### Install unpdf

Source: https://github.com/unjs/unpdf/blob/main/README.md

Install unpdf using pnpm or npm.

```bash
# pnpm
pnpm add unpdf

# npm
npm install unpdf
```

--------------------------------

### Override PDF.js Build with definePDFJSModule

Source: https://context7.com/unjs/unpdf/llms.txt

Register a custom PDF.js resolver before any other unpdf function. This example shows how to use the official pdfjs-dist package instead of the bundled serverless build. Ensure you have installed `pdfjs-dist`.

```typescript
import { definePDFJSModule, extractText, getDocumentProxy } from 'unpdf'

// Use the official pdfjs-dist instead of the bundled serverless build.
// Requires: npm install pdfjs-dist
// WARNING: pdfjs-dist v5.x requires Promise.withResolvers (Node >= 22).
await definePDFJSModule(() => import('pdfjs-dist'))

const buffer = await fetch('https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf')
  .then(res => res.arrayBuffer())

const pdf = await getDocumentProxy(new Uint8Array(buffer))
const { totalPages, text } = await extractText(pdf, { mergePages: true })

console.log(`Pages: ${totalPages}`)
// => Pages: 1
console.log(text)
// => "Dummy PDF file"
```

--------------------------------

### Load PDF and Reuse Proxy with getDocumentProxy

Source: https://context7.com/unjs/unpdf/llms.txt

Parses binary PDF data into a `PDFDocumentProxy` for efficient multi-operation workflows. This example demonstrates reusing the same proxy for metadata extraction and text extraction. Custom options can override default behaviors like font rendering.

```typescript
import { readFile } from 'node:fs/promises'
import { extractLinks, extractText, getDocumentProxy, getMeta } from 'unpdf'

const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Reuse the same proxy for multiple operations
const { info } = await getMeta(pdf)
const { text } = await extractText(pdf, { mergePages: true })
const { links } = await extractLinks(pdf)

console.log(`Author: ${info.Author}`)
// => Author: Evangelos Vlachogiannis
console.log(`Text preview: ${text.slice(0, 50)}`)
// => Text preview: Dummy PDF file
console.log(`Links found: ${links.length}`)

// Custom options – override font rendering for accurate output
const pdfCustom = await getDocumentProxy(new Uint8Array(buffer), {
  disableFontFace: false,
  standardFontDataUrl: 'https://unpkg.com/pdfjs-dist@latest/standard_fonts/'
})
```

--------------------------------

### Render PDF Page to Image (Node.js/Browser)

Source: https://github.com/unjs/unpdf/blob/main/README.md

Renders a specific page of a PDF document into an ArrayBuffer representing a PNG image. Ensure the official PDF.js build is used and `@napi-rs/canvas` is installed for Node.js environments. The `definePDFJSModule` function must be called to set up PDF.js.

```typescript
import { definePDFJSModule, renderPageAsImage } from 'unpdf'

// Use the official PDF.js build
await definePDFJSModule(() => import('pdfjs-dist'))

const pdf = await readFile('./dummy.pdf')
const buffer = new Uint8Array(pdf)
const pageNumber = 1

const result = await renderPageAsImage(buffer, pageNumber, {
  canvasImport: () => import('@napi-rs/canvas'),
  scale: 2,
})
await writeFile('dummy-page-1.png', new Uint8Array(result))
```

--------------------------------

### Get Resolved PDF.js Module

Source: https://github.com/unjs/unpdf/blob/main/README.md

Provides access to the resolved PDF.js module, enabling direct interaction with the PDF.js API. It defaults to the serverless build if no custom module has been defined.

```APIDOC
## Get Resolved PDF.js Module

### Description
Returns the resolved PDF.js module, allowing direct use of the PDF.js API. It defaults to the serverless build if no custom PDF.js module has been defined using `definePDFJSModule`.

### Method
`getResolvedPDFJS(): Promise<PDFJS>`

### Parameters
None

### Request Example
```ts
import { getResolvedPDFJS } from 'unpdf'

const { version } = await getResolvedPDFJS()
console.log(`Using PDF.js version: ${version}`)
```

### Response
#### Success Response (200)
- **PDFJS** (object) - The resolved PDF.js module, providing access to its API.

#### Response Example
```json
{
  "version": "5.6.205"
}
```
```

--------------------------------

### Get PDF Metadata Using PDF.js API

Source: https://github.com/unjs/unpdf/blob/main/README.md

Loads a PDF file using the PDF.js API and retrieves its metadata. Ensure the file path is correct.

```typescript
import { readFile } from 'node:fs/promises'
import { getResolvedPDFJS } from 'unpdf'

const { getDocument } = await getResolvedPDFJS()
const data = await readFile('./dummy.pdf')
const document = await getDocument(new Uint8Array(data)).promise

console.log(await document.getMetadata())
```

--------------------------------

### Access PDF.js API Directly

Source: https://github.com/unjs/unpdf/blob/main/README.md

Get the resolved PDF.js module to use its API directly. If no custom build is defined, the serverless build is used by default.

```typescript
import { getResolvedPDFJS } from 'unpdf'

const { version } = await getResolvedPDFJS()
```

