vwoo has joined #opal
vwoo has quit [Quit: Computer has gone to sleep.]
vwoo has joined #opal
vwoo has quit [Client Quit]
vwoo has joined #opal
fkchang has quit [Ping timeout: 264 seconds]
vwoo has quit [Quit: Computer has gone to sleep.]
fkchang has joined #opal
fkchang has quit [Ping timeout: 260 seconds]
davidboy has quit [Ping timeout: 245 seconds]
lectrick has quit [Ping timeout: 256 seconds]
denysonique has quit [Ping timeout: 268 seconds]
lectrick has joined #opal
davidboy has joined #opal
adambeynon has joined #opal
chap has joined #opal
GitHub57 has joined #opal
GitHub57 has left #opal [#opal]
<GitHub57> [opal] adambeynon pushed 1 new commit to master: http://git.io/Eunu9Q
<GitHub57> opal/master b9d8517 Adam Beynon: Use $ as runtime/helpers prefix instead of __ (fixes #337)
denysonique has joined #opal
elia has joined #opal
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] opal/opal#794 (master - b9d8517 : Adam Beynon): The build passed.
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9766127
GitHub91 has joined #opal
<GitHub91> opal/master 20d3593 Adam Beynon: Add very simple spec for constants inside singleton classes (fixes #340)
GitHub91 has left #opal [#opal]
<GitHub91> [opal] adambeynon pushed 1 new commit to master: http://git.io/l52MVA
davidboy has quit [Ping timeout: 256 seconds]
chap has quit [Quit: bb]
travis-ci has joined #opal
<travis-ci> [travis-ci] opal/opal#795 (master - 20d3593 : Adam Beynon): The build passed.
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9766343
travis-ci has left #opal [#opal]
<meh`> adambeynon, yo
<adambeynon> good mornin' !
GitHub83 has joined #opal
GitHub83 has left #opal [#opal]
<GitHub83> [opal-jquery] adambeynon pushed 1 new commit to master: http://git.io/Gh38ZA
<GitHub83> opal-jquery/master 5e14389 Adam Beynon: Use Element#expose in preference to method_missing
<meh`> adambeynon, what's new?
GitHub129 has joined #opal
<GitHub129> [opalrb.org] adambeynon pushed 1 new commit to master: http://git.io/OV5PbA
GitHub129 has left #opal [#opal]
<GitHub129> opalrb.org/master d9ceccc Adam Beynon: Use latest jquery
<adambeynon> meh`: today is a day of bug fixing
<adambeynon> trying to get through some github issues
<adambeynon> add some wiki docs at the same time
GitHub124 has joined #opal
<GitHub124> [opalrb.org] adambeynon pushed 1 new commit to gh-pages: http://git.io/74vtLw
<GitHub124> opalrb.org/gh-pages 38c74e6 Adam Beynon: Site updated to d9ceccc
GitHub124 has left #opal [#opal]
<adambeynon> meh`: Also, Im trying out a possible solution to subclassable Arrays
<adambeynon> array subclasses, internally, would inherit from BasicObject, and pass all methods through #method_missing to their internal array
<adambeynon> so methods defined on subclasses would be regular methods
<adambeynon> but a subclassed array actually has a self._array variable
<adambeynon> there is no way around subclassing native arrays, it just doesnt work
<adambeynon> a javascript "feature"… or something like that
<adambeynon> well
<adambeynon> either that, or we have to wrap all arrays, including Array itself
<meh`> -1 to wrapping all arrays
<adambeynon> yep
<adambeynon> third possibility, which again isn't great, is what we used to do - adding subclassed methods to each instance upon #allocate
<adambeynon> fine if there are only a few methods
<adambeynon> but we could be talking about a lot
<meh`> OR, get method tables going :P
<adambeynon> well, that as well
<meh`> adambeynon, so, when are you we getting destructuring in blocks? :)
<meh`> I hope it's today
GitHub175 has joined #opal
<GitHub175> [opal] adambeynon pushed 1 new commit to master: http://git.io/WpFuaw
GitHub175 has left #opal [#opal]
<GitHub175> opal/master 8c43aa4 Adam Beynon: Allow super to be called in singleton object method (fixes #335)
<adambeynon> meh`: as in the |a, (b, c)|
<adambeynon> ?
<meh`> yes
<meh`> it shouldn't be hard
<adambeynon> yep
<adambeynon> was it 1.9 or 1.8 feature?
<meh`> 1.9 I think
travis-ci has joined #opal
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9767310
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] opal/opal#796 (master - 8c43aa4 : Adam Beynon): The build passed.
