Solr カスタムanalyzerもどきを作る

gistfile1.txt

schema.xml は schema version 1.5 に合わせて修正された版を、以下から頂きました。
http://johtani.jugem.jp/?eid=44 

--text_ja の analyzer の箇所だけを抜粋--

<analyzer type="index">
    <charFilter class="solr.MappingCharFilterFactory" mapping="mapping-ja.txt"/>
    <tokenizer class="solr.JapaneseTokenizerFactory" mode="search"/>
    <filter class="solr.JapaneseKatakanaStemFilterFactory" minimumLength="4"/>
    <filter class="solr.SynonymFilterFactory" synonyms="synonyms-ja.txt" ignoreCase="true" expand="true" 
            tokenizerFactory="solr.JapaneseTokenizerFactory"/>
    <filter class="solr.JapanesePartOfSpeechStopFilterFactory" tags="pos-deny.txt" enablePositionIncrements="true"/>
    <filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords-ja.txt"/>
    <filter class="solr.LowerCaseFilterFactory"/>
</analyzer>

-------------------------------------

SolrbookAnalyzer.java

package test.solrbook;

import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileReader;
import java.io.IOException;
import java.io.InputStream;
import java.io.Reader;
import java.io.StringReader;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

import org.apache.lucene.analysis.CharReader;
import org.apache.lucene.analysis.CharStream;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.ja.tokenattributes.BaseFormAttribute;
import org.apache.lucene.analysis.ja.tokenattributes.InflectionAttribute;
import org.apache.lucene.analysis.ja.tokenattributes.PartOfSpeechAttribute;
import org.apache.lucene.analysis.ja.tokenattributes.ReadingAttribute;
import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
import org.apache.lucene.analysis.util.ResourceLoader;
import org.apache.lucene.util.Version;
import org.apache.solr.analysis.JapaneseKatakanaStemFilterFactory;
import org.apache.solr.analysis.JapanesePartOfSpeechStopFilterFactory;
import org.apache.solr.analysis.JapaneseTokenizerFactory;
import org.apache.solr.analysis.LowerCaseFilterFactory;
import org.apache.solr.analysis.MappingCharFilterFactory;
import org.apache.solr.analysis.StopFilterFactory;
import org.apache.solr.analysis.SynonymFilterFactory;

/**
 * 「Apache Solr入門」2章
 * text_ja型のanalyzerと同等の動きをする（はずの）クラス
 * Lucene/Solr 4.0 alphaで動作します。
 */
public class SolrbookAnalyzer {

	private ResourceLoader resourceLoader = new MyResourceLoader();
	
	public static void main(String[] args) {
		String content = "ｿｰﾗｰは検索ｴﾝｼﾞﾝです。";
		SolrbookAnalyzer analyzer = new SolrbookAnalyzer();
		analyzer.analyze(new StringReader(content));
	}
	
	private void analyze(Reader reader) {
		CharStream charStream = charFilter(reader);
		TokenStream tokenizer = tokenizer(charStream);
		TokenStream stream = filter(tokenizer);
		try {
			while(stream.incrementToken()) {
				CharTermAttribute charAtt = stream.getAttribute(CharTermAttribute.class);
				BaseFormAttribute bfAtt = stream.getAttribute(BaseFormAttribute.class);
				InflectionAttribute infAtt = stream.getAttribute(InflectionAttribute.class);
				ReadingAttribute readAtt = stream.getAttribute(ReadingAttribute.class);
				PartOfSpeechAttribute posAtt = stream.getAttribute(PartOfSpeechAttribute.class);
				System.out.println(
						charAtt + "\t" + // トークン
						bfAtt.getBaseForm() + "\t" + // 基本形
						infAtt.getInflectionForm() + "\t" + // 活用形
						readAtt.getReading() + "\t" + // 読み
						readAtt.getPronunciation() + "\t" + // 発音
						posAtt.getPartOfSpeech() // 品詞
						);
			}
		} catch (IOException e) {
			e.printStackTrace();
		}
	}
	
	/** charFilter もどき */
	private CharStream charFilter(Reader reader) {
		// solr.MappingCharFilterFactory
		MappingCharFilterFactory factory = new MappingCharFilterFactory();
		Map<String, String> args = new HashMap<String, String>();
		args.put("mapping", "mapping-ja.txt");
		factory.init(args);
		factory.inform(resourceLoader);
		CharStream stream = factory.create(CharReader.get(reader));
		return stream;
	}
	
