Search in sources :

Example 61 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class AbstractStringFieldDataTestCase method testActualMissingValue.

public void testActualMissingValue(boolean reverse) throws IOException {
    // missing value is set to an actual value
    final String[] values = new String[randomIntBetween(2, 30)];
    for (int i = 1; i < values.length; ++i) {
        values[i] = TestUtil.randomUnicodeString(random());
    }
    final int numDocs = scaledRandomIntBetween(10, 3072);
    for (int i = 0; i < numDocs; ++i) {
        final String value = RandomPicks.randomFrom(random(), values);
        if (value == null) {
            writer.addDocument(new Document());
        } else {
            Document d = new Document();
            addField(d, "value", value);
            writer.addDocument(d);
        }
        if (randomInt(10) == 0) {
            writer.commit();
        }
    }
    final IndexFieldData indexFieldData = getForField("value");
    final String missingValue = values[1];
    IndexSearcher searcher = new IndexSearcher(DirectoryReader.open(writer));
    XFieldComparatorSource comparator = indexFieldData.comparatorSource(missingValue, MultiValueMode.MIN, null);
    TopFieldDocs topDocs = searcher.search(new MatchAllDocsQuery(), randomBoolean() ? numDocs : randomIntBetween(10, numDocs), new Sort(new SortField("value", comparator, reverse)));
    assertEquals(numDocs, topDocs.totalHits);
    BytesRef previousValue = reverse ? UnicodeUtil.BIG_TERM : new BytesRef();
    for (int i = 0; i < topDocs.scoreDocs.length; ++i) {
        final String docValue = searcher.doc(topDocs.scoreDocs[i].doc).get("value");
        final BytesRef value = new BytesRef(docValue == null ? missingValue : docValue);
        if (reverse) {
            assertTrue(previousValue.compareTo(value) >= 0);
        } else {
            assertTrue(previousValue.compareTo(value) <= 0);
        }
        previousValue = value;
    }
    searcher.getIndexReader().close();
}
Also used : IndexSearcher(org.apache.lucene.search.IndexSearcher) GlobalOrdinalsIndexFieldData(org.elasticsearch.index.fielddata.ordinals.GlobalOrdinalsIndexFieldData) TopFieldDocs(org.apache.lucene.search.TopFieldDocs) Sort(org.apache.lucene.search.Sort) SortField(org.apache.lucene.search.SortField) XFieldComparatorSource(org.elasticsearch.index.fielddata.IndexFieldData.XFieldComparatorSource) Document(org.apache.lucene.document.Document) MatchAllDocsQuery(org.apache.lucene.search.MatchAllDocsQuery) BytesRef(org.apache.lucene.util.BytesRef)

Example 62 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class CodecTests method assertCompressionEquals.

// write some docs with it, inspect .si to see this was the used compression
private void assertCompressionEquals(Mode expected, Codec actual) throws Exception {
    Directory dir = newDirectory();
    IndexWriterConfig iwc = newIndexWriterConfig(null);
    iwc.setCodec(actual);
    IndexWriter iw = new IndexWriter(dir, iwc);
    iw.addDocument(new Document());
    iw.commit();
    iw.close();
    DirectoryReader ir = DirectoryReader.open(dir);
    SegmentReader sr = (SegmentReader) ir.leaves().get(0).reader();
    String v = sr.getSegmentInfo().info.getAttribute(Lucene50StoredFieldsFormat.MODE_KEY);
    assertNotNull(v);
    assertEquals(expected, Mode.valueOf(v));
    ir.close();
    dir.close();
}
Also used : IndexWriter(org.apache.lucene.index.IndexWriter) DirectoryReader(org.apache.lucene.index.DirectoryReader) Document(org.apache.lucene.document.Document) SegmentReader(org.apache.lucene.index.SegmentReader) Directory(org.apache.lucene.store.Directory) IndexWriterConfig(org.apache.lucene.index.IndexWriterConfig)

Example 63 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class HalfFloatFielddataTests method testSingleValued.

