org.apache.lucene.util
Class NumericUtils

java.lang.Object
  org.apache.lucene.util.NumericUtils

public final class NumericUtils
extends Object

This is a helper class to generate prefix-encoded representations for numerical values and supplies converters to represent float/double values as sortable integers/longs.

To quickly execute range queries in Apache Lucene, a range is divided recursively into multiple intervals for searching: The center of the range is searched only with the lowest possible precision in the trie, while the boundaries are matched more exactly. This reduces the number of terms dramatically.

This class generates terms to achieve this: First the numerical integer values need to be converted to strings. For that integer values (32 bit or 64 bit) are made unsigned and the bits are converted to ASCII chars with each 7 bit. The resulting string is sortable like the original integer value. Each value is also prefixed (in the first char) by the shift value (number of bits removed) used during encoding.

To also index floating point numbers, this class supplies two methods to convert them to integer values by changing their bit layout: doubleToSortableLong(double), floatToSortableInt(float). You will have no precision loss by converting floating point numbers to integers and back (only that the integer form is not usable). Other data types like dates can easily converted to longs or ints (e.g. date to long: Date.getTime()).

For easy usage, the trie algorithm is implemented for indexing inside NumericTokenStream that can index int, long, float, and double. For querying, NumericRangeQuery and NumericRangeFilter implement the query part for the same data types.

This class can also be used, to generate lexicographically sortable (according String.compareTo(String)) representations of numeric data types for other usages (e.g. sorting).

Since:

2.9

NOTE: This API is for internal purposes only and might change in incompatible ways in the next release.

Nested Class Summary

static class	NumericUtils.IntRangeBuilder Expert: Callback for splitIntRange(org.apache.lucene.util.NumericUtils.IntRangeBuilder, int, int, int).
static class	NumericUtils.LongRangeBuilder Expert: Callback for splitLongRange(org.apache.lucene.util.NumericUtils.LongRangeBuilder, int, long, long).

Field Summary

static int	BUF_SIZE_INT Expert: The maximum term length (used for char[] buffer size) for encoding int values.
static int	BUF_SIZE_LONG Expert: The maximum term length (used for char[] buffer size) for encoding long values.
static int	PRECISION_STEP_DEFAULT The default precision step used by NumericField, NumericTokenStream, NumericRangeQuery, and NumericRangeFilter as default
static char	SHIFT_START_INT Expert: Integers are stored at lower precision by shifting off lower bits.
static char	SHIFT_START_LONG Expert: Longs are stored at lower precision by shifting off lower bits.

Method Summary

static String	doubleToPrefixCoded(double val)
static long	doubleToSortableLong(double val) Converts a double value to a sortable signed long.
static String	floatToPrefixCoded(float val)
static int	floatToSortableInt(float val) Converts a float value to a sortable signed int.
static String	intToPrefixCoded(int val)
static String	intToPrefixCoded(int val, int shift)
static int	intToPrefixCoded(int val, int shift, char[] buffer) Expert: Returns prefix coded bits after reducing the precision by shift bits.
static String	longToPrefixCoded(long val)
static String	longToPrefixCoded(long val, int shift)
static int	longToPrefixCoded(long val, int shift, char[] buffer) Expert: Returns prefix coded bits after reducing the precision by shift bits.
static double	prefixCodedToDouble(String val)
static float	prefixCodedToFloat(String val)
static int	prefixCodedToInt(String prefixCoded)
static long	prefixCodedToLong(String prefixCoded)
static float	sortableIntToFloat(int val) Converts a sortable int back to a float.
static double	sortableLongToDouble(long val) Converts a sortable long back to a double.
static void	splitIntRange(NumericUtils.IntRangeBuilder builder, int precisionStep, int minBound, int maxBound) Expert: Splits an int range recursively.
static void	splitLongRange(NumericUtils.LongRangeBuilder builder, int precisionStep, long minBound, long maxBound) Expert: Splits a long range recursively.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

PRECISION_STEP_DEFAULT

public static final int PRECISION_STEP_DEFAULT

The default precision step used by NumericField, NumericTokenStream, NumericRangeQuery, and NumericRangeFilter as default

See Also:

Constant Field Values

SHIFT_START_LONG

public static final char SHIFT_START_LONG

Expert: Longs are stored at lower precision by shifting off lower bits. The shift count is stored as SHIFT_START_LONG+shift in the first character

See Also:

Constant Field Values

BUF_SIZE_LONG

public static final int BUF_SIZE_LONG

Expert: The maximum term length (used for char[] buffer size) for encoding long values.

