Pass Node.js streams to exiftool and stop reading as soon as tags are found
MIT License
Stream to ExifTool
Exposes a Street
class that spawns and keeps ExifTool open,
using the -stay_open option. Respawns it on failure.
It is much faster than spawning exiftool on each request.
It features an option to limit the amount of data read from the readable stream passed as argument.
It also features the ability to try parsing a readable stream by chunks, and return the tags as soon as ExifTool had enough data.
This is very useful for quick inspection of remote resources, and acts as exiftool "-fast" option - but for all formats, and with more cpu load.
import Streat from 'streat';
// init code
const streat = new Street({
step: 32768 // the default step for trying decoding by chunks
});
// start exiftool
streat.start();
// runtime code
streat.run(readableStream, params, function(err, tags) {
});
// multiple times... calls are queued
// close exiftool
streat.stop()
It is also possible to spawn several instances of exiftool, and maintain them using a pool.
The params
can be either:
limit
and step
keysstep
is the number of bytes to read for each pass.
See LICENSE file.
ExifTool by Phil Harvey - Read, Write and Edit Meta Information!