GATE Pipeline池资源管理（ 在Web多线程环境下建立SerialAnalyserController池分配资源）

gate.env.cfg.xml

<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
	xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:gate="http://gate.ac.uk/ns/spring"
	xsi:schemaLocation="http://www.springframework.org/schema/beans
  http://www.springframework.org/schema/beans/spring-beans.xsd
  http://gate.ac.uk/ns/spring http://gate.ac.uk/ns/spring.xsd">
	<description>GATE 池资源</description>  

	<!--Pipeline Pool Config-->
	<bean id="pipelinePool" class="org.apache.commons.pool2.impl.GenericObjectPool">
		<constructor-arg index="0" ref="pipelinePooledObjectFactory"></constructor-arg>
		<constructor-arg index="1" ref="genericObjectPoolConfig"></constructor-arg>
	</bean> 
	 
	<!-- Generic PoolConfig -->
	<bean id="genericObjectPoolConfig" class="org.apache.commons.pool2.impl.GenericObjectPoolConfig"> 
		<property name="maxIdle" value="5"></property>
		<property name="maxTotal" value="100"></property>
		<property name="maxWaitMillis" value="1000"></property>
	</bean> 
</beans>

GATEPoolTest.java

package gto.uadb.test;

import gate.Corpus;
import gate.common.service.IGATECommonService;
import gate.creole.ResourceInstantiationException;
import gate.creole.SerialAnalyserController;
import gto.uadb.service.IAddressProcessPipeline;
import gto.uadb.test.util.AnnotationTestUtil;

import java.net.MalformedURLException;

import javax.annotation.Resource;

import org.apache.commons.pool2.impl.GenericObjectPool;
import org.junit.After;
import org.junit.Before;
import org.junit.Test;
import org.junit.runner.RunWith;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.test.context.ContextConfiguration;
import org.springframework.test.context.junit4.SpringJUnit4ClassRunner;

/**
 * GATE Pool测试示例
 * 
 * @author geosmart
 * 
 */
@RunWith(SpringJUnit4ClassRunner.class)
@ContextConfiguration(locations = {"classpath:app.etl.extract.cfg.xml"})
public class GATEPoolTest {

  @Resource(name = "pipelinePool")
  GenericObjectPool<SerialAnalyserController> pipelinePool;

  @Autowired
  IGATECommonService gateCommonService;

  @Resource(name = "extractPipelinService")
  IAddressProcessPipeline pipelineService;

  String currentPlugin = "gto.uadb.etl.extract";

  String gateDir = System.getProperty("user.dir") + "//src//test//resources//gate";
  private static boolean initialized = false;

  @Before
  public void setup() {
	System.out.println("---setup");

	if (!initialized) {
	  // 初始化GATE环境
	  gateCommonService.setGateDir(gateDir);
	  gateCommonService.initGate();
	  gateCommonService.registerGatePlugins(new String[] {"ANNIE", currentPlugin});

	  // 初始化Pipeline
	  pipelineService.initPipeline();
	  initialized = true;
	}
  }

  /**
   * JAPE和Gazetteer测试
   */
  @Test
  public void testAnnotation() throws ResourceInstantiationException, MalformedURLException {
	Corpus sourceCorpus = AnnotationTestUtil.getCorpus();
	loadPipeline(sourceCorpus);
  }

  /**
   * 装载Pipeline管道
   * 
   * @param corpus 语料库
   */
  private Corpus loadPipeline(Corpus corpus) {
	try {
	  SerialAnalyserController pipeline = pipelinePool.borrowObject();
	  pipeline.setCorpus(corpus);
	  pipeline.execute();
	  pipelinePool.returnObject(pipeline);
	} catch (Exception e) {
	  e.printStackTrace();
	}
	return corpus;
  }

  @After
  public void teardown() {
	System.out.println("---teardown");
  }
}

PipelinePooledObjectFactory.java

package gto.uadb.service.impl;

import gate.Gate;
import gate.creole.SerialAnalyserController;
import gate.util.persistence.PersistenceManager;

import java.io.File;

import org.apache.commons.pool2.BasePooledObjectFactory;
import org.apache.commons.pool2.PooledObject;
import org.apache.commons.pool2.impl.DefaultPooledObject;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

/**
 * 面向Web多线程的SerialAnalyserController池分配 PooledObjectFactory
 * 
 * @author geosmart
 * 
 */
public class PipelinePooledObjectFactory extends BasePooledObjectFactory<SerialAnalyserController> {
  private static Logger log = LoggerFactory.getLogger(PipelinePooledObjectFactory.class);

  String gappFilePath;

  // borrow:

  @Override
  public void passivateObject(PooledObject<SerialAnalyserController> arg0) throws Exception {
	cleanupProcessor(arg0.getObject());
  }

  @Override
  public void activateObject(PooledObject<SerialAnalyserController> arg0) throws Exception {
	cleanupProcessor(arg0.getObject());
  }

  @Override
  public void destroyObject(PooledObject<SerialAnalyserController> arg0) throws Exception {
	System.out.println("how to destry?");
  }

  @Override
  public boolean validateObject(PooledObject<SerialAnalyserController> arg0) {
	return true;
  }

  @Override
  public SerialAnalyserController create() throws Exception {
	File file = new File(Gate.getGateHome().getPath() + gappFilePath);
	SerialAnalyserController processsor = (SerialAnalyserController) PersistenceManager.loadObjectFromFile(file);
	return processsor;
  }

  private void cleanupProcessor(SerialAnalyserController processsor) {
	processsor.cleanup();
  }

  @Override
  public PooledObject<SerialAnalyserController> wrap(SerialAnalyserController obj) {
	return new DefaultPooledObject<SerialAnalyserController>(obj);
  }

  public String getGappFilePath() {
	return gappFilePath;
  }

  public void setGappFilePath(String gappFilePath) {
	this.gappFilePath = gappFilePath;
  }

}

pom.xml

<!-- pom部分dependency -->
  <dependencies>
    <dependency>
        <groupId>commons-pool</groupId>
        <artifactId>commons-pool</artifactId>
        <version>1.5.4</version>
    </dependency>
    <dependency>
    	<groupId>org.apache.commons</groupId>
	<artifactId>commons-pool2</artifactId>
	<version>2.4.2</version>
    </dependency>  
  </dependencies>