@ideadesignmedia/convert-pdf-md

A tiny TypeScript CLI that converts Markdown <-> PDF with zero browser/runtime baggage. It includes a rich renderer/extractor for best-effort formatting plus a lean core fallback for quick text-only conversion.

Features

• Autodetects input (.md, .markdown, .pdf) and infers output paths.
• CLI supports --out, --stdout, --force, and --engine to swap between rich/core modes.
• Markdown -> PDF powered by pdfkit (headings, lists, code fences, blockquotes, tables, basic image handling).
• PDF -> Markdown powered by pdfjs-dist with heuristics (headings from font size, bold/italic, bullets, links, placeholders for images).
• Ships compiled CommonJS (dist/) only; .npmignore keeps sources and tooling out of the package.
• Tooling: ESLint, Prettier, Vitest with end-to-end style tests against the built bundle.

WARNING: PDF -> Markdown uses text extraction and heuristics. Layout, custom fonts, and complex graphics cannot be fully preserved.

Install (local dev)

yarn install
yarn build

Node.js 18+ is required.

CLI usage

Once published:

npx @ideadesignmedia/convert-pdf-md <input> [--out path] [--stdout] [--force] [--engine rich|core]

Local development (after yarn build):

node dist/cli.js sample/sample.md -o sample/out.pdf
node dist/cli.js sample/sample.pdf -o sample/out.md

Engines

• rich (default)
  • Markdown -> PDF: styled output via pdfkit (bold/italic, headings, numbered/bullet lists, code blocks with shading, blockquotes, tables, links, image embedding).
  • PDF -> Markdown: structure-aware extraction using pdfjs-dist (font-size based heading detection, bold/italic, bullet inference, link recovery, image placeholders).
• core
  • Markdown -> PDF: same renderer as rich but without image lookups.
  • PDF -> Markdown: quick text extraction via pdfjs-dist without layout heuristics.

Switch engines on demand:

npx @ideadesignmedia/convert-pdf-md input.pdf --engine core
npx @ideadesignmedia/convert-pdf-md input.md --engine rich

Programmatic usage

import {
  convertFile,
  convertMarkdownToPdf,
  convertPdfToMarkdown,
  mdToPdf,
  pdfToMd,
} from '@ideadesignmedia/convert-pdf-md';

// Auto-detect by extension / magic
await convertFile('path/to/input.md');
await convertFile('path/to/input.pdf', { out: 'output.md', engine: 'core' });

// Direct helpers
const pdfBuffer = await convertMarkdownToPdf('# Hello', { engine: 'rich' });
const markdown = await convertPdfToMarkdown(pdfBuffer);
await mdToPdf('input.md', 'output.pdf', { margin: 48 });
await pdfToMd('input.pdf', 'output.md');

Repository

https://github.com/ideadesignmedia/convert-pdf-md.git

Development workflow

yarn lint
yarn build
yarn test

Vitest runs against the compiled output in dist/ to mirror the published package.

License

MIT � Idea Design Media