class Float - Documentation for Ruby 2.3.0 (original) (raw)

BigDecimal extends the native Float class to provide the to_d method.

When you require BigDecimal in your application, this method will be available on Float objects.

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:

• docs.sun.com/source/806-3568/ncg_goldberg.html
• wiki.github.com/rdp/ruby_tutorials_core/ruby-talk-faq#wiki-floats_imprecise
• en.wikipedia.org/wiki/Floating_point#Accuracy_problems

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 posable 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

Represents the rounding mode for floating point addition.

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

Return the modulo after division of float by other.

6543.21.modulo(137)
6543.21.modulo(137.24)

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.

static VALUE flo_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

static VALUE flo_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_funcall(rb_complex_raw1(x), idPow, 1, y); } 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.

static VALUE flo_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.

static VALUE flo_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.

static VALUE flo_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.

static VALUE flo_div(VALUE x, VALUE y) { long f_y; double d;

if (RB_TYPE_P(y, T_FIXNUM)) {
    f_y = FIX2LONG(y);
    return DBL2NUM(RFLOAT_VALUE(x) / (double)f_y);
}
else if (RB_TYPE_P(y, T_BIGNUM)) {
    d = rb_big2dbl(y);
    return DBL2NUM(RFLOAT_VALUE(x) / d);
}
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 < real → true or false click to toggle source

Returns true if float is less than real.

The result of NaN < NaN is undefined, so the 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 -FIX2INT(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 the 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 -FIX2INT(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, +1 or nil depending on whether float is less than, equal to, or greater than real. This is the basis for the tests in Comparable.

The result of NaN <=> NaN is undefined, so the 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 INT2FIX(-FIX2INT(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.

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

1.0 == 1

static VALUE flo_eq(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.

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

1.0 == 1

static VALUE flo_eq(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 the implementation-dependent value is returned.

static VALUE flo_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 -FIX2INT(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 the 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 -FIX2INT(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.abs

static VALUE flo_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 → integer click to toggle source

Returns the smallest Integer greater than or equal to float.

1.2.ceil
2.0.ceil
(-1.2).ceil
(-2.0).ceil

static VALUE flo_ceil(VALUE num) { double f = ceil(RFLOAT_VALUE(num)); long val;

if (!FIXABLE(f)) {
    return rb_dbl2big(f);
}
val = (long)f;
return LONG2FIX(val);

}

coerce(numeric) → array click to toggle source

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

This is achieved by converting a numeric to a Float.

1.2.coerce(3)
2.5.coerce(1.1)

static VALUE flo_coerce(VALUE x, VALUE y) { return rb_assoc_new(rb_Float(y), x); }

dclone() click to toggle source

denominator → integer click to toggle source

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

See numerator.

static VALUE float_denominator(VALUE self) { double d = RFLOAT_VALUE(self); if (isinf(d) || isnan(d)) return INT2FIX(1); return rb_call_super(0, 0); }

divmod(numeric) → array click to toggle source

See Numeric#divmod.

42.0.divmod 6 42.0.divmod 5

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.

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

1.0.eql?(1)

static VALUE flo_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 rb_funcall(x, '/', 1, y); }

finite? → true or false click to toggle source

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

static VALUE 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 → integer click to toggle source

Returns the largest integer less than or equal to float.

1.2.floor
2.0.floor
(-1.2).floor
(-2.0).floor

static VALUE flo_floor(VALUE num) { double f = floor(RFLOAT_VALUE(num)); long val;

if (!FIXABLE(f)) {
    return rb_dbl2big(f);
}
val = (long)f;
return LONG2FIX(val);

}

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? → nil, -1, +1 click to toggle source

Return values corresponding to the value of float:

finite

nil

-Infinity

-1

+Infinity

1

For example:

(0.0).infinite?
(-1.0/0.0).infinite?
(+1.0/0.0).infinite?

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

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

return Qnil;

}

magnitude → float click to toggle source

Returns the absolute value of float.

