[VII] Float details of the underlying data type

Float basic data type float's wrapper class float type object contains a float type of field Introduction to Properties

The number of bits used to represent the float value in twos complement form	public static final int SIZE = 32;
The number of bytes in the binary complement form that represents the float value	public static final int BYTES = size/byte.size;
Class instance representing the base type float	public static final class<float> TYPE = (class<float>) class.getprimitiveclass ("Float");
The maximum value that can be represented Only one form of standardization, that is, mentioned earlier	public static final float max_value = 0x1.fffffep+127f;
Minimum value for normalization	public static final float min_normal = 0x1.0p-126f;
The minimum value is also non-standardized form	public static final float min_value = 0x0.000002p-126f;
Positive Infinity It equals the value returned by Float.intbitstofloat (0x7f800000)	public static final float positive_infinity = 1.0f/0.0f;
Negative infinity It equals the value returned by Float.intbitstofloat (0xff800000)	public static final float negative_infinity = -1.0f/0.0f;
NaN Not a number It equals the value returned by Float.intbitstofloat (0x7fc00000)	public static final float NaN = 0.0f/0.0f;
The effective maximum of the exponential truth value	public static final int max_exponent = 127;
The effective minimum value of the exponential truth	public static final int Min_exponent =-126;

These properties, read the previous floating point introduction, can be very clear understanding
Construction MethodFloat still provides a string form that is based on the basic type float and float, and is still aided by the Parsexxx form parsefloat In addition, it also provides a way to construct a double based on the basic type, with a direct strong internal

Float (float value)
Float (String s)
Float (Double value)	Direct Strong turn

Common MethodsFor floating-point numbers, there are some additional property methods in our floating-point number introduction, for floating-point representation of the description of float provides a representation of the specified value of the acquisition method, which is the representation of a 32-bit bits sequence

Float Gets the representation Provides two forms of methods for getting representations, mainly for different representations of non-numeric Nan They can be converted to each other with intbitstofloat. floattointbits If the parameter is positive infinity, the result is 0x7f800000 if the argument is negative infinity, the result is 0xff800000 if the argument is NaN, the result is 0x7fc00000   in all cases, the result It's all an integer. Assigning it to the intbitstofloat (int) method generates a floating-point value that is the same as the Floattointbits parameter (and all Nan values generate a specification NaN value)   Dependent floattorawintbits

Converts the representation to float, returning a float value corresponding to the positional representation Local method In fact, the float is calculated according to the layout If the argument is 0x7f800000, the result is positive infinity if the argument is 0xff800000, the result is negative infinity if the parameter value is between 0x7f800001 to 0x7FFFFFFF or 0xff800001 to 0xFFFFFFFF, the result is NaN None of the IEEE 754 floating-point operations provided by Java can distinguish between two homogeneous Nan values with different bit patterns, and only use the Float.floattorawintbits method to differentiate between Nan values

There are several special representations of floating-point numbers, such as Infinite Nan, which also provide some related methods.

Static Boolean IsNaN (float v)	Static methods Whether a non-numeric (NaN) value Non-numeric true
Static Boolean isfinite (float f)	Static methods Whether it is a limited floating-point number Limited true
Static Boolean isinfinite (float v)	Static methods Whether it is infinity Is infinitely large true
Boolean isinfinite ()	Instance method Rely on static methods
Boolean IsNaN ()	Instance methods rely on static methods

Comparison

static int compare (float f1, float F2)	Static methods Comparison of two float F1 < F2 less than 0 F1 = F2 equals 0 F1 > F2 greater than 0
int CompareTo (Float anotherfloat)	Instance method Two objects for size comparison, dependent on static methods

PARSEXXX series strings are parsed as basic types and do not require objects, so they are static methods

Returns a string representation of the base type of float The performance is the same as valueof (String), but valueof returns the object

If string is null or does not contain a string that can be parsed, an exception will be thrown

Bottom-dependent sun.misc.FloatingDecimal

The ValueOf Series wraps the basic basic types as objects used to create the acquisition object, so no objects are required, all static methods are different from the integer values described earlier, they all have buffer float does not exist cache, valueof is also a direct new object

static float valueOf (float f)
Static Float valueOf (String s)	Dependent parsefloat Method So it says it works the same as valueof (String), which is the same thing.

Float No Decode method Xxxvalue Series

Similar to other numeric types described earlier All of them are internal value of strong turn Return (XXX) value;

Bytevalue () Shortvalue () Intvalue () Longvalue () Floatvalue () Doublevalue ()

toString toxxxstring Series

static String toString (float f)
string toString ()	Internal call   Static String toString (float f)
static String tohexstring (float f)	Returns the hexadecimal string representation of the float parameter

The ToString series seems to have nothing to say, and it seems like there's a lot to be said about the rules of the format characters have doubts directly view APIequals

boolean equals (Object obj) &nbs P

Compare this object to a specified object when and only if the parameter is not NULL but a Float object, and represents a float value that is the same as the float value represented by this object, the result is true if and only if the method #floatToLongBits (double) is applied to the value of the int returned by the two value, it is considered that the two float values are the same   attention, In most cases, the value of F1.equals (F2) is true for two instances of the Float class F1 and F2, when and only if the value of f1.floatvalue () = = F2.floatvalue () is true. However, there are two exceptions: if both F1 and F2 represent Float.nan, the False,equals method will return true even if the value of Float.nan==float.nan is used, so this is floattointbits instead of RA W   If F1 represents +0.0f, and F2 represents -0.0f, or vice versa, even 0.0f==-0.0f test will return FALSE if the value is true,equal the definition allows the hash table to operate correctly.

hashcode

static int hashcode (float value)	Static methods Get the hashcode value of a value
int Hashcode ()	Instance method Rely on static methods

Other methods

sum (float, float)
Max (float, float)
Min (float, float)

SummaryIn fact, the expression of floating-point numbers and the use of norms is the focus of float is just float packaging, float is only a IEEE754 standard implementation, is still in the standard understanding

[VII] Float details of the underlying data type