use-segmenter

Split text into graphemes, words, or sentences with Intl.Segmenter.


Package	`@vizejs/composable/use-segmenter`
Own the source	`vize lib pull composable:use-segmenter`
Runtime exports	`useSegmenter`
Gzip budget	2816 B

Usage

import { useSegmenter } from "@vizejs/composable/use-segmenter";

Runtime contract

Utility	Category	Stability	SSR	Hydration	Cleanup	Targets	Host globals	Uses
`useSegmenter`	i18n	experimental	deterministic-fallback	caller-managed	none	web, server, worker, native, desktop, terminal	`Intl`, `window`	`useLocale`

API

`useSegmenter`

Split text into graphemes, words, or sentences with Intl.Segmenter. Counts reflect what users perceive: "👨‍👩‍👧" is one grapheme, and word counts skip spaces and punctuation while handling languages without spaces (Japanese, Thai) through the locale. truncate shortens text without splitting clusters. Derived state only; deterministic during server rendering when the locale is explicit.

function useSegmenter( text: MaybeRefOrGetter<string>, options: MaybeRefOrGetter<UseSegmenterOptions> = {}, ): SegmenterControls

const { graphemeCount, truncate } = useSegmenter(bio, { locale: "en" });
const remaining = computed(() => 160 - graphemeCount.value);

Types

`TextSegment`

One segment of the text.

Member	Type	Description
`segment`	`string`	Segment text.
`index`	`number`	UTF-16 code-unit offset of the segment in the source text.
`isWordLike`	`boolean`	Word granularity: whether the segment is a word (not spaces or punctuation).

`UseSegmenterOptions`

Options for useSegmenter.

Member	Type	Description
`granularity`?	`SegmenterGranularity`	Segmentation unit exposed through `segments` and `count`.
`locale`?	`string \| Intl.Locale`	Locale whose segmentation rules are used (word boundaries in Japanese or Thai depend on it). Pass it explicitly for hydration-stable output.

`SegmenterControls`

Reactive segmentation returned by useSegmenter.

Member	Type	Description
`locale`	`ComputedRef<string>`	Canonical locale in use.
`segments`	`ComputedRef<readonly TextSegment[]>`	Segments at the configured granularity.
`count`	`ComputedRef<number>`	Number of segments at the configured granularity.
`graphemeCount`	`ComputedRef<number>`	User-perceived characters (grapheme clusters), independent of granularity.
`wordCount`	`ComputedRef<number>`	Word-like segments, independent of granularity.
`truncate`	`(maxGraphemes: number, ellipsis?: string) => string`	Truncate the text to at most `maxGraphemes` user-perceived characters, never splitting an emoji or combining sequence.