(-34.56).abs
-34.56.abs

static VALUE flo_abs(VALUE flt) { double val = fabs(RFLOAT_VALUE(flt)); return DBL2NUM(val); }

modulo(other) → float click to toggle source

Return the modulo after division of float by other.

6543.21.modulo(137)
6543.21.modulo(137.24)

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
a.nan?
a = 0.0/0.0
a.nan?

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:

p 0.01.next_float
p 1.0.next_float
p 100.0.next_float

p 0.01.next_float - 0.01
p 1.0.next_float - 1.0
p 100.0.next_float - 100.0

f = 0.01; 20.times { printf "%-20a %s\n", f, f.to_s; f = f.next_float }

f = 0.0 100.times { f += 0.1 } p f
p 10-f
p(10.0.next_float-10)
p((10-f)/(10.0.next_float-10)) p((10-f)/(10*Float::EPSILON))
p "%a" % f

static VALUE flo_next_float(VALUE vx) { double x, y; x = NUM2DBL(vx); y = nextafter(x, INFINITY); return DBL2NUM(y); }

numerator → integer click to toggle source

Returns the numerator. The result is machine dependent.

n = 0.3.numerator
d = 0.3.denominator
n.fdiv(d)

static VALUE float_numerator(VALUE self) { double d = RFLOAT_VALUE(self); if (isinf(d) || isnan(d)) return self; return rb_call_super(0, 0); }

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:

p 0.01.prev_float
p 1.0.prev_float
p 100.0.prev_float

p 0.01 - 0.01.prev_float
p 1.0 - 1.0.prev_float
p 100.0 - 100.0.prev_float

f = 0.01; 20.times { printf "%-20a %s\n", f, f.to_s; f = f.prev_float }

static VALUE flo_prev_float(VALUE vx) { double x, y; x = NUM2DBL(vx); y = nextafter(x, -INFINITY); 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 rb_funcall(x, '/', 1, y); }

rationalize([eps]) → rational click to toggle source

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

0.3.rationalize
1.333.rationalize
1.333.rationalize(0.01)

See to_r.

static VALUE float_rationalize(int argc, VALUE *argv, VALUE self) { VALUE e;

if (f_negative_p(self))
    return f_negate(float_rationalize(argc, argv, f_abs(self)));

rb_scan_args(argc, argv, "01", &e);

if (argc != 0) {
    return rb_flt_rationalize_with_prec(self, e);
}
else {
    return rb_flt_rationalize(self);
}

}

round([ndigits]) → integer or float click to toggle source

Rounds float to a given precision in decimal digits (default 0 digits).

Precision may be negative. Returns a floating point number when ndigits is more than zero.

1.4.round
1.5.round
1.6.round
(-1.5).round

1.234567.round(2)
1.234567.round(3)
1.234567.round(4)
1.234567.round(5)

34567.89.round(-5) 34567.89.round(-4) 34567.89.round(-3) 34567.89.round(-2) 34567.89.round(-1) 34567.89.round(0)
34567.89.round(1)
34567.89.round(2)
34567.89.round(3)

static VALUE flo_round(int argc, VALUE *argv, VALUE num) { VALUE nd; double number, f, x; int ndigits = 0; int binexp; enum {float_dig = DBL_DIG+2};

if (argc > 0 && rb_scan_args(argc, argv, "01", &nd) == 1) {
    ndigits = NUM2INT(nd);
}
if (ndigits < 0) {
    return int_round_0(flo_truncate(num), ndigits);
}
number  = RFLOAT_VALUE(num);
if (ndigits == 0) {
    return dbl2ival(number);
}
frexp(number, &binexp);

