@iyulab/undoc (WASM)

Browser/Node.js WebAssembly package for undoc.

Extracts DOCX, XLSX, and PPTX files to Markdown, plain text, or JSON — no server required.

Install

npm install @iyulab/undoc

Usage (ESM)

This package is built with wasm-pack --target bundler. It is an ES module and the WebAssembly binary is initialised for you, so there is no init() to await — import the functions and call them. Use it through a bundler (webpack, Vite, Rollup, esbuild).

import { parse } from '@iyulab/undoc';

const bytes = new Uint8Array(await file.arrayBuffer());
const doc = parse(bytes);

console.log(doc.format());       // "docx" | "xlsx" | "pptx"
console.log(doc.toMarkdown());   // Markdown string
console.log(doc.toText());       // Plain text string
console.log(doc.toJson());       // JSON string

API

supportedFormats(): string

JSON array of the formats this package can parse:

import { supportedFormats } from '@iyulab/undoc';

JSON.parse(supportedFormats());
// [{ extension: "docx", name: "Word Document" },
//  { extension: "xlsx", name: "Excel Workbook" },
//  { extension: "pptx", name: "PowerPoint Presentation" }]

Ask rather than hardcoding the extensions on your side — a local copy of the list goes stale the moment this package learns a new format, and nothing tells you it has.

parse(data: Uint8Array): OfficeDocument

Parse a DOCX, XLSX, or PPTX byte array. Throws if the format is unrecognized.

OfficeDocument

| Method | Returns | Description | |--------|---------|-------------| | fromBytes(data) | OfficeDocument | Alias for module-level parse() | | format() | string | "docx" | "xlsx" | "pptx" | | toMarkdown() | string | Full document as Markdown | | toText() | string | Plain text extraction | | toJson() | string | Structured JSON | | metadata() | string | JSON of title/author/subject etc. |

Playground

Live demo: https://iyulab.github.io/undoc/