<adambeynon> ok, it is on my list of bugs, so should get it done
GitHub191 has joined #opal
<GitHub191> [opal] adambeynon pushed 1 new commit to master: http://git.io/TEBaOg
GitHub191 has left #opal [#opal]
<GitHub191> opal/master e9a8db5 Adam Beynon: Cleanup dispatch_super runtime helper
<adambeynon> will also remove lots of "pending" specs in our spec/ dir :)
<meh`> great
<meh`> adambeynon, we also need a bunch of Enumerable/Enumerator specss
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9767458
<travis-ci> [travis-ci] opal/opal#797 (master - e9a8db5 : Adam Beynon): The build passed.
GitHub16 has joined #opal
<GitHub16> opal/master a93c18c Adam Beynon: Support class variables in defined? (fixes #217)
GitHub16 has left #opal [#opal]
<GitHub16> [opal] adambeynon pushed 1 new commit to master: http://git.io/9nCSRA
travis-ci has joined #opal
<travis-ci> [travis-ci] opal/opal#798 (master - a93c18c : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9767617
davidboy has joined #opal
GitHub45 has joined #opal
<GitHub45> [opal] adambeynon pushed 1 new commit to master: http://git.io/SCzkEQ
GitHub45 has left #opal [#opal]
<GitHub45> opal/master 0125bb1 Adam Beynon: Support Module#remove_const (fixes #314)
travis-ci has joined #opal
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9769676
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] opal/opal#799 (master - 0125bb1 : Adam Beynon): The build passed.
<meh`> adambeynon, can you give some priority to the block argument destructuring?
<meh`> it's a blocker to get parslet working
<meh`> and deadlines are approaching, I really need lissio to be usable before next week
<meh`> and it's unusable without templates
GitHub59 has joined #opal
<GitHub59> opal/master 86381ef Adam Beynon: Clean up '&' parsing in lexer
<GitHub59> [opal] adambeynon pushed 2 new commits to master: http://git.io/m05GPw
GitHub59 has left #opal [#opal]
<GitHub59> opal/master fa54812 Adam Beynon: Clean up '|' parsing in lexer
<adambeynon> meh`: yep, should get onto it today, hopefully just after lunch
<meh`> thanks
<adambeynon> it is a mustache based templating thing?
<meh`> yeah
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9771259
<travis-ci> [travis-ci] opal/opal#800 (master - fa54812 : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
travis-ci has joined #opal
<meh`> but I'd rather keep it with parslet
<meh`> easier to deal with
<adambeynon> isnt it better to precompile? rather than parsing it inside opal?
<meh`> adambeynon, nah
<meh`> I don't like the approach
<meh`> of erb and friends
<meh`> and I think it will be fast enough
<meh`> but I can't find out until I try :D
<adambeynon> ha, of course!
GitHub54 has joined #opal
GitHub54 has left #opal [#opal]
<GitHub54> [opal] adambeynon pushed 1 new commit to master: http://git.io/1UQSxA
<GitHub54> opal/master 1640269 Adam Beynon: Add real modules to runtime and corelib (fixes #344)
GitHub1 has joined #opal
<GitHub1> [opal] adambeynon pushed 1 new commit to master: http://git.io/_Nyy4w
<GitHub1> opal/master e865f62 Adam Beynon: Stop recursive hash breaking on to_s/inspect (fixes #222)
GitHub1 has left #opal [#opal]
travis-ci has joined #opal
<travis-ci> [travis-ci] opal/opal#802 (master - e865f62 : Adam Beynon): The build passed.
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9772625
travis-ci has left #opal [#opal]
<adambeynon> meh`: considering Native has only #initialize, and #to_n, is it really needed as a base class for inheriting?
<meh`> #is_a? Native has some value
<meh`> and it "standardizes" @native as instance var for the native value
<adambeynon> I know, but having Native::Object is quite awkward, as most of the time when wanting to use is_a?(Native) I actually have to do is_a?(Native::Object)
<adambeynon> If we have a custom wrapper
<adambeynon> aka opal-browser
<adambeynon> do we really want to know if it is a wrapped native?
<meh`> that's not the main reason
<meh`> it's boilerplate
<meh`> boilerplate I don't want to see in my class
<adambeynon> I see that, yes, but in general day-to-day code, I imagine people using Native::Object more than Native, atleast directly
<adambeynon> dom makes sense for wrapping
<adambeynon> but, we're not going to wrap every library out there
<adambeynon> well, not initally anyway ;)
<meh`> well, yeah, if you're using a library you don't even have to know about Native
<adambeynon> I do see the need for 2 classes
<meh`> also I don't think #is_a? would work on BasicObject
<meh`> adambeynon, I also changed my mind on wrapping stuff for the block calls in Native::Object
<meh`> I'm adding the proper support
<adambeynon> for what now?
