class Float

Float objects represent inexact real numbers using the native architecture's double-precision floating point representation.

Floating point has a different arithmetic and is an inexact number. So you should know its esoteric system. See following:

Constants

DIG

The minimum number of significant decimal digits in a double-precision floating point.

Usually defaults to 15.

EPSILON

The difference between 1 and the smallest double-precision floating point number greater than 1.

Usually defaults to 2.2204460492503131e-16.

INFINITY

An expression representing positive infinity.

MANT_DIG

The number of base digits for the double data type.

Usually defaults to 53.

MAX

The largest possible integer in a double-precision floating point number.

Usually defaults to 1.7976931348623157e+308.

MAX_10_EXP

The largest positive exponent in a double-precision floating point where 10 raised to this power minus 1.

Usually defaults to 308.

MAX_EXP

The largest possible exponent value in a double-precision floating point.

Usually defaults to 1024.

MIN

The smallest positive normalized number in a double-precision floating point.

Usually defaults to 2.2250738585072014e-308.

If the platform supports denormalized numbers, there are numbers between zero and Float::MIN. 0.0.next_float returns the smallest positive floating point number including denormalized numbers.

MIN_10_EXP

The smallest negative exponent in a double-precision floating point where 10 raised to this power minus 1.

Usually defaults to -307.

MIN_EXP

The smallest possible exponent value in a double-precision floating point.

Usually defaults to -1021.

NAN

An expression representing a value which is “not a number”.

RADIX

The base of the floating point, or number of unique digits used to represent the number.

Usually defaults to 2 on most systems, which would represent a base-10 decimal.

ROUNDS

Deprecated, do not use.

Represents the rounding mode for floating point addition at the start time.

Usually defaults to 1, rounding to the nearest number.

Other modes include:

-1

Indeterminable

0

Rounding towards zero

1

Rounding to the nearest number

2

Rounding towards positive infinity

3

Rounding towards negative infinity

Public Instance Methods

float % other → float click to toggle source

Returns the modulo after division of float by other.

6543.21.modulo(137)      #=> 104.21000000000004
6543.21.modulo(137.24)   #=> 92.92999999999961

static VALUE
flo_mod(VALUE x, VALUE y)
{
    double fy;

    if (RB_TYPE_P(y, T_FIXNUM)) {
        fy = (double)FIX2LONG(y);
    }
    else if (RB_TYPE_P(y, T_BIGNUM)) {
        fy = rb_big2dbl(y);
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        fy = RFLOAT_VALUE(y);
    }
    else {
        return rb_num_coerce_bin(x, y, '%');
    }
    return DBL2NUM(ruby_float_mod(RFLOAT_VALUE(x), fy));
}
float * other → float click to toggle source

Returns a new Float which is the product of float and other.

VALUE
rb_float_mul(VALUE x, VALUE y)
{
    if (RB_TYPE_P(y, T_FIXNUM)) {
        return DBL2NUM(RFLOAT_VALUE(x) * (double)FIX2LONG(y));
    }
    else if (RB_TYPE_P(y, T_BIGNUM)) {
        return DBL2NUM(RFLOAT_VALUE(x) * rb_big2dbl(y));
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        return DBL2NUM(RFLOAT_VALUE(x) * RFLOAT_VALUE(y));
    }
    else {
        return rb_num_coerce_bin(x, y, '*');
    }
}
float ** other → float click to toggle source

Raises float to the power of other.

2.0**3   #=> 8.0

VALUE
rb_float_pow(VALUE x, VALUE y)
{
    double dx, dy;
    if (RB_TYPE_P(y, T_FIXNUM)) {
        dx = RFLOAT_VALUE(x);
        dy = (double)FIX2LONG(y);
    }
    else if (RB_TYPE_P(y, T_BIGNUM)) {
        dx = RFLOAT_VALUE(x);
        dy = rb_big2dbl(y);
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        dx = RFLOAT_VALUE(x);
        dy = RFLOAT_VALUE(y);
        if (dx < 0 && dy != round(dy))
            return rb_dbl_complex_new_polar_pi(pow(-dx, dy), dy);
    }
    else {
        return rb_num_coerce_bin(x, y, idPow);
    }
    return DBL2NUM(pow(dx, dy));
}
float + other → float click to toggle source

Returns a new Float which is the sum of float and other.

VALUE
rb_float_plus(VALUE x, VALUE y)
{
    if (RB_TYPE_P(y, T_FIXNUM)) {
        return DBL2NUM(RFLOAT_VALUE(x) + (double)FIX2LONG(y));
    }
    else if (RB_TYPE_P(y, T_BIGNUM)) {
        return DBL2NUM(RFLOAT_VALUE(x) + rb_big2dbl(y));
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        return DBL2NUM(RFLOAT_VALUE(x) + RFLOAT_VALUE(y));
    }
    else {
        return rb_num_coerce_bin(x, y, '+');
    }
}
float - other → float click to toggle source

Returns a new Float which is the difference of float and other.

