<RageLtMan>
RBX 2.3 spin up time seems to have gone up
|jemc| has quit [Ping timeout: 240 seconds]
jnh has joined #rubinius
yipstar has quit [Ping timeout: 265 seconds]
elia has quit [Quit: Computer has gone to sleep.]
jnh has quit [Ping timeout: 250 seconds]
|jemc| has joined #rubinius
havenwood has quit []
amsi has quit [Quit: Leaving]
jnh has joined #rubinius
amclain has joined #rubinius
jnh has quit [Ping timeout: 244 seconds]
amclain has quit [Quit: Leaving]
jnh has joined #rubinius
amclain has joined #rubinius
jnh has quit [Quit: Leaving...]
josh-k has joined #rubinius
josh-k has quit [Ping timeout: 245 seconds]
yorickpeterse has quit [Ping timeout: 255 seconds]
yorickpeterse has joined #rubinius
brixen has quit [Ping timeout: 244 seconds]
pd has quit [Ping timeout: 265 seconds]
pd has joined #rubinius
brixen has joined #rubinius
tenderlove has quit [Quit: Leaving...]
zacts has joined #rubinius
<zacts>
hi
<brixen>
hi zacts
<zacts>
can rubinius compile ruby code to native machine code via llvm? or is it only a JIT compiler/interpreter?
<|jemc|>
zacts: currently it cannot
<zacts>
are there plans for it to eventually be possible? :-)
meh` has quit [Ping timeout: 245 seconds]
<brixen>
zacts: you mean, ahead-of-time compile?
<brixen>
it already does compile Ruby to machine code as part of the JIT
<zacts>
oh yes, ahead-of-time compile. for speed.
<zacts>
running speed
<brixen>
you won't get much speed out of that
<zacts>
oh really, interesting.
<brixen>
the reason a JIT makes Ruby fast is because it runs *after* the types in the program have been identified
<brixen>
you can't know that statically
<zacts>
oh, I see
<brixen>
there are somethings you can know, but not nearly enough
<zacts>
thanks for clarifying that for me
<brixen>
no problem
<|jemc|>
brixen: trying to see metrics on graphite; sanity check - I point -Xsystem.metrics.statsd.server to the statsd UDP port, NOT the carbon port, correct?
<brixen>
yes, to the statsd port
<brixen>
btw, that's pretty sweet that there's a docker container for graphite :)
<brixen>
|jemc|: statsd has a port you can telnet to and see the stats it's getting
<|jemc|>
yeah - it makes that part of the process a lot easier - I think it's a good thing to recommend to users if they don't want to sully their system and waste their time
<brixen>
totally
<|jemc|>
the docker container that is
<brixen>
we'll have to do a post on this
<brixen>
want to write it? :)
<|jemc|>
I was actually thinking of doing just that - or a gist at least - as soon as I get it working myself
<|jemc|>
anyway, I'll check on the telnet thing
<|jemc|>
I think I need to forward the other docker port - but that's not too difficult to do
noop has joined #rubinius
<|jemc|>
ah, it was working the whole time - I just don't know how to use graphite apparently :P
<locks>
yorickpeterse: this is definitely the worst erik meijer talk I've seen
noop has quit [Ping timeout: 264 seconds]
amclain has quit [Quit: Leaving]
|jemc| has quit [Read error: Connection reset by peer]
|jemc| has joined #rubinius
Gibheer_ has joined #rubinius
Gibheer has quit [Read error: Connection reset by peer]
JohnBat26 has joined #rubinius
noop has joined #rubinius
|jemc| has quit [Read error: Connection reset by peer]
diegoviola has quit [Remote host closed the connection]
Ori_P has quit [Quit: Computer has gone to sleep.]
josh-k has joined #rubinius
Ori_P has joined #rubinius
josh-k has quit [Remote host closed the connection]
josh-k has joined #rubinius
Ori_P has quit [Quit: Computer has gone to sleep.]
elia has quit [Read error: Connection reset by peer]
josh-k has quit [Ping timeout: 244 seconds]
elia has joined #rubinius
dimday has joined #rubinius
cyndis has quit [Ping timeout: 265 seconds]
dimday has quit [Ping timeout: 244 seconds]
benlovell has joined #rubinius
benlovell has quit [Ping timeout: 255 seconds]
goyox86 has quit [Ping timeout: 240 seconds]
dimday has joined #rubinius
dimday has quit [Client Quit]
cyndis has joined #rubinius
Ori_P has joined #rubinius
houhoulis has quit [Remote host closed the connection]
benlovell has joined #rubinius
benlovell has quit [Ping timeout: 264 seconds]
zacts has quit [Ping timeout: 245 seconds]
zacts has joined #rubinius
zacts has joined #rubinius
Ori_P has quit [Quit: Computer has gone to sleep.]
