Class PrecisionModel

java.lang.Object
org.locationtech.jts.geom.PrecisionModel
All Implemented Interfaces:
Serializable, Comparable

public class PrecisionModel extends Object implements Serializable, Comparable
Specifies the precision model of the Coordinates in a Geometry. In other words, specifies the grid of allowable points for all Geometrys.

The makePrecise(Coordinate) method allows rounding a coordinate to a "precise" value; that is, one whose precision is known exactly.

Coordinates are assumed to be precise in geometries. That is, the coordinates are assumed to be rounded to the precision model given for the geometry. JTS input routines automatically round coordinates to the precision model before creating Geometries. All internal operations assume that coordinates are rounded to the precision model. Constructive methods (such as boolean operations) always round computed coordinates to the appropriate precision model.

Currently three types of precision model are supported:

  • FLOATING - represents full double precision floating point. This is the default precision model used in JTS
  • FLOATING_SINGLE - represents single precision floating point.
  • FIXED - represents a model with a fixed number of decimal places. A Fixed Precision Model is specified by a scale factor. The scale factor specifies the size of the grid which numbers are rounded to. Input coordinates are mapped to fixed coordinates according to the following equations:
    • jtsPt.x = round( (inputPt.x * scale ) / scale
    • jtsPt.y = round( (inputPt.y * scale ) / scale
For example, to specify 3 decimal places of precision, use a scale factor of 1000. To specify -3 decimal places of precision (i.e. rounding to the nearest 1000), use a scale factor of 0.001.

Coordinates are represented internally as Java double-precision values. Since Java uses the IEEE-394 floating point standard, this provides 53 bits of precision. (Thus the maximum precisely representable integer is 9,007,199,254,740,992 - or almost 16 decimal digits of precision).

JTS binary methods currently do not handle inputs which have different precision models. The precision model of any constructed geometric value is undefined.

Version:
1.7
See Also:
  • Field Details

    • serialVersionUID

      private static final long serialVersionUID
      See Also:
    • FIXED

      public static final PrecisionModel.Type FIXED
      Fixed Precision indicates that coordinates have a fixed number of decimal places. The number of decimal places is determined by the log10 of the scale factor.
    • FLOATING

      public static final PrecisionModel.Type FLOATING
      Floating precision corresponds to the standard Java double-precision floating-point representation, which is based on the IEEE-754 standard
    • FLOATING_SINGLE

      public static final PrecisionModel.Type FLOATING_SINGLE
      Floating single precision corresponds to the standard Java single-precision floating-point representation, which is based on the IEEE-754 standard
    • maximumPreciseValue

      public static final double maximumPreciseValue
      The maximum precise value representable in a double. Since IEE754 double-precision numbers allow 53 bits of mantissa, the value is equal to 2^53 - 1. This provides almost 16 decimal digits of precision.
      See Also:
    • modelType

      private PrecisionModel.Type modelType
      The type of PrecisionModel this represents.
    • scale

      private double scale
      The scale factor which determines the number of decimal places in fixed precision.
  • Constructor Details

    • PrecisionModel

      public PrecisionModel()
      Creates a PrecisionModel with a default precision of FLOATING.
    • PrecisionModel

      public PrecisionModel(PrecisionModel.Type modelType)
      Creates a PrecisionModel that specifies an explicit precision model type. If the model type is FIXED the scale factor will default to 1.
      Parameters:
      modelType - the type of the precision model
    • PrecisionModel

      public PrecisionModel(double scale, double offsetX, double offsetY)
      Deprecated.
      offsets are no longer supported, since internal representation is rounded floating point
      Creates a PrecisionModel that specifies Fixed precision. Fixed-precision coordinates are represented as precise internal coordinates, which are rounded to the grid defined by the scale factor.
      Parameters:
      scale - amount by which to multiply a coordinate after subtracting the offset, to obtain a precise coordinate
      offsetX - not used.
      offsetY - not used.
    • PrecisionModel

      public PrecisionModel(double scale)
      Creates a PrecisionModel that specifies Fixed precision. Fixed-precision coordinates are represented as precise internal coordinates, which are rounded to the grid defined by the scale factor.
      Parameters:
      scale - amount by which to multiply a coordinate after subtracting the offset, to obtain a precise coordinate
    • PrecisionModel

      public PrecisionModel(PrecisionModel pm)
      Copy constructor to create a new PrecisionModel from an existing one.
  • Method Details

    • mostPrecise

      public static PrecisionModel mostPrecise(PrecisionModel pm1, PrecisionModel pm2)
      Determines which of two PrecisionModels is the most precise (allows the greatest number of significant digits).
      Parameters:
      pm1 - a PrecisionModel
      pm2 - a PrecisionModel
      Returns:
      the PrecisionModel which is most precise
    • isFloating

      public boolean isFloating()
      Tests whether the precision model supports floating point
      Returns:
      true if the precision model supports floating point
    • getMaximumSignificantDigits

      public int getMaximumSignificantDigits()
      Returns the maximum number of significant digits provided by this precision model. Intended for use by routines which need to print out decimal representations of precise values (such as WKTWriter).