<meh`> adambeynon, you'll see the diff
<adambeynon> meh`: hang on
<adambeynon> its more complex than that
<adambeynon> we cant just wrap a block in an anonymous function
<meh`> oh?
<meh`> why not?
<adambeynon> it is common that certain libraries use functions for reference
<adambeynon> e.g.
<adambeynon> promises and events in jquery
<adambeynon> you need to pass the same function to #off() as you do to #on()
<adambeynon> otherwise it wont remove it
<adambeynon> same with 1.js
<adambeynon> sorry, q.js
<adambeynon> I think its q.js anyway
<meh`> I don't think it's going to be a problem
<meh`> if you're calling it with the block syntax you aren't storing the function in the first place
<meh`> if you're storing it, you're passing -> or proc{} directly
<meh`> so you have it already
<adambeynon> #on() returns the function
<adambeynon> so you can do
<meh`> then it's going to return the wrapper
<adambeynon> element.on { … }
<meh`> and everything will still work
<adambeynon> my_proc = proc { … }
<adambeynon> element.on(&my_proc)
<meh`> yeah, you shouldn't do that
<meh`> we'll make it clear
<adambeynon> but thats how Vienna::View works
<meh`> I doubt anyone will do something like that sincerely
<adambeynon> you define class level event handlers
<meh`> but vienna is using opal-jquery
<meh`> isn't it?
<adambeynon> meh`: for now
<adambeynon> if we can move native, we will
<meh`> adambeynon, just drop the &
<meh`> and it will work the same
<meh`> I don't see the problem
<adambeynon> it seems another "everything works, oh yeah, you cant send blocks to native libraries"
<meh`> isn't it worse to have to wrap and unwrap by hand all natives?
GitHub7 has joined #opal
GitHub7 has left #opal [#opal]
<GitHub7> opal/master ce780e4 Adam Beynon: Add stub method Parser#source_map for parser to lazily generate source maps
<GitHub7> [opal] adambeynon pushed 1 new commit to master: http://git.io/EDVfFw
GitHub25 has joined #opal
<GitHub25> opal/master 81249b0 Adam Beynon: Fix bug when singleton_class objects were not generated as a method receiver
<GitHub25> [opal] adambeynon pushed 1 new commit to master: http://git.io/7hTBKA
GitHub25 has left #opal [#opal]
travis-ci has joined #opal
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9776839
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] opal/opal#804 (master - 81249b0 : Adam Beynon): The build was fixed.
<adambeynon> elia: a small amount of work left and then fragments will enable real source maps
<elia> adambeynon: cool!
<elia> adambeynon: im sorry for being so inactive lately
<elia> m( _ _ )m
<adambeynon> lol, no problem!
<adambeynon> its a fun day of bug fixing
GitHub146 has joined #opal
<GitHub146> opal/master b276a9d Adam Beynon: Remove old source map comments from parser
<GitHub146> [opal] adambeynon pushed 1 new commit to master: http://git.io/b_YTEw
GitHub146 has left #opal [#opal]
travis-ci has joined #opal
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9777477
<travis-ci> [travis-ci] opal/opal#805 (master - b276a9d : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
<meh`> adambeynon, any update on block argument destructuring?
<adambeynon> meh`: havent started them yet
<adambeynon> well
<adambeynon> iv read the specs
<adambeynon> not implemented though
<meh`> do you think you'll be able to do it for tonight?
GitHub130 has joined #opal
<GitHub130> [opal] adambeynon pushed 1 new commit to master: http://git.io/GMaO-g
GitHub130 has left #opal [#opal]
<GitHub130> opal/master ec91c4c Adam Beynon: Generate source maps from fragments
<adambeynon> meh`: yep
<meh`> awesome, thanks
<adambeynon> :D
<meh`> hoefully there won't be other blockers for parslet that I can't fix myself
travis-ci has joined #opal
<travis-ci> [travis-ci] opal/opal#806 (master - ec91c4c : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9780431
fkchang has joined #opal
<fkchang> adambeynon: were u ever successful in embedding opal-irb into your demo?