See Also:

longToPrefixCoded(long,int,char[]), Constant Field Values

SHIFT_START_INT

public static final char SHIFT_START_INT

Expert: Integers are stored at lower precision by shifting off lower bits. The shift count is stored as SHIFT_START_INT+shift in the first character

See Also:

Constant Field Values

BUF_SIZE_INT

public static final int BUF_SIZE_INT

Expert: The maximum term length (used for char[] buffer size) for encoding int values.

See Also:

intToPrefixCoded(int,int,char[]), Constant Field Values

Method Detail

longToPrefixCoded

public static int longToPrefixCoded(long val,
                                    int shift,
                                    char[] buffer)

Expert: Returns prefix coded bits after reducing the precision by shift bits. This is method is used by NumericTokenStream.

Parameters:

val - the numeric value

shift - how many bits to strip from the right

buffer - that will contain the encoded chars, must be at least of BUF_SIZE_LONG length

Returns:

number of chars written to buffer

longToPrefixCoded

public static String longToPrefixCoded(long val,
                                       int shift)

longToPrefixCoded

public static String longToPrefixCoded(long val)

intToPrefixCoded

public static int intToPrefixCoded(int val,
                                   int shift,
                                   char[] buffer)

Expert: Returns prefix coded bits after reducing the precision by shift bits. This is method is used by NumericTokenStream.

Parameters:

val - the numeric value

shift - how many bits to strip from the right

buffer - that will contain the encoded chars, must be at least of BUF_SIZE_INT length

Returns:

number of chars written to buffer

intToPrefixCoded

public static String intToPrefixCoded(int val,
                                      int shift)

intToPrefixCoded

public static String intToPrefixCoded(int val)

prefixCodedToLong

public static long prefixCodedToLong(String prefixCoded)

prefixCodedToInt

public static int prefixCodedToInt(String prefixCoded)

doubleToSortableLong

public static long doubleToSortableLong(double val)

Converts a double value to a sortable signed long. The value is converted by getting their IEEE 754 floating-point "double format" bit layout and then some bits are swapped, to be able to compare the result as long. By this the precision is not reduced, but the value can easily used as a long. The sort order (including Double.NaN) is defined by Double.compareTo(java.lang.Double); NaN is greater than positive infinity.

See Also:

sortableLongToDouble(long)

doubleToPrefixCoded

public static String doubleToPrefixCoded(double val)

sortableLongToDouble

public static double sortableLongToDouble(long val)

Converts a sortable long back to a double.

See Also:

doubleToSortableLong(double)

prefixCodedToDouble

public static double prefixCodedToDouble(String val)

floatToSortableInt

public static int floatToSortableInt(float val)

Converts a float value to a sortable signed int. The value is converted by getting their IEEE 754 floating-point "float format" bit layout and then some bits are swapped, to be able to compare the result as int. By this the precision is not reduced, but the value can easily used as an int. The sort order (including Float.NaN) is defined by Float.compareTo(java.lang.Float); NaN is greater than positive infinity.

See Also:

sortableIntToFloat(int)

floatToPrefixCoded

public static String floatToPrefixCoded(float val)

sortableIntToFloat

public static float sortableIntToFloat(int val)

Converts a sortable int back to a float.

See Also:

floatToSortableInt(float)

prefixCodedToFloat

public static float prefixCodedToFloat(String val)

splitLongRange

public static void splitLongRange(NumericUtils.LongRangeBuilder builder,
                                  int precisionStep,
                                  long minBound,
                                  long maxBound)

Expert: Splits a long range recursively. You may implement a builder that adds clauses to a BooleanQuery for each call to its NumericUtils.LongRangeBuilder.addRange(String,String) method.

This method is used by NumericRangeQuery.

splitIntRange

public static void splitIntRange(NumericUtils.IntRangeBuilder builder,
                                 int precisionStep,
                                 int minBound,
                                 int maxBound)

Expert: Splits an int range recursively. You may implement a builder that adds clauses to a BooleanQuery for each call to its NumericUtils.IntRangeBuilder.addRange(String,String) method.

This method is used by NumericRangeQuery.