@node-media-library/pdf

PDF-backed ImageGenerator using poppler's pdftoppm for rendering pages as PNG, then applying conversions through the sharp pipeline.

Installation

pnpm add @node-media-library/pdf sharp

sharp is a required optional peer. This package rasterizes into core's sharpImageGenerator(), so sharp must be installed even though it is declared optional. It is marked optional only because core declares it that way.

System Requirements

Requires the pdftoppm binary from the poppler utilities:

• macOS: brew install poppler
• Linux (Debian/Ubuntu): apt install poppler-utils
• Linux (Fedora/RHEL): dnf install poppler-utils

Usage

Append the PDF generator to your imageGenerators array alongside sharpImageGenerator:

import { createMediaLibrary, sharpImageGenerator } from '@node-media-library/core'
import { pdfImageGenerator } from '@node-media-library/pdf'

createMediaLibrary({
  // ...
  imageGenerators: [sharpImageGenerator(), pdfImageGenerator()],
})

Configuration

pdfImageGenerator(options?)

Options

• pdftoppmPath (string): Path to the pdftoppm binary. Default: 'pdftoppm' (assumes it's on PATH)
• dpi (number): Render resolution in DPI. Default: 150

Example

pdfImageGenerator({
  pdftoppmPath: '/usr/local/bin/pdftoppm',
  dpi: 300,
})

Selecting Pages

Use conversion().pdfPageNumber(n) to select which page to render (default is page 1):

import { collection, conversion } from '@node-media-library/core'

collection().conversions({
  // Render page 2 at 800px wide
  cover: conversion().pdfPageNumber(2).width(800),
})

conversions() takes the builders themselves — don't call .toDefinition(), which is applied for you.

Testing

Tests include binary-gated suites that skip when pdftoppm is not available. Pure tests (those not requiring the binary) always run:

pnpm --filter @node-media-library/pdf test

• ✓ Pure tests (MIME type support) run without pdftoppm
• ⊘ Binary-gated tests skip if pdftoppm not found
• Install poppler to run all tests