Examples with IndexWriterConfig org.apache.lucene.index.IndexWriterConfig used on opensource projects

Example 41 with IndexWriterConfig

use of org.apache.lucene.index.IndexWriterConfig in project elasticsearch by elastic.

the class CustomUnifiedHighlighterTests method assertHighlightOneDoc.

private void assertHighlightOneDoc(String fieldName, String[] inputs, Analyzer analyzer, Query query, Locale locale, BreakIterator breakIterator, int noMatchSize, String[] expectedPassages) throws Exception {
    Directory dir = newDirectory();
    IndexWriterConfig iwc = newIndexWriterConfig(analyzer);
    iwc.setMergePolicy(newTieredMergePolicy(random()));
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, iwc);
    FieldType ft = new FieldType(TextField.TYPE_STORED);
    ft.setIndexOptions(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS);
    ft.freeze();
    Document doc = new Document();
    for (String input : inputs) {
        Field field = new Field(fieldName, "", ft);
        field.setStringValue(input);
        doc.add(field);
    }
    iw.addDocument(doc);
    DirectoryReader reader = iw.getReader();
    IndexSearcher searcher = newSearcher(reader);
    iw.close();
    TopDocs topDocs = searcher.search(new MatchAllDocsQuery(), 1, Sort.INDEXORDER);
    assertThat(topDocs.totalHits, equalTo(1));
    String rawValue = Strings.arrayToDelimitedString(inputs, String.valueOf(MULTIVAL_SEP_CHAR));
    CustomUnifiedHighlighter highlighter = new CustomUnifiedHighlighter(searcher, analyzer, new CustomPassageFormatter("<b>", "</b>", new DefaultEncoder()), locale, breakIterator, rawValue, noMatchSize);
    highlighter.setFieldMatcher((name) -> "text".equals(name));
    final Snippet[] snippets = highlighter.highlightField("text", query, topDocs.scoreDocs[0].doc, expectedPassages.length);
    assertEquals(snippets.length, expectedPassages.length);
    for (int i = 0; i < snippets.length; i++) {
        assertEquals(snippets[i].getText(), expectedPassages[i]);
    }
    reader.close();
    dir.close();
}

Example 42 with IndexWriterConfig

use of org.apache.lucene.index.IndexWriterConfig in project elasticsearch by elastic.

the class CustomPostingsHighlighterTests method testCustomPostingsHighlighter.

public void testCustomPostingsHighlighter() throws Exception {
    Directory dir = newDirectory();
    IndexWriterConfig iwc = newIndexWriterConfig(new MockAnalyzer(random()));
    iwc.setMergePolicy(newLogMergePolicy());
    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, iwc);
    FieldType offsetsType = new FieldType(TextField.TYPE_STORED);
    offsetsType.setIndexOptions(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS);
    //good position but only one match
    final String firstValue = "This is a test. Just a test1 highlighting from postings highlighter.";
    Field body = new Field("body", "", offsetsType);
    Document doc = new Document();
    doc.add(body);
    body.setStringValue(firstValue);
    //two matches, not the best snippet due to its length though
    final String secondValue = "This is the second highlighting value to perform highlighting on a longer text that gets scored lower.";
    Field body2 = new Field("body", "", offsetsType);
    doc.add(body2);
    body2.setStringValue(secondValue);
    //two matches and short, will be scored highest
    final String thirdValue = "This is highlighting the third short highlighting value.";
    Field body3 = new Field("body", "", offsetsType);
    doc.add(body3);
    body3.setStringValue(thirdValue);
    //one match, same as first but at the end, will be scored lower due to its position
    final String fourthValue = "Just a test4 highlighting from postings highlighter.";
    Field body4 = new Field("body", "", offsetsType);
    doc.add(body4);
    body4.setStringValue(fourthValue);
    iw.addDocument(doc);
    IndexReader ir = iw.getReader();
    iw.close();
    String firstHlValue = "Just a test1 <b>highlighting</b> from postings highlighter.";
    String secondHlValue = "This is the second <b>highlighting</b> value to perform <b>highlighting</b> on a longer text that gets scored lower.";
    String thirdHlValue = "This is <b>highlighting</b> the third short <b>highlighting</b> value.";
    String fourthHlValue = "Just a test4 <b>highlighting</b> from postings highlighter.";
    IndexSearcher searcher = newSearcher(ir);
    Query query = new TermQuery(new Term("body", "highlighting"));
    TopDocs topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    assertThat(topDocs.totalHits, equalTo(1));
    int docId = topDocs.scoreDocs[0].doc;
    String fieldValue = firstValue + HighlightUtils.PARAGRAPH_SEPARATOR + secondValue + HighlightUtils.PARAGRAPH_SEPARATOR + thirdValue + HighlightUtils.PARAGRAPH_SEPARATOR + fourthValue;
    CustomPostingsHighlighter highlighter = new CustomPostingsHighlighter(null, new CustomPassageFormatter("<b>", "</b>", new DefaultEncoder()), fieldValue, false);
    Snippet[] snippets = highlighter.highlightField("body", query, searcher, docId, 5);
    assertThat(snippets.length, equalTo(4));
    assertThat(snippets[0].getText(), equalTo(firstHlValue));
    assertThat(snippets[1].getText(), equalTo(secondHlValue));
    assertThat(snippets[2].getText(), equalTo(thirdHlValue));
    assertThat(snippets[3].getText(), equalTo(fourthHlValue));
    ir.close();
    dir.close();
}