public void testSingleValued() throws IOException {
    Directory dir = newDirectory();
    // we need the default codec to check for singletons
    IndexWriter w = new IndexWriter(dir, new IndexWriterConfig(null).setCodec(TestUtil.getDefaultCodec()));
    Document doc = new Document();
    for (IndexableField f : NumberFieldMapper.NumberType.HALF_FLOAT.createFields("half_float", 3f, false, true, false)) {
        doc.add(f);
    }
    w.addDocument(doc);
    final DirectoryReader dirReader = DirectoryReader.open(w);
    LeafReader reader = getOnlyLeafReader(dirReader);
    SortedNumericDoubleValues values = new SortedNumericDVIndexFieldData.SortedNumericHalfFloatFieldData(reader, "half_float").getDoubleValues();
    assertNotNull(FieldData.unwrapSingleton(values));
    values.setDocument(0);
    assertEquals(1, values.count());
    assertEquals(3f, values.valueAt(0), 0f);
    IOUtils.close(dirReader, w, dir);
}
Also used : IndexableField(org.apache.lucene.index.IndexableField) LeafReader(org.apache.lucene.index.LeafReader) IndexWriter(org.apache.lucene.index.IndexWriter) DirectoryReader(org.apache.lucene.index.DirectoryReader) Document(org.apache.lucene.document.Document) SortedNumericDoubleValues(org.elasticsearch.index.fielddata.SortedNumericDoubleValues) Directory(org.apache.lucene.store.Directory) IndexWriterConfig(org.apache.lucene.index.IndexWriterConfig)

Example 64 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class ScaledFloatFieldTypeTests method testStats.

public void testStats() throws IOException {
    ScaledFloatFieldMapper.ScaledFloatFieldType ft = new ScaledFloatFieldMapper.ScaledFloatFieldType();
    ft.setName("scaled_float");
    ft.setScalingFactor(0.1 + randomDouble() * 100);
    Directory dir = newDirectory();
    IndexWriter w = new IndexWriter(dir, new IndexWriterConfig(null));
    try (DirectoryReader reader = DirectoryReader.open(w)) {
        assertNull(ft.stats(reader));
    }
    Document doc = new Document();
    doc.add(new StoredField("scaled_float", -1));
    w.addDocument(doc);
    try (DirectoryReader reader = DirectoryReader.open(w)) {
        // field exists, but has no point values
        FieldStats<?> stats = ft.stats(reader);
        assertFalse(stats.hasMinMax());
        assertNull(stats.getMinValue());
        assertNull(stats.getMaxValue());
    }
    LongPoint point = new LongPoint("scaled_float", -1);
    doc.add(point);
    w.addDocument(doc);
    point.setLongValue(10);
    w.addDocument(doc);
    try (DirectoryReader reader = DirectoryReader.open(w)) {
        FieldStats<?> stats = ft.stats(reader);
        assertEquals(-1 / ft.getScalingFactor(), stats.getMinValue());
        assertEquals(10 / ft.getScalingFactor(), stats.getMaxValue());
        assertEquals(3, stats.getMaxDoc());
    }
    w.deleteAll();
    try (DirectoryReader reader = DirectoryReader.open(w)) {
        assertNull(ft.stats(reader));
    }
    IOUtils.close(w, dir);
}
Also used : StoredField(org.apache.lucene.document.StoredField) IndexWriter(org.apache.lucene.index.IndexWriter) DirectoryReader(org.apache.lucene.index.DirectoryReader) LongPoint(org.apache.lucene.document.LongPoint) Document(org.apache.lucene.document.Document) Directory(org.apache.lucene.store.Directory) IndexWriterConfig(org.apache.lucene.index.IndexWriterConfig)

Example 65 with Document

use of org.apache.lucene.document.Document in project elasticsearch by elastic.

the class ScaledFloatFieldTypeTests method testFieldData.

