"0u8.." panics

[Christopher22]

The range for u8 does not work properly, it panics at the (valid) value 255. If this behavior is expected for bytes it should be documentated.

I tried this code to check correctness and it panics:
fn main() { assert_eq!((0u8..).last(), Some(255)); }

thread 'main' panicked at 'arithmetic operation overflowed', ../src/libcore/ops.rs:211
note: Run with RUST_BACKTRACE=1 for a backtrace.

Rusty greedings, Christopher
#24119

[TimNN]

Mh, this issue looks interesting. Frist of all, in my opinition, (0..).last() would always overflow / be an "endless loop" since you're trying to get the last element of an open ended sequence.

I've been trying some things and found the following "interesting behaviour":

The code println!("{:?}", (255u8..).next());:

• tiggers arithmetic operation overflowed on stable/debug
• causes timeout triggered! (playpen) on beta/debug
• prints Some(255) on nightly/debug
• prints Some(255) on */release

[nagisa]

prints Some(255) on nightly/debug

That’s amusing. I feel like overflow on debug was intentional (i.e. the stable/debug behaviour is the intended one).

[nagisa]

Duplicate of #25708 by the way, but @TimNN points out a regression, so probably shouldn’t be closed?

[Mark-Simulacrum]

Looks like another regression that slipped through. cc @alexcrichton @brson; I don't know what we want to do about this issue, but if the regression is no longer a priority, it can be closed as a duplicate. Nightly output today, without release optimizations, is below. With them, it simply hangs.

thread 'main' panicked at 'attempt to add with overflow', /home/mark/BuildDisk/rust/src/libcore/ops.rs:264
note: Run with `RUST_BACKTRACE=1` for a backtrace.

[nagisa]

This is what the documentation for last has to say:

This method will evaluate the iterator until it returns None. While doing so, it keeps track of the current element. After None is returned, last() will then return the last element it saw.

Iterator 0u8.. can never return a None by definition, making last on this iterator divergent.

The fact that 0u8.. cannot yield 255 at all is tracked at issue #25708. Therefore closing as a duplicate.