fkchang has quit [Ping timeout: 268 seconds]
elia has quit [Quit: Computer has gone to sleep.]
meh` has quit [Ping timeout: 260 seconds]
elia has joined #opal
elia has quit [Remote host closed the connection]
Nick____ has quit [Quit: Page closed]
<
shawn42>
anyone still around?
eventualbuddha has joined #opal
lupine has quit [Ping timeout: 245 seconds]
lectrick has quit [Ping timeout: 272 seconds]
lectrick has joined #opal
eventualbuddha has quit [Ping timeout: 256 seconds]
e_dub has joined #opal
e_dub has quit [Ping timeout: 245 seconds]
e_dub has joined #opal
e_dub has quit [Quit: It's a hard knock life]
GitHub76 has joined #opal
<
GitHub76>
opal/master edddc0f Adam Beynon: Explicitly add compiler options to compiler and cli runner
GitHub76 has left #opal [#opal]
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<
travis-ci>
[travis-ci] opal/opal#1211 (master - edddc0f : Adam Beynon): The build passed.
GitHub51 has joined #opal
<
GitHub51>
opal/master e6d2a7d Adam Beynon: Fix compiler flag for dynamic require errors
GitHub51 has left #opal [#opal]
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<
travis-ci>
[travis-ci] opal/opal#1212 (master - e6d2a7d : Adam Beynon): The build passed.
adambeynon has joined #opal
lupine has joined #opal
elia has joined #opal
<
adambeynon>
elia: on master, Opal::Parser is now Opal::Compiler and Opal.parse() is now Opal.compile()
<
adambeynon>
just a heads up
<
adambeynon>
I can fix those
<
adambeynon>
but I didnt know if you were running an apps off master
elia has quit [Quit: Computer has gone to sleep.]
elia has joined #opal
<
elia>
adambeynon, thanks, perhaps we can add a deprecated.rb which defines const_missing to print warnings…
GitHub12 has joined #opal
GitHub12 has left #opal [#opal]
<
GitHub12>
opal/master 689a501 Adam Beynon: Seperate grammar(parser) and lexer into 2 classes
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1213 (master - 689a501 : Adam Beynon): The build was broken.
travis-ci has left #opal [#opal]
GitHub71 has joined #opal
<
GitHub71>
opal/master c510d49 Adam Beynon: Fix private method error (Parser#do_parse)
GitHub71 has left #opal [#opal]
GitHub192 has joined #opal
<
GitHub192>
opal/master 3299ba3 Adam Beynon: Use self (instead of this) for obj reference in all super calls
GitHub192 has left #opal [#opal]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1214 (master - c510d49 : Adam Beynon): The build was fixed.
travis-ci has left #opal [#opal]
GitHub32 has joined #opal
<
GitHub32>
opal/master cda720f Dan Allen: implemented Kernel#warn...
GitHub32 has left #opal [#opal]
<
GitHub32>
opal/master 1eef4d8 Adam Beynon: Merge pull request #410 from mojavelinux/kernel-warn...
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1215 (master - 3299ba3 : Adam Beynon): The build was fixed.
travis-ci has left #opal [#opal]
GitHub12 has joined #opal
<
GitHub12>
opal/master 515efb9 Adam Beynon: Add Template.paths as list of registered templates (#412)
GitHub12 has left #opal [#opal]
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<
travis-ci>
[travis-ci] opal/opal#1216 (master - 1eef4d8 : Adam Beynon): The build passed.
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1217 (master - 515efb9 : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
elia has quit [Quit: (IRC Client: textualapp.com)]
DouweM has joined #opal
meh` has joined #opal
<
meh`>
adambeynon, yo
elia has joined #opal
<
meh`>
adambeynon, I just realized I actually need array inheritance in opal-browser
<
adambeynon>
meh`: I think method_missing might be our answer
<
adambeynon>
with the inherited swizziling
<
meh`>
adambeynon, yes
<
meh`>
that's what I'm going to do
<
meh`>
but your trick doesn't work
<
adambeynon>
what was wrong with it?
<
meh`>
adambeynon, does you trick work on class methods too?
