expo-media-metadata

A native Expo module for reading rich media metadata from images and videos — including EXIF, GPS, TIFF, IPTC, XMP, PNG, C2PA/JUMBF, file checksums, and more.

Unlike JavaScript-based solutions, this module reads directly from the original file bytes on the native layer:

• On iOS, supports ph:// URIs so you can read the original PHAsset bytes from the gallery — bypassing the re-encoding that expo-image-picker performs, which would otherwise strip C2PA provenance data.
• On Android, reads directly via content:// and file:// URIs using the metadata-extractor library.

Platform Support

| Platform | Minimum Version | Notes | |---|---|---| | iOS | 16.0+ | Supports file://, ph:// (PHAsset) URIs | | Android | API 24+ | Supports content://, file:// URIs |

Installation

npx expo install expo-media-metadata

iOS (CocoaPods)

cd ios && pod install

Android

No extra steps needed — the Gradle plugin handles it automatically.

API

getMediaMetadata(uri: string): Promise<MediaMetadataResult>

Reads all available metadata from a media file.

Parameters

• uri — File URI of the media. Accepts:
  • file:///... — Direct file path
  • ph://<assetId> — PHAsset URI (iOS only — reads original, unmodified bytes)
  • content://... — Android content URI

Returns a MediaMetadataResult object with structured metadata fields.

import { getMediaMetadata } from 'expo-media-metadata';

const metadata = await getMediaMetadata('file:///path/to/image.jpg');
console.log(metadata.exif);      // EXIF key-value pairs
console.log(metadata.metadata);  // All metadata (GPS, IPTC, C2PA, checksums, etc.)

getMediaRawBase64(uri: string): Promise<string>

Reads the raw bytes of a media file and returns a base64-encoded string. Useful for C2PA verification where the complete binary is needed.

import { getMediaRawBase64 } from 'expo-media-metadata';

const base64 = await getMediaRawBase64('file:///path/to/image.jpg');

Types

type MediaMetadataResult = {
  /** EXIF metadata as key-value pairs */
  exif: Record<string, any> | null;
  /**
   * All metadata including TIFF, GPS, IPTC, XMP, PNG, C2PA/JUMBF,
   * file checksums (MD5, SHA256), file size, MIME type, etc.
   */
  metadata: Record<string, any> | null;
};

Example metadata keys

| Key | Description | |---|---| | file_name | Original filename | | file_size | File size in bytes | | mime_type | e.g. "image/jpeg" | | file_type_extension | e.g. "jpg" | | checksum_md5 | MD5 hex digest of file bytes | | checksum_sha256 | SHA-256 hex digest of file bytes | | Exif.* | All EXIF fields (e.g. Exif.DateTimeOriginal) | | GPS.* | GPS fields (e.g. GPS.Latitude, GPS.Longitude) | | TIFF.* | TIFF fields (e.g. TIFF.Make, TIFF.Model) | | IPTC.* | IPTC fields | | c2pa:has_c2pa | true if C2PA manifest is present | | c2pa:c2pa_claim_found | true if a C2PA claim box was found | | c2pa:digital_source_type | e.g. "trainedAlgorithmicMedia" |

Real-World Usage Example

[!WARNING]
Metadata Loss Warning: When using libraries like expo-image-picker, do NOT lower the image quality (e.g., avoid quality: 0.8) or enable cropping/editing (allowsEditing: true). Doing so forces the OS to re-encode the file, which irreversibly strips out rich metadata like C2PA, GPS, and sensitive EXIF. Always request the unmodified image.

This example shows how to use expo-media-metadata with expo-image-picker to extract metadata from a picked image, including reading original PHAsset bytes on iOS (preserving C2PA provenance data):

import { launchImageLibraryAsync } from 'expo-image-picker';
import { getMediaMetadata } from 'expo-media-metadata';
import { Platform } from 'react-native';

async function pickAndReadMetadata() {
  const result = await launchImageLibraryAsync({
    mediaTypes: ['images'],
    exif: true,
    quality: 1,
  });

  if (result.canceled) return;

  const asset = result.assets[0];

  // On iOS, use ph://<assetId> to read the original PHAsset bytes
  // (bypasses re-encoding that strips C2PA data)
  const metadataUri =
    Platform.OS === 'ios' && asset.assetId
      ? `ph://${asset.assetId}`
      : asset.uri;

  const metadata = await getMediaMetadata(metadataUri);

  console.log('EXIF:', metadata.exif);
  console.log('All metadata:', metadata.metadata);

  // Check for C2PA content credentials
  if (metadata.metadata?.['c2pa:has_c2pa']) {
    console.log('✅ Image has C2PA provenance data');
    console.log('Source type:', metadata.metadata['c2pa:digital_source_type']);
  }
}

What metadata is extracted?

iOS (via ImageIO framework)

• Full EXIF dictionary (camera settings, timestamps, etc.)
• GPS coordinates and altitude
• TIFF fields (Make, Model, Software, etc.)
• IPTC (caption, keywords, copyright)
• XMP metadata
• PNG metadata
• HEIC/HEIF properties
• C2PA / JUMBF manifest detection (APP11 in JPEG, caBX chunk in PNG)
• File-level: size, MD5 checksum, SHA-256 checksum, MIME type

Android (via metadata-extractor library)

• EXIF, GPS, IPTC, XMP
• JPEG, PNG, WebP container metadata
• File-level: size, MIME type

Why this module?

| Approach | EXIF | C2PA | Original bytes | ph:// support | |---|---|---|---|---| | expo-image-picker (built-in exif: true) | ⚠️ Partial | ❌ Lost on copy | ❌ | ❌ | | JavaScript EXIF parsers | ✅ | ❌ | Depends | ❌ | | expo-media-metadata | ✅ | ✅ | ✅ | ✅ (iOS) |

License

MIT © Arun Kumar Bhardwaj