--------------------------------

### Extract Plain Text from PDF with extractText

Source: https://context7.com/unjs/unpdf/llms.txt

Use extractText to get all text content from a PDF. By default, it returns an array of per-page strings. Set mergePages to true to get a single concatenated string with collapsed whitespace.

```typescript
import { readFile } from 'node:fs/promises'
import { extractText, getDocumentProxy } from 'unpdf'

const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Per-page text array (default)
const { totalPages, text: pages } = await extractText(pdf)
console.log(`Total pages: ${totalPages}`) // => Total pages: 3
pages.forEach((pageText, i) => {
  console.log(`Page ${i + 1}:`, pageText.slice(0, 80))
})

// Merged into a single string (ideal for AI summarization)
const { text } = await extractText(pdf, { mergePages: true })
console.log(typeof text) // => "string"
console.log(text.slice(0, 120))
// => "Dummy PDF file"

// Works equally on remote PDFs
const remoteBuffer = await fetch('https://example.com/report.pdf').then(r => r.arrayBuffer())
const { text: remoteText } = await extractText(new Uint8Array(remoteBuffer), { mergePages: true })
```

--------------------------------

### Get PDF Metadata

Source: https://github.com/unjs/unpdf/blob/main/README.md

Extracts metadata information from a PDF document. Optionally, date properties within the metadata can be parsed into JavaScript Date objects.

```APIDOC
## Get PDF Metadata

### Description
Extracts metadata and information from a PDF document. Supports parsing date fields into `Date` objects if the `parseDates` option is enabled.

### Method
`getMeta(data: DocumentInitParameters['data'] | PDFDocumentProxy, options?: { parseDates?: boolean }): Promise<{ info: Record<string, any>; metadata: Record<string, any> }>`

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Request Example
```ts
import { getMeta } from 'unpdf'
import { readFile } from 'node:fs/promises'

const data = await readFile('./document.pdf')
const { info, metadata } = await getMeta(new Uint8Array(data), { parseDates: true })

console.log('Info:', info)
console.log('Metadata:', metadata)
```

### Response
#### Success Response (200)
- **info** (Record<string, any>) - General information about the PDF.
- **metadata** (Record<string, any>) - Extracted metadata from the PDF.

#### Response Example
```json
{
  "info": {
    "PDFFormatVersion": "1.5",
    "IsLinearized": true
  },
  "metadata": {
    "CreationDate": "D:20230101T100000Z",
    "ModDate": "D:20230101T110000Z"
  }
}
```
```

--------------------------------

### Extract Structured Text Items with extractTextItems

Source: https://context7.com/unjs/unpdf/llms.txt

Use extractTextItems to get rich text item objects per page, including position, dimensions, font details, and reading direction. This is useful for layout analysis or building bounding-box overlays.

```typescript
import { readFile } from 'node:fs/promises'
import { extractTextItems, getDocumentProxy } from 'unpdf'
import type { StructuredTextItem } from 'unpdf'

const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

const { totalPages, items } = await extractTextItems(pdf)
console.log(`Pages: ${totalPages}`) // => Pages: 1

// items is a StructuredTextItem[][] — one array per page
const firstPageItems: StructuredTextItem[] = items[0]!
const firstItem = firstPageItems[0]!

console.log(firstItem)
// => {
//   str: "Dummy PDF file",
//   x: 56.8,
//   y: 758.1,
//   width: 123.41,
//   height: 16.1,
//   fontSize: 16.1,
//   fontFamily: "sans-serif",
//   dir: "ltr",
//   hasEOL: false
// }

// Example: collect all text larger than 14pt (likely headings)
const headings = firstPageItems.filter(item => item.fontSize > 14)
console.log(headings.map(h => h.str))
// => ["Dummy PDF file"]
```

--------------------------------

### Customizing getDocumentProxy Options

Source: https://github.com/unjs/unpdf/blob/main/README.md

Demonstrates how to provide custom options to `getDocumentProxy` for fine-grained control over PDF.js behavior, such as disabling font face loading or specifying a custom URL for standard font data.

```typescript
const pdf = await getDocumentProxy(buffer, {
  disableFontFace: false,
  standardFontDataUrl: 'https://unpkg.com/pdfjs-dist@latest/standard_fonts/',
})
```

--------------------------------

### renderPageAsImage

Source: https://context7.com/unjs/unpdf/llms.txt

Renders a complete PDF page using a canvas backend and returns the result as an ArrayBuffer (default) or base64 data URL. Requires the official `pdfjs-dist` build and the `@napi-rs/canvas` package in Node.js. Scale can be set explicitly or derived from a target `width` or `height`.

```APIDOC
## renderPageAsImage

### Description
Renders a complete PDF page using a canvas backend and returns the result as an `ArrayBuffer` (default) or base64 data URL (`toDataURL: true`). Requires the official `pdfjs-dist` build (not the serverless bundle) and the `@napi-rs/canvas` package in Node.js. Scale can be set explicitly or derived from a target `width` or `height`.

### Usage
```ts
import { readFile, writeFile } from 'node:fs/promises'
import { definePDFJSModule, getDocumentProxy, renderPageAsImage } from 'unpdf'

