches a dropped role or accessible name, so
until now a regression could land unnoticed.

Adds an `include-e2e` input to the reusable package-check workflow and an
`e2e-tests` composite action that installs chromium and webkit (cached by
lockfile hash) and runs `test:e2e`. Kept as its own job: it needs browsers
and takes minutes, and its failures read differently from a unit test's.
`voiceover.spec.ts` stays excluded — screen readers cannot run headless.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* fix(ui): announce the inline toolbar again after it is closed with Escape

Escape closes the popover inside ui-kit, without going through #hide(),
so #isVisible stayed true and every later appearance was taken for a
reselection while already open: the toolbar was never announced again.

Listen for PopoverEvent.Closed to reset it. The listener is removed
before the popover is destroyed, since destroy() emits Closed as well
and a rebuild would otherwise re-announce the toolbar on every
selection change.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci(editorjs): write the Playwright html report the e2e action uploads

With only the list reporter nothing was written to playwright-report/,
so the upload step had nothing to upload and the traces recorded on a
retry were lost with the runner.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci: key the Playwright browser cache by the Playwright version

Hashing yarn.lock dropped the ~400MB browser cache on any unrelated
dependency bump. Each Playwright release pins its own browser builds,
so the installed version is all the key needs.

Also bump upload-artifact to v7, in line with the other actions.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* chore(deps): bump @editorjs/dom to ^1.2.0 and dedupe it

@editorjs/ui-kit 2.0.0 depends on @editorjs/dom ^1.2.0, while the workspace
packages still asked for ^1.1.0. Yarn will not re-resolve an already-locked
entry, so the two ranges resolved separately and node_modules ended up with
two copies of @editorjs/dom - including a nested one under @editorjs/caret,
which pins ^1.1.0 transitively. Two copies split module state and break any
identity check across the boundary.

Widen the workspace ranges to ^1.2.0 and run `yarn dedupe @editorjs/dom` so
both ranges collapse onto a single 1.2.0 resolution. The lockfile also gains
the @editorjs/ui-kit 2.0.0 resolution it was missing.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* refactor(ui,editorjs): drop dead popover field and reuse the e2e helpers

`#popoverElement` was only ever written: assigned from `getElement()`, used as
the receiver for the id and the append, then nulled in the teardown. A local
covers all three, and dropping the field removes a second piece of state that
had to be kept in step with `#popover` for no one's benefit.

`editor.spec.ts` was the only spec reimplementing the mount-and-wait sequence
and the triple-click selection instead of importing `mountDocument` and
`selectParagraphText`. The readiness signal it waited on lives in the fixture,
so a change there would have fixed the three specs using the helpers and left
this one timing out.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci(editorjs): gate PRs on the VoiceOver suite when announcements can change

The suite drives a real VoiceOver instance one test at a time, so it cannot sit
in the path of every PR the way the headless suites do. It is wired as a job
condition rather than a `paths:` filter on the workflow: GitHub leaves the
checks of a path-filtered workflow Pending forever, which blocks any PR
required to pass it, whereas a skipped job reports success and does not. The
workflow therefore always runs and the macOS job decides for itself.

`should-run-voiceover.sh` answers a narrower question than `should-run-ci.sh`:
not "could this affect @editorjs/editorjs" - true of almost every change here -
but "could this alter what a screen reader announces". It watches the UI and
paragraph sources, the bundle, the suite itself, the lockfile (@editorjs/ui-kit
owns the popover roles), and its own CI wiring. It fails open on anything it
cannot determine, for the same reason its sibling does.

Runs on a standard macos-15 runner, which is free on a public repository, and
cancels a superseded run so a stale push does not hold one of the five macOS
job slots shared across the account.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci(editorjs): fix the Guidepup setup invocation and drop the change gate

`npx @guidepup/setup` runs the package's `guidepup` binary, which needs a
subcommand - without one it prints its usage and exits 1, which is what failed
the job. The correct call is `setup`, plus the `--ci` flag the README calls for
in CI so the command does not wait on the manual steps a local machine needs.

The change detection goes away with it. Deciding what counts as announcement-
relevant turned out to be guesswork that had already missed the inline tools,
and it is moot once the suite moves off the PR path onto a schedule.

