Examples with BinaryPoint org.apache.lucene.document.BinaryPoint used on opensource projects

Example 6 with BinaryPoint

use of org.apache.lucene.document.BinaryPoint in project lucene-solr by apache.

the class TestBackwardsCompatibility method addDoc.

private void addDoc(IndexWriter writer, int id) throws IOException {
    Document doc = new Document();
    doc.add(new TextField("content", "aaa", Field.Store.NO));
    doc.add(new StringField("id", Integer.toString(id), Field.Store.YES));
    FieldType customType2 = new FieldType(TextField.TYPE_STORED);
    customType2.setStoreTermVectors(true);
    customType2.setStoreTermVectorPositions(true);
    customType2.setStoreTermVectorOffsets(true);
    doc.add(new Field("autf8", "Lu𝄞ce𝅘𝅥𝅮ne  ☠ ab񕰗cd", customType2));
    doc.add(new Field("utf8", "Lu𝄞ce𝅘𝅥𝅮ne  ☠ ab񕰗cd", customType2));
    doc.add(new Field("content2", "here is more content with aaa aaa aaa", customType2));
    doc.add(new Field("fieⱷld", "field with non-ascii name", customType2));
    // add docvalues fields
    doc.add(new NumericDocValuesField("dvByte", (byte) id));
    byte[] bytes = new byte[] { (byte) (id >>> 24), (byte) (id >>> 16), (byte) (id >>> 8), (byte) id };
    BytesRef ref = new BytesRef(bytes);
    doc.add(new BinaryDocValuesField("dvBytesDerefFixed", ref));
    doc.add(new BinaryDocValuesField("dvBytesDerefVar", ref));
    doc.add(new SortedDocValuesField("dvBytesSortedFixed", ref));
    doc.add(new SortedDocValuesField("dvBytesSortedVar", ref));
    doc.add(new BinaryDocValuesField("dvBytesStraightFixed", ref));
    doc.add(new BinaryDocValuesField("dvBytesStraightVar", ref));
    doc.add(new DoubleDocValuesField("dvDouble", (double) id));
    doc.add(new FloatDocValuesField("dvFloat", (float) id));
    doc.add(new NumericDocValuesField("dvInt", id));
    doc.add(new NumericDocValuesField("dvLong", id));
    doc.add(new NumericDocValuesField("dvPacked", id));
    doc.add(new NumericDocValuesField("dvShort", (short) id));
    doc.add(new SortedSetDocValuesField("dvSortedSet", ref));
    doc.add(new SortedNumericDocValuesField("dvSortedNumeric", id));
    doc.add(new IntPoint("intPoint1d", id));
    doc.add(new IntPoint("intPoint2d", id, 2 * id));
    doc.add(new FloatPoint("floatPoint1d", (float) id));
    doc.add(new FloatPoint("floatPoint2d", (float) id, (float) 2 * id));
    doc.add(new LongPoint("longPoint1d", id));
    doc.add(new LongPoint("longPoint2d", id, 2 * id));
    doc.add(new DoublePoint("doublePoint1d", (double) id));
    doc.add(new DoublePoint("doublePoint2d", (double) id, (double) 2 * id));
    doc.add(new BinaryPoint("binaryPoint1d", bytes));
    doc.add(new BinaryPoint("binaryPoint2d", bytes, bytes));
    // a field with both offsets and term vectors for a cross-check
    FieldType customType3 = new FieldType(TextField.TYPE_STORED);
    customType3.setStoreTermVectors(true);
    customType3.setStoreTermVectorPositions(true);
    customType3.setStoreTermVectorOffsets(true);
    customType3.setIndexOptions(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS);
    doc.add(new Field("content5", "here is more content with aaa aaa aaa", customType3));
    // a field that omits only positions
    FieldType customType4 = new FieldType(TextField.TYPE_STORED);
    customType4.setStoreTermVectors(true);
    customType4.setStoreTermVectorPositions(false);
    customType4.setStoreTermVectorOffsets(true);
    customType4.setIndexOptions(IndexOptions.DOCS_AND_FREQS);
    doc.add(new Field("content6", "here is more content with aaa aaa aaa", customType4));
    // TODO: 
    //   index different norms types via similarity (we use a random one currently?!)
    //   remove any analyzer randomness, explicitly add payloads for certain fields.
    writer.addDocument(doc);
}

Example 7 with BinaryPoint

use of org.apache.lucene.document.BinaryPoint in project lucene-solr by apache.

the class TestLucene60PointsFormat method testEstimatePointCount.

public void testEstimatePointCount() throws IOException {
    Directory dir = newDirectory();
    IndexWriterConfig iwc = newIndexWriterConfig();
    // number of points per leaf hard to predict
    while (iwc.getMergePolicy() instanceof MockRandomMergePolicy) {
        iwc.setMergePolicy(newMergePolicy());
    }
    IndexWriter w = new IndexWriter(dir, iwc);
    byte[] pointValue = new byte[3];
    byte[] uniquePointValue = new byte[3];
    random().nextBytes(uniquePointValue);
    // make sure we have several leaves
    final int numDocs = atLeast(10000);
    for (int i = 0; i < numDocs; ++i) {
        Document doc = new Document();
        if (i == numDocs / 2) {
            doc.add(new BinaryPoint("f", uniquePointValue));
        } else {
            do {
                random().nextBytes(pointValue);
            } while (Arrays.equals(pointValue, uniquePointValue));
            doc.add(new BinaryPoint("f", pointValue));
        }
        w.addDocument(doc);
    }
    w.forceMerge(1);
    final IndexReader r = DirectoryReader.open(w);
    w.close();
    final LeafReader lr = getOnlyLeafReader(r);
    PointValues points = lr.getPointValues("f");
    // If all points match, then the point count is numLeaves * maxPointsInLeafNode
    final int numLeaves = (int) Math.ceil((double) numDocs / maxPointsInLeafNode);
    assertEquals(numLeaves * maxPointsInLeafNode, points.estimatePointCount(new IntersectVisitor() {

        @Override
        public void visit(int docID, byte[] packedValue) throws IOException {
        }

        @Override
        public void visit(int docID) throws IOException {
        }

        @Override
        public Relation compare(byte[] minPackedValue, byte[] maxPackedValue) {
            return Relation.CELL_INSIDE_QUERY;
        }
    }));
    // Return 0 if no points match
    assertEquals(0, points.estimatePointCount(new IntersectVisitor() {

        @Override
        public void visit(int docID, byte[] packedValue) throws IOException {
        }

        @Override
        public void visit(int docID) throws IOException {
        }

        @Override
        public Relation compare(byte[] minPackedValue, byte[] maxPackedValue) {
            return Relation.CELL_OUTSIDE_QUERY;
        }
    }));
    // If only one point matches, then the point count is (maxPointsInLeafNode + 1) / 2
    // in general, or maybe 2x that if the point is a split value
    final long pointCount = points.estimatePointCount(new IntersectVisitor() {

        @Override
        public void visit(int docID, byte[] packedValue) throws IOException {
        }

        @Override
        public void visit(int docID) throws IOException {
        }

        @Override
        public Relation compare(byte[] minPackedValue, byte[] maxPackedValue) {
            if (StringHelper.compare(3, uniquePointValue, 0, maxPackedValue, 0) > 0 || StringHelper.compare(3, uniquePointValue, 0, minPackedValue, 0) < 0) {
                return Relation.CELL_OUTSIDE_QUERY;
            }
            return Relation.CELL_CROSSES_QUERY;
        }
    });
    assertTrue("" + pointCount, // common case
    pointCount == (maxPointsInLeafNode + 1) / 2 || // if the point is a split value
    pointCount == 2 * ((maxPointsInLeafNode + 1) / 2));
    r.close();
    dir.close();
}

Example 8 with BinaryPoint

use of org.apache.lucene.document.BinaryPoint in project lucene-solr by apache.

the class TestPointValues method testIllegalDimChangeViaAddIndexesSlowCodecReader.

