<adambeynon>
meh`: away from laptop. Currently on my phone :/
<adambeynon>
Be back home later though
e_dub has joined #opal
<elia>
"adambeynon summoning" lol works every time
<elia>
adambeynon, meh`, I think it would be good to have at least two pushers for each gem, to have backup
<meh`>
elia, yeah
e_dub has quit [Quit: It's a hard knock life]
dfranciosi has quit [Remote host closed the connection]
<meh`>
adambeynon, ping me when you push a new version
dfranciosi has joined #opal
marcandre has joined #opal
<meh`>
I have 13 commits to push that I can't push because rspec :<
<elia>
meh`, can't you git rspec in the gemfile meanwhile?
<meh`>
elia, nope, in the git the rspec.js isn't built
<elia>
meh`, I really dislike the prebuilt rspec.js, I remeber to have fixed that (requiring rspec 3 beta as a dep)
<elia>
adambeynon removed it for the good (the dep) but maybe the harness is still there…
DouweM has joined #opal
fkchang has joined #opal
elia has quit [Ping timeout: 272 seconds]
ryanstout has joined #opal
Liothen has joined #opal
<royvandermeij>
elia: I always find alias hard to read (even though I have been developing in Ruby for 8 years, and have thaught it at a university)
<royvandermeij>
with a method definition it's instantly clear what it does
<royvandermeij>
in my opinion that is totally worth the cost of 2 extra lines
<royvandermeij>
s/thaught/taught
<meh`>
royvandermeij, in naïve implementations an alias is different from a method calling another method
<meh`>
performance wise
<royvandermeij>
ah, I did not know that
<royvandermeij>
interesting
<royvandermeij>
I did already change the pull request to use alias
<meh`>
I think only Rubinius is clever enough to inline those calls
<meh`>
a call is a call, you're going deeper in the stack
<meh`>
and when you think neither MRI nor JavaScript support TCO, it can become a big deal
<royvandermeij>
true but if its marginal then it's a discussion of semantics in stead of performance ;-)
<royvandermeij>
what is TCO?
<meh`>
tail call optimization
<royvandermeij>
aha
<royvandermeij>
(still had to look it up hehe)
<royvandermeij>
I only look for performance optimalisations when it hurts (so I dont optimise premature), but I understand that when your writing a framework you need it as fast as possible
DrShoggoth has joined #opal
wmnnd has quit [Ping timeout: 245 seconds]
<meh`>
well, aliases are idiomic, and they're faster as well, so I don't see any downside in using them
<meh`>
while the same isn't valid for methods just deferring to other methods
<meh`>
now if only adambeynon came back on his laptop
<meh`>
basically, it's whining about type mismatches
<meh`>
the thing is, I want to use the constructor and not the initMouseEvent
<meh`>
this is because constructors are supported
<meh`>
I'm just failing to find the properties it needs to have checked
<meh`>
and I can't find what they are
mneorr_ has quit [Remote host closed the connection]
<meh`>
in the specs I use mouse events
<meh`>
so I'm getting failures only there
<fkchang>
meh`: I think I figured it out, I was copying from the documentation and then gwentoo, I did not see that gwentoo defines REST, and thought it was the same as in the lissio doc, calling Lissio::Adapter::REST gets me pass that
<meh`>
fkchang, yeah, defining your own subclass of REST is useful to add default parameters/behavior
<meh`>
that is app-wide
mneorr has joined #opal
mneorr has quit [Remote host closed the connection]
mneorr has joined #opal
<fkchang>
meh`: should Cards.fetch return a promise?
<meh`>
fkchang, yes
wmnnd has joined #opal
<fkchang>
ok, then I guess I don't know what to do with that promise
<fkchang>
hmm, nothing seems to happen, I run "Models::Cards.fetch.then { |cards| puts cards; alert cards }.rescue{ |error| alert error }" and I don't get alerts of any kind
<meh`>
fkchang, did you define the endpoint?
<meh`>
when setting the adapter
<meh`>
or check the network tab in the inspector
<fkchang>
I might've done it wrong
<fkchang>
I saw a typo,
<fkchang>
can't I just do the following to hit a local sinatra app?