DouweM has quit [Read error: Connection reset by peer]
adambeynon has joined #opal
GitHub152 has joined #opal
GitHub152 has left #opal [#opal]
<
GitHub152>
opal/master e5103ad Adam Beynon: Fix bug so instance_methods of bridged classes acts correctly
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1137 (master - e5103ad : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
GitHub70 has joined #opal
GitHub70 has left #opal [#opal]
<
GitHub70>
opal-rspec/master 010c01c Adam Beynon: Add some colors to text formatter output
kludge` has quit [Ping timeout: 272 seconds]
kludge` has joined #opal
GitHub168 has joined #opal
GitHub168 has left #opal [#opal]
<
GitHub168>
opal-rspec/master a7b6e17 Adam Beynon: Add some simple tests for include/extend in Rspec.configure
GitHub183 has joined #opal
<
GitHub183>
opal-rspec/master f84b578 Adam Beynon: Add simple specs for helper methods and subject
GitHub183 has left #opal [#opal]
DouweM has joined #opal
<
meh`>
adambeynon, fixed that Singleton thing yet?
<
adambeynon>
meh`: nope, not yet
<
adambeynon>
meh`: got 1 bug to fix with rspec, then Im going to have a look at it.
<
adambeynon>
meh`: we have to change Module#name lookups to be on-demand
<
adambeynon>
otherwise Class.new + const_set doesnt work properly
<
meh`>
no way to cache it?
<
DouweM>
adambeynon: how doesn't it work properly with Opal.casgn?
<
adambeynon>
DouweM: rspec brought up a corner case. if you have 2 anonymous classes
<
adambeynon>
if you set the name of the inner class
<
adambeynon>
then set the outer class name
<
adambeynon>
then the inner class will be wrong
<
adambeynon>
as it wont know that its parent has had a proper name set on it
<
DouweM>
And resetting the name for all children in Opal.casgn(parent) wouldn't suffice?
<
adambeynon>
DouweM: just trying that approach now as well
<
adambeynon>
hmm, well, children might be donated children from a superclass
<
adambeynon>
so, we either have to keep an array of both direct children AND inherited children
<
adambeynon>
or have Module#name cache results
<
meh`>
I'd go for the latter
GitHub45 has joined #opal
GitHub45 has left #opal [#opal]
<
GitHub45>
opal/master 1207c2c Adam Beynon: Support anonymous classes in Module#name
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1138 (master - 1207c2c : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
<
DouweM>
adambeynon: is the new spec/opal/module/name_spec necessary? isn't there one in rubyspec already?
<
adambeynon>
DouweM: there is, but opal was struggling to parse it. problem with the grammar for module declarations
<
adambeynon>
need to open an issue to fix that
skofo has joined #opal
GitHub51 has joined #opal
GitHub51 has left #opal [#opal]
<
GitHub51>
opal/master a0b3b74 Adam Beynon: Fix sexp level when processing a wrapped begin; end block
GitHub146 has joined #opal
<
GitHub146>
opal-rspec/master d75d686 Adam Beynon: Add specs to check for subject compliance
GitHub146 has left #opal [#opal]
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<
travis-ci>
[travis-ci] opal/opal#1139 (master - a0b3b74 : Adam Beynon): The build passed.
GitHub128 has joined #opal
<
GitHub128>
opal-rspec/master b4786e8 Adam Beynon: Clarify 'things to fix' in README
GitHub128 has left #opal [#opal]
GitHub46 has joined #opal
GitHub46 has left #opal [#opal]
<
GitHub46>
opal-rspec/master 22d7cd0 Adam Beynon: Clarify building of rspec to run specs
GitHub74 has joined #opal
GitHub74 has left #opal [#opal]
<
GitHub74>
opal/master 2fcf583 Adam Beynon: Add Kernel#eval to stdlib/opal-parser
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1140 (master - 2fcf583 : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
skofo has quit [Ping timeout: 248 seconds]
skofo has joined #opal
skofo has quit [Ping timeout: 272 seconds]
DouweM has quit [Ping timeout: 265 seconds]
DouweM has joined #opal
elia has joined #opal
elia has quit [Client Quit]
<
adambeynon>
meh`: I was having some performance issues with rspec because all anonymous modules were sharing a constant_scope with Module
<
adambeynon>
now I fixed it, and it is damn fast
<
adambeynon>
like really fast
<
meh`>
what about the singleton bug?
<
meh`>
adambeynon, some stuff explodes if you override Class#name
<
adambeynon>
meh`: what stuff?
<
meh`>
adambeynon, I don't know
<
meh`>
some internal stuff
<
meh`>
I had overridden Event.name
<
meh`>
and it raised by itself because it required an argument
<
meh`>
you might want to check the Class#name is the real one
<
adambeynon>
meh`: Class#to_s uses Class#name
<
meh`>
before using it internally
<
meh`>
I wasn't printing it
<
adambeynon>
thats probably a bad idea
<
meh`>
yes yes, I fixed it
<
meh`>
but it puzzled me
<
meh`>
because the error seemed to pop out of nowhere in my code
<
adambeynon>
I mean Class#to_s using Class#name is probably a bad idea
<
adambeynon>
we should make it an internal function
GitHub108 has joined #opal
GitHub108 has left #opal [#opal]
<
GitHub108>
opal/master c91a71e Adam Beynon: Anon modules should have their own constant scope
GitHub91 has joined #opal
<
GitHub91>
opal/master 2e25724 Adam Beynon: Objects do not have private methods (private_methods !== methods)
GitHub91 has left #opal [#opal]
<
meh`>
adambeynon, couldn't we mark methods as private?
