Search in sources :

Example 1 with BINARY

use of org.apache.parquet.schema.PrimitiveType.PrimitiveTypeName.BINARY in project presto by prestodb.

the class TupleDomainParquetPredicate method getDomain.

/**
 * Get a domain for the ranges defined by each pair of elements from {@code minimums} and {@code maximums}.
 * Both arrays must have the same length.
 */
private static Domain getDomain(ColumnDescriptor column, Type type, List<Object> minimums, List<Object> maximums, boolean hasNullValue) {
    checkArgument(minimums.size() == maximums.size(), "Expected minimums and maximums to have the same size");
    List<Range> ranges = new ArrayList<>();
    if (type.equals(BOOLEAN)) {
        boolean hasTrueValues = minimums.stream().anyMatch(value -> (boolean) value) || maximums.stream().anyMatch(value -> (boolean) value);
        boolean hasFalseValues = minimums.stream().anyMatch(value -> !(boolean) value) || maximums.stream().anyMatch(value -> !(boolean) value);
        if (hasTrueValues && hasFalseValues) {
            return Domain.all(type);
        }
        if (hasTrueValues) {
            return Domain.create(ValueSet.of(type, true), hasNullValue);
        }
        if (hasFalseValues) {
            return Domain.create(ValueSet.of(type, false), hasNullValue);
        }
        // All nulls case is handled earlier
        throw new VerifyException("Impossible boolean statistics");
    }
    if ((type.equals(BIGINT) || type.equals(TINYINT) || type.equals(SMALLINT) || type.equals(INTEGER))) {
        for (int i = 0; i < minimums.size(); i++) {
            long min = asLong(minimums.get(i));
            long max = asLong(maximums.get(i));
            if (isStatisticsOverflow(type, min, max)) {
                return Domain.create(ValueSet.all(type), hasNullValue);
            }
            ranges.add(Range.range(type, min, true, max, true));
        }
        checkArgument(!ranges.isEmpty(), "cannot use empty ranges");
        return Domain.create(ValueSet.ofRanges(ranges), hasNullValue);
    }
    if (type.equals(REAL)) {
        for (int i = 0; i < minimums.size(); i++) {
            Float min = (Float) minimums.get(i);
            Float max = (Float) maximums.get(i);
            if (min.isNaN() || max.isNaN()) {
                return Domain.create(ValueSet.all(type), hasNullValue);
            }
            ranges.add(Range.range(type, (long) floatToRawIntBits(min), true, (long) floatToRawIntBits(max), true));
        }
        checkArgument(!ranges.isEmpty(), "cannot use empty ranges");
        return Domain.create(ValueSet.ofRanges(ranges), hasNullValue);
    }
    if (type.equals(DOUBLE)) {
        for (int i = 0; i < minimums.size(); i++) {
            Double min = (Double) minimums.get(i);
            Double max = (Double) maximums.get(i);
            if (min.isNaN() || max.isNaN()) {
                return Domain.create(ValueSet.all(type), hasNullValue);
            }
            ranges.add(Range.range(type, min, true, max, true));
        }
        checkArgument(!ranges.isEmpty(), "cannot use empty ranges");
        return Domain.create(ValueSet.ofRanges(ranges), hasNullValue);
    }
    if (isVarcharType(type)) {
        for (int i = 0; i < minimums.size(); i++) {
            Slice min = Slices.wrappedBuffer(((Binary) minimums.get(i)).toByteBuffer());
            Slice max = Slices.wrappedBuffer(((Binary) maximums.get(i)).toByteBuffer());
            ranges.add(Range.range(type, min, true, max, true));
        }
        checkArgument(!ranges.isEmpty(), "cannot use empty ranges");
        return Domain.create(ValueSet.ofRanges(ranges), hasNullValue);
    }
    if (type.equals(DATE)) {
        for (int i = 0; i < minimums.size(); i++) {
            long min = asLong(minimums.get(i));
            long max = asLong(maximums.get(i));
            if (isStatisticsOverflow(type, min, max)) {
                return Domain.create(ValueSet.all(type), hasNullValue);
            }
            ranges.add(Range.range(type, min, true, max, true));
        }
        checkArgument(!ranges.isEmpty(), "cannot use empty ranges");
        return Domain.create(ValueSet.ofRanges(ranges), hasNullValue);
    }
    return Domain.create(ValueSet.all(type), hasNullValue);
}
Also used : PrimitiveType(org.apache.parquet.schema.PrimitiveType) RichColumnDescriptor(com.facebook.presto.parquet.RichColumnDescriptor) PredicateUtils.isStatisticsOverflow(com.facebook.presto.parquet.predicate.PredicateUtils.isStatisticsOverflow) FilterPredicate(org.apache.parquet.filter2.predicate.FilterPredicate) ColumnIndex(org.apache.parquet.internal.column.columnindex.ColumnIndex) FilterApi(org.apache.parquet.filter2.predicate.FilterApi) ByteBuffer(java.nio.ByteBuffer) ParquetCorruptionException(com.facebook.presto.parquet.ParquetCorruptionException) Preconditions.checkArgument(com.google.common.base.Preconditions.checkArgument) Slices(io.airlift.slice.Slices) Map(java.util.Map) Varchars.isVarcharType(com.facebook.presto.common.type.Varchars.isVarcharType) UserDefinedPredicate(org.apache.parquet.filter2.predicate.UserDefinedPredicate) DOUBLE(com.facebook.presto.common.type.DoubleType.DOUBLE) BINARY(org.apache.parquet.schema.PrimitiveType.PrimitiveTypeName.BINARY) ColumnIndexStore(org.apache.parquet.internal.filter2.columnindex.ColumnIndexStore) String.format(java.lang.String.format) Range(com.facebook.presto.common.predicate.Range) Binary(org.apache.parquet.io.api.Binary) Serializable(java.io.Serializable) INT64(org.apache.parquet.schema.PrimitiveType.PrimitiveTypeName.INT64) List(java.util.List) LITTLE_ENDIAN(java.nio.ByteOrder.LITTLE_ENDIAN) ColumnDescriptor(org.apache.parquet.column.ColumnDescriptor) INTEGER(com.facebook.presto.common.type.IntegerType.INTEGER) Optional(java.util.Optional) DictionaryPage(com.facebook.presto.parquet.DictionaryPage) ColumnPath(org.apache.parquet.hadoop.metadata.ColumnPath) Slice(io.airlift.slice.Slice) TINYINT(com.facebook.presto.common.type.TinyintType.TINYINT) HashMap(java.util.HashMap) INT32(org.apache.parquet.schema.PrimitiveType.PrimitiveTypeName.INT32) Function(java.util.function.Function) DATE(com.facebook.presto.common.type.DateType.DATE) REAL(com.facebook.presto.common.type.RealType.REAL) ArrayList(java.util.ArrayList) Float.floatToRawIntBits(java.lang.Float.floatToRawIntBits) ImmutableList(com.google.common.collect.ImmutableList) Objects.requireNonNull(java.util.Objects.requireNonNull) ParquetDataSourceId(com.facebook.presto.parquet.ParquetDataSourceId) PrimitiveTypeName(org.apache.parquet.schema.PrimitiveType.PrimitiveTypeName) BOOLEAN(com.facebook.presto.common.type.BooleanType.BOOLEAN) Type(com.facebook.presto.common.type.Type) VerifyException(com.google.common.base.VerifyException) Statistics(org.apache.parquet.column.statistics.Statistics) BIGINT(com.facebook.presto.common.type.BigintType.BIGINT) Domain(com.facebook.presto.common.predicate.Domain) TupleDomain(com.facebook.presto.common.predicate.TupleDomain) Dictionary(com.facebook.presto.parquet.dictionary.Dictionary) FLOAT(org.apache.parquet.schema.PrimitiveType.PrimitiveTypeName.FLOAT) SMALLINT(com.facebook.presto.common.type.SmallintType.SMALLINT) VisibleForTesting(com.google.common.annotations.VisibleForTesting) ValueSet(com.facebook.presto.common.predicate.ValueSet) VerifyException(com.google.common.base.VerifyException) Slice(io.airlift.slice.Slice) ArrayList(java.util.ArrayList) Range(com.facebook.presto.common.predicate.Range)

