Java 在lucene中搜索UUID不工作

我有一个UUID字段，我将以以下格式添加到文档中：372d325c-e01b-432f-98bd-bc4c949f15b8。但是，当我尝试通过UUID查询文档时，无论我如何尝试转义表达式，它都不会返回它们。例如：

+uuid:372d325c-e01b-432f-98bd-bc4c949f15b8
+uuid:"372d325c-e01b-432f-98bd-bc4c949f15b8"
+uuid:372d325c\-e01b\-432f\-98bd\-bc4c949f15b8
+uuid:(372d325c-e01b-432f-98bd-bc4c949f15b8)
+uuid:("372d325c-e01b-432f-98bd-bc4c949f15b8")

甚至使用TermQuery完全跳过QueryParser，如下所示：

new TermQuery(new Term("uuid", uuid.toString()))

IndexWriterConfig config = new IndexWriterConfig(Version.LUCENE_36, new WhitespaceAnalyzer(Version.LUCENE_36))
            .setOpenMode(IndexWriterConfig.OpenMode.CREATE_OR_APPEND);
writer = new IndexWriter( directory, config);

或

这些搜索都不会返回文档，但如果搜索UUID的部分，它将返回文档。例如，这些将返回一些内容：

+uuid:372d325c
+uuid:e01b
+uuid:432f

我应该做些什么来索引这些文档，以便通过它们的UUID将它们拉回来？我已经考虑过重新格式化UUID以删除连字符，但还没有实现

+uuid:372d325c\-e01b\-432f\-98bd\-bc4c949f15b8

应该有用

---

我猜如果没有，那是因为

uuid

字段没有像文档插入索引时那样填充。你能确定这个字段到底插入了什么吗？您可以使用对索引进行爬网，并查找为

uuid

字段存储的实际值。

实现此功能的唯一方法是使用WhitespaceAnalyzer而不是StandardAnalyzer。然后使用这样的术语查询：

new TermQuery(new Term("uuid", uuid.toString()))

IndexWriterConfig config = new IndexWriterConfig(Version.LUCENE_36, new WhitespaceAnalyzer(Version.LUCENE_36))
            .setOpenMode(IndexWriterConfig.OpenMode.CREATE_OR_APPEND);
writer = new IndexWriter( directory, config);

然后搜索：

TopDocs docs = searcher.search(new TermQuery(new Term("uuid", uuid.toString())), 1);

---

WhitespaceAnalyzer阻止Lucene通过连字符拆分UUID。另一种选择是从UUID中删除破折号，但出于我的目的，使用WhitespaceAnalyzer也同样有效。

如果计划将UUID字段作为查找键，则需要要求Lucene将整个字段作为单个字符串进行索引，而无需进行标记化。这是通过为UUID字段设置正确的字段类型来完成的。在Lucene 4+中，可以使用StringField

import java.io.IOException;
import java.util.UUID;
import junit.framework.Assert;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.standard.StandardAnalyzer;
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.lucene.document.StringField;
import org.apache.lucene.document.TextField;
import org.apache.lucene.index.DirectoryReader;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.index.IndexWriterConfig;
import org.apache.lucene.index.Term;
import org.apache.lucene.queryparser.classic.ParseException;
import org.apache.lucene.queryparser.classic.QueryParser;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.TopDocs;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.RAMDirectory;
import org.apache.lucene.util.Version;
import org.junit.Test;

/**
 * Using Lucene 4.7 on Java 7.
 */
public class LuceneUUIDFieldLookupTest {

    private Directory directory;
    private Analyzer analyzer;

    @Test
    public void testUsingUUIDAsLookupKey() throws IOException, ParseException {

        directory = new RAMDirectory();
        analyzer = new StandardAnalyzer(Version.LUCENE_47);

        UUID docUUID = UUID.randomUUID();
        String docContentText1 = "Stack Overflow is a question and answer site for professional and enthusiast programmers.";

        index(docUUID, docContentText1);

        QueryParser parser = new QueryParser(Version.LUCENE_47, MyIndexedFields.DOC_TEXT_FIELD.name(), analyzer);
        Query queryForProgrammer = parser.parse("programmers");

