```fun main() { val num: Number = 25 println(num > 10) }``` Shouldn't something like this...

# stdlib

[jeggy]

fun main() {
    val num: Number = 25
    println(num > 10)
}

Shouldn't something like this be possible by the stdlib?

[jw]

How would it be implemented?

[jw]

I can create an infinite-precision Number subtype and represent a number outside the valid range of all available conversion functions

[jeggy]

operator fun Number.compareTo(i: Int) = toLong().compareTo(i.toLong())
operator fun Number.compareTo(i: Long) = toLong().compareTo(i)

I created these

[jw]

i think that means 0.2 > 0.1 would return false

[jw]

It also will return false for really high values of BigInteger given its longValue() implementation states:

if this BigInteger is too big to fit in a long, only the low-order 64 bits are returned.

[andries.fc]

Problem is the number class has not to indicate the precision of the number. To make a comparison you have to convert to the lowest precision of the two numbers.