<
meh`>
it could be useful for debugging and stuff like private_methods
<
meh`>
we just iterate and add only the private ones
<
meh`>
we can't enforce private calls in production mode, that doesn't mean we shouldn't implement it
<
meh`>
I mean, we already do it for stubs
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1141 (master - c91a71e : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
<
adambeynon>
meh`: well we could I suppose
<
meh`>
I don't think it would slow down anything
travis-ci has joined #opal
travis-ci has left #opal [#opal]
<
travis-ci>
[travis-ci] opal/opal#1142 (master - 2e25724 : Adam Beynon): The build passed.
<
adambeynon>
meh`: what does the demo app do?
<
adambeynon>
re rubyconf
<
meh`>
adambeynon, money tracking, money you owe to people, money that people owe to you, money spent for groceries and the like
<
meh`>
adambeynon, please, can you fix that Enumerable#detect bug?
<
meh`>
I'm writing so much code with a FIXME
<
meh`>
adambeynon, why don't we have JSON.dump?
<
adambeynon>
meh`: I've always just used #to_json, we do need it though
<
meh`>
adambeynon, I'm working on it
<
meh`>
adambeynon, I also think Kernel#to_json should be as_json.to_json, not to_s.to_json
<
meh`>
I'm also working on json_create support
<
meh`>
because I need it
<
DouweM>
meh`: I'm curious, what is the Enumerable#detect bug?
<
meh`>
DouweM, it doesn't do the proper block destructuring
<
meh`>
{ a: 3, b: 4 }.find {|a, b| b == 4 }
<
meh`>
adambeynon, the actual JSON stdlib doesn't seem to define #as_json at all on Kernel
<
DouweM>
meh`: ah, okay
<
adambeynon>
meh`: I dont think as_json should really be in stdlib, should it?
<
meh`>
adambeynon, it is
<
meh`>
adambeynon, or wait
<
meh`>
yes, there are no #as_json on the core classes
<
meh`>
I'm fixing that up too
<
adambeynon>
meh`: leave them in for now
<
meh`>
adambeynon, oh, why?
<
adambeynon>
until I have time to update my apps
<
meh`>
are you using them?
<
adambeynon>
yeah. we need to add it to opal-activesupport
<
meh`>
are you also using Kernel#as_json?
<
adambeynon>
ummm, not sure
<
meh`>
I'll keep at anyway
<
meh`>
I'll just add # FIXME: remove this on them
<
meh`>
so we don't forget
<
adambeynon>
meh`: yep. I will have a look at my apps later to try and fix them up
kludge` has quit [Ping timeout: 248 seconds]
kludge` has joined #opal
<
meh`>
adambeynon, what's the easiest way to get a constant with a full path constant?
GitHub2 has joined #opal
<
GitHub2>
opal/master bd9daa9 meh: Some compliancy fixes for the JSON stdlib
<
GitHub2>
opal/master 68bd1be meh: Fix Array#to_n and Hash#to_n
GitHub2 has left #opal [#opal]
<
GitHub2>
opal/master a563cd4 meh: Add Opal.cget to get a fully scoped constant
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1143 (master - bd9daa9 : meh): The build passed.
travis-ci has left #opal [#opal]
<
meh`>
adambeynon, is inheriting from Hash broken?
<
meh`>
doesn't look like it
<
adambeynon>
meh`: it should be working. rspec is using it quite a bit
<
meh`>
adambeynon, yeah, something else is fucking it up
<
meh`>
quite puzzling
<
DouweM>
related: shouldnt Hash.[] return an instance of #{self} too, instead of Hash hardcoded?
<
meh`>
DouweM, it should
<
meh`>
I smell super breakage, again
kludge` has quit [Ping timeout: 248 seconds]
kludge` has joined #opal
elia has joined #opal
<
meh`>
adambeynon, we have a problem
<
meh`>
adambeynon, super inside define_method with no arguments is broken
<
meh`>
and splats in the block arguments
<
adambeynon>
meh`: ok, open a github issue with a small code sample of it
<
meh`>
adambeynon, already done
<
adambeynon>
just noticed ;)
GitHub142 has joined #opal
<
GitHub142>
opal/master 53064ef Adam Beynon: Kernel#instance_variables should not include private _klass and _id vars
GitHub142 has left #opal [#opal]
GitHub143 has joined #opal
GitHub143 has left #opal [#opal]
<
GitHub143>
opal-rspec/master 9a899ec Adam Beynon: Small spec to check let() still works in nested groups
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/opal#1144 (master - 53064ef : Adam Beynon): The build passed.
travis-ci has left #opal [#opal]
GitHub62 has joined #opal
<
GitHub62>
vienna/rspec f2371e9 Adam Beynon: Initial commit using rspec (opal-rspec)
GitHub62 has left #opal [#opal]
<
adambeynon>
now that commit is
*awesome*
<
adambeynon>
meh`: working on the super() bug now
travis-ci has joined #opal
<
travis-ci>
[travis-ci] opal/vienna#114 (rspec - f2371e9 : Adam Beynon): The build has errored.
travis-ci has left #opal [#opal]
<
adambeynon>
yeh, should have skipped that one..
elia has quit [Ping timeout: 240 seconds]
elia has joined #opal
adambeynon has joined #opal
elia has quit [Quit: Computer has gone to sleep.]
elia has joined #opal