Examples with DefaultEncoder org.apache.lucene.search.highlight.DefaultEncoder used on opensource projects

Example 6 with DefaultEncoder

use of org.apache.lucene.search.highlight.DefaultEncoder in project elasticsearch by elastic.

the class CustomPostingsHighlighterTests method testNoMatchSize.

public void testNoMatchSize() 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);
    Field body = new Field("body", "", offsetsType);
    Field none = new Field("none", "", offsetsType);
    Document doc = new Document();
    doc.add(body);
    doc.add(none);
    String firstValue = "This is a test. Just a test highlighting from postings. Feel free to ignore.";
    body.setStringValue(firstValue);
    none.setStringValue(firstValue);
    iw.addDocument(doc);
    IndexReader ir = iw.getReader();
    iw.close();
    Query query = new TermQuery(new Term("none", "highlighting"));
    IndexSearcher searcher = newSearcher(ir);
    TopDocs topDocs = searcher.search(query, 10, Sort.INDEXORDER);
    assertThat(topDocs.totalHits, equalTo(1));
    int docId = topDocs.scoreDocs[0].doc;
    CustomPassageFormatter passageFormatter = new CustomPassageFormatter("<b>", "</b>", new DefaultEncoder());
    CustomPostingsHighlighter highlighter = new CustomPostingsHighlighter(null, passageFormatter, firstValue, false);
    Snippet[] snippets = highlighter.highlightField("body", query, searcher, docId, 5);
    assertThat(snippets.length, equalTo(0));
    highlighter = new CustomPostingsHighlighter(null, passageFormatter, firstValue, true);
    snippets = highlighter.highlightField("body", query, searcher, docId, 5);
    assertThat(snippets.length, equalTo(1));
    assertThat(snippets[0].getText(), equalTo("This is a test."));
    ir.close();
    dir.close();
}

Example 7 with DefaultEncoder

use of org.apache.lucene.search.highlight.DefaultEncoder in project lucene-solr by apache.

the class FastVectorHighlighterTest method testMultiValuedSortByScore.

public void testMultiValuedSortByScore() throws IOException {
    Directory dir = newDirectory();
    IndexWriter writer = new IndexWriter(dir, newIndexWriterConfig(new MockAnalyzer(random())));
    Document doc = new Document();
    FieldType type = new FieldType(TextField.TYPE_STORED);
    type.setStoreTermVectorOffsets(true);
    type.setStoreTermVectorPositions(true);
    type.setStoreTermVectors(true);
    type.freeze();
    // The first two fields contain the best match
    doc.add(new Field("field", "zero if naught", type));
    // but total a lower score (3) than the bottom
    doc.add(new Field("field", "hero of legend", type));
    // two fields (4)
    doc.add(new Field("field", "naught of hero", type));
    doc.add(new Field("field", "naught of hero", type));
    writer.addDocument(doc);
    FastVectorHighlighter highlighter = new FastVectorHighlighter();
    ScoreOrderFragmentsBuilder fragmentsBuilder = new ScoreOrderFragmentsBuilder();
    fragmentsBuilder.setDiscreteMultiValueHighlighting(true);
    IndexReader reader = DirectoryReader.open(writer);
    String[] preTags = new String[] { "<b>" };
    String[] postTags = new String[] { "</b>" };
    Encoder encoder = new DefaultEncoder();
    int docId = 0;
    BooleanQuery.Builder query = new BooleanQuery.Builder();
    query.add(clause("field", "hero"), Occur.SHOULD);
    query.add(clause("field", "of"), Occur.SHOULD);
    query.add(clause("field", "legend"), Occur.SHOULD);
    FieldQuery fieldQuery = highlighter.getFieldQuery(query.build(), reader);
    for (FragListBuilder fragListBuilder : new FragListBuilder[] { new SimpleFragListBuilder(), new WeightedFragListBuilder() }) {
        String[] bestFragments = highlighter.getBestFragments(fieldQuery, reader, docId, "field", 20, 1, fragListBuilder, fragmentsBuilder, preTags, postTags, encoder);
        assertEquals("<b>hero</b> <b>of</b> <b>legend</b>", bestFragments[0]);
        bestFragments = highlighter.getBestFragments(fieldQuery, reader, docId, "field", 28, 1, fragListBuilder, fragmentsBuilder, preTags, postTags, encoder);
        assertEquals("<b>hero</b> <b>of</b> <b>legend</b>", bestFragments[0]);
        bestFragments = highlighter.getBestFragments(fieldQuery, reader, docId, "field", 30000, 1, fragListBuilder, fragmentsBuilder, preTags, postTags, encoder);
        assertEquals("<b>hero</b> <b>of</b> <b>legend</b>", bestFragments[0]);
    }
    reader.close();
    writer.close();
    dir.close();
}