	/** tokenizer もどき */
	private TokenStream tokenizer(CharStream charStream) {
		// solr.JapaneseTokenizerFactory
		JapaneseTokenizerFactory factory = new JapaneseTokenizerFactory();
		Map<String, String> args = new HashMap<String, String>();
		args.put("mode", "search");
		factory.init(args);
		factory.inform(resourceLoader);

		TokenStream stream = factory.create(charStream);
		return stream;
	}
	
	/** filter もどき */
	private TokenStream filter(TokenStream tokenizer) {
		TokenStream stream = null;

		// solr.JapaneseKatakanaStemFilterFactory
		JapaneseKatakanaStemFilterFactory factory1 = new JapaneseKatakanaStemFilterFactory();
		Map<String, String> args1 = new HashMap<String, String>();
		args1.put("minimumLength", "4");
		factory1.init(args1);
		stream = factory1.create(tokenizer);

		// solr.SynonymFilterFactory
		SynonymFilterFactory factory2 = new SynonymFilterFactory();
		Map<String, String> args2 = new HashMap<String, String>();
		args2.put("synonyms", "synonyms-ja.txt");
		args2.put("ignoreCase", "true");
		args2.put("expand", "true");
		args2.put("tokenizerFactory", "org.apache.solr.analysis.JapaneseTokenizerFactory");
		factory2.setLuceneMatchVersion(Version.LUCENE_40);
		factory2.init(args2);
		factory2.inform(resourceLoader);
		stream = factory2.create(stream);
		
		// solr.JapanesePartOfSpeechStopFilterFactory
		JapanesePartOfSpeechStopFilterFactory factory3 = new JapanesePartOfSpeechStopFilterFactory();
		Map<String, String> args3 = new HashMap<String, String>();
		args3.put("tags", "pos-deny.txt");
		args3.put("enablePositionIncrements", "true");
		factory3.setLuceneMatchVersion(Version.LUCENE_40);
		factory3.init(args3);
		factory3.inform(resourceLoader);
		stream = factory3.create(stream);

		// solr.StopFilterFactory
		StopFilterFactory factory4 = new StopFilterFactory();
		Map<String, String> args4 = new HashMap<String, String>();
		args4.put("ignoreCase", "true");
		args4.put("words", "stopwords-ja.txt");
		factory4.setLuceneMatchVersion(Version.LUCENE_40);
		factory4.init(args4);
		factory4.inform(resourceLoader);
		stream = factory4.create(stream);
		
		// solr.LowerCaseFilterFactory
		LowerCaseFilterFactory factory5 = new LowerCaseFilterFactory();
		factory5.setLuceneMatchVersion(Version.LUCENE_40);
		stream = factory5.create(stream);
		
		return stream;
	}
	
	/** リソースローダーもどき */
	class MyResourceLoader implements ResourceLoader {
		// 設定ファイルを置くディレクトリ
		// ここに、mapping-ja.txt, pos-deny.txt, stopwords-ja.txt, synonyms-ja.txt を置いておく
		// 本のサンプル（サポートサイトからダウンロード）をそのまま使用
		private static final String confdir = "solrbook";
		
		@Override
		public InputStream openResource(String resource) throws IOException {
			return new FileInputStream(new File(confdir, resource));
		}

		@Override
		public List<String> getLines(String resource) throws IOException {
			List<String> lines = new ArrayList<String>();
			File file = new File(confdir, resource);
			BufferedReader reader = new BufferedReader(new FileReader(file));
			String line = null;
			while ((line = reader.readLine()) != null) {
				if (line.startsWith("#")) continue;
				if (line.length() == 0) continue;
				lines.add(line);
			}
			reader.close();
			return lines;
		}

		@Override
		public <T> T newInstance(String cname, Class<T> expectedType,
				String... subpackages) {
			// !! とりあえずなにかnewするだけのダメコード
			T obj = null;
			try {
				ClassLoader classLoader = MyResourceLoader.class.getClassLoader();
				Class<? extends T> clazz = Class.forName(cname, true, classLoader).asSubclass(expectedType);
				obj = clazz.newInstance();
			} catch (Exception e) {
				e.printStackTrace();
			}
			return obj;
		}
		
	}
}

実行結果

solr	null	null	null	null	null
ソーラ	null	null	null	null	null
検索	null	null	ケンサク	ケンサク	名詞-サ変接続
エンジン	null	null	エンジン	エンジン	名詞-一般

（実際にSolrを起動して、管理画面から「Field Analysis」にかけた時と同じ出力となります。）