goyox86 has joined #rubinius
benlovell has joined #rubinius
benlovell has quit [Ping timeout: 255 seconds]
benlovell has joined #rubinius
benlovell has quit [Client Quit]
nirvdrum has joined #rubinius
nirvdrum has quit [Changing host]
nirvdrum has joined #rubinius
houhoulis has joined #rubinius
noop has quit [Quit: Leaving]
Ori_P has joined #rubinius
Gibheer_ is now known as Gibheer
Ori_P has quit [Ping timeout: 240 seconds]
Ori_P has joined #rubinius
elia has quit [Quit: Computer has gone to sleep.]
<dreinull>
evening
|jemc| has joined #rubinius
Ori_P has quit [Ping timeout: 244 seconds]
diegoviola has joined #rubinius
tenderlove has joined #rubinius
havenwood has joined #rubinius
goyox86_ has joined #rubinius
goyox86 has quit [Ping timeout: 264 seconds]
<chrisseaton>
how do I attribute the Rubinius logo? The license is here https://github.com/rubinius/collateral. It says to attribute - but doesn't say the copyright owner or anything so not sure _how_ to attribute.
<brixen>
good question
<brixen>
shane becker was the author of most of that work
<brixen>
the conditions of use don't actually require knowing who owns the copyright
<brixen>
it could be an anonymous person and that would not change the conditions
<brixen>
in other words "Author of [referenced work]"
<brixen>
I'll ping shane, though
<chrisseaton>
cheers
<headius>
the license on github says CC-BY-ND
<chrisseaton>
headius: yeah - so I have to "attribute the work in the manner specified by the author", but no such manner is specified - we went through this for JRuby and I did a PR with an agreed upon attribution - remember?
<headius>
yup yup
<headius>
just wasn't sure you found the rubinius/collateral repo
<chrisseaton>
ah yeah
amsi has joined #rubinius
flavio has quit [Quit: WeeChat 1.0]
lbianc has quit [Quit: ChatZilla 0.9.91 [Firefox 24.8.1/20140923194127]]
DanielVartanov has quit [Remote host closed the connection]
<goyox86_>
dude, getting graphite running is freaking tedious
<yorickpeterse>
goyox86_: haha yeah
<yorickpeterse>
hm, no part 5 yeat
<yorickpeterse>
* yet
<brixen>
lol
<brixen>
so so close
<brixen>
stupid clocks, hours, time
goyox86_ has quit [Ping timeout: 240 seconds]
<|jemc|>
goyox86_: have you tried using the docker container?
<yorickpeterse>
brixen: what's our ETA on 2.3.1? I might want to clean up my dtrace stuff (so it's actually useful) and get that in, unless you were intending on releasing this weekend
<yorickpeterse>
(well might, I will clean it up)
<brixen>
tomorrow
<yorickpeterse>
Ah ok, that I can't make
<yorickpeterse>
since it will need some feedback and all that
<brixen>
it's ok, we'll release again next week
<brixen>
maybe twice or three times :)
<yorickpeterse>
heh
<brixen>
pretty soon you'll just release by pushing to master
<brixen>
maybe :)
<yorickpeterse>
:O
<brixen>
heh
<yorickpeterse>
I might want to stop pushing to master in that case
<brixen>
what's the worst that can happen
<yorickpeterse>
then again I already release weekly builds every Sunday
<yorickpeterse>
but nobody uses them
<yorickpeterse>
I think
arrubin has quit []
meh` has joined #rubinius
GitHub23 has joined #rubinius
<GitHub23>
[rubinius] YorickPeterse force-pushed dtrace-boxing from 2535035 to 1df5a70: http://git.io/855tFQ
<zacts>
I'm hoping to be able to run the sabotage distro on my laptop, but want the option to choose a ruby implementation
<brixen>
zacts: someone mentioned trying to link with musl
<brixen>
I haven't done so
<zacts>
ah ok
<zacts>
and is rubinius production ready in your opinion?
<brixen>
depends on what you're doing
<brixen>
Rubinius is in production
<zacts>
ah ok cool
<zacts>
what would you recommend not using rubinius for?
<zacts>
and, I'm just learning ruby right now, so nothing fancy yet.
<yorickpeterse>
zacts: asking that the maintainers might produce biased opinions :P
<zacts>
yeah, I mainly wanted to know if there are truly official releases of rubinius
<yorickpeterse>
Yes
<yorickpeterse>
e.g. I'm using it in productin
<yorickpeterse>
* production
<yorickpeterse>
Granted it's only 1 app out of 20 so far
<yorickpeterse>
brixen: wait, can I add line comments in Draftin?