      This method would be more correctly called getMinimumDecimalPlaces, since it actually computes the number of decimal places that is required to correctly display the full precision of an ordinate value.

      Since it is difficult to compute the required number of decimal places for scale factors which are not powers of 10, the algorithm uses a very rough approximation in this case. This has the side effect that for scale factors which are powers of 10 the value returned is 1 greater than the true value.

      Returns:
      the maximum number of decimal places provided by this precision model
    • getScale

      public double getScale()
      Returns the scale factor used to specify a fixed precision model. The number of decimal places of precision is equal to the base-10 logarithm of the scale factor. Non-integral and negative scale factors are supported. Negative scale factors indicate that the places of precision is to the left of the decimal point.
      Returns:
      the scale factor for the fixed precision model
    • getType

      public PrecisionModel.Type getType()
      Gets the type of this precision model
      Returns:
      the type of this precision model
      See Also:
    • setScale

      private void setScale(double scale)
      Sets the multiplying factor used to obtain a precise coordinate. This method is private because PrecisionModel is an immutable (value) type.
    • getOffsetX

      public double getOffsetX()
      Deprecated.
      Offsets are no longer used
      Returns the x-offset used to obtain a precise coordinate.
      Returns:
      the amount by which to subtract the x-coordinate before multiplying by the scale
    • getOffsetY

      public double getOffsetY()
      Deprecated.
      Offsets are no longer used
      Returns the y-offset used to obtain a precise coordinate.
      Returns:
      the amount by which to subtract the y-coordinate before multiplying by the scale
    • toInternal

      public void toInternal(Coordinate external, Coordinate internal)
      Deprecated.
      use makePrecise instead
      Sets internal to the precise representation of external.
      Parameters:
      external - the original coordinate
      internal - the coordinate whose values will be changed to the precise representation of external
    • toInternal

      public Coordinate toInternal(Coordinate external)
      Deprecated.
      use makePrecise instead
      Returns the precise representation of external.
      Parameters:
      external - the original coordinate
      Returns:
      the coordinate whose values will be changed to the precise representation of external
    • toExternal

      public Coordinate toExternal(Coordinate internal)
      Deprecated.
      no longer needed, since internal representation is same as external representation
      Returns the external representation of internal.
      Parameters:
      internal - the original coordinate
      Returns:
      the coordinate whose values will be changed to the external representation of internal
    • toExternal

      public void toExternal(Coordinate internal, Coordinate external)
      Deprecated.
      no longer needed, since internal representation is same as external representation
      Sets external to the external representation of internal.
      Parameters:
      internal - the original coordinate
      external - the coordinate whose values will be changed to the external representation of internal
    • makePrecise

      public double makePrecise(double val)
      Rounds a numeric value to the PrecisionModel grid. Asymmetric Arithmetic Rounding is used, to provide uniform rounding behaviour no matter where the number is on the number line.

      This method has no effect on NaN values.

      Note: Java's Math#rint uses the "Banker's Rounding" algorithm, which is not suitable for precision operations elsewhere in JTS.

    • makePrecise

      public void makePrecise(Coordinate coord)
      Rounds a Coordinate to the PrecisionModel grid.
    • toString

      public String toString()
      Overrides:
      toString in class Object
    • equals

      public boolean equals(Object other)
      Overrides:
      equals in class Object
    • compareTo

      public int compareTo(Object o)
      Compares this PrecisionModel object with the specified object for order. A PrecisionModel is greater than another if it provides greater precision. The comparison is based on the value returned by the getMaximumSignificantDigits() method. This comparison is not strictly accurate when comparing floating precision models to fixed models; however, it is correct when both models are either floating or fixed.
      Specified by:
      compareTo in interface Comparable
      Parameters:
      o - the PrecisionModel with which this PrecisionModel is being compared
      Returns:
      a negative integer, zero, or a positive integer as this PrecisionModel is less than, equal to, or greater than the specified PrecisionModel