How to deal with non-integer Range limits

[dloscutoff]

Currently, if you construct a Range from non-integers, both values are truncated to ints. For example, 1.2,3.9 gives the same result as 1,3, and -3.9,-1.2 gives the same result as -3,-1. Is this the desired behavior?

• Do we always want to convert Range bounds to integers, or would it ever be useful to store a Range object with float bounds (regardless of how that is interpreted)?
  • In Python, you can't have a range with floats, but you can have a slice with floats... not sure what the use cases for that are.
• Intuitively, it seems like the upper bound should represent "up to but not including"; shouldn't that mean that a Range with an upper bound of 3.9 should include 3, since it's up to but not including 3.9?