<yorickpeterse>
brixen: Oga release this weekend, so you can start preparing the lawyers at Enova :P
<|jemc|>
heh
<yorickpeterse>
and with that I'm off because "oh lord it's already 01:15"
<postmodern>
yorickpeterse, oops thanks, was an old failure
<headius>
yorickpeterse: performance?!
<headius>
if I have some cycles tonight I'll give it a shot...you still have good benchmarks in there?
<yorickpeterse>
headius: "meh" for now
<yorickpeterse>
best to benchmark on the "css" branch as I made a bunch of changes to the XML lexer there
<headius>
ok
<heftig>
hm, so will rbx 3 step away from the bison parser?
<yorickpeterse>
heftig: No
<yorickpeterse>
at least I highly doubt we'll be re-writing the Ruby parser in LPEG
<yorickpeterse>
That would be _a lot_ of work
postmodern has quit [Quit: Leaving]
<heftig>
"Adding parsing instructions also enables Rubinius to read pre-compiled bytecode describing parsing operations and execute them. This is extremely useful when bootstrapping, where we need to parse Ruby code to execute it, but we need to execute Ruby code to be able to parse it."
<heftig>
his paragraph seems to suggest that, though
<yorickpeterse>
That might've been an example, though I could be mistaken
<yorickpeterse>
But considering it took whitequark well over a year to do this in Ruby, and his parser isn't 100% compatible, I'd be surprised if we were going to rewrite it
<yorickpeterse>
and I don't think there are Bison -> LPEG converters either
<headius>
very unlikely :-)
<yorickpeterse>
MRI also randomly changes (and breaks) syntax on patch levels, keeping up with that would be mad
<headius>
in my 9 years working on Ruby, there has not been a 100% compatible (or even close) parser that wasn't basically just the bison grammar ported to some other setup
<headius>
the closest is probably an ANTLR grammar that Xruby and later Rhodes used
<yorickpeterse>
XRuby?
* yorickpeterse
is probably too young for this
<yorickpeterse>
Ah, Ruby to Java
<headius>
XRuby was a JVM Ruby that started out with a bytecode compiler and worked from there
<headius>
yeah, it never took off but the Rhomobile folks adopted that for their blackberry and android support
<yorickpeterse>
right, I'm off, toodles
<headius>
have a good one
houhoulis has joined #rubinius
josh-k has joined #rubinius
josh-k has quit [Remote host closed the connection]
|jemc| has quit [Ping timeout: 245 seconds]
<locks>
automatic updates is something else :D (reading the posts)
<headius>
I like it
<headius>
but I like ruby installers too
<locks>
from what I understood rbx will use whatever mechanism you used to install
<headius>
yeah, seems good if that's the case
<johnmuhl>
headius: What do you like so much about ruby installers? My experience has been that they add pain to a pretty simple process; especially for JRuby (which I feel has the best install experience).
<headius>
johnmuhl: for me it's a matter of managing many rubies reasonably well, rather than a few rubies perfectly
<locks>
johnmuhl: so far the chruby+ruby-install combo as served me well
<headius>
I have never had trouble installing or managing dozens of installs with rvm, but I'm not a heavy user
<johnmuhl>
Ah. I only need the latest of whatever Ruby the project requires so maintaining a bunch of versions has never been something I needed.
<locks>
different strokes ;)
dimday has joined #rubinius
goyox86 has joined #rubinius
jeremyevans has quit [Read error: Connection reset by peer]
goyox86 has quit [Ping timeout: 265 seconds]
<johnmuhl>
locks: I use ruby-install and a ~10 line fish function for auto-switching but using ruby-install is more about laziness rather than actually liking the thing.
<johnmuhl>
it’s just brew install … afterall
<locks>
why don't you brew install then?
<johnmuhl>
laziness
<locks>
haha
<johnmuhl>
ruby-install remembers the brew install … part better than me
<locks>
I had to install a bunch of versions to test mustache
<johnmuhl>
it’s the … part I prefer to not remember
<locks>
For example, some people are very interested in Rust and have asked if we're going to rewrite the virtual machine in Rust. We have no plans to do so, but if someone is interested, all they would need to do is implement the instruction set. The same goes for Haskell or Go or even asm.js.
