subtlex-ch-chr

Typed JSON distribution of the Chinese character frequency dataset SUBTLEX-CH-CHR.

This package publishes a parsed JSON version of data/SUBTLEX-CH-CHR (tab-separated source) and exposes it as a default ESM export with TypeScript definitions.

What this package contains

• SUBTLEX-CH-CHR.json: Parsed dataset payload
• index.js: ESM entrypoint that default-exports the JSON
• index.d.ts: Handwritten type definitions for the exported structure

The exported object shape is:

• metadata.totalCharacterCount: total character tokens in the corpus
• metadata.contextNumber: number of subtitle contexts
• headers: original source headers (preserved exactly)
• data: array of rows with fields:
  • Character
  • CHRCount
  • CHR/million
  • logCHR
  • CHR-CD
  • CHR-CD%
  • logCHR-CD

Install

npm install subtlex-ch-chr

Usage

import subtlexChChr from "subtlex-ch-chr";

console.log(subtlexChChr.metadata.totalCharacterCount);
console.log(subtlexChChr.data[0]);

You can also import the raw JSON subpath export:

import dataset from "subtlex-ch-chr/SUBTLEX-CH-CHR.json";

Regenerating the JSON

The source TSV-like file lives at data/SUBTLEX-CH-CHR. To regenerate the published JSON:

bun run convert:data

This runs scripts/convert-subtlex-ch-chr.ts and writes SUBTLEX-CH-CHR.json to the repository root.

Repository

• Source: github.com/leonsilicon/subtlex-ch-chr