Runs on every PR for now, only so the job can be proven green before that move.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci(editorjs): install the Guidepup screen reader assets

`setup` and `install` are separate commands and both are required. `setup`
configures the OS; `install` puts the preferences disk image into the project.
With only the first, every test failed in VoiceOver.start() with "Failed to
mount Guidepup preferences" out of resolveDmgPath - the error names both
commands, and only one of them had been run.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* test(editorjs): report what VoiceOver said when Case 18's baseline fails

The baseline assertion guards the real one, but it fails on an *empty* match
list - and the matches alone say nothing in that case. On CI it failed with
"Expected: > 0, Received: 0" and no indication whether the cursor never reached
the menu or VoiceOver worded the item differently than /menu item/ expects.
`collectReachable` now carries the whole sweep alongside the matches so the
message can distinguish the two.

Retries follow playwright.config.ts, which already uses `isCI ? 2 : 0`. A real
screen reader is driven by real keystrokes whose timing shifts under a loaded
runner, and resetCursor's own comment records Case 18 as having been
non-deterministic before. This does not hide a regression: a case that fails
three times running is not a timing artefact, and the message now says what was
announced each time.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* test(editorjs): enter the toolbox before sweeping it in Case 18

The sweep never reached a menu item because it started from the top of the web
content. VO+Right moves between siblings rather than into them, and the block
actions toolbar is the last sibling, so `next()` returned that same item for all
twenty-five stops: ["hello world paragraph...", "block actions toolbar" x24].
The menu was open and correct the whole time - the page snapshot in the failure
artifact shows menuitem "Text" present.

Walking to the button that opens the toolbox first puts the cursor inside that
container, which is where Case 4 reaches menu items from.

Worth noting the baseline assertion earned its keep exactly as its comment said
it would: without it, "no menu item is reachable after filtering" would have
passed because no menu item was reachable before filtering either.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* test(editorjs): keep the cursor inside the toolbox while the menu is open

Three CI runs agree on the constraint: once the toolbox menu is open, VoiceOver
will not enter the block actions toolbar from the web content root. A forward
walk from there reports "block actions toolbar" at all twenty-five stops and
never descends, and the previous attempt showed even the button that opened the
menu is unreachable that way - "did not reach an item matching /add block/i
within 40 next steps". Case 4 reaches the items only because it never leaves the
container.

So neither sweep resets now. The first runs from where `act()` left the cursor.
The second walks backwards to the search field, which re-anchors without going
to the root and doubles as the navigation command VoiceOver needs before it will
see the `fill()`.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* refactor: trim the comments added while getting the VoiceOver suite green

The investigation notes were useful while the cause was unknown and are noise
now that it is settled. What is left is one line per non-obvious decision.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* ci(editorjs): run the VoiceOver suite on the merge queue only

Eight minutes driving a real screen reader is too slow to sit in front of every
PR push. The workflow still runs on pull_request for the rest of its jobs; the
VoiceOver job skips there, and a skipped job reports success, so it never
blocks a pull request.

Inert until a merge queue is enabled on main - `merge_group` does not fire
without one.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* docs(openspec): record the ui-kit publish as done and fix a stale claim

5a.3 was still written as the blocking, not-yet-done publish. @editorjs/ui-kit
2.0.0 is published, the lockfile resolves it, and the suite was re-run against
the published package: build, 973 unit tests, 80 e2e on both engines, lint. The
task now also records the duplicate @editorjs/dom the regeneration surfaced and
the dedupe that resolved it, since @editorjs/caret still pins the old range.

The verification summary listed the same publish as an outstanding blocker and
its verdict turned on it; both now say otherwise.

Case 18's comment credited the CSS class with removing filtered items from the
accessibility tree. ui-kit sets `hidden` alongside the class and that is what
does it - as the spec and aria.spec.ts both already said.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* docs(openspec): drop the three deferred findings from tasks.md

10.5b, 10.7b and 10.8 were records of pre-existing bugs the change deliberately
did not fix, not work items. Every archived change here has zero unchecked
tasks, so as encoded they would have blocked archiving and reported 69/72.

The findings themselves are preserved outside the repo. What referenced them by
number - the proposal, the design note, 10.1c, 10.7a and the verification
summary - now describes the bug instead, so nothing points at a task that is
gone.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* docs(openspec): drop the unarchived add-editor-aria-semantics change

Main already archives it under archive/2026-10-08-add-editor-aria-semantics; the copies are identical.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants