Examples with BytesRef org.apache.lucene.util.BytesRef used on opensource projects

Example 96 with BytesRef

use of org.apache.lucene.util.BytesRef in project elasticsearch by elastic.

the class MultiValueModeTests method verify.

private void verify(SortedBinaryDocValues values, int maxDoc) {
    for (BytesRef missingValue : new BytesRef[] { new BytesRef(), new BytesRef(RandomStrings.randomAsciiOfLength(random(), 8)) }) {
        for (MultiValueMode mode : new MultiValueMode[] { MultiValueMode.MIN, MultiValueMode.MAX }) {
            final BinaryDocValues selected = mode.select(values, missingValue);
            for (int i = 0; i < maxDoc; ++i) {
                final BytesRef actual = selected.get(i);
                BytesRef expected = null;
                values.setDocument(i);
                int numValues = values.count();
                if (numValues == 0) {
                    expected = missingValue;
                } else {
                    for (int j = 0; j < numValues; ++j) {
                        if (expected == null) {
                            expected = BytesRef.deepCopyOf(values.valueAt(j));
                        } else {
                            if (mode == MultiValueMode.MIN) {
                                expected = expected.compareTo(values.valueAt(j)) <= 0 ? expected : BytesRef.deepCopyOf(values.valueAt(j));
                            } else if (mode == MultiValueMode.MAX) {
                                expected = expected.compareTo(values.valueAt(j)) > 0 ? expected : BytesRef.deepCopyOf(values.valueAt(j));
                            }
                        }
                    }
                    if (expected == null) {
                        expected = missingValue;
                    }
                }
                assertEquals(mode.toString() + " docId=" + i, expected, actual);
            }
        }
    }
}

Example 97 with BytesRef

use of org.apache.lucene.util.BytesRef in project elasticsearch by elastic.

the class DocValueFormatTests method testRawParse.

public void testRawParse() {
    assertEquals(-1L, DocValueFormat.RAW.parseLong("-1", randomBoolean(), null));
    assertEquals(1L, DocValueFormat.RAW.parseLong("1", randomBoolean(), null));
    // not checking exception messages as they could depend on the JVM
    expectThrows(IllegalArgumentException.class, () -> DocValueFormat.RAW.parseLong("", randomBoolean(), null));
    expectThrows(IllegalArgumentException.class, () -> DocValueFormat.RAW.parseLong("abc", randomBoolean(), null));
    assertEquals(-1d, DocValueFormat.RAW.parseDouble("-1", randomBoolean(), null), 0d);
    assertEquals(1d, DocValueFormat.RAW.parseDouble("1", randomBoolean(), null), 0d);
    assertEquals(.5, DocValueFormat.RAW.parseDouble("0.5", randomBoolean(), null), 0d);
    // not checking exception messages as they could depend on the JVM
    expectThrows(IllegalArgumentException.class, () -> DocValueFormat.RAW.parseLong("", randomBoolean(), null));
    expectThrows(IllegalArgumentException.class, () -> DocValueFormat.RAW.parseLong("abc", randomBoolean(), null));
    assertEquals(new BytesRef("abc"), DocValueFormat.RAW.parseBytesRef("abc"));
}

Example 98 with BytesRef

use of org.apache.lucene.util.BytesRef in project elasticsearch by elastic.

the class MultiValueModeTests method testSingleValuedStrings.

public void testSingleValuedStrings() throws Exception {
    final int numDocs = scaledRandomIntBetween(1, 100);
    final BytesRef[] array = new BytesRef[numDocs];
    final FixedBitSet docsWithValue = randomBoolean() ? null : new FixedBitSet(numDocs);
    for (int i = 0; i < array.length; ++i) {
        if (randomBoolean()) {
            array[i] = new BytesRef(RandomStrings.randomAsciiOfLength(random(), 8));
            if (docsWithValue != null) {
                docsWithValue.set(i);
            }
        } else {
            array[i] = new BytesRef();
            if (docsWithValue != null && randomBoolean()) {
                docsWithValue.set(i);
            }
        }
    }
    final BinaryDocValues singleValues = new BinaryDocValues() {