Example 43 with IndexWriterConfig

use of org.apache.lucene.index.IndexWriterConfig in project ansj_seg by NLPchina.

the class HeightLightTest method index.

private static void index(Analyzer analysis, String content) throws CorruptIndexException, IOException {
    Document doc = new Document();
    IndexWriter iwriter = new IndexWriter(directory, new IndexWriterConfig(analysis));
    doc.add(new TextField("text", content, Field.Store.YES));
    iwriter.addDocument(doc);
    iwriter.commit();
    iwriter.close();
}

Example 44 with IndexWriterConfig

use of org.apache.lucene.index.IndexWriterConfig in project ansj_seg by NLPchina.

the class IndexAndTest method test.

@Test
public void test() throws Exception {
    DicLibrary.put(DicLibrary.DEFAULT, "../../library/default.dic");
    PerFieldAnalyzerWrapper analyzer = new PerFieldAnalyzerWrapper(new AnsjAnalyzer(TYPE.index_ansj));
    Directory directory = null;
    IndexWriter iwriter = null;
    IndexWriterConfig ic = new IndexWriterConfig(analyzer);
    String text = "旅游和服务是最好的";
    System.out.println(IndexAnalysis.parse(text));
    // 建立内存索引对象
    directory = new RAMDirectory();
    iwriter = new IndexWriter(directory, ic);
    addContent(iwriter, text);
    iwriter.commit();
    iwriter.close();
    System.out.println("索引建立完毕");
    Analyzer queryAnalyzer = new AnsjAnalyzer(AnsjAnalyzer.TYPE.index_ansj);
    System.out.println("index ok to search!");
    for (Term t : IndexAnalysis.parse(text)) {
        System.out.println(t.getName());
        search(queryAnalyzer, directory, "\"" + t.getName() + "\"");
    }
}

Example 45 with IndexWriterConfig

use of org.apache.lucene.index.IndexWriterConfig in project ansj_seg by NLPchina.

the class IndexTest method indexTest.

@Test
public void indexTest() throws CorruptIndexException, LockObtainFailedException, IOException, ParseException {
    PerFieldAnalyzerWrapper analyzer = new PerFieldAnalyzerWrapper(new AnsjAnalyzer(TYPE.index_ansj));
    Directory directory = null;
    IndexWriter iwriter = null;
    IndexWriterConfig ic = new IndexWriterConfig(analyzer);
    // 建立内存索引对象
    directory = new RAMDirectory();
    iwriter = new IndexWriter(directory, ic);
    addContent(iwriter, "助推企业转型升级提供强有力的技术支持和服保障。中心的建成将使青岛的服务器承载能力突破10万台，达到世界一流水平。");
    addContent(iwriter, "涉及民生的部分商品和服务成本监审政策");
    addContent(iwriter, "我穿着和服");
    iwriter.commit();
    iwriter.close();
    System.out.println("索引建立完毕");
    Analyzer queryAnalyzer = new AnsjAnalyzer(AnsjAnalyzer.TYPE.dic_ansj);
    System.out.println("index ok to search!");
    search(queryAnalyzer, directory, "\"和服\"");
}