// Step 1: Switch to the official PDF.js build
await definePDFJSModule(() => import('pdfjs-dist'))

const buffer = new Uint8Array(await readFile('./document.pdf'))

// Render page 1 as PNG ArrayBuffer (2× scale)
const pngBuffer = await renderPageAsImage(buffer, 1, {
  canvasImport: () => import('@napi-rs/canvas'), // Required in Node.js
  scale: 2,
})
await writeFile('page1.png', new Uint8Array(pngBuffer))
// => page1.png written (valid PNG, header: [137,80,78,71,13,10,26,10])

// Render with fixed width (scale is calculated automatically)
const fixedWidthBuffer = await renderPageAsImage(buffer, 1, {
  canvasImport: () => import('@napi-rs/canvas'),
  width: 1200,
})
await writeFile('page1-1200w.png', new Uint8Array(fixedWidthBuffer))

// Return a data URL instead (e.g., for embedding in HTML)
const dataURL = await renderPageAsImage(buffer, 1, {
  canvasImport: () => import('@napi-rs/canvas'),
  scale: 1.5,
  toDataURL: true,
})
console.log(dataURL.startsWith('data:image/png;base64,')) // => true

const html = `<!DOCTYPE html><html><body><img src="${dataURL}"></body></html>`
await writeFile('page1.html', html)
```
```

--------------------------------

### Define Custom PDF.js Build

Source: https://github.com/unjs/unpdf/blob/main/README.md

Allows defining a custom PDF.js build before using other unpdf methods. This is useful for specifying official or legacy builds. Note that older Node.js versions might not support Promise.withResolvers.

```typescript
import { definePDFJSModule, extractText, getDocumentProxy } from 'unpdf'

// Define the PDF.js build before using any other unpdf method
await definePDFJSModule(() => import('pdfjs-dist'))

// Now, you can use all unpdf methods with the official PDF.js build
const pdf = await getDocumentProxy(/* … */)
const { text } = await extractText(pdf)
```

--------------------------------

### Extract PDF Metadata with getMeta

Source: https://context7.com/unjs/unpdf/llms.txt

Use getMeta to retrieve the /Info dictionary and XMP metadata. Set parseDates to true to convert date fields into Date objects.

```typescript
import { readFile } from 'node:fs/promises'
import { getDocumentProxy, getMeta } from 'unpdf'

const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Basic metadata (dates remain as raw strings)
const { info, metadata } = await getMeta(pdf)
console.log(info)
// => {
//   Author: "Evangelos Vlachogiannis",
//   CreationDate: "D:20070223175637+02'00'",
//   Creator: "Writer",
//   PDFFormatVersion: "1.4",
//   Producer: "OpenOffice.org 2.1",
//   IsLinearized: false,
//   ...
// }

// Parse date fields into Date objects
const { info: parsedInfo, metadata: parsedMeta } = await getMeta(pdf, { parseDates: true })
console.log(parsedInfo.CreationDate instanceof Date) // => true
console.log(parsedInfo.CreationDate.getFullYear())   // => 2007

// Access XMP dates (also parsed when parseDates: true)
const createDate = parsedMeta.get('xmp:createdate')
console.log(createDate instanceof Date) // => true
```

--------------------------------

### Create Isomorphic Canvas Factory

Source: https://context7.com/unjs/unpdf/llms.txt

Provides an environment-aware canvas factory for browser (DOM) or Node.js (@napi-rs/canvas). Used internally by renderPageAsImage for custom rendering.

```typescript
import { createIsomorphicCanvasFactory, definePDFJSModule, getResolvedPDFJS } from 'unpdf'

await definePDFJSModule(() => import('pdfjs-dist'))

// Obtain the correct CanvasFactory for this runtime
const CanvasFactory = await createIsomorphicCanvasFactory(
  () => import('@napi-rs/canvas'), // required in Node.js; omit in browser
)

// Use it directly with getDocument for custom rendering pipelines
const { getDocument } = await getResolvedPDFJS()
const doc = await getDocument({
  data: pdfData,
  CanvasFactory,
  isEvalSupported: false,
}).promise

const page = await doc.getPage(1)
const viewport = page.getViewport({ scale: 2 })
const { canvas, context } = (new CanvasFactory()).create(viewport.width, viewport.height)

await page.render({ canvas, canvasContext: context, viewport }).promise
```

--------------------------------

### Extract Text From PDF

Source: https://github.com/unjs/unpdf/blob/main/README.md

Demonstrates how to extract text content from a PDF file. It involves fetching the PDF, obtaining a PDF document proxy, and then extracting text, with an option to merge text from all pages.

```APIDOC
## Extract Text From PDF

### Description
Extracts text content from a PDF document. Supports merging text from all pages into a single string.

### Method
`extractText(pdf: PDFDocumentProxy, options?: { mergePages?: boolean }): Promise<{ totalPages: number; text: string }>`

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Request Example
```ts
import { extractText, getDocumentProxy } from 'unpdf'

const buffer = await fetch('https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf')
  .then(res => res.arrayBuffer())

const pdf = await getDocumentProxy(new Uint8Array(buffer))
const { totalPages, text } = await extractText(pdf, { mergePages: true })

console.log(`Total pages: ${totalPages}`)
console.log(text)
```

### Response
#### Success Response (200)
- **totalPages** (number) - The total number of pages in the PDF.
- **text** (string) - The extracted text content from the PDF.

#### Response Example
```json
{
  "totalPages": 1,
  "text": "This is a sample PDF document."
}
```
```