        @Override
        public BytesRef get(int docID) {
            return BytesRef.deepCopyOf(array[docID]);
        }
    };
    final SortedBinaryDocValues multiValues = FieldData.singleton(singleValues, docsWithValue);
    verify(multiValues, numDocs);
    final FixedBitSet rootDocs = randomRootDocs(numDocs);
    final FixedBitSet innerDocs = randomInnerDocs(rootDocs);
    verify(multiValues, numDocs, rootDocs, innerDocs);
}

Example 99 with BytesRef

use of org.apache.lucene.util.BytesRef in project elasticsearch by elastic.

the class MultiValueModeTests method verify.

private void verify(SortedBinaryDocValues values, int maxDoc, FixedBitSet rootDocs, FixedBitSet innerDocs) throws IOException {
    for (BytesRef missingValue : new BytesRef[] { new BytesRef(), new BytesRef(RandomStrings.randomAsciiOfLength(random(), 8)) }) {
        for (MultiValueMode mode : new MultiValueMode[] { MultiValueMode.MIN, MultiValueMode.MAX }) {
            final BinaryDocValues selected = mode.select(values, missingValue, rootDocs, new BitSetIterator(innerDocs, 0L), maxDoc);
            int prevRoot = -1;
            for (int root = rootDocs.nextSetBit(0); root != -1; root = root + 1 < maxDoc ? rootDocs.nextSetBit(root + 1) : -1) {
                final BytesRef actual = selected.get(root);
                BytesRef expected = null;
                for (int child = innerDocs.nextSetBit(prevRoot + 1); child != -1 && child < root; child = innerDocs.nextSetBit(child + 1)) {
                    values.setDocument(child);
                    for (int j = 0; j < values.count(); ++j) {
                        if (expected == null) {
                            expected = BytesRef.deepCopyOf(values.valueAt(j));
                        } else {
                            if (mode == MultiValueMode.MIN) {
                                expected = expected.compareTo(values.valueAt(j)) <= 0 ? expected : BytesRef.deepCopyOf(values.valueAt(j));
                            } else if (mode == MultiValueMode.MAX) {
                                expected = expected.compareTo(values.valueAt(j)) > 0 ? expected : BytesRef.deepCopyOf(values.valueAt(j));
                            }
                        }
                    }
                }
                if (expected == null) {
                    expected = missingValue;
                }
                assertEquals(mode.toString() + " docId=" + root, expected, actual);
                prevRoot = root;
            }
        }
    }
}

Example 100 with BytesRef

use of org.apache.lucene.util.BytesRef in project elasticsearch by elastic.

the class SignificantTermsTests method createTestAggregatorBuilder.

