11:05 <wmnnd> Hey there, my implementation of named captures using XRegExp seems to be working fine now :-)

11:06 <wmnnd> So now I wonder what I have to do in order to do the PR, especially since I still haven't figured out how/where I should include the XRegExp library in the Opal runtime and where/how the author of XRE needs to be mentioned.

11:07 <wmnnd> Also, in the master branch, using rackup to get (I suppose) opal-rspec up doesn't work, so I'm also not quite sure how to verify whether my changes pass all required tests.

11:33 <adambeynon> wmnnd: so are all generated regexps, by the compiler, being passed through XRegExp in javascript?

11:54 <wmnnd> adambeynon: Yes.

11:54 <wmnnd> adambeynon: I made it so that the compiler has all Regexps go through Opal's Regexp.new method.

12:58 <wmnnd> Oh well, I have created the PR anyways and you guys can decide how including XRegExp should be handled :-)

15:50 <elia> wmnnd, good job :)

18:19 <opal2> !balance

21:10 <elia> adambeynon, I don't think I'll ever understand the core relations between classes and prototypes in runtime.js :(

21:12 <adambeynon> elia: lol

21:12 <adambeynon> i hate having to go back and understand it all again when something needs fixing

21:12 <elia> lol indeed

21:13 <adambeynon> elia: what part are you trying to work through

21:13 <adambeynon> where classes inherit from?

21:13 <elia> yeh, core relations between classes, their alloc / prototypes

21:14 <elia> and stuff stored in the class objects (which is silly to say, bc it means everything)

21:14 <elia> what I mean is $$alloc vs. $$parent vs. $$proto

21:15 <elia> I'm trying to get module inheritance done

21:16 <elia> adambeynon, I even tried to split runtime.js in the process

21:16 <adambeynon> $$alloc is a function, whose prototype is $$proto

21:16 <adambeynon> new Opal.Object.$$alloc()

21:16 <adambeynon> ===

21:16 <adambeynon> Object.new

21:16 <elia> k

21:16 <adambeynon> $$parent is

21:16 <adambeynon> er

21:16 <adambeynon> no idea

21:16 <elia> lol

21:17 <adambeynon> I think it points to the superclass

21:17 <adambeynon> but it might also include modules

21:17 <elia> (I think we'll need to precisely document that stuff)

21:17 <adambeynon> Opal.Object.$$parent = Opal.Kernel

21:17 <adambeynon> probably

21:17 <elia> you remember my plan for modules?

21:17 <adambeynon> 2 prototypes yeh?

21:18 <elia> yes, I kinda refined it, bc it's like a linked list and it probably doesn't need the top one and to bottom proto can be the class itself (until #prepend at least)

21:19 <elia> so I want to switch the next prototype in the class chain and inject the module (via a ghost class)

21:20 <elia> and make the ghost class parent proto point to the previous A-class parent

21:20 <elia> does it make sense?

21:20 <elia> then the Opal.donate syncronization must update the ghost classes instead of the actual classes prototypes

21:21 <elia> (adambeynon)

21:32 <elia> adambeynon, well, it's good to know that I'm not alone in finding that stuff difficult, I'll sleep it off

21:32 <elia> :)