原文链接 http://codepub.cn/2017/10/10/several-ways-to-get-the-field-values-that-are-not-stored-in-lucene/
注：以下为加速网络访问所做的原文缓存，经过重新格式化，可能存在格式方面的问题，或偶有遗漏信息，请以原文为准。
一般来说，如果想要从Lucene索引中获取Field的值，那么需要在索引阶段设置Field.Store.YES才可以，然后在搜索阶段得到TopDocs对象之后，用它去获取ScoreDoc再取出Document，使用Document获取存储在索引中的值。但是我们都知道，存储字段是需要硬盘空间的，如果想要追求极致的存储空间并且获取Field的值，那么在不存储的情况下，如何获取呢？其实仔细思索一下，在我们只索引不存储的情况下，Lucene依然可以判断搜索是否命中，这说明在Lucene索引中依然存有一份Field的值，这样在搜索阶段才能判断是否匹配。本文就是探讨在这种情形下，使用Lucene的核心包获取没有存储的Field的值的几种方法，如果你还有其它不同的方法请留言。
- testGetFieldByStore 演示存储Field值时如何获取
- testGetFieldByTerms 演示通过Terms获取没有存储的Field值
- testGetFieldByFieldDocWithSorted 演示通过FieldDoc获取没有存储的值
- testGetFieldByTermVector 演示通过TermVector获取没有存储的值
- testGetFieldByTermVectors 演示通过TermVectors获取没有存储的值
这里补充一下，在lucene-suggest包中，有LuceneDictionary类，通过该类的getEntryIterator方法也能获取没有存储的Field的值，不过其本质和通过Terms获取方式一样，在此不再列举。源码示例如下
import org.apache.lucene.analysis.core.WhitespaceAnalyzer;
import org.apache.lucene.document.*;
import org.apache.lucene.index.*;
import org.apache.lucene.search.*;
import org.apache.lucene.store.RAMDirectory;
import org.apache.lucene.util.BytesRef;
import org.junit.Test;
import java.io.IOException;
import static org.apache.lucene.search.SortField.Type.STRING;
/**
* <p>
* Created by wangxu on 2017/10/10 17:33.
* </p>
* <p>
* Description: Lucene 6.5.0
* </p>
*
* @author Wang Xu
* @version V1.0.0
* @since V1.0.0 <br/>
* WebSite: http://codepub.cn <br>
* Licence: Apache v2 License
*/
public class GetNonStoredFieldDemo {
private RAMDirectory ramDirectory = new RAMDirectory();
private IndexWriter indexWriter = new IndexWriter(ramDirectory, new IndexWriterConfig(new WhitespaceAnalyzer()));
public GetNonStoredFieldDemo() throws IOException {
}
@Test
public void testGetFieldByStore() throws IOException {
initIndexForStore();
IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(ramDirectory));
int count = indexSearcher.count(new MatchAllDocsQuery());
TopDocs search = indexSearcher.search(new MatchAllDocsQuery(), count);
ScoreDoc[] scoreDocs = search.scoreDocs;
for (ScoreDoc scoreDoc : scoreDocs) {
Document doc = indexSearcher.doc(scoreDoc.doc);
System.out.println(doc.get("IDX") + "=>" + doc.get("title"));
}
ramDirectory.close();
}
@Test
public void testGetFieldByTerms() throws IOException {
initIndexForTerms();
Fields fields = MultiFields.getFields(DirectoryReader.open(ramDirectory));
Terms idx = fields.terms("IDX");
Terms title = fields.terms("title");
//or you can use like this
//TermsEnum idxIter = MultiFields.getTerms(DirectoryReader.open(ramDirectory), "IDX").iterator();
TermsEnum idxIter = idx.iterator();
TermsEnum titleIter = title.iterator();
BytesRef bytesRef;
while ((bytesRef = idxIter.next()) != null) {
System.out.println(bytesRef.utf8ToString() + "=>" + titleIter.next().utf8ToString());
}
ramDirectory.close();
}
@Test
public void testGetFieldByFieldDocWithSorted() throws IOException {
initIndexForFieldDocWithSorted();
IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(ramDirectory));
int count = indexSearcher.count(new MatchAllDocsQuery());
//must use method which returns TopFieldDocs
TopFieldDocs search = indexSearcher.search(new MatchAllDocsQuery(), count, new Sort(new SortField("IDX", STRING)));
ScoreDoc[] scoreDocs = search.scoreDocs;
for (ScoreDoc scoreDoc : scoreDocs) {
FieldDoc fieldDoc = (FieldDoc) scoreDoc;
Object[] fields = fieldDoc.fields;
if (fields[0] instanceof BytesRef) {
BytesRef temp = (BytesRef) fields[0];
System.out.println(temp.utf8ToString() + "=>" + indexSearcher.doc(scoreDoc.doc).get("title"));
}
}
ramDirectory.close();
}
@Test
public void testGetFieldByTermVector() throws IOException {
initIndexForTermVector();
IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(ramDirectory));
int count = indexSearcher.count(new MatchAllDocsQuery());
TopDocs search = indexSearcher.search(new MatchAllDocsQuery(), count);
ScoreDoc[] scoreDocs = search.scoreDocs;
for (ScoreDoc scoreDoc : scoreDocs) {
int doc = scoreDoc.doc;
Terms idx = indexSearcher.getIndexReader().getTermVector(doc, "IDX");
TermsEnum iterator = idx.iterator();
BytesRef bytesRef;
while ((bytesRef = iterator.next()) != null) {