--------------------------------

### Define Custom PDF.js Module

Source: https://github.com/unjs/unpdf/blob/main/README.md

Allows users to specify a custom PDF.js build to be used by unpdf. This is useful for compatibility or if a specific PDF.js version is required. This method must be called before any other unpdf method.

```APIDOC
## Define Custom PDF.js Module

### Description
Allows defining a custom PDF.js module to be used by `unpdf`. This method should be invoked before any other `unpdf` functions. If not called, `unpdf` defaults to its bundled serverless build.

### Method
`definePDFJSModule(pdfjs: () => Promise<PDFJS>): Promise<void>`

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Request Example
```ts
import { definePDFJSModule, extractText, getDocumentProxy } from 'unpdf'

// Define the PDF.js build before using any other unpdf method
await definePDFJSModule(() => import('pdfjs-dist'))

// Now, you can use all unpdf methods with the official PDF.js build
const pdf = await getDocumentProxy(/* ... */)
const { text } = await extractText(pdf)
```

### Response
#### Success Response (200)
This method returns a Promise that resolves when the PDF.js module has been successfully defined.

#### Response Example
None (void return type)
```

--------------------------------

### getDocumentProxy — Load a PDF into a reusable proxy

Source: https://context7.com/unjs/unpdf/llms.txt

Parses binary PDF data into a `PDFDocumentProxy` that can be passed to any extraction function, avoiding repeated parsing when multiple operations are needed on the same file. Automatically applies sensible defaults.

```APIDOC
## getDocumentProxy

### Description
Parses binary PDF data into a `PDFDocumentProxy` that can be passed to any extraction function, avoiding repeated parsing when multiple operations are needed on the same file. Automatically applies sensible defaults (`isEvalSupported: false`, `useSystemFonts: true`), and in Node.js also sets `disableFontFace: true` and resolves `standardFontDataUrl` from the local `pdfjs-dist` installation.

### Method
`getDocumentProxy(data: Uint8Array | ArrayBuffer, options?: PDFProxyOptions)`

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Request Example
```ts
import { readFile } from 'node:fs/promises'
import { extractLinks, extractText, getDocumentProxy, getMeta } from 'unpdf'

const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Reuse the same proxy for multiple operations
const { info } = await getMeta(pdf)
const { text } = await extractText(pdf, { mergePages: true })
const { links } = await extractLinks(pdf)

console.log(`Author: ${info.Author}`)
// => Author: Evangelos Vlachogiannis
console.log(`Text preview: ${text.slice(0, 50)}`)
// => Text preview: Dummy PDF file
console.log(`Links found: ${links.length}`)

// Custom options – override font rendering for accurate output
const pdfCustom = await getDocumentProxy(new Uint8Array(buffer), {
  disableFontFace: false,
  standardFontDataUrl: 'https://unpkg.com/pdfjs-dist@latest/standard_fonts/',
})
```

### Response
#### Success Response (200)
- **pdf**: `PDFDocumentProxy` - A proxy object representing the loaded PDF document.
- **info**: `object` - Metadata about the PDF document.

#### Response Example
```json
{
  "pdf": { /* PDFDocumentProxy object */ },
  "info": { "Author": "Evangelos Vlachogiannis", ... }
}
```
```

--------------------------------

### createIsomorphicCanvasFactory

Source: https://context7.com/unjs/unpdf/llms.txt

A lower-level helper exported for advanced usage that returns the appropriate canvas factory class for the current runtime: `DOMCanvasFactory` in browsers and `NodeCanvasFactory` in Node.js (using `@napi-rs/canvas`). Throws in unsupported environments. Used internally by `renderPageAsImage`.

```APIDOC
## createIsomorphicCanvasFactory

### Description
A lower-level helper exported for advanced usage that returns the appropriate canvas factory class for the current runtime: `DOMCanvasFactory` in browsers and `NodeCanvasFactory` in Node.js (using `@napi-rs/canvas`). Throws in unsupported environments. Used internally by `renderPageAsImage`.

### Usage
```ts
import { createIsomorphicCanvasFactory, definePDFJSModule, getResolvedPDFJS } from 'unpdf'

await definePDFJSModule(() => import('pdfjs-dist'))

// Obtain the correct CanvasFactory for this runtime
const CanvasFactory = await createIsomorphicCanvasFactory(
  () => import('@napi-rs/canvas'), // required in Node.js; omit in browser
)

