01:03 <travis-ci> [travis-ci] Build details : http://travis-ci.org/manastech/crystal/builds/30022858

01:03 <travis-ci> [travis-ci] Change view : https://github.com/manastech/crystal/compare/463bd4b6db0e...7873ddc0755d

01:03 <travis-ci> [travis-ci] manastech/crystal#1262 (master - 7873ddc : Ary Borenszweig): The build passed.

01:15 <travis-ci> [travis-ci] Build details : http://travis-ci.org/manastech/crystal/builds/30025916

01:15 <travis-ci> [travis-ci] Change view : https://github.com/manastech/crystal/compare/7873ddc0755d...a5953849b363

01:15 <travis-ci> [travis-ci] manastech/crystal#1263 (master - a595384 : Ary Borenszweig): The build passed.

04:04 <travis-ci> [travis-ci] Build details : http://travis-ci.org/manastech/crystal/builds/30041963

04:04 <travis-ci> [travis-ci] Change view : https://github.com/manastech/crystal/commit/25196bcdfc8f

04:04 <travis-ci> [travis-ci] manastech/crystal#1264 (llvm-config - 25196bc : Juan Wajnerman): The build passed.

14:24 <kostya> hi, did you see swift? https://developer.apple.com/library/prerelease/ios/documentation/Swift/Conceptual/Swift_Programming_Language/GuidedTour.html#//apple_ref/doc/uid/TP40014097-CH2-XID_1

14:24 <kostya> looks like crystal

14:44 <asterite> Definitely :)

14:44 <asterite> The other day I read this article: http://www.mrspeaker.net/2014/06/04/selling-swift/

14:53 <asterite> So Swift = Python + Ruby + Javascript + Haskell + Go + Kotlin + Lua + Typescript + Coffeescript + Groovy + Rust + ...

15:47 <farleyknight> kostya, I'm not seeing it

15:47 <farleyknight> Way more like ECMAScript than anything

19:14 <asterite> hi emmanueloga!

20:32 <farleyknight> asterite, Can you explain this behavior? https://gist.github.com/farleyknight/4a7fd7ef1330ac9ca431

20:33 <asterite> Yes

20:33 <asterite> Crystal determines that @x is nilable, because there's one initialize that doesn't initialize @x to any value

20:33 <asterite> You can do: if x = @x; (x + 3).to_s; end

20:34 <asterite> Because doing "if" with an instance variable doesn't make the instance variable not-nil inside the "if" (unlike regular variables)

20:34 <farleyknight> So you can't infer that the @x inside the if statement is not nil?

20:34 <asterite> You can read more here: http://crystal-lang.org/docs/syntax_and_semantics/if_var.html

20:35 <asterite> No, because maybe between that statement and the next one another thread makes @x nil

20:35 <asterite> At least that's our excuse for not implementing it :-P

20:35 <asterite> But another thread can't change a local variable

20:36 <farleyknight> Okay, I've used try a few times.. It seems the the appropriate thing here..

20:37 <bcardiff> farleyknight might be expecting the if @x to restrict @x to be non nillable. right?

20:37 <asterite> Yes, but that's not safe, at least not in a concurrent world

20:37 <farleyknight> Will there ever be "locked" classes where instance variables can't be modified by outside threads?

20:38 <waj> I think it would be fairly complicated for the compiler to deduce that

20:41 <emmanueloga> hey asterite :)

20:41 <farleyknight> I think the try idiom is a fair enough replacement for that bit of inference. I was just a bit confused why it worked fine for local variables but not instance variables!

20:43 <asterite> Yes… `try` and `if var = @var` are the two alternatives that we have right now