Float
Talk0this wiki
Float is the class of floating point numbers. It uses the double
type of the underlying C environment.
Inherits: Numeric
.
Mixins: Comparable
(from Numeric
), Precision
.
ConstantsEdit
All constants are equivalent to ones defined by Standard C in its limits.h header, but without the DBL_
prefix. That is, Float::MIN_EXP
is defined as the value of DBL_MIN_EXP
in the underlying C environment. The single exception is Float::ROUNDS
, because this is common to all C floatingpoint types and only FLT_ROUNDS
is defined (there is no DBL_ROUNDS
). As such, some of the descriptions are taken from the ISO C Standard, ISO/IEC 9899:1999. Many are paraphrased for clarity. Where any conflict exists between the descriptions given here and that standard, the standard shall prevail.
Constant  Description 

ROUNDS  An integer constant giving the rounding mode for floating point arithmetic:

RADIX  Radix of the floating point type 
MANT_DIG  Number of baseRADIX digits in the floating point significand

DIG  Maximum number of decimal digits that can be converted to a Float , converted back again, and rounded to DIG digits using round to nearest, without changing the decimal digits

MIN_EXP  Minimum negative integer such that RADIX to that power is a normalized floating point number

MAX_EXP  Maximum integer such that RADIX to that power is a representable finite floating point number

MIN_10_EXP  Minimum negative integer such that 10 to that power is a normalized floating point number 
MAX_10_EXP  Maximum integer such that 10 to that power is a representable finite floating point number 
MIN  Minimum normalized finite floating point number 
MAX  Maximum representable finite floating point number 
EPSILON  The difference between 1 and the next higher representable floating point number 
MethodsEdit
Method  Parameters  Description 

==  1  Returns true if self is equal to the parameter

>  1  Returns true if self is greater than the parameter

>=  1  Returns true if self is greater than or equal to the parameter

<  1  Returns true if self is less than the parameter

<=  1  Returns true if self is less than or equal to the parameter

<=>  1  "Spaceship" operator. Returns 1 if self is less than the parameter, 0 if equal, +1 if greater.

@  0  Unary minus. Returns the negation of self .

+@  0  Unary plus. Returns self unchanged.

+  1  Addition operator. Returns the sum of self and the single parameter.

  1  Subtraction operator. Returns the difference of self and the single parameter.

*  1  Multiplication operator. Returns the product of self and the single parameter.

/  1  Division operator. Returns the quotient of self and the single parameter.

%  1  Modulo operator. Returns the remainder from dividing self by the single paramter. The remainder is determined by rounding the quotient downward and has the same sign as the modulus.

**  1  Power operator. Returns self raised to the power of the single parameter.

abs  0  Returns the absolute value of self .

between?  2  self.between(min, max) returns true if self is not less than min and not greater than max . The limits are part of the range, so 2.between?(2, 6) returns true .

ceil  0  Returns the least integer that is not less than self .

coerce  1  Because self is always a Float in this case, self.coerce(other) returns [ other.to_f, self ] .

divmod  1  self.divmod(other) returns an Array containing the integer quotient and the remainder, in that order, resulting from dividing self by other . The quotient is an Integer and is rounded downward; the remainder is either an Integer or a Float and satisfies (quo*other+rem) == self . This is equivalent to [self.div(other), self.modulo(other)] , except that self and other are evaluated only once.

div  1  self.div(other) returns the integer quotient that results from dividing self by other . The quotient is an Integer and is rounded down. Equivalent to self.quo(other).floor and to self.divmod(other)[0] .

eql?  1  self.eql?(other) returns true if self and other have the same type and equal values.

finite?  0  Returns true if self is a finite floating point number (neither an infinity nor a NaN).

floor  0  Returns the greatest integer that is not greater than self .

hash  0  Returns a hash for use by Hash objects.

infinite?  0  Returns true if self is a floatingpoint infinity.

integer?  0  Returns false .

modulo  1  self.modulo(other) returns the remainder that results from dividing self by other . The remainder has the same sign as other .

nan?  0  Returns true if self is a NaN.

nonzero?  0  Returns true if self is not equal to zero.

prec  1  self.prec(klass) converts self to an object of class klass , using the induced_from singleton method of that class .

prec_f  1  self.prec_f is equivalent to self.prec(Float) .

prec_i  1  self.prec_i is equivalent to self.prec(Integer) .

quo  1  self.quo(other) returns the result of self divided by other as a Float .

remainder  1  self.remainder(other) returns the remainder that results from dividing self by other . Different from modulo in that the remainder is derived by rounding the quotient toward zero rather than downward and has the same sign as self ; the sign of other is ignored.

round  0  Rounds self to the nearest integer, with ties rounding away from zero.

step  1 or 2  self.step(limit, stride) calls a passed block, passing it self the first time, and then self+stride , self+2*stride , and so on until the limit is exceeded. A step value that is equal to the limit will call the block one last time. If stride is not present, it is taken to be 1.

to_i  0  Converts self to an Integer .

to_int  0  Synonym for to_i .

to_f  0  Returns self unchanged.

to_s  0  Returns a string representation of self .

truncate  0  Rounds toward zero; that is, returns the Integer with greatest absolute value that has the same sign as self and absolute value not greater than that of self .

zero?  0  Returns true if self is equal to zero.

New in Ruby 1.9Edit
Method  Parameters  Description 

fdiv  1  A synonym for quo .

round  0 or 1  With no parameter, round behaves as in Ruby 1.8. An optional numeric parameter gives the number of digits to the right of the decimal point to which self is to be rounded. This number may be negative, in which case self will be rounded to a power of ten equal to the negation of the digit count. If the number of digits is positive, the return value is a Float ; otherwise it is an Integer .

scalar?  0  Returns true .