VALUE
rb_float_minus(VALUE x, VALUE y)
{
    if (RB_TYPE_P(y, T_FIXNUM)) {
        return DBL2NUM(RFLOAT_VALUE(x) - (double)FIX2LONG(y));
    }
    else if (RB_TYPE_P(y, T_BIGNUM)) {
        return DBL2NUM(RFLOAT_VALUE(x) - rb_big2dbl(y));
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        return DBL2NUM(RFLOAT_VALUE(x) - RFLOAT_VALUE(y));
    }
    else {
        return rb_num_coerce_bin(x, y, '-');
    }
}
-float → float click to toggle source

Returns float, negated.

VALUE
rb_float_uminus(VALUE flt)
{
    return DBL2NUM(-RFLOAT_VALUE(flt));
}
float / other → float click to toggle source

Returns a new Float which is the result of dividing float by other.

VALUE
rb_float_div(VALUE x, VALUE y)
{
    double num = RFLOAT_VALUE(x);
    double den;
    double ret;

    if (RB_TYPE_P(y, T_FIXNUM)) {
        den = FIX2LONG(y);
    }
    else if (RB_TYPE_P(y, T_BIGNUM)) {
        den = rb_big2dbl(y);
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        den = RFLOAT_VALUE(y);
    }
    else {
        return rb_num_coerce_bin(x, y, '/');
    }

    ret = double_div_double(num, den);
    return DBL2NUM(ret);
}
float < real → true or false click to toggle source

Returns true if float is less than real.

The result of NaN < NaN is undefined, so an implementation-dependent value is returned.

static VALUE
flo_lt(VALUE x, VALUE y)
{
    double a, b;

    a = RFLOAT_VALUE(x);
    if (RB_TYPE_P(y, T_FIXNUM) || RB_TYPE_P(y, T_BIGNUM)) {
        VALUE rel = rb_integer_float_cmp(y, x);
        if (FIXNUM_P(rel))
            return -FIX2LONG(rel) < 0 ? Qtrue : Qfalse;
        return Qfalse;
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        b = RFLOAT_VALUE(y);
#if defined(_MSC_VER) && _MSC_VER < 1300
        if (isnan(b)) return Qfalse;
#endif
    }
    else {
        return rb_num_coerce_relop(x, y, '<');
    }
#if defined(_MSC_VER) && _MSC_VER < 1300
    if (isnan(a)) return Qfalse;
#endif
    return (a < b)?Qtrue:Qfalse;
}
float <= real → true or false click to toggle source

Returns true if float is less than or equal to real.

The result of NaN <= NaN is undefined, so an implementation-dependent value is returned.

static VALUE
flo_le(VALUE x, VALUE y)
{
    double a, b;

    a = RFLOAT_VALUE(x);
    if (RB_TYPE_P(y, T_FIXNUM) || RB_TYPE_P(y, T_BIGNUM)) {
        VALUE rel = rb_integer_float_cmp(y, x);
        if (FIXNUM_P(rel))
            return -FIX2LONG(rel) <= 0 ? Qtrue : Qfalse;
        return Qfalse;
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        b = RFLOAT_VALUE(y);
#if defined(_MSC_VER) && _MSC_VER < 1300
        if (isnan(b)) return Qfalse;
#endif
    }
    else {
        return rb_num_coerce_relop(x, y, idLE);
    }
#if defined(_MSC_VER) && _MSC_VER < 1300
    if (isnan(a)) return Qfalse;
#endif
    return (a <= b)?Qtrue:Qfalse;
}
float <=> real → -1, 0, +1, or nil click to toggle source

Returns -1, 0, or +1 depending on whether float is less than, equal to, or greater than real. This is the basis for the tests in the Comparable module.

The result of NaN <=> NaN is undefined, so an implementation-dependent value is returned.

nil is returned if the two values are incomparable.

static VALUE
flo_cmp(VALUE x, VALUE y)
{
    double a, b;
    VALUE i;

    a = RFLOAT_VALUE(x);
    if (isnan(a)) return Qnil;
    if (RB_TYPE_P(y, T_FIXNUM) || RB_TYPE_P(y, T_BIGNUM)) {
        VALUE rel = rb_integer_float_cmp(y, x);
        if (FIXNUM_P(rel))
            return LONG2FIX(-FIX2LONG(rel));
        return rel;
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        b = RFLOAT_VALUE(y);
    }
    else {
        if (isinf(a) && (i = rb_check_funcall(y, rb_intern("infinite?"), 0, 0)) != Qundef) {
            if (RTEST(i)) {
                int j = rb_cmpint(i, x, y);
                j = (a > 0.0) ? (j > 0 ? 0 : +1) : (j < 0 ? 0 : -1);
                return INT2FIX(j);
            }
            if (a > 0.0) return INT2FIX(1);
            return INT2FIX(-1);
        }
        return rb_num_coerce_cmp(x, y, id_cmp);
    }
    return rb_dbl_cmp(a, b);
}
float == obj → true or false click to toggle source

Returns true only if obj has the same value as float. Contrast this with Float#eql?, which requires obj to be a Float.

