Package org.webpki.cbor

Class CBORFloat

java.lang.Object

org.webpki.cbor.CBORObject

org.webpki.cbor.CBORFloat

All Implemented Interfaces:

Cloneable, Comparable<CBORObject>

public class CBORFloat
extends CBORObject

Class for holding CBOR float objects.

Numbers are constrained to the IEEE 754 notation using the length 16, 32, and 64 bit on the "wire". Which length to use is governed by the size and precision required to (minimally) correctly represent a number.

Constructor Summary

Constructors

Constructor	Description
CBORFloat(double value)	Creates a CBOR float object.

Method Summary

Modifier and Type	Method	Description
static String	formatDouble(Double value)	Get number in diagnostic notation.
int	length()	Get length of the optimized IEEE 754 type.
static void	setNonFiniteFloatsMode(boolean reject)	Globally disable NaN and Infinity.

Methods inherited from class org.webpki.cbor.CBORObject

checkForUnread, clone, compareTo, encode, equals, getArray, getBigInteger, getBoolean, getBytes, getDateTime, getEpochTime, getFloat16, getFloat32, getFloat64, getInt16, getInt32, getInt64, getInt8, getMap, getSimple, getString, getTag, getUint16, getUint32, getUint64, getUint8, hashCode, isNull, scan, toDiagnosticNotation, toString

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, wait, wait, wait

Constructor Details

CBORFloat

public CBORFloat(double value)

Creates a CBOR float object.

Note that this implementation does not provide a specific constructor for Java float values. Due to the CBOR normalization algorithm, numbers are still correctly encoded.

See also CBORObject.getFloat64() and CBORObject.getFloat32()

For NaN and Infinity support see CBORDecoder.REJECT_NON_FINITE_FLOATS and setNonFiniteFloatsMode(boolean).

Parameters:

value - Java double

Throws:

CBORException

Method Details

setNonFiniteFloatsMode

public static void setNonFiniteFloatsMode(boolean reject)

Globally disable NaN and Infinity.

Note that this method unlike CBORDecoder.REJECT_NON_FINITE_FLOATS, also affects encoding of NaN and Infinity values. Since this is a global setting. you need to consider how it could affect other applications running in the same JVM.

Parameters:

reject - If true, disable NaN and Infinity support.

formatDouble

public static String formatDouble(Double value)

Get number in diagnostic notation.

Floating point numbers are serialized using at least one integer digit (may be 0), a decimal point, and one or more fractional digits.

Possible exponents are written as e±n, where n != 0.

This method also supports NaN, Infinity, and -Infinity.

Parameters:

value - The double

Returns:

The double in string format

length

public int length()

Get length of the optimized IEEE 754 type.

Note that you must cast a CBORObject to CBORFloat in order to access length().

Returns:

Length in bytes: 2, 4, or 8.