What's the difference between Float#floor and Float#to_i in Ruby?

Question

I'm learning Ruby, and I'm currently in numbers. So as I understand it, there are five ways (perhaps more) to coerce integers and floats to each other:

• Integer#to_f: Coerce to a new float
• Float#ceil: Round up to the nearest integer
• Float#floor: Round down to the nearest integer
• Float#round: Round to the nearest integer
• Float#to_i: Truncate to the nearest integer

What's the difference between "rounding down" and "truncating" to the nearest integer?

When I tested it...

puts 34.4.to_i()
puts 34.4.floor()

... it resulted in the same values:

34
34

Answer 1

• floor (without arguments) returns the next integer less than or equal to the receiver
• ceil (without arguments) returns the next integer greater than or equal to the receiver
• to_i discards the fractional part of the receiver and returns the integer part

f	f.to_i	f.floor	f.ceil	f.round
11.8	11	11	12	12
11.5	11	11	12	12
11.2	11	11	12	11
11.0	11	11	11	11
-11.2	-11	-12	-11	-11
-11.5	-11	-12	-11	-12
-11.8	-11	-12	-11	-12

to_i behaves like floor for positive numbers and like ceil for negative numbers.

In fact, that's actually how Float#to_i is implemented in numeric.c:

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);
}

There's also truncate which behaves like to_i but takes an optional number of digits (like floor, ceil, and round).