1.0 == 1   #=> true

The result of NaN == NaN is undefined, so an implementation-dependent value is returned.

MJIT_FUNC_EXPORTED VALUE
rb_float_equal(VALUE x, VALUE y)
{
    volatile double a, b;

    if (RB_TYPE_P(y, T_FIXNUM) || RB_TYPE_P(y, T_BIGNUM)) {
        return rb_integer_float_eq(y, x);
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        b = RFLOAT_VALUE(y);
#if defined(_MSC_VER) && _MSC_VER < 1300
        if (isnan(b)) return Qfalse;
#endif
    }
    else {
        return num_equal(x, y);
    }
    a = RFLOAT_VALUE(x);
#if defined(_MSC_VER) && _MSC_VER < 1300
    if (isnan(a)) return Qfalse;
#endif
    return (a == b)?Qtrue:Qfalse;
}
float == obj → true or false click to toggle source

Returns true only if obj has the same value as float. Contrast this with Float#eql?, which requires obj to be a Float.

1.0 == 1   #=> true

The result of NaN == NaN is undefined, so an implementation-dependent value is returned.

MJIT_FUNC_EXPORTED VALUE
rb_float_equal(VALUE x, VALUE y)
{
    volatile double a, b;

    if (RB_TYPE_P(y, T_FIXNUM) || RB_TYPE_P(y, T_BIGNUM)) {
        return rb_integer_float_eq(y, x);
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        b = RFLOAT_VALUE(y);
#if defined(_MSC_VER) && _MSC_VER < 1300
        if (isnan(b)) return Qfalse;
#endif
    }
    else {
        return num_equal(x, y);
    }
    a = RFLOAT_VALUE(x);
#if defined(_MSC_VER) && _MSC_VER < 1300
    if (isnan(a)) return Qfalse;
#endif
    return (a == b)?Qtrue:Qfalse;
}
float > real → true or false click to toggle source

Returns true if float is greater than real.

The result of NaN > NaN is undefined, so an implementation-dependent value is returned.

VALUE
rb_float_gt(VALUE x, VALUE y)
{
    double a, b;

    a = RFLOAT_VALUE(x);
    if (RB_TYPE_P(y, T_FIXNUM) || RB_TYPE_P(y, T_BIGNUM)) {
        VALUE rel = rb_integer_float_cmp(y, x);
        if (FIXNUM_P(rel))
            return -FIX2LONG(rel) > 0 ? Qtrue : Qfalse;
        return Qfalse;
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        b = RFLOAT_VALUE(y);
#if defined(_MSC_VER) && _MSC_VER < 1300
        if (isnan(b)) return Qfalse;
#endif
    }
    else {
        return rb_num_coerce_relop(x, y, '>');
    }
#if defined(_MSC_VER) && _MSC_VER < 1300
    if (isnan(a)) return Qfalse;
#endif
    return (a > b)?Qtrue:Qfalse;
}
float >= real → true or false click to toggle source

Returns true if float is greater than or equal to real.

The result of NaN >= NaN is undefined, so an implementation-dependent value is returned.

static VALUE
flo_ge(VALUE x, VALUE y)
{
    double a, b;

    a = RFLOAT_VALUE(x);
    if (RB_TYPE_P(y, T_FIXNUM) || RB_TYPE_P(y, T_BIGNUM)) {
        VALUE rel = rb_integer_float_cmp(y, x);
        if (FIXNUM_P(rel))
            return -FIX2LONG(rel) >= 0 ? Qtrue : Qfalse;
        return Qfalse;
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        b = RFLOAT_VALUE(y);
#if defined(_MSC_VER) && _MSC_VER < 1300
        if (isnan(b)) return Qfalse;
#endif
    }
    else {
        return rb_num_coerce_relop(x, y, idGE);
    }
#if defined(_MSC_VER) && _MSC_VER < 1300
    if (isnan(a)) return Qfalse;
#endif
    return (a >= b)?Qtrue:Qfalse;
}
abs → float click to toggle source

Returns the absolute value of float.

(-34.56).abs   #=> 34.56
-34.56.abs     #=> 34.56
34.56.abs      #=> 34.56

Float#magnitude is an alias for Float#abs.

VALUE
rb_float_abs(VALUE flt)
{
    double val = fabs(RFLOAT_VALUE(flt));
    return DBL2NUM(val);
}
angle → 0 or float click to toggle source

Returns 0 if the value is positive, pi otherwise.

static VALUE
float_arg(VALUE self)
{
    if (isnan(RFLOAT_VALUE(self)))
        return self;
    if (f_tpositive_p(self))
        return INT2FIX(0);
    return rb_const_get(rb_mMath, id_PI);
}
arg → 0 or float click to toggle source

Returns 0 if the value is positive, pi otherwise.

static VALUE
float_arg(VALUE self)
{
    if (isnan(RFLOAT_VALUE(self)))
        return self;
    if (f_tpositive_p(self))
        return INT2FIX(0);
    return rb_const_get(rb_mMath, id_PI);
}
ceil([ndigits]) → integer or float click to toggle source

Returns the smallest number greater than or equal to float with a precision of ndigits decimal digits (default: 0).

