WebAssembly API

Package

npm: @edgeparse/edgeparse-wasm · Source: crates/edgeparse-wasm/

Initialization

Before calling any conversion function, initialize the WASM module:

import init from '@edgeparse/edgeparse-wasm';

// Async initialization — fetches and compiles the .wasm binary
await init();

Call init() once at application startup. Subsequent calls are no-ops.

convert()

Parses PDF bytes and returns a structured JavaScript object representing the full document.

function convert(
  pdf_bytes: Uint8Array,
  format?: string | null,
  pages?: string | null,
  reading_order?: string | null,
  table_method?: string | null,
): any;

Parameters

Parameter	Type	Default	Description
pdf_bytes	Uint8Array	(required)	Raw PDF file bytes
format	string | null	"json"	Output format hint: "json", "markdown", "html", "text"
pages	string | null	"all"	Page range: "all", "1-5", "1,3,7"
reading_order	string | null	"auto"	Reading order algorithm: "auto" (XY-Cut++) or "off"
table_method	string | null	"default"	Table detection: "default" (ruling lines) or "cluster" (borderless)

Return value

A JavaScript object matching the PdfDocument structure:

{
  pages: [
    {
      page_number: 1,
      width: 612.0,
      height: 792.0,
      elements: [
        {
          type: "heading",       // "heading" | "paragraph" | "table" | "list" | "image" | ...
          text: "Introduction",
          level: 1,              // heading level (1-6)
          bbox: {
            x0: 72.0,
            y0: 700.0,
            x1: 300.0,
            y1: 720.0,
          },
        },
        // ... more elements
      ],
    },
  ],
}

Example

import init, { convert } from '@edgeparse/edgeparse-wasm';

await init();

const bytes = new Uint8Array(await file.arrayBuffer());
const doc = convert(bytes, 'json');

// Iterate pages and elements
for (const page of doc.pages) {
  for (const el of page.elements) {
    if (el.type === 'table') {
      console.log('Table found on page', page.page_number);
    }
  }
}

convert_to_string()

Parses PDF bytes and returns a formatted string.

function convert_to_string(
  pdf_bytes: Uint8Array,
  format?: string | null,
  pages?: string | null,
  reading_order?: string | null,
  table_method?: string | null,
): string;

Parameters

Same as convert().

Return value

A string in the requested format:

Format	Output
"json"	JSON string with bounding boxes and element types
"markdown"	Standard Markdown with GFM tables
"html"	HTML5 with semantic elements
"text"	Plain UTF-8 text with reading order preserved

Example

import init, { convert_to_string } from '@edgeparse/edgeparse-wasm';

await init();

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

const markdown = convert_to_string(bytes, 'markdown');
const html = convert_to_string(bytes, 'html');
const text = convert_to_string(bytes, 'text');
const json = convert_to_string(bytes, 'json');

version()

Returns the EdgeParse version string.

function version(): string;

Example

import { version } from '@edgeparse/edgeparse-wasm';
console.log(version()); // "0.1.1"

Error handling

Both convert() and convert_to_string() throw JsError on failure (corrupted PDF, invalid page range, etc.):

try {
  const markdown = convert_to_string(bytes, 'markdown');
} catch (err) {
  console.error('PDF parsing failed:', err.message);
}

Browser compatibility

Browser	Minimum version
Chrome	57+
Firefox	52+
Safari	11+
Edge	16+

Requires WebAssembly and ES module support. The build.target: 'esnext' setting in Vite/Webpack ensures compatibility.

TypeScript support

Full .d.ts type definitions ship with the package. Your IDE will provide autocomplete and type checking automatically.

Live demo

Try the interactive demo: edgeparse.com/demo/