fix(plugin-sdk-value): drop remote normalize-merge mutation echoes

[ryanbonial]

Description

Receiving editors were pushing their own normalize fallout after applying a collaborator's formatting (adjacent same-mark span merges). Those competing merges truncated or duplicated text on the originator's block under concurrent edit (Axios format-while-peer-types).

Studio FormBuilder does not show this class for the same gesture; the SDK plugin path does because it eagerly pushes every mutation flush. This change arms a short gate after remote apply and drops a flush only when its patches look like a normalize merge (unset of a keyed children member plus text diffMatchPatch/set). Ordinary typing flushes are unaffected.

Supersedes the approach in #2999 (core withoutPatching nesting), which broke self-solving tests. Prefer fixing at the plugin push boundary.

Not in scope: paste-count loss, same-block typing scramble, overlapping-format SILENT-LOSS. Those remain a separate roadmap; Phase 0 research will decide whether follow-ups branch from this PR or from main.

What to review

• packages/plugin-sdk-value/src/plugin.sdk-value.tsx: remoteApplyEchoGate / isNormalizeMergeEcho / drop in listenToEditor
• packages/plugin-sdk-value/src/plugin.value-sync.browser.test.tsx: echo not pushed; unpushed typing still pushed after remote normalize

Testing

• Plugin browser suite: 24/24 including new regression tests
• External harness (pte-sdk-concurrent-repro on local plugin build):
  • duo-typing.mjs format: block A text integrity restored vs released (released still mangled)
  • duo-typing.mjs blocks / solo-typing.mjs: green after shape-based gate (value-equality gate had caused blocks scramble; fixed before this PR)
  • Full matrix still has known reds (paste counts, same-block, overlapping SILENT-LOSS); not claimed fixed here

Notes for release

Customer-facing: concurrent bold/format while another user types in another block should no longer scramble the formatted block's text via competing normalize merges when using @portabletext/plugin-sdk-value.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
portable-text-example-basic	Ready	Preview, Comment	Jul 22, 2026 3:17am

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
portable-text-playground	Skipped		Jul 22, 2026 3:17am

[changeset-bot]

⚠️ No Changeset found

Latest commit: 16925b9

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[github-actions]

📦 Bundle Stats — @portabletext/editor

Compared against main (11d72d02)

@portabletext/editor

Metric	Value	vs main (11d72d0)
Internal (raw)	806.6 KB	-
Internal (gzip)	154.9 KB	-
Bundled (raw)	1.42 MB	-
Bundled (gzip)	319.2 KB	-
Import time	90ms	+0ms, +0.5%

@portabletext/editor/behaviors

Metric	Value	vs main (11d72d0)
Internal (raw)	467 B	-
Internal (gzip)	207 B	-
Bundled (raw)	424 B	-
Bundled (gzip)	171 B	-
Import time	2ms	+0ms, +3.1%

@portabletext/editor/plugins

Metric	Value	vs main (11d72d0)
Internal (raw)	2.7 KB	-
Internal (gzip)	894 B	-
Bundled (raw)	2.5 KB	-
Bundled (gzip)	827 B	-
Import time	7ms	+0ms, +1.1%

@portabletext/editor/selectors

Metric	Value	vs main (11d72d0)
Internal (raw)	82.7 KB	-
Internal (gzip)	15.4 KB	-
Bundled (raw)	78.4 KB	-
Bundled (gzip)	14.3 KB	-
Import time	7ms	-0ms, -1.0%

@portabletext/editor/traversal

Metric	Value	vs main (11d72d0)
Internal (raw)	28.1 KB	-
Internal (gzip)	5.6 KB	-
Bundled (raw)	28.1 KB	-
Bundled (gzip)	5.5 KB	-
Import time	6ms	+0ms, +0.6%

@portabletext/editor/utils

Metric	Value	vs main (11d72d0)
Internal (raw)	30.6 KB	-
Internal (gzip)	6.4 KB	-
Bundled (raw)	28.2 KB	-
Bundled (gzip)	6.1 KB	-
Import time	6ms	-0ms, -0.6%

🗺️ . · ./behaviors · ./plugins · ./selectors · ./traversal · ./utils · Artifacts

Details

• Import time regressions over 10% are flagged with ⚠️
• Sizes shown as raw / gzip 🗜️. Internal bytes = own code only. Total bytes = with all dependencies. Import time = Node.js cold-start median.

📦 Bundle Stats — @portabletext/markdown

Compared against main (11d72d02)

Metric	Value	vs main (11d72d0)
Internal (raw)	53.8 KB	-
Internal (gzip)	9.8 KB	-
Bundled (raw)	348.9 KB	-
Bundled (gzip)	96.3 KB	-
Import time	35ms	-2ms, -5.5%

🗺️ View treemap · Artifacts

Details

• Import time regressions over 10% are flagged with ⚠️
• Sizes shown as raw / gzip 🗜️. Internal bytes = own code only. Total bytes = with all dependencies. Import time = Node.js cold-start median.

[ryanbonial]

Confirmation vs #3001 / @portabletext/editor@7.10.11

Harness rebaseline on released @portabletext/editor@7.10.11 + @portabletext/plugin-sdk-value@7.1.3 (no push-boundary echo gate from this PR):

Cell	Result
duo-typing.mjs format	4/4 SAFE (first run); confirming second run
Completed trials MANGLED	0
Plugin contains remoteApplyEchoGate	No (7.1.3)

Agreed with the #3001 rationale: skipping cosmetic normalize on remote apply removes the echo at the source, so this plugin push gate has nothing left to drop. Keeping a local-only merge would still fork span structure from the store.

Recommendation: close this draft as superseded by #3001. Residual roadmap work (paste counts, same-block scramble, overlapping-format SILENT-LOSS) is separate and does not need this branch.