When the precision is negative, the returned value is an integer with at least ndigits.abs trailing zeros.

Returns a floating point number when ndigits is positive, otherwise returns an integer.

1.2.ceil      #=> 2
2.0.ceil      #=> 2
(-1.2).ceil   #=> -1
(-2.0).ceil   #=> -2

1.234567.ceil(2)   #=> 1.24
1.234567.ceil(3)   #=> 1.235
1.234567.ceil(4)   #=> 1.2346
1.234567.ceil(5)   #=> 1.23457

34567.89.ceil(-5)  #=> 100000
34567.89.ceil(-4)  #=> 40000
34567.89.ceil(-3)  #=> 35000
34567.89.ceil(-2)  #=> 34600
34567.89.ceil(-1)  #=> 34570
34567.89.ceil(0)   #=> 34568
34567.89.ceil(1)   #=> 34567.9
34567.89.ceil(2)   #=> 34567.89
34567.89.ceil(3)   #=> 34567.89

Note that the limited precision of floating point arithmetic might lead to surprising results:

(2.1 / 0.7).ceil  #=> 4 (!)

static VALUE
flo_ceil(int argc, VALUE *argv, VALUE num)
{
    int ndigits = 0;

    if (rb_check_arity(argc, 0, 1)) {
        ndigits = NUM2INT(argv[0]);
    }
    return rb_float_ceil(num, ndigits);
}
coerce(numeric) → array click to toggle source

Returns an array with both numeric and float represented as Float objects.

This is achieved by converting numeric to a Float.

1.2.coerce(3)       #=> [3.0, 1.2]
2.5.coerce(1.1)     #=> [1.1, 2.5]

static VALUE
flo_coerce(VALUE x, VALUE y)
{
    return rb_assoc_new(rb_Float(y), x);
}
denominator → integer click to toggle source

Returns the denominator (always positive). The result is machine dependent.

See also Float#numerator.

VALUE
rb_float_denominator(VALUE self)
{
    double d = RFLOAT_VALUE(self);
    VALUE r;
    if (isinf(d) || isnan(d))
        return INT2FIX(1);
    r = float_to_r(self);
    if (canonicalization && k_integer_p(r)) {
        return ONE;
    }
    return nurat_denominator(r);
}
divmod(numeric) → array click to toggle source

See Numeric#divmod.

42.0.divmod(6)   #=> [7, 0.0]
42.0.divmod(5)   #=> [8, 2.0]

static VALUE
flo_divmod(VALUE x, VALUE y)
{
    double fy, div, mod;
    volatile VALUE a, b;

    if (RB_TYPE_P(y, T_FIXNUM)) {
        fy = (double)FIX2LONG(y);
    }
    else if (RB_TYPE_P(y, T_BIGNUM)) {
        fy = rb_big2dbl(y);
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        fy = RFLOAT_VALUE(y);
    }
    else {
        return rb_num_coerce_bin(x, y, id_divmod);
    }
    flodivmod(RFLOAT_VALUE(x), fy, &div, &mod);
    a = dbl2ival(div);
    b = DBL2NUM(mod);
    return rb_assoc_new(a, b);
}
eql?(obj) → true or false click to toggle source

Returns true only if obj is a Float with the same value as float. Contrast this with Float#==, which performs type conversions.

1.0.eql?(1)   #=> false

The result of NaN.eql?(NaN) is undefined, so an implementation-dependent value is returned.

MJIT_FUNC_EXPORTED VALUE
rb_float_eql(VALUE x, VALUE y)
{
    if (RB_TYPE_P(y, T_FLOAT)) {
        double a = RFLOAT_VALUE(x);
        double b = RFLOAT_VALUE(y);
#if defined(_MSC_VER) && _MSC_VER < 1300
        if (isnan(a) || isnan(b)) return Qfalse;
#endif
        if (a == b)
            return Qtrue;
    }
    return Qfalse;
}
fdiv(numeric) → float click to toggle source

Returns float / numeric, same as Float#/.

static VALUE
flo_quo(VALUE x, VALUE y)
{
    return num_funcall1(x, '/', y);
}
finite? → true or false click to toggle source

Returns true if float is a valid IEEE floating point number, i.e. it is not infinite and Float#nan? is false.

VALUE
rb_flo_is_finite_p(VALUE num)
{
    double value = RFLOAT_VALUE(num);

#ifdef HAVE_ISFINITE
    if (!isfinite(value))
        return Qfalse;
#else
    if (isinf(value) || isnan(value))
        return Qfalse;
#endif

    return Qtrue;
}
floor([ndigits]) → integer or float click to toggle source

Returns the largest number less than or equal to float with a precision of ndigits decimal digits (default: 0).

When the precision is negative, the returned value is an integer with at least ndigits.abs trailing zeros.

Returns a floating point number when ndigits is positive, otherwise returns an integer.