// Use it directly with getDocument for custom rendering pipelines
const { getDocument } = await getResolvedPDFJS()
const doc = await getDocument({
  data: pdfData,
  CanvasFactory,
  isEvalSupported: false,
}).promise

const page = await doc.getPage(1)
const viewport = page.getViewport({ scale: 2 })
const { canvas, context } = (new CanvasFactory()).create(viewport.width, viewport.height)

await page.render({ canvas, canvasContext: context, viewport }).promise
```
```

--------------------------------

### Extract Text From PDF

Source: https://github.com/unjs/unpdf/blob/main/README.md

Fetches a PDF, extracts text, and logs the total pages and extracted text. Ensure the PDF buffer is correctly formatted.

```typescript
import { extractText, getDocumentProxy } from 'unpdf'

// Fetch a PDF from the web or load it from the file system
const buffer = await fetch('https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf')
  .then(res => res.arrayBuffer())

const pdf = await getDocumentProxy(new Uint8Array(buffer))
const { totalPages, text } = await extractText(pdf, { mergePages: true })

console.log(`Total pages: ${totalPages}`)
console.log(text)
```

--------------------------------

### getMeta

Source: https://context7.com/unjs/unpdf/llms.txt

Extracts metadata from a PDF, including the /Info dictionary and XMP metadata. Dates can be optionally parsed into Date objects.

```APIDOC
## getMeta

### Description
Extracts the `/Info` dictionary and XMP metadata for a PDF. When `parseDates` is `true`, `CreationDate` and `ModDate` fields in `/Info` are converted to `Date` objects, and XMP date properties are also parsed.

### Method
```javascript
getMeta(pdf: PDFProxy, options?: { parseDates?: boolean }): Promise<{ info: Record<string, any>, metadata: Map<string, any> }>
```

### Parameters
#### `pdf`
- **pdf** (PDFProxy) - Required - The PDF document proxy object.

#### `options`
- **parseDates** (boolean) - Optional - If true, date fields will be parsed into `Date` objects.

### Request Example
```javascript
import { readFile } from 'node:fs/promises'
import { getDocumentProxy, getMeta } from 'unpdf'

const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Basic metadata
const { info, metadata } = await getMeta(pdf)
console.log(info)

// Parse date fields into Date objects
const { info: parsedInfo, metadata: parsedMeta } = await getMeta(pdf, { parseDates: true })
console.log(parsedInfo.CreationDate instanceof Date) // => true
```

### Response
#### Success Response
- **info** (Record<string, any>) - An object containing PDF information.
- **metadata** (Map<string, any>) - A Map containing XMP metadata.

#### Response Example
```json
{
  "info": {
    "Author": "Evangelos Vlachogiannis",
    "CreationDate": "D:20070223175637+02'00'",
    "Creator": "Writer",
    "PDFFormatVersion": "1.4",
    "Producer": "OpenOffice.org 2.1",
    "IsLinearized": false
  },
  "metadata": {
    "xmp:createdate": "2007-02-23T17:56:37+02:00",
    "xmp:modifydate": "2007-02-23T17:56:37+02:00",
    "xmp:metadatadate": "2007-02-23T17:56:37+02:00"
  }
}
```
```

--------------------------------

### Render PDF Page as Image

Source: https://context7.com/unjs/unpdf/llms.txt

Renders a PDF page to a raster image (ArrayBuffer or data URL). Requires 'pdfjs-dist' and '@napi-rs/canvas' in Node.js. Scale can be explicit or derived from width/height.

```typescript
import { readFile, writeFile } from 'node:fs/promises'
import { definePDFJSModule, getDocumentProxy, renderPageAsImage } from 'unpdf'

// Step 1: Switch to the official PDF.js build
await definePDFJSModule(() => import('pdfjs-dist'))

const buffer = new Uint8Array(await readFile('./document.pdf'))

// Render page 1 as PNG ArrayBuffer (2× scale)
const pngBuffer = await renderPageAsImage(buffer, 1, {
  canvasImport: () => import('@napi-rs/canvas'), // Required in Node.js
  scale: 2,
})
await writeFile('page1.png', new Uint8Array(pngBuffer))
// => page1.png written (valid PNG, header: [137,80,78,71,13,10,26,10])

// Render with fixed width (scale is calculated automatically)
const fixedWidthBuffer = await renderPageAsImage(buffer, 1, {
  canvasImport: () => import('@napi-rs/canvas'),
  width: 1200,
})
await writeFile('page1-1200w.png', new Uint8Array(fixedWidthBuffer))

// Return a data URL instead (e.g., for embedding in HTML)
const dataURL = await renderPageAsImage(buffer, 1, {
  canvasImport: () => import('@napi-rs/canvas'),
  scale: 1.5,
  toDataURL: true,
})
console.log(dataURL.startsWith('data:image/png;base64,')) // => true

const html = `<!DOCTYPE html><html><body><img src="${dataURL}"></body></html>`
await writeFile('page1.html', html)
```

--------------------------------

### Access Raw PDF.js Module with getResolvedPDFJS

Source: https://context7.com/unjs/unpdf/llms.txt

Returns the fully resolved PDF.js module, providing direct access to low-level PDF.js APIs like `getDocument`. This is useful for advanced use cases not covered by unpdf's higher-level helpers. It also allows checking the currently used PDF.js version.

```typescript
import { readFile } from 'node:fs/promises'
import { getResolvedPDFJS } from 'unpdf'

// Check which PDF.js version is in use
const { version } = await getResolvedPDFJS()
console.log(version)
// => "5.6.205"

// Use the low-level getDocument API directly
const { getDocument } = await getResolvedPDFJS()
const data = await readFile('./document.pdf')
const document = await getDocument(new Uint8Array(data)).promise

const meta = await document.getMetadata()
console.log(meta.info)
// => { Author: "...", CreationDate: "D:20070223...", PDFFormatVersion: "1.4", ... }
```

