編譯Ansj之Solr插件

　　Ansj是一個比較優秀的中文分詞組件，具體狀況就不在本文介紹了。ansj做者在其官方代碼中，提供了對lucene接口的支持。若是用在Solr下，還須要簡單的擴展一下。

一、基於maven管理

   ansj是基於maven進行開發管理的。咱們首先修改一下其pom.xml，具體以下所示：

　　

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
	xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
	<modelVersion>4.0.0</modelVersion>

	<parent>
		<groupId>org.ansj</groupId>
		<artifactId>MavenAccount-aggregator</artifactId>
		<version>0.0.1</version>
		<relativePath>../pom.xml</relativePath>
	</parent>

	<artifactId>ansj_lucene4_plug</artifactId>
	<version>2.0.2</version>
	<packaging>jar</packaging>

	<name>ansj_lucene4_plug</name>
	
 	<properties>
        <solr.version>4.8.0</solr.version>
    </properties>

	<dependencies>
		<dependency>
			<groupId>org.ansj</groupId>
			<artifactId>ansj_seg</artifactId>
			<version>2.0.5</version>
			<classifier>min</classifier>
			<scope>provided</scope>
		</dependency>
		
		<dependency>
			<groupId>org.apache.lucene</groupId>
			<artifactId>lucene-core</artifactId>
			<version>${solr.version}</version>
			<scope>provided</scope>
		</dependency>
		<dependency>
			<groupId>org.apache.lucene</groupId>
			<artifactId>lucene-highlighter</artifactId>
			<version>${solr.version}</version>
			<scope>provided</scope>
		</dependency>

		<dependency>
			<groupId>org.apache.lucene</groupId>
			<artifactId>lucene-queries</artifactId>
			<version>${solr.version}</version>
			<scope>provided</scope>
		</dependency>

		<dependency>
			<groupId>org.apache.lucene</groupId>
			<artifactId>lucene-queryparser</artifactId>
			<version>${solr.version}</version>
			<scope>provided</scope>
		</dependency>
		
	   <dependency>
			<groupId>org.apache.solr</groupId>
			<artifactId>solr-dataimporthandler</artifactId>
			<version>${solr.version}</version>
			<scope>provided</scope>
	  </dependency>

		<dependency>
			<groupId>junit</groupId>
			<artifactId>junit</artifactId>
			<version>4.4</version>
			<scope>test</scope>
		</dependency>
	</dependencies>
</project>

　　其中，代碼依賴的配置項：<scope>provided</scope> 表示只用於代碼編譯階段。依賴關係整理好之後，寫一個TokenizerFactory類，用於solr中配置使用，代碼以下：

package org.ansj.solr;

import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InputStreamReader;
import java.io.Reader;
import java.util.HashSet;
import java.util.Map;
import java.util.Set;

import org.ansj.lucene.util.AnsjTokenizer;
import org.ansj.splitWord.analysis.IndexAnalysis;
import org.ansj.splitWord.analysis.ToAnalysis;
import org.apache.lucene.analysis.Tokenizer;
import org.apache.lucene.analysis.util.TokenizerFactory;
import org.apache.lucene.util.AttributeSource.AttributeFactory;

public class AnsjTokenizerFactory extends TokenizerFactory{
    boolean pstemming;
    boolean isQuery;
    private String stopwordsDir;
    public Set<String> filter;  

    public AnsjTokenizerFactory(Map<String, String> args) {
        super(args);
        assureMatchVersion();
        isQuery = getBoolean(args, "isQuery", true);
        pstemming = getBoolean(args, "pstemming", false);
        stopwordsDir = get(args,"words");
        addStopwords(stopwordsDir);
    }
    //add stopwords list to filter
    private void addStopwords(String dir) {
        if (dir == null){
            System.out.println("no stopwords dir");
            return;
        }
        //read stoplist
        System.out.println("stopwords: " + dir);
        filter = new HashSet<String>();
        File file = new File(dir); 
        InputStreamReader reader;
        try {
            reader = new InputStreamReader(new FileInputStream(file),"UTF-8");
            BufferedReader br = new BufferedReader(reader); 
            String word = br.readLine();  
            while (word != null) {
                filter.add(word);
                word = br.readLine(); 
            }  
        } catch (FileNotFoundException e) {
            System.out.println("No stopword file found");
        } catch (IOException e) {
            System.out.println("stopword file io exception");
        }      
    }
    @Override
    public Tokenizer create(AttributeFactory factory, Reader input) {
        if(isQuery == true){
            //query
            return new AnsjTokenizer(new ToAnalysis(new BufferedReader(input)), input, filter, pstemming);
        } else {
            //index
            return new AnsjTokenizer(new IndexAnalysis(new BufferedReader(input)), input, filter, pstemming);
        }
    }       
}

