Supported Formats

undms supports extracting text and metadata from a wide variety of document formats. This guide provides a complete overview of all supported formats and their capabilities.

Overview

Format	MIME Type	Text	Metadata	OCR
Plain Text	text/*	✅	Lines, words, characters	-
JSON	application/json	✅	Lines, words, characters	-
XML	application/xml	✅	Lines, words, characters	-
DOCX	application/vnd.openxmlformats-officedocument.wordprocessingml.document	✅	Paragraphs, tables, images, hyperlinks	-
XLSX	application/vnd.openxmlformats-officedocument.spreadsheetml.sheet	✅	Sheets, rows, columns, cells	-
PDF	application/pdf	✅	Title, author, subject, pages	-
Images	image/*	OCR	EXIF, GPS, dimensions	✅

Text Files

Text-based formats including plain text, JSON, and XML.

Supported MIME Types

• text/plain
• text/*
• application/json
• application/xml
• application/javascript

Extracted Metadata

interface TextMetadata {
  lineCount: number; // Number of lines in the document
  wordCount: number; // Total word count
  characterCount: number; // Total characters including whitespace
  nonWhitespaceCharacterCount: number; // Characters without whitespace
}

Example

const result = extract([
  {
    name: 'sample.txt',
    size: 1024,
    type: 'text/plain',
    lastModified: Date.now(),
    webkitRelativePath: '',
    buffer: Buffer.from('Line 1\nLine 2\nLine 3'),
  },
]);

console.log(result[0].documents[0].metadata?.text);
// {
//   lineCount: 3,
//   wordCount: 3,
//   characterCount: 22,
//   nonWhitespaceCharacterCount: 15
// }

DOCX (Microsoft Word)

Extracts content and structure from Microsoft Word documents.

Supported MIME Type

application/vnd.openxmlformats-officedocument.wordprocessingml.document

Extracted Metadata

interface DocxMetadata {
  paragraphCount: number; // Total paragraphs in the document
  tableCount: number; // Number of tables
  imageCount: number; // Embedded images
  hyperlinkCount: number; // Hyperlinks in the document
}

Example

const result = extract([
  {
    name: 'document.docx',
    size: 5000,
    type: 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
    lastModified: Date.now(),
    webkitRelativePath: '',
    buffer: Buffer.from(docxBuffer),
  },
]);

console.log(result[0].documents[0].metadata?.docx);
// {
//   paragraphCount: 15,
//   tableCount: 2,
//   imageCount: 3,
//   hyperlinkCount: 5
// }

Features

• Extracts paragraph text
• Handles nested tables
• Processes inline images
• Extracts hyperlinks

XLSX (Microsoft Excel)

Extracts content and structure from Microsoft Excel spreadsheets.

Supported MIME Type

application/vnd.openxmlformats-officedocument.spreadsheetml.sheet

Extracted Metadata

interface XlsxMetadata {
  sheetCount: number; // Number of worksheets
  sheetNames: string[]; // Names of all sheets
  rowCount: number; // Total rows across all sheets
  columnCount: number; // Maximum columns in any sheet
  cellCount: number; // Total cells with content
}

Example

const result = extract([
  {
    name: 'spreadsheet.xlsx',
    size: 8000,
    type: 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
    lastModified: Date.now(),
    webkitRelativePath: '',
    buffer: Buffer.from(xlsxBuffer),
  },
]);

console.log(result[0].documents[0].metadata?.xlsx);
// {
//   sheetCount: 3,
//   sheetNames: ['Sales', 'Expenses', 'Summary'],
//   rowCount: 150,
//   columnCount: 10,
//   cellCount: 1200
// }

PDF (Portable Document Format)

Extracts text content and document properties from PDF files.

Supported MIME Type

application/pdf

Extracted Metadata

interface PdfMetadata {
  title?: string; // Document title
  author?: string; // Document author
  subject?: string; // Document subject
  producer?: string; // PDF producer application
  pageSize?: PdfPageSize; // First page dimensions
  pageCount: number; // Total number of pages
}

interface PdfPageSize {
  width: number; // Page width in points
  height: number; // Page height in points
}

Example

const result = extract([
  {
    name: 'document.pdf',
    size: 15000,
    type: 'application/pdf',
    lastModified: Date.now(),
    webkitRelativePath: '',
    buffer: Buffer.from(pdfBuffer),
  },
]);

console.log(result[0].documents[0].metadata?.pdf);
// {
//   title: 'Annual Report 2024',
//   author: 'John Doe',
//   subject: 'Financial Summary',
//   producer: 'Microsoft Word',
//   pageSize: { width: 612, height: 792 },
//   pageCount: 24
// }

Limitations

• Image-only PDFs (scanned documents) will return empty text content
• OCR is not currently applied to PDFs

Images

Extracts text from images using OCR and retrieves EXIF metadata.

Supported MIME Types

• image/jpeg
• image/png
• image/gif
• image/bmp
• image/tiff
• image/webp

Extracted Metadata

interface ImageMetadata {
  width: number; // Image width in pixels
  height: number; // Image height in pixels
  format?: string; // Image format (JPEG, PNG, etc.)
  cameraMake?: string; // Camera manufacturer
  cameraModel?: string; // Camera model
  datetimeOriginal?: string; // Date/time when photo was taken
  location: ImageLocation; // GPS coordinates
}

interface ImageLocation {
  latitude?: number; // GPS latitude
  longitude?: number; // GPS longitude
}

OCR Text Extraction

When text is detected in an image, it's included in the extracted content:

const result = extract([
  {
    name: 'photo.jpg',
    size: 200000,
    type: 'image/jpeg',
    lastModified: Date.now(),
    webkitRelativePath: '',
    buffer: Buffer.from(imageBuffer),
  },
]);

console.log(result[0].documents[0].content);
// Extracted text from OCR

console.log(result[0].documents[0].metadata?.image);
// {
//   width: 1920,
//   height: 1080,
//   format: 'JPEG',
//   cameraMake: 'Apple',
//   cameraModel: 'iPhone 14 Pro',
//   datetimeOriginal: '2024:01:15 10:30:00',
//   location: { latitude: 40.7128, longitude: -74.0060 }
// }

Example: Photo with Location

const result = extract([
  {
    name: 'vacation_photo.jpg',
    size: 350000,
    type: 'image/jpeg',
    lastModified: Date.now(),
    webkitRelativePath: '',
    buffer: Buffer.from(jpegBuffer),
  },
]);

const imageMeta = result[0].documents[0].metadata?.image;

if (imageMeta?.location?.latitude && imageMeta?.location?.longitude) {
  console.log(`Photo location: ${imageMeta.location.latitude}, ${imageMeta.location.longitude}`);
}

if (imageMeta?.cameraMake && imageMeta?.cameraModel) {
  console.log(`Taken with: ${imageMeta.cameraMake} ${imageMeta.cameraModel}`);
}

MIME Type Detection

undms uses the type field you provide to determine which handler to use. Make sure to provide the correct MIME type for each document.

Common MIME Types

File Extension	MIME Type
.txt	text/plain
.json	application/json
.xml	application/xml
.docx	application/vnd.openxmlformats-officedocument.wordprocessingml.document
.xlsx	application/vnd.openxmlformats-officedocument.spreadsheetml.sheet
.pdf	application/pdf
.jpg / .jpeg	image/jpeg
.png	image/png
.gif	image/gif
.bmp	image/bmp
.tiff	image/tiff
.webp	image/webp

TIP

For web applications, you can use the File object's type property which automatically provides the correct MIME type.