<
adambeynon>
meh`: well, it can do
<
adambeynon>
just need to add the 2 extra properties
<
adambeynon>
meh`: actually
<
adambeynon>
forget that
<
adambeynon>
it works fine
<
adambeynon>
it doesnt matter about class methods
<
meh`>
it doesn't :(
<
adambeynon>
super() will just skip the super and go to Arrat
<
adambeynon>
ARRAY*
<
adambeynon>
I cant type in the cold weather :'(
<
meh`>
adambeynon, with the swizzling, when I call Child.new it doesn't call the Array::Wrapper.new
<
meh`>
but Array.new
<
adambeynon>
of course
<
adambeynon>
class methods wont work
<
meh`>
any way around it?
<
adambeynon>
meh`: would copying Array::Wrapper.new work?
<
adambeynon>
as in, manually set the .new() method to the one of the subclass?
<
adambeynon>
any other class methods wrapper overrides?
<
meh`>
only .new and .[] are overridden
<
adambeynon>
well, without having to do major runtime hacks, we could just copy those two methods onto the subclass
<
meh`>
yeah, already did it :)
<
meh`>
now there's another problem
<
adambeynon>
also, for super() to work, you might want to add : subclass._super = #{Array::Wrapper}
<
meh`>
yeah, that was the problem
<
adambeynon>
good guess ayee
<
meh`>
the .class on the create object is nil
<
meh`>
or the name is nil
<
adambeynon>
subclass._name = real_subclass._name
<
adambeynon>
no thats wrong
<
adambeynon>
wtf, the name should be right? we arent changing it
<
meh`>
does the name setting depend on alloc or proto?
<
meh`>
because we are creating an anonymous class
<
adambeynon>
meh`: no, it gets set on subclass itself, which is the class we are keeping
<
adambeynon>
real_subclass will be nil
<
adambeynon>
but we arent copying the name over
<
meh`>
the rest seems to work
<
meh`>
only class name is borked
<
meh`>
Array::Wrapper is now in the ancestors chain
<
adambeynon>
subclass.__parent = #{Array}
<
adambeynon>
__parent is used for modules
<
adambeynon>
_super points to the next class up
<
adambeynon>
__parent points to the next module/class
<
meh`>
all that inconsistency
<
meh`>
why __parent and not _parent?
<
meh`>
adambeynon, even with __parent overloaded Array::Wrapper is still the superclass
<
meh`>
_super fucks it up
<
meh`>
I guess I'll reimplement new
<
adambeynon>
meh`: rspec adds @_parent to classes, which was breaking it
<
adambeynon>
we need a prefix
<
adambeynon>
yay prefixes
<
meh`>
I vote for $$
gawwo has joined #opal
<
meh`>
shawn42, but I wouldn't trust it, the documentation needs a good revamp
<
adambeynon>
shawn42: Im going to spend some time this afternoon bringing the docs up to date
<
adambeynon>
let me know if there is anything specific you were looking for
<
adambeynon>
or anything that is unclear
<
adambeynon>
which is a lot of things looking at what are docs are currently filled with
<
shawn42>
adambeynon: meh` thanks
<
shawn42>
I figured a lot of stuff out via the examples in the code and opal-jquery
GitHub45 has joined #opal
<
GitHub45>
opal/master 0a30293 Adam Beynon: Move require() and auotload() resolve logic into DependencyResolver class
GitHub45 has left #opal [#opal]
GitHub61 has joined #opal
<
GitHub61>
opal/master 33e9ade Adam Beynon: link to all documentation files from home.md
GitHub61 has left #opal [#opal]
GitHub155 has joined #opal
<
GitHub155>
opal/master 45f2c92 Adam Beynon: formatting fix
GitHub155 has left #opal [#opal]
<
adambeynon>
and we now have a basic table of contents
<
adambeynon>
yay for github rendering markdown
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1218 (master - 0a30293 : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1219 (master - 33e9ade : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1220 (master - 45f2c92 : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
<
meh`>
adambeynon, shouldn't #allocate use the actual _alloc?
<
meh`>
it should mmh
<
adambeynon>
meh`: yes
<
meh`>
adambeynon, ok, the ancestors are right now
<
meh`>
adambeynon, but there's still the name error
<
meh`>
it's nil for some reason
<
adambeynon>
meh`: check what `the_class._name` property is
<
meh`>
adambeynon, it's nil
<
meh`>
LOL._name is "LOL"
<
meh`>
LOL.new.class._name is nil
<
adambeynon>
LOL.new is still creating a wrapper instance then
<
adambeynon>
or atleast its ._klass is pointing towards the wrapper
<
adambeynon>
.allocate is still creating an instance of ArrayWrapper
<
meh`>
but that shouldn't be the case, should it?
