18:45 <SeanTAllen> Hi all, those of us over at Wallaroo Labs officially opened up the Pony codebase of our fast, elastic data processing engine Wallaroo. If you've wanted to peer into the guts of a Pony application, now is your chance. Blog post announcing the release is at https://blog.wallaroolabs.com/2017/09/open-sourcing-wallaroo/. Enjoy!

20:30 <tokenrove> SeanTAllen: this is great! I think the biggest problem for Pony right now is just the lack of large-scale examples so this makes a big difference.

21:03 <vaninwagen> SeanTAllen let me cite yourself to express how i feel about wallaroo: ᕕ( ᐛ )ᕗ

21:04 <SeanTAllen> ha

21:04 <SeanTAllen> We are very excited and nervous. Lots of good stuff there. Lots more work to do.

21:04 <SeanTAllen> I was about to say "there's an IRC channel" if you want to join, but I see you already did.

21:07 <vaninwagen> i dont want to miss out

21:55 <endformationage> Am I able to use `iftype` to specialize a generic function `get_length` to return a F32 when T is of type I32?

21:55 <endformationage> http://playground.ponylang.org/?gist=0d66de9ff26c4a89d14eb22855038d00

21:56 <SeanTAllen> i havent used iftype endformationage so someone else will have to answer that.

21:58 <endformationage> No problem! Thanks for checking anyhow.

21:58 <endformationage> ... my attempt from what I could gather from the RFC.

22:03 <jemc> endformationage: that `iftype` RFC is only half-implemented - we have `iftype` as a branching condition that can be used inside a method body, but we don't have iftype as a method signature guard

22:03 <jemc> that's why the parser is complaining at you

22:06 <endformationage> Ah, that makes sense!

22:06 <jemc> so in other words you can't yet vary the method signature based on an iftype condition, but you can vary the logic insie

22:06 <jemc> s/insie/inside/

22:06 <endformationage> Got it.

22:07 <endformationage> Sounds like it wil be pretty useful

22:07 <jemc> if you want to share more about what you're trying to accomplish at a higher level I may be able to give another suggestion

22:07 <jemc> but don't feel obligated to do so ;)

22:08 <endformationage> I will simply expand my types, they are not super complicated.

22:08 <endformationage> I thought I'd just use the feature if it were available

22:11 <jemc> yeah, I'm super excited about getting that feature available as well

22:11 <endformationage> I am writing some Vector types to use with my pony-kore wrapper around Kode's Kore framework, as seen https://github.com/Kode/Kore/blob/master/Sources/Kore/Math/Vector.h

22:11 <jemc> we'll be able to do lots of nifty stuff, like make `Array[A]` a subtype of `Stringable` iff `A` is a subtype of `Stringable`

22:12 <endformationage> Those are written with C++ templates, hence my question regarding type expression params the other day

22:13 <jemc> you could actually do something very similar to what they're doing there for `LengthType`

22:14 <endformationage> Ah, yeah I'll give it a go

22:14 <endformationage> Just a second type param I suppose

22:17 <endformationage> Lol, I just didn't notice before as I'm referencing Kode's other framework which doesn't use gernerics in the same way.

22:18 <jemc> http://playground.ponylang.org/?gist=735e9f63c30877e0a79a94dd720cf46c

22:19 <jemc> yeah, there's a quick example using a second type param

22:20 <endformationage> Thanks.

22:23 <endformationage> Hmm, cannot open the gist.

22:23 <endformationage> There it goes