        IndexSearcher indexSearcher = getIndexSearcher();
        TopDocs hits = indexSearcher.search(queryForProgrammer, Integer.MAX_VALUE);
        Assert.assertTrue(hits.scoreDocs.length == 1);

        Integer internalDocId1 = hits.scoreDocs[0].doc;
        Document docRetrieved1 = indexSearcher.doc(internalDocId1);
        indexSearcher.getIndexReader().close();

        String docText1 = docRetrieved1.get(MyIndexedFields.DOC_TEXT_FIELD.name());
        Assert.assertEquals(docText1, docContentText1);

        String docContentText2 = "TechCrunch is a leading technology media property, dedicated to ... according to a new report from the Wall Street Journal confirmed by Google to TechCrunch.";
        reindex(docUUID, docContentText2);

        Query queryForTechCrunch = parser.parse("technology");
        indexSearcher = getIndexSearcher(); //you must reopen directory because the previous IndexSearcher only sees a snapshoted directory.
        hits = indexSearcher.search(queryForTechCrunch, Integer.MAX_VALUE);
        Assert.assertTrue(hits.scoreDocs.length == 1);

        Integer internalDocId2 = hits.scoreDocs[0].doc;
        Document docRetrieved2 = indexSearcher.doc(internalDocId2);
        indexSearcher.getIndexReader().close();

        String docText2 = docRetrieved2.get(MyIndexedFields.DOC_TEXT_FIELD.name());
        Assert.assertEquals(docText2, docContentText2);
    }

    private void reindex(UUID myUUID, String docContentText) throws IOException {
        try (IndexWriter indexWriter = new IndexWriter(directory, getIndexWriterConfig())) {
            Term term = new Term(MyIndexedFields.MY_UUID_FIELD.name(), myUUID.toString());
            indexWriter.updateDocument(term, buildDoc(myUUID, docContentText));
        }//auto-close
    }

    private void index(UUID myUUID, String docContentText) throws IOException {
        try (IndexWriter indexWriter = new IndexWriter(directory, getIndexWriterConfig())) {
            indexWriter.addDocument(buildDoc(myUUID, docContentText));
        }//auto-close
    }

    private IndexWriterConfig getIndexWriterConfig() {
        return new IndexWriterConfig(Version.LUCENE_47, analyzer).setOpenMode(IndexWriterConfig.OpenMode.CREATE_OR_APPEND);
    }

    private Document buildDoc(UUID myUUID, String docContentText) {
        Document doc = new Document();
        doc.add(new Field(
                MyIndexedFields.MY_UUID_FIELD.name(),
                myUUID.toString(),
                StringField.TYPE_STORED));//use TYPE_STORED if you want to read it back in search result.

        doc.add(new Field(
                MyIndexedFields.DOC_TEXT_FIELD.name(),
                docContentText,
                TextField.TYPE_STORED));

        return doc;
    }

    private IndexSearcher getIndexSearcher() throws IOException {
        DirectoryReader ireader = DirectoryReader.open(directory);
        IndexSearcher indexSearcher = new IndexSearcher(ireader);
        return indexSearcher;
    }

    enum MyIndexedFields {

        MY_UUID_FIELD,
        DOC_TEXT_FIELD
    }
}

---

您是否检查了字段的索引方式？uuid是否有可能被lucene标记器分离？现在我将如何将uuid添加到索引中：doc.add（新字段（“uuid”，id.toString（），Field.Store.YES，Field.index.NOT_analysis））。我在另一个项目中使用了完全相同的方案，效果非常好，但区别在于另一个项目中的ID不是UUID，并且不包含连字符。如果未分析该字段（因此未标记），则以下查询不应返回空+UUID:372d325c。一般规则是确保在索引和搜索时使用相同的分析器。您能否确认使用Field.Index.NOT_分析的索引，然后使用新的TermQuery（new Term（“uuid”，uuid.toString（））进行搜索）返回为空？我已在Luke中确认字段中的值已保存并显示在文档中。正如我在最初的问题中提到的，我可以通过搜索uuid的一个片段来收回文档，这进一步证实了这一点。