<adambeynon> fkchang: didnt get around to it in the end
<adambeynon> used your actual version
<adambeynon> on github.io
<fkchang> related to a demo, and some feature requests I've seen, we should have a way to require external files into opal-irb, then you could go to the github.io site, require some files into it and show that
<fkchang> I'd been playing w/a require_js that you could point to say cdn'd js files to include those into opal-irb and then you could start messing w/them through opal
<fkchang> I also have been thinking about adding jsfiddle type capability to opal-inspector, so you could mess w/markup and css too, so the ability to include arbitrary js or ruby code would fit nicely
<fkchang> get all that together with enabling the class inspector w/source code and we'd have something sweet. Then I can try to see if I can duplicate Smalltalk's ability to code from the debugger
<meh`> adambeynon, any news?
elia has quit [Ping timeout: 246 seconds]
chap has joined #opal
meh` has quit [Ping timeout: 240 seconds]
GitHub184 has joined #opal
<GitHub184> [vienna] adambeynon pushed 1 new commit to master: http://git.io/cpumrg
<GitHub184> vienna/master 4dc3b99 Adam Beynon: Use latest opal/opal-jquery
GitHub184 has left #opal [#opal]
chap has quit [Quit: bb]
GitHub141 has joined #opal
<GitHub141> opalrb.org/master 8fa1173 Adam Beynon: Add docs link to site
GitHub141 has left #opal [#opal]
<GitHub141> [opalrb.org] adambeynon pushed 1 new commit to master: http://git.io/HeTNIQ
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/vienna/builds/9785651
<travis-ci> [travis-ci] opal/vienna#99 (master - 4dc3b99 : Adam Beynon): The build passed.
<adambeynon> fkchang: sorry, missed the notification
<adambeynon> yes
GitHub94 has joined #opal
GitHub94 has left #opal [#opal]
<GitHub94> [opalrb.org] adambeynon pushed 1 new commit to gh-pages: http://git.io/vY4-IQ
<GitHub94> opalrb.org/gh-pages a12648b Adam Beynon: Site updated to 8fa1173
<adambeynon> some remote js require would be useful
<adambeynon> we can already gather the requires() by the parser, so we should be able to hook into that as well
elia has joined #opal
meh` has joined #opal
elia has quit [Ping timeout: 276 seconds]
markluffel has quit [*.net *.split]
harrow` has quit [*.net *.split]
meh` has quit [*.net *.split]
kludge` has quit [*.net *.split]
fkchang has quit [*.net *.split]
davidboy has quit [*.net *.split]
lectrick has quit [*.net *.split]
lupine has quit [*.net *.split]
brixen has quit [*.net *.split]
adambeynon has quit [*.net *.split]
denysonique has quit [*.net *.split]
fkchang has joined #opal
meh` has joined #opal
adambeynon has joined #opal
brixen has joined #opal
lupine has joined #opal
markluffel has joined #opal
lectrick has joined #opal
davidboy has joined #opal
denysonique has joined #opal
kludge` has joined #opal
harrow` has joined #opal
<meh`> adambeynon, is there a way to fail a spec without an assert?
<adambeynon> meh`: you can just "raise 'foo'"
<adambeynon> or not doing any should() will also fail it
<adambeynon> (assuming we are talking about mspec here, not opal-spec)
<meh`> nope, talking about opal-spec
<meh`> adambeynon, I think we have a problem
<adambeynon> another? ;)
<meh`> yeah
<meh`> super alone doesn't pass the block over
<adambeynon> that is true, I noticed that issue earlier today
<meh`> fast fix, shouldn't it?
<adambeynon> as in, just plain `super`, no args, no parans?
<adambeynon> parens*
<meh`> yes
<adambeynon> attempting to fix now
<adambeynon> hmmm, it might be an awakrd fix
<adambeynon> well
<adambeynon> a fix for `def foo; end` sufficient for now?
<meh`> better awkward than non working
<adambeynon> super from blocks is a bit more tricky
<meh`> took me a while it was an opal bug
<meh`> *to realize
<meh`> adambeynon, it's a def self.new(*)
<meh`> in Native
<meh`> it's not from a block, so it's fine
<meh`> adambeynon, any luck with the destructuring?
<adambeynon> meh`: It wont be finished quickly enough - need to rework masgn in the code generator completely
<meh`> :(
<adambeynon> sorry :(
<adambeynon> I did try on a positive Iv got the super stuff working
vwoo has joined #opal
<adambeynon> (apart from iters)
<adambeynon> also, I just wasted 5 minutes trying to debug a spec that was failing
<adambeynon> turns out, I had be_true instead of be_false
<meh`> lol
GitHub162 has joined #opal
GitHub162 has left #opal [#opal]
<GitHub162> opal/master 775d2c6 Adam Beynon: Auto-magically pass blocks in zsuper calls (super with no args)
<GitHub162> [opal] adambeynon pushed 1 new commit to master: http://git.io/5QIR5Q
* adambeynon should probably make a note of what all these implementation javascript variables are for
travis-ci has joined #opal
<travis-ci> [travis-ci] opal/opal#807 (master - 775d2c6 : Adam Beynon): The build passed.