public void testIllegalDimChangeViaAddIndexesSlowCodecReader() throws Exception {
    Directory dir = newDirectory();
    IndexWriterConfig iwc = new IndexWriterConfig(new MockAnalyzer(random()));
    IndexWriter w = new IndexWriter(dir, iwc);
    Document doc = new Document();
    doc.add(new BinaryPoint("dim", new byte[4]));
    w.addDocument(doc);
    w.close();
    Directory dir2 = newDirectory();
    iwc = new IndexWriterConfig(new MockAnalyzer(random()));
    IndexWriter w2 = new IndexWriter(dir2, iwc);
    doc = new Document();
    doc.add(new BinaryPoint("dim", new byte[4], new byte[4]));
    w2.addDocument(doc);
    DirectoryReader r = DirectoryReader.open(dir);
    IllegalArgumentException expected = expectThrows(IllegalArgumentException.class, () -> {
        TestUtil.addIndexesSlowly(w2, r);
    });
    assertEquals("cannot change point dimension count from 2 to 1 for field=\"dim\"", expected.getMessage());
    IOUtils.close(r, w2, dir, dir2);
}

Example 9 with BinaryPoint

use of org.apache.lucene.document.BinaryPoint in project lucene-solr by apache.

the class TestPointValues method doTestMergedStats.

private void doTestMergedStats() throws IOException {
    final int numDims = TestUtil.nextInt(random(), 1, 8);
    final int numBytesPerDim = TestUtil.nextInt(random(), 1, 16);
    Directory dir = new RAMDirectory();
    IndexWriter w = new IndexWriter(dir, new IndexWriterConfig(null));
    final int numDocs = TestUtil.nextInt(random(), 10, 20);
    for (int i = 0; i < numDocs; ++i) {
        Document doc = new Document();
        final int numPoints = random().nextInt(3);
        for (int j = 0; j < numPoints; ++j) {
            doc.add(new BinaryPoint("field", randomBinaryValue(numDims, numBytesPerDim)));
        }
        w.addDocument(doc);
        if (random().nextBoolean()) {
            DirectoryReader.open(w).close();
        }
    }
    final IndexReader reader1 = DirectoryReader.open(w);
    w.forceMerge(1);
    final IndexReader reader2 = DirectoryReader.open(w);
    final PointValues expected = getOnlyLeafReader(reader2).getPointValues("field");
    if (expected == null) {
        assertNull(PointValues.getMinPackedValue(reader1, "field"));
        assertNull(PointValues.getMaxPackedValue(reader1, "field"));
        assertEquals(0, PointValues.getDocCount(reader1, "field"));
        assertEquals(0, PointValues.size(reader1, "field"));
    } else {
        assertArrayEquals(expected.getMinPackedValue(), PointValues.getMinPackedValue(reader1, "field"));
        assertArrayEquals(expected.getMaxPackedValue(), PointValues.getMaxPackedValue(reader1, "field"));
        assertEquals(expected.getDocCount(), PointValues.getDocCount(reader1, "field"));
        assertEquals(expected.size(), PointValues.size(reader1, "field"));
    }
    IOUtils.close(w, reader1, reader2, dir);
}

Example 10 with BinaryPoint

use of org.apache.lucene.document.BinaryPoint in project lucene-solr by apache.

the class TestPointValues method testIllegalDimChangeTwoDocs.

public void testIllegalDimChangeTwoDocs() throws Exception {
    Directory dir = newDirectory();
    IndexWriterConfig iwc = new IndexWriterConfig(new MockAnalyzer(random()));
    IndexWriter w = new IndexWriter(dir, iwc);
    Document doc = new Document();
    doc.add(new BinaryPoint("dim", new byte[4]));
    w.addDocument(doc);
    Document doc2 = new Document();
    doc2.add(new BinaryPoint("dim", new byte[4], new byte[4]));
    IllegalArgumentException expected = expectThrows(IllegalArgumentException.class, () -> {
        w.addDocument(doc2);
    });
    assertEquals("cannot change point dimension count from 1 to 2 for field=\"dim\"", expected.getMessage());
    w.close();
    dir.close();
}