@Override
protected SignificantTermsAggregationBuilder createTestAggregatorBuilder() {
    String name = randomAsciiOfLengthBetween(3, 20);
    SignificantTermsAggregationBuilder factory = new SignificantTermsAggregationBuilder(name, null);
    String field = randomAsciiOfLengthBetween(3, 20);
    int randomFieldBranch = randomInt(2);
    switch(randomFieldBranch) {
        case 0:
            factory.field(field);
            break;
        case 1:
            factory.field(field);
            factory.script(new Script("_value + 1"));
            break;
        case 2:
            factory.script(new Script("doc[" + field + "] + 1"));
            break;
        default:
            fail();
    }
    if (randomBoolean()) {
        factory.missing("MISSING");
    }
    if (randomBoolean()) {
        factory.bucketCountThresholds().setRequiredSize(randomIntBetween(1, Integer.MAX_VALUE));
    }
    if (randomBoolean()) {
        factory.bucketCountThresholds().setShardSize(randomIntBetween(1, Integer.MAX_VALUE));
    }
    if (randomBoolean()) {
        int minDocCount = randomInt(4);
        switch(minDocCount) {
            case 0:
                break;
            case 1:
            case 2:
            case 3:
            case 4:
                minDocCount = randomIntBetween(0, Integer.MAX_VALUE);
                break;
        }
        factory.bucketCountThresholds().setMinDocCount(minDocCount);
    }
    if (randomBoolean()) {
        int shardMinDocCount = randomInt(4);
        switch(shardMinDocCount) {
            case 0:
                break;
            case 1:
            case 2:
            case 3:
            case 4:
                shardMinDocCount = randomIntBetween(0, Integer.MAX_VALUE);
                break;
            default:
                fail();
        }
        factory.bucketCountThresholds().setShardMinDocCount(shardMinDocCount);
    }
    if (randomBoolean()) {
        factory.executionHint(randomFrom(executionHints));
    }
    if (randomBoolean()) {
        factory.format("###.##");
    }
    if (randomBoolean()) {
        IncludeExclude incExc = null;
        switch(randomInt(5)) {
            case 0:
                incExc = new IncludeExclude(new RegExp("foobar"), null);
                break;
            case 1:
                incExc = new IncludeExclude(null, new RegExp("foobaz"));
                break;
            case 2:
                incExc = new IncludeExclude(new RegExp("foobar"), new RegExp("foobaz"));
                break;
            case 3:
                SortedSet<BytesRef> includeValues = new TreeSet<>();
                int numIncs = randomIntBetween(1, 20);
                for (int i = 0; i < numIncs; i++) {
                    includeValues.add(new BytesRef(randomAsciiOfLengthBetween(1, 30)));
                }
                SortedSet<BytesRef> excludeValues = null;
                incExc = new IncludeExclude(includeValues, excludeValues);
                break;
            case 4:
                SortedSet<BytesRef> includeValues2 = null;
                SortedSet<BytesRef> excludeValues2 = new TreeSet<>();
                int numExcs2 = randomIntBetween(1, 20);
                for (int i = 0; i < numExcs2; i++) {
                    excludeValues2.add(new BytesRef(randomAsciiOfLengthBetween(1, 30)));
                }
                incExc = new IncludeExclude(includeValues2, excludeValues2);
                break;
            case 5:
                SortedSet<BytesRef> includeValues3 = new TreeSet<>();
                int numIncs3 = randomIntBetween(1, 20);
                for (int i = 0; i < numIncs3; i++) {
                    includeValues3.add(new BytesRef(randomAsciiOfLengthBetween(1, 30)));
                }
                SortedSet<BytesRef> excludeValues3 = new TreeSet<>();
                int numExcs3 = randomIntBetween(1, 20);
                for (int i = 0; i < numExcs3; i++) {
                    excludeValues3.add(new BytesRef(randomAsciiOfLengthBetween(1, 30)));
                }
                incExc = new IncludeExclude(includeValues3, excludeValues3);
                break;
            default:
                fail();
        }
        factory.includeExclude(incExc);
    }
    if (randomBoolean()) {
        SignificanceHeuristic significanceHeuristic = null;
        switch(randomInt(5)) {
            case 0:
                significanceHeuristic = new PercentageScore();
                break;
            case 1:
                significanceHeuristic = new ChiSquare(randomBoolean(), randomBoolean());
                break;
            case 2:
                significanceHeuristic = new GND(randomBoolean());
                break;
            case 3:
                significanceHeuristic = new MutualInformation(randomBoolean(), randomBoolean());
                break;
            case 4:
                significanceHeuristic = new ScriptHeuristic(new Script("foo"));
                break;
            case 5:
                significanceHeuristic = new JLHScore();
                break;
            default:
                fail();
        }
        factory.significanceHeuristic(significanceHeuristic);
    }
    if (randomBoolean()) {
        factory.backgroundFilter(QueryBuilders.termsQuery("foo", "bar"));
    }
    return factory;
}