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9789314
travis-ci has left #opal [#opal]
<meh`> adambeynon, when are we getting escaped HTML in opal-spec browser results?
<meh`> and an updated gem
<meh`> brb
<adambeynon> well, all we really need is ERB::Util
<adambeynon> which should be simple enough
meh` has quit [Ping timeout: 248 seconds]
GitHub88 has joined #opal
<GitHub88> opal/master c018aed Adam Beynon: Generate less code inside def/defs to gather block from method
<GitHub88> [opal] adambeynon pushed 1 new commit to master: http://git.io/d53ZiQ
GitHub88 has left #opal [#opal]
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] opal/opal#808 (master - c018aed : Adam Beynon): The build passed.
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9789771
GitHub123 has joined #opal
GitHub123 has left #opal [#opal]
<GitHub123> opal/master d68c2e2 Adam Beynon: Replace some Opal references with $opal in generated code
<GitHub123> [opal] adambeynon pushed 1 new commit to master: http://git.io/Zo2inw
travis-ci has joined #opal
<travis-ci> [travis-ci] opal/opal#809 (master - d68c2e2 : Adam Beynon): The build passed.
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9789853
travis-ci has left #opal [#opal]
meh` has joined #opal
<meh`> adambeynon, two problems
<meh`> 1) now the block is passed to super no matter what
<meh`> 2) Hash subclassing is completely broken
GitHub38 has joined #opal
GitHub38 has left #opal [#opal]
<GitHub38> [opal] adambeynon pushed 1 new commit to master: http://git.io/tRosNg
<GitHub38> opal/master 8e47a09 Adam Beynon: Slight tidy up of process_args()
<adambeynon> isnt it meant to be passed to super always?
<adambeynon> or even if you specify args?
<meh`> adambeynon, if you have super(something)
<meh`> the block gets passed anyway
<adambeynon> right
<adambeynon> ok
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9790147
<travis-ci> [travis-ci] opal/opal#810 (master - 8e47a09 : Adam Beynon): The build passed.
vwoo has quit [Quit: Computer has gone to sleep.]
GitHub124 has joined #opal
GitHub124 has left #opal [#opal]
<GitHub124> opal/master 2404795 Adam Beynon: Only pass blocks to super automatically for zsuper
<GitHub124> [opal] adambeynon pushed 1 new commit to master: http://git.io/6QDrjQ
<adambeynon> meh`: ok, so, subclassing Hash, as in some methods dont return subclasses? or….?
<meh`> adambeynon, yes
<meh`> for instance, #[]
<meh`> maybe completely was too strong of a word :P
<adambeynon> lol
travis-ci has joined #opal
<travis-ci> [travis-ci] opal/opal#811 (master - 2404795 : Adam Beynon): The build passed.
<travis-ci> [travis-ci] Build details : http://travis-ci.org/opal/opal/builds/9790434
travis-ci has left #opal [#opal]
<adambeynon> although
<adambeynon> what is wrong with #[] ?
<meh`> adambeynon, it doesn't return the subclass
<meh`> it returns a raw Hash
<adambeynon> ohhhhh
<adambeynon> Hash.[]
<meh`> yes, sorry
<meh`> ◕ ◡ ◔
<adambeynon> makes more sense now ;)
<adambeynon> sure
<meh`> also I think some stuff in Enumerable is still broken
<meh`> we *really* need more specs for Enumerable and Hash
<adambeynon> broken, as in yielding multiple values?
<meh`> adambeynon, I don't know yet
<meh`> I'm seeing something weird in Hash#map
vwoo has joined #opal
<adambeynon> I need to study up on yielding multiple block args to clear my mind on how/what/when args get handled
<adambeynon> obsiouly the deconstructed values comes into this as well
<adambeynon> obviosuly*
<meh`> adambeynon, yeah
<meh`> I fixed what I could in Enumerable
<meh`> but I used empirical tests to know what to do
<meh`> so specs would be nice :)
<adambeynon> looking though rubyspecs now
<adambeynon> yield is weird
vwoo has quit [Quit: Computer has gone to sleep.]
<adambeynon> yeah, yield is going to get very ugly
adambeynon has quit [Quit: ["Textual IRC Client: www.textualapp.com"]]
vwoo has joined #opal
vwoo has quit [Ping timeout: 268 seconds]
vwoo has joined #opal
vwoo has quit [Quit: Computer has gone to sleep.]
vwoo has joined #opal
meh` has quit [Ping timeout: 264 seconds]