1.2.floor      #=> 1
2.0.floor      #=> 2
(-1.2).floor   #=> -2
(-2.0).floor   #=> -2

1.234567.floor(2)   #=> 1.23
1.234567.floor(3)   #=> 1.234
1.234567.floor(4)   #=> 1.2345
1.234567.floor(5)   #=> 1.23456

34567.89.floor(-5)  #=> 0
34567.89.floor(-4)  #=> 30000
34567.89.floor(-3)  #=> 34000
34567.89.floor(-2)  #=> 34500
34567.89.floor(-1)  #=> 34560
34567.89.floor(0)   #=> 34567
34567.89.floor(1)   #=> 34567.8
34567.89.floor(2)   #=> 34567.89
34567.89.floor(3)   #=> 34567.89

Note that the limited precision of floating point arithmetic might lead to surprising results:

(0.3 / 0.1).floor  #=> 2 (!)

static VALUE
flo_floor(int argc, VALUE *argv, VALUE num)
{
    int ndigits = 0;
    if (rb_check_arity(argc, 0, 1)) {
        ndigits = NUM2INT(argv[0]);
    }
    return rb_float_floor(num, ndigits);
}
hash → integer click to toggle source

Returns a hash code for this float.

See also Object#hash.

static VALUE
flo_hash(VALUE num)
{
    return rb_dbl_hash(RFLOAT_VALUE(num));
}
infinite? → -1, 1, or nil click to toggle source

Returns nil, -1, or 1 depending on whether the value is finite, -Infinity, or +Infinity.

(0.0).infinite?        #=> nil
(-1.0/0.0).infinite?   #=> -1
(+1.0/0.0).infinite?   #=> 1

VALUE
rb_flo_is_infinite_p(VALUE num)
{
    double value = RFLOAT_VALUE(num);

    if (isinf(value)) {
        return INT2FIX( value < 0 ? -1 : 1 );
    }

    return Qnil;
}
inspect()
Alias for: to_s
magnitude → float click to toggle source

Returns the absolute value of float.

(-34.56).abs   #=> 34.56
-34.56.abs     #=> 34.56
34.56.abs      #=> 34.56

Float#magnitude is an alias for Float#abs.

VALUE
rb_float_abs(VALUE flt)
{
    double val = fabs(RFLOAT_VALUE(flt));
    return DBL2NUM(val);
}
modulo(other) → float click to toggle source

Returns the modulo after division of float by other.

6543.21.modulo(137)      #=> 104.21000000000004
6543.21.modulo(137.24)   #=> 92.92999999999961

static VALUE
flo_mod(VALUE x, VALUE y)
{
    double fy;

    if (RB_TYPE_P(y, T_FIXNUM)) {
        fy = (double)FIX2LONG(y);
    }
    else if (RB_TYPE_P(y, T_BIGNUM)) {
        fy = rb_big2dbl(y);
    }
    else if (RB_TYPE_P(y, T_FLOAT)) {
        fy = RFLOAT_VALUE(y);
    }
    else {
        return rb_num_coerce_bin(x, y, '%');
    }
    return DBL2NUM(ruby_float_mod(RFLOAT_VALUE(x), fy));
}
nan? → true or false click to toggle source

Returns true if float is an invalid IEEE floating point number.

a = -1.0      #=> -1.0
a.nan?        #=> false
a = 0.0/0.0   #=> NaN
a.nan?        #=> true

static VALUE
flo_is_nan_p(VALUE num)
{
    double value = RFLOAT_VALUE(num);

    return isnan(value) ? Qtrue : Qfalse;
}
negative? → true or false click to toggle source

Returns true if float is less than 0.

static VALUE
flo_negative_p(VALUE num)
{
    double f = RFLOAT_VALUE(num);
    return f < 0.0 ? Qtrue : Qfalse;
}
next_float → float click to toggle source

Returns the next representable floating point number.

Float::MAX.next_float and Float::INFINITY.next_float is Float::INFINITY.

Float::NAN.next_float is Float::NAN.

For example:

0.01.next_float    #=> 0.010000000000000002
1.0.next_float     #=> 1.0000000000000002
100.0.next_float   #=> 100.00000000000001

0.01.next_float - 0.01     #=> 1.734723475976807e-18
1.0.next_float - 1.0       #=> 2.220446049250313e-16
100.0.next_float - 100.0   #=> 1.4210854715202004e-14

f = 0.01; 20.times { printf "%-20a %s\n", f, f.to_s; f = f.next_float }
#=> 0x1.47ae147ae147bp-7 0.01
#   0x1.47ae147ae147cp-7 0.010000000000000002
#   0x1.47ae147ae147dp-7 0.010000000000000004
#   0x1.47ae147ae147ep-7 0.010000000000000005
#   0x1.47ae147ae147fp-7 0.010000000000000007
#   0x1.47ae147ae148p-7  0.010000000000000009
#   0x1.47ae147ae1481p-7 0.01000000000000001
#   0x1.47ae147ae1482p-7 0.010000000000000012
#   0x1.47ae147ae1483p-7 0.010000000000000014
#   0x1.47ae147ae1484p-7 0.010000000000000016
#   0x1.47ae147ae1485p-7 0.010000000000000018
#   0x1.47ae147ae1486p-7 0.01000000000000002
#   0x1.47ae147ae1487p-7 0.010000000000000021
#   0x1.47ae147ae1488p-7 0.010000000000000023
#   0x1.47ae147ae1489p-7 0.010000000000000024
#   0x1.47ae147ae148ap-7 0.010000000000000026
#   0x1.47ae147ae148bp-7 0.010000000000000028
#   0x1.47ae147ae148cp-7 0.01000000000000003
#   0x1.47ae147ae148dp-7 0.010000000000000031
#   0x1.47ae147ae148ep-7 0.010000000000000033