--------------------------------

### renderPageAsImage

Source: https://github.com/unjs/unpdf/blob/main/README.md

Renders a specific page of a PDF document as an image. It can return an ArrayBuffer or a data URL string. This method is compatible with Node.js and browser environments.

```APIDOC
## `renderPageAsImage`

### Description

To render a PDF page as an image, you can use the `renderPageAsImage` method. This method will return an `ArrayBuffer` of the rendered image. It can also return a data URL (`string`) if `toDataURL` option is set to `true`.

> [!NOTE]
> This method will only work in Node.js and browser environments.

### Method Signature

```ts
function renderPageAsImage(
  data: DocumentInitParameters['data'] | PDFDocumentProxy,
  pageNumber: number,
  options?: {
    canvasImport?: () => Promise<typeof import('@napi-rs/canvas')>
    /** @default 1.0 */
    scale?: number
    width?: number
    height?: number
    toDataURL?: false
  },
): Promise<ArrayBuffer>

function renderPageAsImage(
  data: DocumentInitParameters['data'] | PDFDocumentProxy,
  pageNumber: number,
  options: {
    canvasImport?: () => Promise<typeof import('@napi-rs/canvas')>
    /** @default 1.0 */
    scale?: number
    width?: number
    height?: number
    toDataURL: true
  },
): Promise<string>
```

### Parameters

- **data** (DocumentInitParameters['data'] | PDFDocumentProxy) - The PDF data or a PDFDocumentProxy object.
- **pageNumber** (number) - The number of the page to render.
- **options** (object) - Optional configuration for rendering.
  - **canvasImport** (function) - A function that imports the canvas module, required for Node.js environments.
  - **scale** (number) - The scale factor for rendering the page. Defaults to 1.0.
  - **width** (number) - The desired width of the rendered image.
  - **height** (number) - The desired height of the rendered image.
  - **toDataURL** (boolean) - If true, the method returns a data URL string; otherwise, it returns an ArrayBuffer. Defaults to false.

### Examples

**Rendering to ArrayBuffer:**

```ts
import { definePDFJSModule, renderPageAsImage } from 'unpdf'

// Use the official PDF.js build
await definePDFJSModule(() => import('pdfjs-dist'))

const pdf = await readFile('./dummy.pdf')
const buffer = new Uint8Array(pdf)
const pageNumber = 1

const result = await renderPageAsImage(buffer, pageNumber, {
  canvasImport: () => import('@napi-rs/canvas'),
  scale: 2,
})
await writeFile('dummy-page-1.png', new Uint8Array(result))
```

**Rendering to Data URL:**

```ts
import { definePDFJSModule, renderPageAsImage } from 'unpdf'

await definePDFJSModule(() => import('pdfjs-dist'))

const pdf = await readFile('./dummy.pdf')
const buffer = new Uint8Array(pdf)
const pageNumber = 1

const result = await renderPageAsImage(buffer, pageNumber, {
  canvasImport: () => import('@napi-rs/canvas'),
  scale: 2,
  toDataURL: true,
})

const html = `<!DOCTYPE html>
<html lang="en">
  <head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Dummy Page</title>
  </head>
  <body>
    <img alt="Example Page" src="${result}">
  </body>
</html>`

await writeFile('dummy-page-1.html', html)
```
```

--------------------------------

### definePDFJSModule — Override the PDF.js build

Source: https://context7.com/unjs/unpdf/llms.txt

Registers a custom PDF.js resolver before any other unpdf function is called. By default unpdf uses its bundled serverless build; call this to swap in the official pdfjs-dist package or a legacy version. Must be awaited before any extraction calls.

```APIDOC
## definePDFJSModule

### Description
Registers a custom PDF.js resolver before any other `unpdf` function is called. By default `unpdf` uses its bundled serverless build; call this to swap in the official `pdfjs-dist` package or a legacy version. Must be awaited before any extraction calls.

### Method
`definePDFJSModule(resolver: () => Promise<any>)`

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Request Example
```ts
import { definePDFJSModule, extractText, getDocumentProxy } from 'unpdf'

// Use the official pdfjs-dist instead of the bundled serverless build.
// Requires: npm install pdfjs-dist
// WARNING: pdfjs-dist v5.x requires Promise.withResolvers (Node >= 22).
await definePDFJSModule(() => import('pdfjs-dist'))

const buffer = await fetch('https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf')
  .then(res => res.arrayBuffer())

const pdf = await getDocumentProxy(new Uint8Array(buffer))
const { totalPages, text } = await extractText(pdf, { mergePages: true })

console.log(`Pages: ${totalPages}`)
// => Pages: 1
console.log(text)
// => "Dummy PDF file"
```

### Response
#### Success Response
None (This function is for configuration and does not return a value directly usable in the same way as extraction functions).

#### Response Example
None
```