/* Let exp be such that number is written as:"0.#{digits}e#{exp}", i.e. such that 10 ** (exp - 1) <= |number| < 10 ** exp Recall that up to float_dig digits can be needed to represent a double, so if ndigits + exp >= float_dig, the intermediate value (number * 10 ** ndigits) will be an integer and thus the result is the original number. If ndigits + exp <= 0, the result is 0 or "1e#{exp}", so if ndigits + exp < 0, the result is 0. We have: 2 ** (binexp-1) <= |number| < 2 ** binexp 10 ** ((binexp-1)/log_2(10)) <= |number| < 10 ** (binexp/log_2(10)) If binexp >= 0, and since log_2(10) = 3.322259: 10 ** (binexp/4 - 1) < |number| < 10 ** (binexp/3) floor(binexp/4) <= exp <= ceil(binexp/3) If binexp <= 0, swap the /4 and the /3 So if ndigits + floor(binexp/(4 or 3)) >= float_dig, the result is number If ndigits + ceil(binexp/(3 or 4)) < 0 the result is 0 */ if (isinf(number) || isnan(number) || (ndigits >= float_dig - (binexp > 0 ? binexp / 4 : binexp / 3 - 1))) { return num; } if (ndigits < - (binexp > 0 ? binexp / 3 + 1 : binexp / 4)) { return DBL2NUM(0); } f = pow(10, ndigits); x = round(number * f); if (x > 0) { if ((double)((x + 0.5) / f) <= number) x += 1; } else { if ((double)((x - 0.5) / f) >= number) x -= 1; } return DBL2NUM(x / f);}

/*

• call-seq:

• float.to_i      ->  integer

• float.to_int    ->  integer

• float.truncate  ->  integer

• Returns the +float+ truncated to an Integer.
• Synonyms are #to_i, #to_int, and #truncate. */

static VALUE flo_truncate(VALUE num) { double f = RFLOAT_VALUE(num); long val;

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

if (!FIXABLE(f)) {
    return rb_dbl2big(f);
}
val = (long)f;
return LONG2FIX(val);

}

to_d → bigdecimal click to toggle source

Convert flt to a BigDecimal and return it.

require 'bigdecimal' require 'bigdecimal/util'

0.5.to_d

def to_d(precision=nil) BigDecimal(self, precision || Float::DIG) end

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.

Synonyms are to_i, to_int, and truncate.

static VALUE flo_truncate(VALUE num) { double f = RFLOAT_VALUE(num); long val;

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

if (!FIXABLE(f)) {
    return rb_dbl2big(f);
}
val = (long)f;
return LONG2FIX(val);

}

to_int → integer click to toggle source

Returns the float truncated to an Integer.

Synonyms are to_i, to_int, and truncate.

static VALUE flo_truncate(VALUE num) { double f = RFLOAT_VALUE(num); long val;

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

if (!FIXABLE(f)) {
    return rb_dbl2big(f);
}
val = (long)f;
return LONG2FIX(val);

}

to_r → rational click to toggle source

Returns the value as a rational.

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.

2.0.to_r
2.5.to_r
-0.75.to_r
0.0.to_r

See rationalize.

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

float_decode_internal(self, &f, &n);

#if FLT_RADIX == 2 { long ln = FIX2LONG(n);

    if (ln == 0)
        return f_to_r(f);
    if (ln > 0)
        return f_to_r(f_lshift(f, n));
    ln = -ln;
    return rb_rational_new2(f, f_lshift(ONE, INT2FIX(ln)));
}

#else return f_to_r(f_mul(f, f_expt(INT2FIX(FLT_RADIX), n))); #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))
    return rb_usascii_str_new2(value < 0 ? "-Infinity" : "Infinity");
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;

}

truncate → integer click to toggle source

Returns the float truncated to an Integer.

Synonyms are to_i, to_int, and truncate.

static VALUE flo_truncate(VALUE num) { double f = RFLOAT_VALUE(num); long val;

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

if (!FIXABLE(f)) {
    return rb_dbl2big(f);
}
val = (long)f;
return LONG2FIX(val);

}

zero? → true or false click to toggle source

Returns true if float is 0.0.

static VALUE flo_zero_p(VALUE num) { if (RFLOAT_VALUE(num) == 0.0) { return Qtrue; } return Qfalse; }