02:26 <FromGitter> <watzon> Well I broke crystal play with this ⏎ ⏎ ```(1..10).step(-1).each { |i| puts i }``` [https://gitter.im/crystal-lang/crystal?at=5def0248f65fec1c8ef3067f]

02:26 <FromGitter> <watzon> Actually just broke crystal in general haha

02:27 <FromGitter> <watzon> Has to be a bug

02:31 <FromGitter> <watzon> #8572

02:31 <DeBot> https://github.com/crystal-lang/crystal/issues/8572 (Negative step size causes infinite loop)

02:32 <FromGitter> <Blacksmoke16> https://play.crystal-lang.org/#/r/869y

02:32 <FromGitter> <Blacksmoke16> did i find my own bug? :p

02:33 <FromGitter> <watzon> I'd think that should work

02:34 <FromGitter> <Blacksmoke16> indeed

02:36 <FromGitter> <tenebrousedge> since when does `step` take negative integers?

02:39 <FromGitter> <watzon> That's the point, it doesn't seem like it does

02:39 <FromGitter> <watzon> But it doesn't throw an error

02:39 <FromGitter> <watzon> Also I don't think it would be too hard to add support for negative integers if we wanted

02:40 <FromGitter> <tenebrousedge> hmm, maybe. What should `0` do?

02:40 <FromGitter> <tenebrousedge> infinite loop?

02:40 <FromGitter> <tenebrousedge> error?

02:41 <FromGitter> <watzon> I'd think that should always error, regardless of what negatives do

02:41 <FromGitter> <watzon> You can't step by zero

02:42 <FromGitter> <tenebrousedge> hmm, `(1.0..1.3).step(0.1)` doesn't work

02:42 <FromGitter> <watzon> Weird, with a float range it should

02:43 <FromGitter> <tenebrousedge> apparently it uses `times`

02:43 <FromGitter> <tenebrousedge> and you can't do `0.1.times do`

02:43 <FromGitter> <watzon> I have to say, the way Python handles slices is kinda nice

02:44 <FromGitter> <tenebrousedge> no o___o it's not allowed to say nice things about Python

02:44 <FromGitter> <watzon> `arr[1:10:2]` from 1 to 10 step by 2

02:44 <FromGitter> <watzon> I know I'm gonna burn in developer hell now

02:44 <FromGitter> <tenebrousedge> flaaaaames! 🔥

02:45 <FromGitter> <watzon> I'll be forced to work with Java for the rest of my life, whilst running windows xp on an Intel Atom processor and my clients will never pay me

02:46 <FromGitter> <tenebrousedge> and your web target will be IE 6 on Windows 10

02:47 <FromGitter> <tenebrousedge> what about `arr[1..10, 2]` ?

02:47 <FromGitter> <watzon> It would be nice to have that imo

02:47 <FromGitter> <tenebrousedge> no, it's already taken

02:48 <FromGitter> <tenebrousedge> wait, that's for assignment

02:48 <FromGitter> <watzon> Yeah it's not taken yet

02:49 <FromGitter> <watzon> It would be nice if it could do what python does with negative indexes too

02:49 <FromGitter> <tenebrousedge> so does that grab every second element, or should it do `each_slice(2)` with the given range of elements?

02:50 <FromGitter> <watzon> Well if it's doing what `step` does it should grab every second element

02:50 <FromGitter> <tenebrousedge> is that the least surprising behavior?

02:52 <FromGitter> <watzon> It would be for me, but idk

02:52 <FromGitter> <watzon> I don't think it's likely to get implemented anyway

04:11 <FromGitter> <wontruefree> This might seem like a crazy question be is there a list of small tasks that can be done in the compiler?

08:02 <FromGitter> <bew> What kind of tasks?

09:29 <FromGitter> <yxhuvud> I'd guess he want bite sized tasks that he can learn the ropes on. I'd just browse the open tickets of tickets tagged as compiler and see if something might seem simple.

14:50 <FromGitter> <christopherzimmerman> @watzon I'm running into the same thing with slices right now... There is no good syntax for strided ranges. Right now, for strides != 1, I use `{..., -1}` or `{1...-1, 2}`, but it's so messy.

14:50 <FromGitter> <christopherzimmerman> I want Python syntax :(