f = 0.0
100.times { f += 0.1 }
f                           #=> 9.99999999999998       # should be 10.0 in the ideal world.
10-f                        #=> 1.9539925233402755e-14 # the floating point error.
10.0.next_float-10          #=> 1.7763568394002505e-15 # 1 ulp (unit in the last place).
(10-f)/(10.0.next_float-10) #=> 11.0                   # the error is 11 ulp.
(10-f)/(10*Float::EPSILON)  #=> 8.8                    # approximation of the above.
"%a" % 10                   #=> "0x1.4p+3"
"%a" % f                    #=> "0x1.3fffffffffff5p+3" # the last hex digit is 5.  16 - 5 = 11 ulp.

static VALUE
flo_next_float(VALUE vx)
{
    double x, y;
    x = NUM2DBL(vx);
    y = nextafter(x, HUGE_VAL);
    return DBL2NUM(y);
}
numerator → integer click to toggle source

Returns the numerator. The result is machine dependent.

n = 0.3.numerator    #=> 5404319552844595
d = 0.3.denominator  #=> 18014398509481984
n.fdiv(d)            #=> 0.3

See also Float#denominator.

VALUE
rb_float_numerator(VALUE self)
{
    double d = RFLOAT_VALUE(self);
    VALUE r;
    if (isinf(d) || isnan(d))
        return self;
    r = float_to_r(self);
    if (canonicalization && k_integer_p(r)) {
        return r;
    }
    return nurat_numerator(r);
}
phase → 0 or float click to toggle source

Returns 0 if the value is positive, pi otherwise.

static VALUE
float_arg(VALUE self)
{
    if (isnan(RFLOAT_VALUE(self)))
        return self;
    if (f_tpositive_p(self))
        return INT2FIX(0);
    return rb_const_get(rb_mMath, id_PI);
}
positive? → true or false click to toggle source

Returns true if float is greater than 0.

static VALUE
flo_positive_p(VALUE num)
{
    double f = RFLOAT_VALUE(num);
    return f > 0.0 ? Qtrue : Qfalse;
}
prev_float → float click to toggle source

Returns the previous representable floating point number.

(-Float::MAX).prev_float and (-Float::INFINITY).prev_float is -Float::INFINITY.

Float::NAN.prev_float is Float::NAN.

For example:

0.01.prev_float    #=> 0.009999999999999998
1.0.prev_float     #=> 0.9999999999999999
100.0.prev_float   #=> 99.99999999999999

0.01 - 0.01.prev_float     #=> 1.734723475976807e-18
1.0 - 1.0.prev_float       #=> 1.1102230246251565e-16
100.0 - 100.0.prev_float   #=> 1.4210854715202004e-14

f = 0.01; 20.times { printf "%-20a %s\n", f, f.to_s; f = f.prev_float }
#=> 0x1.47ae147ae147bp-7 0.01
#   0x1.47ae147ae147ap-7 0.009999999999999998
#   0x1.47ae147ae1479p-7 0.009999999999999997
#   0x1.47ae147ae1478p-7 0.009999999999999995
#   0x1.47ae147ae1477p-7 0.009999999999999993
#   0x1.47ae147ae1476p-7 0.009999999999999992
#   0x1.47ae147ae1475p-7 0.00999999999999999
#   0x1.47ae147ae1474p-7 0.009999999999999988
#   0x1.47ae147ae1473p-7 0.009999999999999986
#   0x1.47ae147ae1472p-7 0.009999999999999985
#   0x1.47ae147ae1471p-7 0.009999999999999983
#   0x1.47ae147ae147p-7  0.009999999999999981
#   0x1.47ae147ae146fp-7 0.00999999999999998
#   0x1.47ae147ae146ep-7 0.009999999999999978
#   0x1.47ae147ae146dp-7 0.009999999999999976
#   0x1.47ae147ae146cp-7 0.009999999999999974
#   0x1.47ae147ae146bp-7 0.009999999999999972
#   0x1.47ae147ae146ap-7 0.00999999999999997
#   0x1.47ae147ae1469p-7 0.009999999999999969
#   0x1.47ae147ae1468p-7 0.009999999999999967

static VALUE
flo_prev_float(VALUE vx)
{
    double x, y;
    x = NUM2DBL(vx);
    y = nextafter(x, -HUGE_VAL);
    return DBL2NUM(y);
}
quo(numeric) → float click to toggle source

Returns float / numeric, same as Float#/.

static VALUE
flo_quo(VALUE x, VALUE y)
{
    return num_funcall1(x, '/', y);
}
rationalize([eps]) → rational click to toggle source

Returns a simpler approximation of the value (flt-|eps| <= result <= flt+|eps|). If the optional argument eps is not given, it will be chosen automatically.