--------------------------------

### getResolvedPDFJS — Access the raw PDF.js module

Source: https://context7.com/unjs/unpdf/llms.txt

Returns the fully resolved PDF.js module, giving direct access to the low-level `getDocument`, `PDFDateString`, `OPS`, and all other PDF.js exports. Useful when `unpdf`'s higher-level helpers do not cover a specific use case.

```APIDOC
## getResolvedPDFJS

### Description
Returns the fully resolved PDF.js module, giving direct access to the low-level `getDocument`, `PDFDateString`, `OPS`, and all other PDF.js exports. Useful when `unpdf`'s higher-level helpers do not cover a specific use case.

### Method
`getResolvedPDFJS(): Promise<any>`

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
None

### Request Example
```ts
import { readFile } from 'node:fs/promises'
import { getResolvedPDFJS } from 'unpdf'

// Check which PDF.js version is in use
const { version } = await getResolvedPDFJS()
console.log(version)
// => "5.6.205"

// Use the low-level getDocument API directly
const { getDocument } = await getResolvedPDFJS()
const data = await readFile('./document.pdf')
const document = await getDocument(new Uint8Array(data)).promise

const meta = await document.getMetadata()
console.log(meta.info)
// => { Author: "...", CreationDate: "D:20070223...", PDFFormatVersion: "1.4", ... }
```

### Response
#### Success Response (200)
- **pdfjsModule**: `object` - The resolved PDF.js module, containing exports like `getDocument`, `OPS`, `PDFDateString`, etc.

#### Response Example
```json
{
  "pdfjsModule": {
    "version": "5.6.205",
    "getDocument": "function",
    "OPS": { ... },
    "PDFDateString": "function",
    ...
  }
}
```
```

--------------------------------

### Render PDF Page to Data URL (Node.js/Browser)

Source: https://github.com/unjs/unpdf/blob/main/README.md

Renders a PDF page and returns a data URL string, suitable for embedding directly into HTML. This requires setting the `toDataURL` option to `true`. Similar to image rendering, ensure PDF.js is defined and the canvas package is available for Node.js.

```typescript
import { definePDFJSModule, renderPageAsImage } from 'unpdf'

await definePDFJSModule(() => import('pdfjs-dist'))

const pdf = await readFile('./dummy.pdf')
const buffer = new Uint8Array(pdf)
const pageNumber = 1

const result = await renderPageAsImage(buffer, pageNumber, {
  canvasImport: () => import('@napi-rs/canvas'),
  scale: 2,
  toDataURL: true,
})

const html = `<!DOCTYPE html>
<html lang="en">
  <head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Dummy Page</title>
  </head>
  <body>
    <img alt="Example Page" src="${result}">
  </body>
</html>`

await writeFile('dummy-page-1.html', html)
```

--------------------------------

### Extract Images from PDF Page

Source: https://context7.com/unjs/unpdf/llms.txt

Extracts raster images from a specific PDF page as pixel buffers. Requires Node.js environment with 'sharp' for image processing.

```typescript
import { readFile, writeFile } from 'node:fs/promises'
import sharp from 'sharp'
import { extractImages, getDocumentProxy } from 'unpdf'

const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Extract images from page 1 (1-indexed; throws if out of range)
const images = await extractImages(pdf, 1)
console.log(`Found ${images.length} images on page 1`)

for (let i = 0; i < images.length; i++) {
  const img = images[i]!
  console.log(`Image ${i + 1}: key=${img.key}, ${img.width}x${img.height}, ${img.channels}ch`)
  // => Image 1: key=img_p0_1, 800x600, 3ch

  // Convert raw pixel buffer to PNG with sharp
  await sharp(img.data, {
    raw: {
      width: img.width,
      height: img.height,
      channels: img.channels,
    },
  })
    .png()
    .toFile(`page1-image-${i + 1}.png`)
}
```

--------------------------------

### extractImages

Source: https://context7.com/unjs/unpdf/llms.txt

Extracts all raster images embedded on a specific page as Uint8ClampedArray pixel buffers, with their dimensions and channel count. Designed for use with image processing libraries such as sharp.

```APIDOC
## extractImages

### Description
Extracts all raster images embedded on a specific page as `Uint8ClampedArray` pixel buffers, with their dimensions and channel count (1 = grayscale, 3 = RGB, 4 = RGBA). Designed for use with image processing libraries such as `sharp`.

### Usage
```ts
import { readFile, writeFile } from 'node:fs/promises'
import sharp from 'sharp'
import { extractImages, getDocumentProxy } from 'unpdf'

const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Extract images from page 1 (1-indexed; throws if out of range)
const images = await extractImages(pdf, 1)
console.log(`Found ${images.length} images on page 1`)

for (let i = 0; i < images.length; i++) {
  const img = images[i]!
  console.log(`Image ${i + 1}: key=${img.key}, ${img.width}x${img.height}, ${img.channels}ch`)
  // => Image 1: key=img_p0_1, 800x600, 3ch

  // Convert raw pixel buffer to PNG with sharp
  await sharp(img.data, {
    raw: {
      width: img.width,
      height: img.height,
      channels: img.channels,
    },
  })
    .png()
    .toFile(`page1-image-${i + 1}.png`)
}
```
```

