feat(plugin-table): add tokens to the reference Table

[christianhg]

The plugin themes through --pt-plugin-table-* custom properties, which works cleanly when the host's theme is CSS: declare overrides in a stylesheet and the zero-specificity defaults yield. But when the theme exists only as a runtime object (a design system resolving colors, fonts, and radii in JavaScript), the host has to write the properties into the DOM itself, and every scope it owns fails one way or another: a wrapper around the editor cannot reach the chrome that portals out of the editor's subtree (the trash chips, the built-in menu's dropdown), and a shared global scope like documentElement makes concurrent tables fight over one set of values, with the last unmount's cleanup stripping properties a surviving table still needs.

Table now accepts the values through its own API:

render: (props) => (
  <Table
    {...props}
    tokens={{
      "--pt-plugin-table-accent": theme.focusRing,
      "--pt-plugin-table-bg": theme.bg,
      // ...any subset; the stylesheet defaults cover the rest
    }}
  />
)

The record's keys are the documented token names, so the README's Theming table remains the single contract and the prop is a second delivery route for the same vocabulary, not a second vocabulary. The plugin applies the values inline to the roots it owns: the in-tree chrome wrapper (covering the table, handles, lanes, dots, and drag ghost by inheritance) and each portal layer (the trash layer and the built-in menu's dropdown carry their own copy, since portals escape the in-tree scope). Inline properties win over the stylesheet defaults exactly where a value is passed and nowhere else, and each table instance carries its own set, so two tables with different tokens theme independently by construction.

This is the same host-facing shape widgets with their own theming systems conventionally expose: live theme values in through the widget's API, applied by the widget to surfaces only it knows about. The application mechanism is inherited custom properties rather than generated stylesheets because the plugin's styling is one static stylesheet already parameterized by these properties, so no rule generation is needed, only values.

The pin sets --pt-plugin-table-bg and --pt-plugin-table-trash-bg through the prop and asserts computed colors in both worlds: a cell resolving the value by inheritance through the in-tree root, and the trash chip resolving it from its portaled layer's own copy (summoned by selecting a row via keyboard, so the scenario also exercises the portal path end to end).

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
portable-text-editor-documentation	Ready	Preview, Comment	Jul 7, 2026 3:08pm
portable-text-example-basic	Ready	Preview, Comment	Jul 7, 2026 3:08pm
portable-text-playground	Ready	Preview, Comment	Jul 7, 2026 3:08pm

[changeset-bot]

🦋 Changeset detected

Latest commit: b81a361

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
@portabletext/plugin-table	Minor

Not sure what this means? Click here to learn what changesets are.

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

[github-actions]

📦 Bundle Stats — @portabletext/editor

Compared against main (9dbc50f5)

@portabletext/editor

Metric	Value	vs main (9dbc50f)
Internal (raw)	803.0 KB	-
Internal (gzip)	153.7 KB	-
Bundled (raw)	1.41 MB	-
Bundled (gzip)	317.9 KB	-
Import time	96ms	+0ms, +0.0%

@portabletext/editor/behaviors

Metric	Value	vs main (9dbc50f)
Internal (raw)	467 B	-
Internal (gzip)	207 B	-
Bundled (raw)	424 B	-
Bundled (gzip)	171 B	-
Import time	2ms	+0ms, +1.4%

@portabletext/editor/plugins

Metric	Value	vs main (9dbc50f)
Internal (raw)	2.7 KB	-
Internal (gzip)	894 B	-
Bundled (raw)	2.5 KB	-
Bundled (gzip)	827 B	-
Import time	7ms	+0ms, +0.1%

@portabletext/editor/selectors

Metric	Value	vs main (9dbc50f)
Internal (raw)	81.2 KB	-
Internal (gzip)	14.9 KB	-
Bundled (raw)	76.7 KB	-
Bundled (gzip)	13.8 KB	-
Import time	8ms	-0ms, -1.2%

@portabletext/editor/traversal

Metric	Value	vs main (9dbc50f)
Internal (raw)	28.1 KB	-
Internal (gzip)	5.6 KB	-
Bundled (raw)	27.9 KB	-
Bundled (gzip)	5.5 KB	-
Import time	6ms	+0ms, +0.3%

@portabletext/editor/utils

Metric	Value	vs main (9dbc50f)
Internal (raw)	29.7 KB	-
Internal (gzip)	6.2 KB	-
Bundled (raw)	27.1 KB	-
Bundled (gzip)	5.8 KB	-
Import time	6ms	-0ms, -0.9%

🗺️ . · ./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 (9dbc50f5)

Metric	Value	vs main (9dbc50f)
Internal (raw)	53.8 KB	-
Internal (gzip)	9.8 KB	-
Bundled (raw)	348.9 KB	-
Bundled (gzip)	96.3 KB	-
Import time	38ms	+0ms, +1.3%

🗺️ 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.