<GitHub80>
jruby/master 9dd4e21 Kevin Menard: [Truffle] Updated the make leaf rope nodes to pass along the character length of the rope if we already have it.
subbu has joined #jruby
camlow325 has joined #jruby
subbu has quit [Quit: Leaving]
yfeldblum has quit [Remote host closed the connection]
camlow325 has quit [Remote host closed the connection]
<lan5432>
enebo, is my solution the interface "method new" issue good enough? Should I do something else with it?
hightower2 has quit [Ping timeout: 260 seconds]
<enebo>
Lan5432: I think kares should review this since he has been working on JI code more recently
<enebo>
Lan5432: I am indifferent on the Interface idea personally but I am so used to interfaces being modules that it does not confuse me. In any case, if you want to try that you should do it as a separate PR
donV has quit [Quit: donV]
<lan5432>
He did comment on that, but I'm unclear as to whether to submit a pull request or not, I tried to reach him lately
camlow32_ has joined #jruby
camlow32_ has quit [Remote host closed the connection]
camlow325 has quit [Ping timeout: 268 seconds]
camlow325 has joined #jruby
tomjoro has joined #jruby
<enebo>
Lan5432: I will add a note but submit a PR…worst thing that can happen is it will need some tweaking
<enebo>
Lan5432: ah I see no code specified here so kares will comment once you make a PR
<enebo>
Lan5432: and definitely run the two targets kares mentions
<kares>
lan5432: hey! sorry I have not really dealt with proposals
<kares>
pretty much what I suggested in the mail ... happy to review what you come up with
<enebo>
kares: my only concern about the feature is some feature we are confused about where somehow we instantiate an interface somehow
<kares>
lan5432: ... and as Tom noted PR is fine - but we're hoping you spec the new behavior :)
<enebo>
kares: I doubt we do but it is stil my fear
<kares>
enebo: yy
<lan5432>
enebo I think the problem JRuby had is that we had a @ with the keyword on the code
<lan5432>
I just made that dissappear and JRuby would not instantiate interfaces
<kares>
lan5432: that's perfect but we still need a spec
<lan5432>
The method was unused, though, so I didn't pay much attention to it at first
<lan5432>
Yes Kares, I'm working on learning that, sorry I did not provide one yet
<kares>
lan5432: NP
<lan5432>
And thanks for the details on mail, with that and GSoC general recomendations I'm working on something
<kares>
lan5432: we're hoping its un-used but we can not tell without running the suite
<kares>
you can submti a PR without a spec as well
<kares>
we should see a failure if its used and the spec can be added later
<enebo>
I was wondering how we define ‘impl()'
<kares>
enebo: its elsewhere - does not use new and requires an argument
<kares>
last time I checked :)
<enebo>
kares: yeah I was fairly sure it required an argument but I do not remember how it is set up
<enebo>
kares: in any case we should have coverage for impl
<lan5432>
So a spec is just a simple test case to show how new behaviour works?
<kares>
lan5432: good thing to realize - we allos instantiating interfaces in a way such as :
<kares>
java.lang.Runnable.impl { puts 'hello'! }
<kares>
* good think to learn
<enebo>
afk a few minutes
<kares>
that should still work - as enebo noted there's spec/test for that somewhere try spec/java_integration
<lan5432>
Ye, that's still working
<kares>
enebo: if you feel like its too much (due users for new to go away) we could do a deprecation warning
<lan5432>
I just repeated the case you last showed in the issue kares
<kares>
lan5432: good just so you learn something about how it works :)
<lan5432>
I was also planning on writing documentation since a user asked about it, so I will have to do that haha
<lan5432>
About that, do we have general programming documentation about JRuby and the other implementations such as the one in C?
<kares>
lan5432: we're always low on doc! esp. Java integration parts
<kares>
most general is the wiki ... I personally was kind of hoping that some day Java/JRuby .rb parts would get a ruby-doc
<lan5432>
Well, I meant more on the sense of programming use, to learn how to use JRuby as a language
<lan5432>
But that can get some work as well
<kares>
there's a Java doc for JRuby's Java APIs than in general for Ruby all of Ruby's documentation applies :)
<lan5432>
So you want docs on .rb files?
<kares>
its just JRuby's extension that are loosely covered
<kares>
lan5432: yy having .rb files documented (even if they're not used) to have some rdoc generated sounds good.
<lan5432>
So far I have Java integration, Java 8 advantages, programming documentation and general issue taclking on my proposal, we could include a small section about documentation coverage on .rb files
<kares>
lan5432: yeah I think that would be good
<kares>
maybe mention that it would be:
<kares>
documentation for JRuby's extensions such as jrubyc and Java integration
<lan5432>
I will have to use the time that I have from now on to the GSoC start to learn enough to provide at least a minimal documentation
shellac has joined #jruby
<kares>
lan5432: for JRuby?
<lan5432>
The extensions yep
<kares>
OK
<kares>
lan5432: you could also include into the proposal :
<kares>
exploring ways to optimize Java integration parts
<kares>
sounds "hard-core" :)
<lan5432>
I included something on the likes of "general refactoring work" haha
<kares>
in the end we have a few ideas around that so we can definitely guide you around those bits
shellac has quit [Client Quit]
<lan5432>
You mean for the proposal or for carrying out the job? In general the ideas are clear, refactoring the code applying Java8 features and improving what already exists. But I still need to know where to do any of that, so that's what I would be investing the time until GSoC on
<kares>
OK
enebo has quit [Quit: enebo]
<kares>
lan5432: maybe I will push some refactoring out next week that I have locally - its around interfaces
<kares>
but its not yet finished as I have a regression around it and other stuff in the queue ... will cc you when its ready for a review
<lan5432>
Perfect
<lan5432>
I will work on my proposal and creating a spec for the new behaviour, whenever it's ready I will make a pull request
<kares>
sounds good
<lan5432>
kares, how can I add a spec? All the links to the RubySpec project are broken
<kares>
RubySpec is spec-ing standard Ruby behavior - we won't use that as Java integration is JRuby specific
<lan5432>
That's what I thought
<lan5432>
Where should I add it then? Local specs are downloaded, right?
<lan5432>
Oh, I didn't see the link, sorry
<lan5432>
That's the folder where I'm working on, yep. So I just create it and then upload it with the pull request?
<kares>
yep
camlow325 has quit [Remote host closed the connection]
camlow325 has joined #jruby
<lan5432>
kares: the spec is throwing an error saying it does not find "expect" for the test. And also the require statement doesn't find the file. Is there a problem with my rspec?
<kares>
yes very likely
<kares>
`rspec -v`
<kares>
`bin/jruby -S rspec -v`
<kares>
should be 3.x
<kares>
bin/jruby -S gem install rspec to get latest
<lan5432>
Yeah, it's not able to find it
<lan5432>
Now, 3.4.4
<lan5432>
Now it's complaining about "describe" keyword, but I'm not using that...
<lan5432>
Oh, wait, I am, nvm
camlow325 has quit [Remote host closed the connection]