0.3.rationalize          #=> (3/10)
1.333.rationalize        #=> (1333/1000)
1.333.rationalize(0.01)  #=> (4/3)

See also Float#to_r.

static VALUE
float_rationalize(int argc, VALUE *argv, VALUE self)
{
    double d = RFLOAT_VALUE(self);

    if (d < 0.0)
        return rb_rational_uminus(float_rationalize(argc, argv, DBL2NUM(-d)));

    if (rb_check_arity(argc, 0, 1)) {
        return rb_flt_rationalize_with_prec(self, argv[0]);
    }
    else {
        return rb_flt_rationalize(self);
    }
}
round([ndigits] [, half: mode]) → integer or float click to toggle source

Returns float rounded to the nearest value with a precision of ndigits decimal digits (default: 0).

When the precision is negative, the returned value is an integer with at least ndigits.abs trailing zeros.

Returns a floating point number when ndigits is positive, otherwise returns an integer.

1.4.round      #=> 1
1.5.round      #=> 2
1.6.round      #=> 2
(-1.5).round   #=> -2

1.234567.round(2)   #=> 1.23
1.234567.round(3)   #=> 1.235
1.234567.round(4)   #=> 1.2346
1.234567.round(5)   #=> 1.23457

34567.89.round(-5)  #=> 0
34567.89.round(-4)  #=> 30000
34567.89.round(-3)  #=> 35000
34567.89.round(-2)  #=> 34600
34567.89.round(-1)  #=> 34570
34567.89.round(0)   #=> 34568
34567.89.round(1)   #=> 34567.9
34567.89.round(2)   #=> 34567.89
34567.89.round(3)   #=> 34567.89

If the optional half keyword argument is given, numbers that are half-way between two possible rounded values will be rounded according to the specified tie-breaking mode:

  • :up or nil: round half away from zero (default)

  • :down: round half toward zero

  • :even: round half toward the nearest even number

    2.5.round(half: :up)      #=> 3
2.5.round(half: :down)    #=> 2
2.5.round(half: :even)    #=> 2
3.5.round(half: :up)      #=> 4
3.5.round(half: :down)    #=> 3
3.5.round(half: :even)    #=> 4
(-2.5).round(half: :up)   #=> -3
(-2.5).round(half: :down) #=> -2
(-2.5).round(half: :even) #=> -2

static VALUE
flo_round(int argc, VALUE *argv, VALUE num)
{
    double number, f, x;
    VALUE nd, opt;
    int ndigits = 0;
    enum ruby_num_rounding_mode mode;

    if (rb_scan_args(argc, argv, "01:", &nd, &opt)) {
        ndigits = NUM2INT(nd);
    }
    mode = rb_num_get_rounding_option(opt);
    number = RFLOAT_VALUE(num);
    if (number == 0.0) {
        return ndigits > 0 ? DBL2NUM(number) : INT2FIX(0);
    }
    if (ndigits < 0) {
        return rb_int_round(flo_to_i(num), ndigits, mode);
    }
    if (ndigits == 0) {
        x = ROUND_CALL(mode, round, (number, 1.0));
        return dbl2ival(x);
    }
    if (isfinite(number)) {
        int binexp;
        frexp(number, &binexp);
        if (float_round_overflow(ndigits, binexp)) return num;
        if (float_round_underflow(ndigits, binexp)) return DBL2NUM(0);
        f = pow(10, ndigits);
        x = ROUND_CALL(mode, round, (number, f));
        return DBL2NUM(x / f);
    }
    return num;
}
to_f → self click to toggle source

Since float is already a Float, returns self.

static VALUE
flo_to_f(VALUE num)
{
    return num;
}
to_i → integer click to toggle source
to_int → integer

Returns the float truncated to an Integer.

1.2.to_i      #=> 1
(-1.2).to_i   #=> -1

Note that the limited precision of floating point arithmetic might lead to surprising results:

(0.3 / 0.1).to_i  #=> 2 (!)

to_int is an alias for to_i.

static VALUE
flo_to_i(VALUE num)
{
    double f = RFLOAT_VALUE(num);

    if (f > 0.0) f = floor(f);
    if (f < 0.0) f = ceil(f);

    return dbl2ival(f);
}
to_int → integer click to toggle source

Returns the float truncated to an Integer.

1.2.to_i      #=> 1
(-1.2).to_i   #=> -1

Note that the limited precision of floating point arithmetic might lead to surprising results:

(0.3 / 0.1).to_i  #=> 2 (!)

to_int is an alias for to_i.

static VALUE
flo_to_i(VALUE num)
{
    double f = RFLOAT_VALUE(num);

    if (f > 0.0) f = floor(f);
    if (f < 0.0) f = ceil(f);

    return dbl2ival(f);
}
to_r → rational click to toggle source

Returns the value as a rational.

