Fast, cross-platform Node.js access to ExifTool.

1. Best-of-class cross-platform performance.

   Expect an order of magnitude faster performance than other packages.

2. Proper extraction of

   • dates with correct timezone offset encoding, when available)
   • latitudes & longitudes as floats (where negative values indicate west or south of the meridian)
   • embedded images, both Thumbnail and Preview (if they exist)

3. Robust type definitions of the top 99.5% tags used by over 6,000 different camera makes and models

4. Auditable ExifTool source code (the vendored code is checksum verified)

5. Automated updates to ExifTool (as new versions come out monthly)

6. Robust test suite, performed with Node v4, v6 and v7 on Linux, Mac, & Windows.

npm install --save exiftool-vendored

Note that exiftool-vendored provides an installation of ExifTool relevant for your platform via platform-dependent-modules. You shouldn't include either the exiftool-vendored.exe or exiftool-vendored.pl as direct dependencies to your project.

import { exiftool } from "exiftool-vendored";

// Read all metadata tags in `path/to/image.jpg`. 
// Returns a `Promise<Tags>`.
exiftool
  .read("path/to/image.jpg")
  .then((tags /*: Tags */) => console.log(`Make: ${tags.Make}, Model: ${tags.Model}`))
  .catch(err => console.error("Something terrible happened: ", err))

// Extract the low-resolution thumbnail in `path/to/image.jpg`,
// write it to `path/to/thumbnail.jpg`, and return a Promise<void>
// that is fulfilled when the image is extracted:
exiftool.extractThumbnail("path/to/image.jpg", "path/to/thumbnail.jpg");

// Extract the "Preview" image (found in some images):
exiftool.extractPreview("path/to/image.jpg", "path/to/preview.jpg");

// Extract the "JpgFromRaw" image (found in some RAW images):
exiftool.extractJpgFromRaw("path/to/image.cr2", "path/to/fromRaw.jpg");

// Extract the binary value from "tagname" tag in `path/to/image.jpg`
// and write it to `dest.bin` (which cannot exist already 
// and whose parent directory must already exist):
exiftool.extractBinaryTag("tagname", "path/to/file.exf", "path/to/dest.bin");

Generally, EXIF tags encode dates and times with no timezone offset. Presumably the time is captured in local time, but this means parsing the same file in different parts of the world results in a different absolute timestamp for the same file.

Rather than returning a Date which always includes a timezone, this library returns classes that encode the date, the time of day, or both, with an optional tzoffset. It's up to you, then, to do what's right.

In many cases, though, a tzoffset can be determined, either by the composite TimeZone tag, or by looking at the difference between the local DateTimeOriginal and GPSDateTime tags. GPSDateTime is present in most smartphone images.

If a tzoffset can be determined, it is encoded in all related ExifDateTime tags for those files.

Note also that some smartphones record timestamps with microsecond precision (not just millis!), and both ExifDateTime and ExifTime have floating point milliseconds.

Official EXIF tag names are PascalCased, like AFPointSelected and ISO. ("Fixing" the field names to be camelCase, would result in ungainly aFPointSelected and iSO atrocities).

The Tags.ts file is autogenerated by parsing through images of more than 3,000 different camera makes and models taken from the ExifTool site. It groups tags, their type, popularity, and example values such that your IDE can autocomplete.

Note that if parsing fails (for, example, a datetime string), the raw string will be returned. Consuming code should verify both existence and type as reasonable for safety.

The npm run mktags target reads all tags found in a batch of sample images and parses the results.

Using exiftool-vendored:

Read 2236 unique tags from 3011 files.
Parsing took 16191ms (5.4ms / file) # win32, core i7, maxProcs 4
Parsing took 27141ms (9.0ms / file) # ubuntu, core i3, maxProcs 1
Parsing took 12545ms (4.2ms / file) # ubuntu, core i3, maxProcs 4

Using the exiftool npm package takes 7x longer:

Reading 3011 files...
Parsing took 85654ms (28.4ms / file) # ubuntu, core i3

Starting the perl version of ExifTool is expensive, and is especially expensive on the Windows version of ExifTool.

On Windows, for every invocation, exiftool installs a distribution of Perl and extracts the ~1000 files that make up ExifTool, and then runs the script. Windows virus scanners prevent reads on these files while they are scanned, which makes this approach even more costly.

Using ExifTool's -stay_open batch mode, we can reuse a single instance of ExifTool across many requests, dropping response latency dramatically and reducing system load.

The exiftool singleton is configured with a maxProcs set to the number of CPUs on the current system; no more than maxProcs instances of exiftool will be spawned.

Note that each child process consumes between 10 and 50 MB of RAM. If you have limited system resources you may want to use a smaller maxProcs value.

debuglog is used with the exiftool prefix. To enable logging, set the environment flag NODE_DEBUG=exiftool.

See CHANGELOG.md.