--------------------------------

### Extract Hyperlinks from PDF Annotations with extractLinks

Source: https://context7.com/unjs/unpdf/llms.txt

Use extractLinks to iterate through PDF pages and collect all URL strings from Link annotations. The result is a flat array of all URLs found in the document.

```typescript
import { readFile } from 'node:fs/promises'
import { extractLinks, getDocumentProxy } from 'unpdf'

const buffer = await readFile('./document-with-links.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

const { totalPages, links } = await extractLinks(pdf)

console.log(`Total pages: ${totalPages}`) // => Total pages: 2
console.log(`Links found: ${links.length}`) // => Links found: 4
links.forEach(url => console.log(url))
// => https://www.antennahouse.com/
// => https://example.com/page2
// ...

// Filter to only external HTTPS links
const externalLinks = links.filter(url => url.startsWith('https://'))
console.log(`External HTTPS links: ${externalLinks.length})
```

--------------------------------

### Extract Images from PDF Page

Source: https://github.com/unjs/unpdf/blob/main/README.md

Extracts images from a specified page of a PDF. Requires the 'sharp' library for image processing and saving. The function returns an array of image objects, each containing image data, dimensions, and channel information.

```typescript
import { readFile, writeFile } from 'node:fs/promises'
import sharp from 'sharp'
import { extractImages, getDocumentProxy } from 'unpdf'

async function extractPdfImages() {
  const buffer = await readFile('./document.pdf')
  const pdf = await getDocumentProxy(new Uint8Array(buffer))

  // Extract images from page 1
  const imagesData = await extractImages(pdf, 1)
  console.log(`Found ${imagesData.length} images on page 1`)
```

```typescript
  // Process each image with sharp (optional)
  let totalImagesProcessed = 0
  for (const imgData of imagesData) {
    const imageIndex = ++totalImagesProcessed

    await sharp(imgData.data, {
      raw: {
        width: imgData.width,
        height: imgData.height,
        channels: imgData.channels
      }
    })
      .png()
      .toFile(`image-${imageIndex}.png`)

    console.log(`Saved image ${imageIndex} (${imgData.width}x${imgData.height}, ${imgData.channels} channels)`)
```

```typescript
  }
}

extractPdfImages().catch(console.error)
```

--------------------------------

### extractLinks

Source: https://context7.com/unjs/unpdf/llms.txt

Extracts all hyperlink URLs from PDF annotations across all pages.

```APIDOC
## extractLinks

### Description
Iterates all pages and collects URL strings from `Link` annotations. Returns a flat array of all URLs across the entire document.

### Method
```javascript
extractLinks(pdf: PDFProxy | Uint8Array): Promise<{ totalPages: number, links: string[] }>
```

### Parameters
#### `pdf`
- **pdf** (PDFProxy | Uint8Array) - Required - The PDF document proxy object or a Uint8Array buffer of the PDF.

### Request Example
```javascript
import { readFile } from 'node:fs/promises'
import { extractLinks, getDocumentProxy } from 'unpdf'

const buffer = await readFile('./document-with-links.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

const { totalPages, links } = await extractLinks(pdf)

console.log(`Total pages: ${totalPages}`)
console.log(`Links found: ${links.length}`)
links.forEach(url => console.log(url))
```

### Response
#### Success Response
- **totalPages** (number) - The total number of pages in the PDF.
- **links** (string[]) - An array of all URL strings found in the PDF's link annotations.

#### Response Example
```json
{
  "totalPages": 2,
  "links": [
    "https://www.antennahouse.com/",
    "https://example.com/page2",
    "http://another-link.org"
  ]
}
```
```

--------------------------------

### Extract Links from PDF

Source: https://github.com/unjs/unpdf/blob/main/README.md

Use this function to extract all hyperlinks and external URLs from a PDF document. It returns the total number of pages and an array of found links.

```typescript
import { readFile } from 'node:fs/promises'
import { extractLinks, getDocumentProxy } from 'unpdf'

// Load a PDF file
const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Extract all links from the PDF
const { totalPages, links } = await extractLinks(pdf)

console.log(`Total pages: ${totalPages}`)
console.log(`Found ${links.length} links:`)
```

```typescript
for (const link of links) console.log(link)
```

--------------------------------

### extractLinks

Source: https://github.com/unjs/unpdf/blob/main/README.md

Extracts all links from a PDF document, including hyperlinks and external URLs.

```APIDOC
## extractLinks

### Description
Extracts all links from a PDF document, including hyperlinks and external URLs.

### Method Signature
```ts
function extractLinks(
  data: DocumentInitParameters['data'] | PDFDocumentProxy,
): Promise<{ totalPages: number, links: string[] }>
```

### Example
```ts
import { readFile } from 'node:fs/promises'
import { extractLinks, getDocumentProxy } from 'unpdf'

// Load a PDF file
const buffer = await readFile('./document.pdf')
const pdf = await getDocumentProxy(new Uint8Array(buffer))

// Extract all links from the PDF
const { totalPages, links } = await extractLinks(pdf)

console.log(`Total pages: ${totalPages}`)
console.log(`Found ${links.length} links:`)
for (const link of links) console.log(link)
```
```