@love-rox/tcy-core

Framework-agnostic tokenizer for Japanese tate-chu-yoko (縦中横) span wrapping.

Returns a Segment[] that splits a string into text chunks and target chunks (typically half-width alphanumerics). Framework wrappers build on top of this:

• @love-rox/tcy-react — React <Tcy> component
• @love-rox/tcy-vue — Vue 3 <Tcy> component
• @love-rox/tcy-rehype — rehype plugin for HAST
• @love-rox/tcy-astro — Astro integration + <Tcy> component

Install

pnpm add @love-rox/tcy-core

Usage

import { tokenize } from '@love-rox/tcy-core';

tokenize('第1章 2026年4月');
// [
//   { type: 'text', value: '第' },
//   { type: 'tcy',  value: '1' },
//   { type: 'text', value: '章 ' },
//   { type: 'tcy',  value: '2026' },
//   { type: 'text', value: '年' },
//   { type: 'tcy',  value: '4' },
//   { type: 'text', value: '月' },
// ]

API

type Segment = { type: 'text'; value: string } | { type: 'tcy'; value: string };

function tokenize(input: string, options?: TcyOptions): Segment[];

interface TcyOptions {
  target?: 'alphanumeric' | 'alpha' | 'digit' | 'ascii' | RegExp; // default: 'alphanumeric'
  combine?: boolean; // default: true
  include?: string | string[];
  exclude?: string | string[];
  maxLength?: number;
  excludeWords?: string[];
}

• target: preset or custom RegExp for the characters to wrap. alphanumeric = [0-9A-Za-z]; ascii = full printable ASCII
• combine: if true, consecutive target characters become one tcy segment; if false, each character becomes its own segment
• include / exclude: per-character overrides. exclude wins over include
• maxLength: maximum length for a tcy segment. Segments longer than this are demoted to plain text
• excludeWords: exact words to exclude from tcy wrapping. Matched against the combined segment value

Links

• Full documentation: Love-Rox/tate-chu-yoko (日本語)
• Issues: https://github.com/Love-Rox/tate-chu-yoko/issues

License

MIT