<
meh`>
maybe just copying isn't enough?
<
meh`>
the function uses the old this?
<
adambeynon>
meh`: after klass._proto = replace._proto;
<
adambeynon>
you need to do
<
adambeynon>
klass._proto._klass = klass
<
meh`>
should I clone the object first?
<
adambeynon>
instances need to have their class pointers updated
<
meh`>
it's a new class
<
meh`>
works now :D
<
adambeynon>
yeah, but the instances think that they are instances of Array::Wrapper
<
adambeynon>
meh`: how much code for all this for arrays?
GitHub167 has joined #opal
<
GitHub167>
opal/master c723073 Adam Beynon: Initial commit of foo[bar] += baz, support
GitHub167 has left #opal [#opal]
<
adambeynon>
meh`: once done, pop it on a branch for now, so we can review it to see if its the best option
<
meh`>
it's already on a branch
<
meh`>
I'll push it when it's fine
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1221 (master - c723073 : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
fkchang has joined #opal
GitHub86 has joined #opal
<
GitHub86>
opal/master 33264fb Adam Beynon: Fix Time#+ and Time#-
GitHub86 has left #opal [#opal]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1222 (master - 33264fb : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
elia has quit [Ping timeout: 245 seconds]
<
meh`>
adambeynon, do you know where #be_an_instance_of is implemented?
<
meh`>
(in rubyspecs)
<
meh`>
or is it part of mspec?
<
meh`>
nope, it's not that
<
meh`>
I don't get this
<
meh`>
adambeynon, does a super in an instance method use _super?
<
adambeynon>
meh`: nope
<
adambeynon>
__parent
<
adambeynon>
meh`: and be_an_instance_of is defined by mspec
<
meh`>
it's not the issue
gawwo has quit [Remote host closed the connection]
GitHub136 has joined #opal
GitHub136 has left #opal [#opal]
<
GitHub136>
opal/master 145bf2b Adam Beynon: Partial rewrite of CallNode
elia has joined #opal
elia has quit [Client Quit]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1223 (master - 145bf2b : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
GitHub10 has joined #opal
GitHub10 has left #opal [#opal]
<
GitHub10>
opal/master ce96e67 Adam Beynon: Update docs on compiler output to reflect current status
GitHub107 has joined #opal
GitHub107 has left #opal [#opal]
<
GitHub107>
opalrb.org/master af11c9a Adam Beynon: Update documentation links
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1224 (master - ce96e67 : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
GitHub98 has joined #opal
<
GitHub98>
opalrb.org/gh-pages 8908c53 Adam Beynon: Site updated to af11c9a
GitHub98 has left #opal [#opal]
elia has joined #opal
elia has quit [Ping timeout: 264 seconds]
DrShoggoth has joined #opal
<
meh`>
adambeynon, ping
<
meh`>
adambeynon, is there any way to find out where arity checking is kicking in?
<
adambeynon>
meh`: the error message should give you the name of the object and method that it fails for
<
meh`>
adambeynon, yes, but I need more
<
meh`>
adambeynon, it's not really helping
<
meh`>
unless it's broken
<
meh`>
adambeynon, does the ac take into account splats?
<
meh`>
because I'm getting
<
meh`>
6. Array#last does not return subclass instance on Array subclasses
<
meh`>
ArgumentError: [MyArray#initialize] wrong number of arguments(1 for 2)
<
meh`>
and a bunch of others
<
meh`>
Array::Wrapper#initialize has (*args, &block)
<
adambeynon>
meh`: what about Array#initialize ?
<
meh`>
adambeynon, same, it's (*args)
<
adambeynon>
it must be hitting some other #initialize method somewhere
<
adambeynon>
it gets the class name from the object
<
adambeynon>
so it ignores super
<
meh`>
adambeynon, I think it happens inside the Array::Wrapper #method_missing
<
meh`>
it does a self.class.new(result)
<
meh`>
adambeynon, I think it's some weirdness in the MyArray implementation in mspec
<
meh`>
or well, in rubyspecs
<
meh`>
because if I test it manually it works
<
meh`>
adambeynon, found it
<
adambeynon>
meh` what was it?