public void testFieldData() throws IOException {
    ScaledFloatFieldMapper.ScaledFloatFieldType ft = new ScaledFloatFieldMapper.ScaledFloatFieldType();
    ft.setScalingFactor(0.1 + randomDouble() * 100);
    Directory dir = newDirectory();
    IndexWriter w = new IndexWriter(dir, new IndexWriterConfig(null));
    Document doc = new Document();
    doc.add(new SortedNumericDocValuesField("scaled_float1", 10));
    doc.add(new SortedNumericDocValuesField("scaled_float2", 5));
    doc.add(new SortedNumericDocValuesField("scaled_float2", 12));
    w.addDocument(doc);
    try (DirectoryReader reader = DirectoryReader.open(w)) {
        IndexMetaData indexMetadata = new IndexMetaData.Builder("index").settings(Settings.builder().put("index.version.created", Version.CURRENT).put("index.number_of_shards", 1).put("index.number_of_replicas", 0).build()).build();
        IndexSettings indexSettings = new IndexSettings(indexMetadata, Settings.EMPTY);
        // single-valued
        ft.setName("scaled_float1");
        IndexNumericFieldData fielddata = (IndexNumericFieldData) ft.fielddataBuilder().build(indexSettings, ft, null, null, null);
        assertEquals(fielddata.getNumericType(), IndexNumericFieldData.NumericType.DOUBLE);
        AtomicNumericFieldData leafFieldData = fielddata.load(reader.leaves().get(0));
        SortedNumericDoubleValues values = leafFieldData.getDoubleValues();
        values.setDocument(0);
        assertEquals(1, values.count());
        assertEquals(10 / ft.getScalingFactor(), values.valueAt(0), 10e-5);
        // multi-valued
        ft.setName("scaled_float2");
        fielddata = (IndexNumericFieldData) ft.fielddataBuilder().build(indexSettings, ft, null, null, null);
        leafFieldData = fielddata.load(reader.leaves().get(0));
        values = leafFieldData.getDoubleValues();
        values.setDocument(0);
        assertEquals(2, values.count());
        assertEquals(5 / ft.getScalingFactor(), values.valueAt(0), 10e-5);
        assertEquals(12 / ft.getScalingFactor(), values.valueAt(1), 10e-5);
    }
    IOUtils.close(w, dir);
}
Also used : DirectoryReader(org.apache.lucene.index.DirectoryReader) IndexSettings(org.elasticsearch.index.IndexSettings) IndexNumericFieldData(org.elasticsearch.index.fielddata.IndexNumericFieldData) Document(org.apache.lucene.document.Document) IndexMetaData(org.elasticsearch.cluster.metadata.IndexMetaData) SortedNumericDocValuesField(org.apache.lucene.document.SortedNumericDocValuesField) IndexWriter(org.apache.lucene.index.IndexWriter) SortedNumericDoubleValues(org.elasticsearch.index.fielddata.SortedNumericDoubleValues) AtomicNumericFieldData(org.elasticsearch.index.fielddata.AtomicNumericFieldData) Directory(org.apache.lucene.store.Directory) IndexWriterConfig(org.apache.lucene.index.IndexWriterConfig)

Aggregations

Document (org.apache.lucene.document.Document)2344 Directory (org.apache.lucene.store.Directory)1374 MockAnalyzer (org.apache.lucene.analysis.MockAnalyzer)798 RandomIndexWriter (org.apache.lucene.index.RandomIndexWriter)752 IndexReader (org.apache.lucene.index.IndexReader)598 Field (org.apache.lucene.document.Field)480 IndexSearcher (org.apache.lucene.search.IndexSearcher)470 Term (org.apache.lucene.index.Term)456 BytesRef (org.apache.lucene.util.BytesRef)415 StringField (org.apache.lucene.document.StringField)403 TextField (org.apache.lucene.document.TextField)389 NumericDocValuesField (org.apache.lucene.document.NumericDocValuesField)325 IndexWriter (org.apache.lucene.index.IndexWriter)312 IndexWriterConfig (org.apache.lucene.index.IndexWriterConfig)278 TopDocs (org.apache.lucene.search.TopDocs)270 TermQuery (org.apache.lucene.search.TermQuery)237 FieldType (org.apache.lucene.document.FieldType)231 DirectoryReader (org.apache.lucene.index.DirectoryReader)226 Test (org.junit.Test)222 RAMDirectory (org.apache.lucene.store.RAMDirectory)211