OCRExecutor.java

package strimillinn.core.ocr;

import java.io.File;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.atomic.LongAdder;

import org.apache.cayenne.query.SelectQuery;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import strimillinn.core.StrimillinnCore;
import strimillinn.core.model.Receipt;

public class OCRExecutor {

	private static final Logger logger = LoggerFactory.getLogger( OCRExecutor.class );

	public void run() {
		int numberOfThreads = Runtime.getRuntime().availableProcessors();
		ExecutorService es = Executors.newFixedThreadPool( numberOfThreads );

		SelectQuery<Receipt> query = new SelectQuery<>( Receipt.class );
		query.addPrefetch( Receipt.DOCUMENT.joint() );

		LongAdder count = new LongAdder();

		StrimillinnCore.newContext().iterate( query, receipt -> {
			es.execute( () -> {
				count.increment();
				Receipt localReceipt = StrimillinnCore.newContext().localObject( receipt );
				logger.info( "Processing:" + localReceipt.creationDate() + " : " + count );
				File file = localReceipt.document().file();
				String string = null;

				try {
					string = new OCREngineTesseract().parseFile( file );
				}
				catch( OCRException e ) {
					e.printStackTrace();
				}

				localReceipt.setText( string, "tesseract" );
				localReceipt.getObjectContext().commitChanges();
			} );
		} );

		es.shutdown();

		logger.info( "Done!" );
	}
}