Aggregations

Domain (com.facebook.presto.common.predicate.Domain)1 Range (com.facebook.presto.common.predicate.Range)1 TupleDomain (com.facebook.presto.common.predicate.TupleDomain)1 ValueSet (com.facebook.presto.common.predicate.ValueSet)1 BIGINT (com.facebook.presto.common.type.BigintType.BIGINT)1 BOOLEAN (com.facebook.presto.common.type.BooleanType.BOOLEAN)1 DATE (com.facebook.presto.common.type.DateType.DATE)1 DOUBLE (com.facebook.presto.common.type.DoubleType.DOUBLE)1 INTEGER (com.facebook.presto.common.type.IntegerType.INTEGER)1 REAL (com.facebook.presto.common.type.RealType.REAL)1 SMALLINT (com.facebook.presto.common.type.SmallintType.SMALLINT)1 TINYINT (com.facebook.presto.common.type.TinyintType.TINYINT)1 Type (com.facebook.presto.common.type.Type)1 Varchars.isVarcharType (com.facebook.presto.common.type.Varchars.isVarcharType)1 DictionaryPage (com.facebook.presto.parquet.DictionaryPage)1 ParquetCorruptionException (com.facebook.presto.parquet.ParquetCorruptionException)1 ParquetDataSourceId (com.facebook.presto.parquet.ParquetDataSourceId)1 RichColumnDescriptor (com.facebook.presto.parquet.RichColumnDescriptor)1 Dictionary (com.facebook.presto.parquet.dictionary.Dictionary)1 PredicateUtils.isStatisticsOverflow (com.facebook.presto.parquet.predicate.PredicateUtils.isStatisticsOverflow)1