System.out.println(bytesRef.utf8ToString() + "=>" + indexSearcher.doc(doc).get("title"));
}
}
ramDirectory.close();
}
@Test
public void testGetFieldByTermVectors() throws IOException {
initIndexForTermVector();
IndexSearcher indexSearcher = new IndexSearcher(DirectoryReader.open(ramDirectory));
int count = indexSearcher.count(new MatchAllDocsQuery());
TopDocs search = indexSearcher.search(new MatchAllDocsQuery(), count);
ScoreDoc[] scoreDocs = search.scoreDocs;
for (ScoreDoc scoreDoc : scoreDocs) {
int doc = scoreDoc.doc;
Fields termVectors = indexSearcher.getIndexReader().getTermVectors(doc);
Terms idx = termVectors.terms("IDX");
TermsEnum iterator = idx.iterator();
BytesRef bytesRef;
while ((bytesRef = iterator.next()) != null) {
System.out.println(bytesRef.utf8ToString() + "=>" + indexSearcher.doc(doc).get("title"));
}
}
ramDirectory.close();
}
private void initIndexForStore() throws IOException {
Document document = new Document();
document.add(new StringField("IDX", "TEST01", Field.Store.YES));
document.add(new StringField("title", "TITLE01", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new StringField("IDX", "TEST02", Field.Store.YES));
document.add(new StringField("title", "TITLE02", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new StringField("IDX", "TEST03", Field.Store.YES));
document.add(new StringField("title", "TITLE03", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new StringField("IDX", "TEST04", Field.Store.YES));
document.add(new StringField("title", "TITLE04", Field.Store.YES));
indexWriter.addDocument(document);
indexWriter.close();
}
private void initIndexForTerms() throws IOException {
Document document = new Document();
document.add(new StringField("IDX", "TEST01", Field.Store.NO));
document.add(new StringField("title", "TITLE01", Field.Store.NO));
indexWriter.addDocument(document);
document = new Document();
document.add(new StringField("IDX", "TEST02", Field.Store.NO));
document.add(new StringField("title", "TITLE02", Field.Store.NO));
indexWriter.addDocument(document);
document = new Document();
document.add(new StringField("IDX", "TEST03", Field.Store.NO));
document.add(new StringField("title", "TITLE03", Field.Store.NO));
indexWriter.addDocument(document);
document = new Document();
document.add(new StringField("IDX", "TEST04", Field.Store.NO));
document.add(new StringField("title", "TITLE04", Field.Store.NO));
indexWriter.addDocument(document);
indexWriter.close();
}
private void initIndexForTermVector() throws IOException {
FieldType fieldType = new FieldType();
fieldType.setStoreTermVectors(true);
fieldType.setIndexOptions(IndexOptions.DOCS);
Document document = new Document();
document.add(new Field("IDX", "TEST01", fieldType));
document.add(new StringField("title", "TITLE01", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new Field("IDX", "TEST02", fieldType));
document.add(new StringField("title", "TITLE02", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new Field("IDX", "TEST03", fieldType));
document.add(new StringField("title", "TITLE03", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new Field("IDX", "TEST04", fieldType));
document.add(new StringField("title", "TITLE04", Field.Store.YES));
indexWriter.addDocument(document);
indexWriter.close();
}
private void initIndexForFieldDocWithSorted() throws IOException {
Document document = new Document();
document.add(new SortedDocValuesField("IDX", new BytesRef("TEST01")));
document.add(new StringField("title", "TITLE01", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new SortedDocValuesField("IDX", new BytesRef("TEST02")));
document.add(new StringField("title", "TITLE02", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new SortedDocValuesField("IDX", new BytesRef("TEST03")));
document.add(new StringField("title", "TITLE03", Field.Store.YES));
indexWriter.addDocument(document);
document = new Document();
document.add(new SortedDocValuesField("IDX", new BytesRef("TEST04")));
document.add(new StringField("title", "TITLE04", Field.Store.YES));
indexWriter.addDocument(document);
indexWriter.close();
}
}