<yorickpeterse>
or are they just general comments?
<yorickpeterse>
looks like the latter, hmm
<yorickpeterse>
Ah, you can highlight text and then comment
<yorickpeterse>
ooooooooh I see we're finally getting that-one-thing
<zacts>
I use draftin, I like it
<zacts>
I wish it had vim bindings though
<yorickpeterse>
heh
<yorickpeterse>
That's why I write pretty much everything in Vim :P
<yorickpeterse>
even comments on the interbutts
<zacts>
yeah
Ori_P has joined #rubinius
<brixen>
yeah, draftin is pretty nice
<brixen>
someone keeps sending me manually by email though :p
<brixen>
me edits manually*
<yorickpeterse>
wasn't me
<yorickpeterse>
Instead I use IRC :P
zacts has quit [Quit: leaving]
GitHub42 has joined #rubinius
<GitHub42>
[rubinius] YorickPeterse pushed 1 new commit to dtrace-boxing: http://git.io/Y2Nicw
GitHub42 has left #rubinius [#rubinius]
<GitHub42>
rubinius/dtrace-boxing 3bfd583 Yorick Peterse: Added RBX_DTRACE_CHAR for dtrace hooks....
<brixen>
yorickpeterse: heh, no, it wasn't you
max96at is now known as max96at|off
GitHub160 has joined #rubinius
<GitHub160>
[rubinius] YorickPeterse opened pull request #3214: DTrace/SystemTap object allocation hooks plus some housekeeping (master...dtrace-boxing) http://git.io/8_yIqQ
GitHub160 has left #rubinius [#rubinius]
<yorickpeterse>
ha
<yorickpeterse>
I forgot I also added method cache resets to the PR
<yxhuvud>
brixen: it may be a good idea to put links to earlier parts when you do that kind of series (and add links forward in the old entries). Makes it simpler for a reader that may have missed one part to find them
<GitHub48>
[rubinius] YorickPeterse pushed 1 new commit to dtrace-boxing: http://git.io/efn99Q
<GitHub48>
rubinius/dtrace-boxing effbc58 Yorick Peterse: Use comma for dummy object allocate macro.
GitHub48 has left #rubinius [#rubinius]
<|jemc|>
ah, multiple dispatch - that would be nifty in some cases
<brixen>
yxhuvud: yeah, forward links don't work so hot, but reverse links can be nice
<|jemc|>
also, maybe it's best I won't have time to get back to my myco project for a while - it sounds like a lot of nifty features to come for language implementers :)
<brixen>
yxhuvud: someone is working on a redesign that puts the post index next to the posts, so it'll be pretty hard to miss the links
<brixen>
|jemc|: the niftiest, if I do say so :)
<chrisseaton>
brixen: multiple dispatch is a good idea - Truffle uses something like that to specialise core lib methods to just the cases you need
<brixen>
it's a very old idea, for sure
<|jemc|>
brixen: is there an important reason to use def foo(a=Integer()) instead of def foo(Integer(a)) - it seems like the latter is less ambiguous (the former looks like you use the Integer() function to produce a default value)?
<chrisseaton>
brixen: have you started using MD or fun in your core library implementation on master yet? if so I will fork it now to avoid them for the time being
<brixen>
chrisseaton: it's a 3.0 feature
<chrisseaton>
brixen: ah ok - wasn't sure if 3.0 was now master
<brixen>
|jemc|: yes, there's an important reason and rbx compile -S -e will tell you what it is :)
<|jemc|>
brixen: all it's telling me is that the "def foo(Integer(a)); end" isn't valid syntax - which was my point about it being more desirable for a new semantic
<yorickpeterse>
|jemc|: using something like def foo(num: int64); end; means you can piggy back on the kwarg syntax
<yorickpeterse>
for arguments with default values you'd derive the type based on the default value
nirvdrum_ has quit [Ping timeout: 245 seconds]
<|jemc|>
yorickpeterse: well, I was referring to the multiple dispatch example using '=' instead of keyword arg syntax, but addressing your example: so in that paradigm, arguments with default values cannot be passed a type other than the type of the default value?
<yorickpeterse>
|jemc|: not sure, it would be one option
<yorickpeterse>
the alternative is to only disallow it when an explicit type is given
<|jemc|>
orickpeterse: to me it seems too different from the concept of a default value to share syntax with it - that is, I would think you would want to _not_ piggy back on other syntaxes
<|jemc|>
heh, "syntaces"?
<|jemc|>
anyway, I'm not complaining, just trying to understand the motivation
<yorickpeterse>
If you wouldn't re-use existing syntax you'd have to modify the Ruby parser