2.0.to_r    #=> (2/1)
2.5.to_r    #=> (5/2)
-0.75.to_r  #=> (-3/4)
0.0.to_r    #=> (0/1)
0.3.to_r    #=> (5404319552844595/18014398509481984)

NOTE: 0.3.to_r isn't the same as “0.3”.to_r. The latter is equivalent to “3/10”.to_r, but the former isn't so.

0.3.to_r   == 3/10r  #=> false
"0.3".to_r == 3/10r  #=> true

See also Float#rationalize.

static VALUE
float_to_r(VALUE self)
{
    VALUE f;
    int n;

    float_decode_internal(self, &f, &n);
#if FLT_RADIX == 2
    if (n == 0)
        return rb_rational_new1(f);
    if (n > 0)
        return rb_rational_new1(rb_int_lshift(f, INT2FIX(n)));
    n = -n;
    return rb_rational_new2(f, rb_int_lshift(ONE, INT2FIX(n)));
#else
    f = rb_int_mul(f, rb_int_pow(INT2FIX(FLT_RADIX), n));
    if (RB_TYPE_P(f, T_RATIONAL))
        return f;
    return rb_rational_new1(f);
#endif
}
to_s → string click to toggle source

Returns a string containing a representation of self. As well as a fixed or exponential form of the float, the call may return NaN, Infinity, and -Infinity.

static VALUE
flo_to_s(VALUE flt)
{
    enum {decimal_mant = DBL_MANT_DIG-DBL_DIG};
    enum {float_dig = DBL_DIG+1};
    char buf[float_dig + (decimal_mant + CHAR_BIT - 1) / CHAR_BIT + 10];
    double value = RFLOAT_VALUE(flt);
    VALUE s;
    char *p, *e;
    int sign, decpt, digs;

    if (isinf(value)) {
        static const char minf[] = "-Infinity";
        const int pos = (value > 0); /* skip "-" */
        return rb_usascii_str_new(minf+pos, strlen(minf)-pos);
    }
    else if (isnan(value))
        return rb_usascii_str_new2("NaN");

    p = ruby_dtoa(value, 0, 0, &decpt, &sign, &e);
    s = sign ? rb_usascii_str_new_cstr("-") : rb_usascii_str_new(0, 0);
    if ((digs = (int)(e - p)) >= (int)sizeof(buf)) digs = (int)sizeof(buf) - 1;
    memcpy(buf, p, digs);
    xfree(p);
    if (decpt > 0) {
        if (decpt < digs) {
            memmove(buf + decpt + 1, buf + decpt, digs - decpt);
            buf[decpt] = '.';
            rb_str_cat(s, buf, digs + 1);
        }
        else if (decpt <= DBL_DIG) {
            long len;
            char *ptr;
            rb_str_cat(s, buf, digs);
            rb_str_resize(s, (len = RSTRING_LEN(s)) + decpt - digs + 2);
            ptr = RSTRING_PTR(s) + len;
            if (decpt > digs) {
                memset(ptr, '0', decpt - digs);
                ptr += decpt - digs;
            }
            memcpy(ptr, ".0", 2);
        }
        else {
            goto exp;
        }
    }
    else if (decpt > -4) {
        long len;
        char *ptr;
        rb_str_cat(s, "0.", 2);
        rb_str_resize(s, (len = RSTRING_LEN(s)) - decpt + digs);
        ptr = RSTRING_PTR(s);
        memset(ptr += len, '0', -decpt);
        memcpy(ptr -= decpt, buf, digs);
    }
    else {
      exp:
        if (digs > 1) {
            memmove(buf + 2, buf + 1, digs - 1);
        }
        else {
            buf[2] = '0';
            digs++;
        }
        buf[1] = '.';
        rb_str_cat(s, buf, digs + 1);
        rb_str_catf(s, "e%+03d", decpt - 1);
    }
    return s;
}
Also aliased as: inspect
truncate([ndigits]) → integer or float click to toggle source

Returns float truncated (toward zero) to a precision of ndigits decimal digits (default: 0).

When the precision is negative, the returned value is an integer with at least ndigits.abs trailing zeros.

Returns a floating point number when ndigits is positive, otherwise returns an integer.

2.8.truncate           #=> 2
(-2.8).truncate        #=> -2
1.234567.truncate(2)   #=> 1.23
34567.89.truncate(-2)  #=> 34500

Note that the limited precision of floating point arithmetic might lead to surprising results:

(0.3 / 0.1).truncate  #=> 2 (!)

static VALUE
flo_truncate(int argc, VALUE *argv, VALUE num)
{
    if (signbit(RFLOAT_VALUE(num)))
        return flo_ceil(argc, argv, num);
    else
        return flo_floor(argc, argv, num);
}
zero? → true or false click to toggle source

Returns true if float is 0.0.

static VALUE
flo_zero_p(VALUE num)
{
    return flo_iszero(num) ? Qtrue : Qfalse;
}