GitHub41 has joined #opal
<
GitHub41>
opal/master e895410 Adam Beynon: Add general purpose RakeHelper class for opal gems to use
GitHub41 has left #opal [#opal]
<
meh`>
adambeynon, everything was right all along
<
meh`>
adambeynon, MyArray#initialize(a, b)
<
meh`>
fixed by doing the thing myself
<
meh`>
now onto the real bugs
<
adambeynon>
Im off out now, be back later tonight
DouweM has quit [Ping timeout: 256 seconds]
DouweM has joined #opal
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1225 (master - e895410 : Adam Beynon): The build was broken.
travis-ci has left #opal [#opal]
elia has joined #opal
<
meh`>
now our specs running crash phantomjs
elia has quit [Ping timeout: 256 seconds]
GitHub22 has joined #opal
GitHub22 has left #opal [#opal]
<
GitHub22>
opal/master 2946e9b Adam Beynon: Fix RakeHelper for 1.8
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1226 (master - 2946e9b : Adam Beynon): The build was fixed.
travis-ci has left #opal [#opal]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1227 (array-inheritance - dc73679 : meh): The build passed.
travis-ci has left #opal [#opal]
adambeynon has joined #opal
DrShoggoth has quit [Quit: Leaving]
<
meh`>
adambeynon, hey
<
adambeynon>
meh`: yo yo yo
<
meh`>
adambeynon, everything works with that pull request
<
meh`>
all the bugs were actual real bugs in the corelib
<
meh`>
I'm backporting those fixes as we speak
GitHub57 has joined #opal
GitHub57 has left #opal [#opal]
<
GitHub57>
opal/master e2535fe Adam Beynon: Revert: RakeHelper
GitHub134 has joined #opal
<
GitHub134>
opal/master ca9d03c meh: Cleanup and consistency fixes for Array.new
<
GitHub134>
opal/master 3c5c2a7 meh: Cleanup and consistency fixes for Array#&
GitHub134 has left #opal [#opal]
<
GitHub134>
opal/master 8a66499 meh: Cleanup and consistency fixes for Array.try_convert
GitHub9 has joined #opal
GitHub9 has left #opal [#opal]
<
GitHub9>
opal/master 5aa9af2 meh: Remove leftover code in Array#flatten!
GitHub86 has joined #opal
GitHub86 has left #opal [#opal]
<
GitHub86>
opal/master 6a62d0b meh: Check the whole result only on different length in Array#flatten!
GitHub17 has joined #opal
<
GitHub17>
opal/array-inheritance 6bbce45 meh: WIP
GitHub17 has left #opal [#opal]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1229 (master - e2535fe : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
<
meh`>
adambeynon, would you mind if I cleaned up the runtime and the special attributes?
<
meh`>
basically extend names like casgn and friends and use $$ instead of _xxx or __xxx
<
adambeynon>
meh`: im not sure if $$ if the way to go
<
adambeynon>
we do need to clean up
<
meh`>
adambeynon, what would you rather go with?
<
meh`>
$ is not a valid characters in instance variables
<
meh`>
so it's not going to collide
<
adambeynon>
meh`: problem with $ is that makes it harder to see th magic varibles when debugging - I tried it and didnt like
<
adambeynon>
possibly a postfix '$' could work
<
meh`>
that's even uglier imho
<
meh`>
I'd either go with $$xxx
<
meh`>
or $$.xxx :P
<
meh`>
but then we might break stuff with the latter
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1230 (master - a318a62 : meh): The build passed.
travis-ci has left #opal [#opal]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1231 (master - 5aa9af2 : meh): The build passed.
travis-ci has left #opal [#opal]
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1233 (array-inheritance - 6bbce45 : meh): The build passed.
travis-ci has left #opal [#opal]
adambeynon has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
eventualbuddha has joined #opal
GitHub143 has joined #opal
<
GitHub143>
opal/master 5575cda meh: Fix Integer#=== and Float#===
<
GitHub143>
opal/master 36d2ec2 meh: Allow Enumerator to have infinite elements
GitHub143 has left #opal [#opal]
<
GitHub143>
opal/master 53c5683 meh: Allow empty lines in spec/rubyspecs