　　pstemming 參數是ansj須要的參數。

　　isQuery 是用於判斷是查詢仍是索引，通常搜索index階段分詞比較細，查詢的分詞比較粗。

二、編譯jar包。

    代碼結構以下：

   

　　編寫mavn編譯命令：mvn install -DskipTests=true# 忽略單元測試編譯。

　　

    執行編譯：

[INFO] Scanning for projects...
[INFO]                                                                         
[INFO] ------------------------------------------------------------------------
[INFO] Building ansj_lucene4_plug 2.0.2
[INFO] ------------------------------------------------------------------------
[INFO] 
[INFO] --- maven-clean-plugin:2.4.1:clean (default-clean) @ ansj_lucene4_plug ---
[INFO] Deleting R:\ansj-seg\ansj_seg\plug\ansj_lucene4_plug\target
[INFO] 
[INFO] --- maven-resources-plugin:2.4.3:resources (default-resources) @ ansj_lucene4_plug ---
[INFO] Using 'UTF-8' encoding to copy filtered resources.
[INFO] skip non existing resourceDirectory R:\ansj-seg\ansj_seg\plug\ansj_lucene4_plug\src\main\resources
[INFO] 
[INFO] --- maven-compiler-plugin:2.3.2:compile (default-compile) @ ansj_lucene4_plug ---
[INFO] Compiling 5 source files to R:\ansj-seg\ansj_seg\plug\ansj_lucene4_plug\target\classes
[INFO] 
[INFO] --- maven-resources-plugin:2.4.3:testResources (default-testResources) @ ansj_lucene4_plug ---
[INFO] Using 'UTF-8' encoding to copy filtered resources.
[INFO] skip non existing resourceDirectory R:\ansj-seg\ansj_seg\plug\ansj_lucene4_plug\src\test\resources
[INFO] 
[INFO] --- maven-compiler-plugin:2.3.2:testCompile (default-testCompile) @ ansj_lucene4_plug ---
[INFO] Compiling 3 source files to R:\ansj-seg\ansj_seg\plug\ansj_lucene4_plug\target\test-classes
[INFO] 
[INFO] --- maven-surefire-plugin:2.7.1:test (default-test) @ ansj_lucene4_plug ---
[INFO] Tests are skipped.
[INFO] 
[INFO] --- maven-jar-plugin:2.3.1:jar (default-jar) @ ansj_lucene4_plug ---
[INFO] Building jar: R:\ansj-seg\ansj_seg\plug\ansj_lucene4_plug\target\ansj_lucene4_plug-2.0.2.jar
[INFO] 
[INFO] --- maven-install-plugin:2.3.1:install (default-install) @ ansj_lucene4_plug ---
[INFO] Installing R:\ansj-seg\ansj_seg\plug\ansj_lucene4_plug\target\ansj_lucene4_plug-2.0.2.jar to C:\Users\GCZX-016\.m2\repository\org\ansj\ansj_lucene4_plug\2.0.2\ansj_lucene4_plug-2.0.2.jar
[INFO] Installing R:\ansj-seg\ansj_seg\plug\ansj_lucene4_plug\pom.xml to C:\Users\GCZX-016\.m2\repository\org\ansj\ansj_lucene4_plug\2.0.2\ansj_lucene4_plug-2.0.2.pom
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 8.149s
[INFO] Finished at: Tue May 05 15:29:19 CST 2015
[INFO] Final Memory: 27M/245M
[INFO] ------------------------------------------------------------------------