* locks
perks ears
enebo has joined #rubinius
amsi has quit [Quit: Leaving]
tenderlove has quit [Quit: Leaving...]
amclain has joined #rubinius
goyox86 has joined #rubinius
<cremes>
locks: we need multiple implementations of the instruction set to prove its worth. ready? set? go!
goyox86 has quit [Ping timeout: 240 seconds]
parndt has joined #rubinius
<locks>
cremes: how else could it be, right
parndt has quit [Ping timeout: 264 seconds]
meh` has quit [Ping timeout: 244 seconds]
parndt has joined #rubinius
enebo has quit [Quit: enebo]
<cremes>
exactly. P)
<cremes>
my real hope for rubinius 3 is that there is very little c++ or whatever non-ruby language folks use for SPEED involved in implementation.
<cremes>
i like ruby. i want ruby.
<locks>
ruby's all you need
|jemc| has joined #rubinius
|jemc|_ has joined #rubinius
|jemc| has quit [Read error: Connection reset by peer]
jeremyevans has joined #rubinius
|jemc|_ has quit [Client Quit]
|jemc| has joined #rubinius
havenwood has quit [Ping timeout: 250 seconds]
havenwood has joined #rubinius
amclain has quit [Quit: Leaving]
goyox86 has joined #rubinius
goyox86 has quit [Ping timeout: 264 seconds]
_whitelogger_ has joined #rubinius
stormwind has joined #rubinius
LTe_ has joined #rubinius
jnh_ has joined #rubinius
justinmcp_ has joined #rubinius
the_g_cat has joined #rubinius
erdic_ has joined #rubinius
alexsuraci has quit [Ping timeout: 272 seconds]
dbussink has quit [Ping timeout: 272 seconds]
cyndis has quit [Ping timeout: 272 seconds]
_whitelogger has quit [Ping timeout: 272 seconds]
sbryant has quit [Ping timeout: 272 seconds]
LTe has quit [Ping timeout: 272 seconds]
thegcat has quit [Ping timeout: 272 seconds]
jnh has quit [Ping timeout: 272 seconds]
stormwin1 has quit [Ping timeout: 272 seconds]
halorgium has quit [Ping timeout: 272 seconds]
TheMoonMaster has quit [Ping timeout: 272 seconds]
erdic has quit [Ping timeout: 272 seconds]
justinmcp has quit [Ping timeout: 272 seconds]
dmilith has quit [Ping timeout: 272 seconds]
the_g_cat is now known as thegcat
alexsuraci_ is now known as alexsuraci
dbussink_ has joined #rubinius
dbussink_ is now known as dbussink
erdic_ is now known as erdic
dmilith has joined #rubinius
sbryant has joined #rubinius
halorgium has joined #rubinius
TheMoonMaster has joined #rubinius
amclain has joined #rubinius
noop has joined #rubinius
noop has quit [Ping timeout: 265 seconds]
parndt has quit []
<brixen>
cremes: we do, too :)
JohnBat26 has joined #rubinius
amclain has quit [Quit: Leaving]
|jemc| has quit [Quit: WeeChat 1.0.1]
|jemc| has joined #rubinius
RageLtMan has quit [Ping timeout: 250 seconds]
RageLtMan has joined #rubinius
goyox86 has joined #rubinius
goyox86 has quit [Ping timeout: 240 seconds]
<bennyklotz>
brixen: awesome post :)
dviola has quit [Read error: Connection reset by peer]
|jemc| has quit [Ping timeout: 272 seconds]
jnh_ has quit [Remote host closed the connection]
<RageLtMan>
just found myself wandering through the twilight zone as i realized that RBX 2.1 does a better job of functioning in a manner compatible with 1.9 than MRI 2.1.3. The gremlins are really starting to come out of the woodwork in that codebase
larstobi has joined #rubinius
goyox86 has joined #rubinius
goyox86 has quit [Ping timeout: 255 seconds]
jnh has joined #rubinius
jnh has quit [Ping timeout: 265 seconds]
soulhacker has joined #rubinius
kagaro has quit [Ping timeout: 244 seconds]
josh-k has joined #rubinius
josh-k_ has joined #rubinius
josh-k has quit [Ping timeout: 265 seconds]
dimday has quit [Ping timeout: 265 seconds]
noop has joined #rubinius
flavio has joined #rubinius
havenwood has quit [Remote host closed the connection]
<mpapis>
yorickpeterse, so rbx comes with RG-2.4.2 and can not be updated to 2.4.3?
<yorickpeterse>
mpapis: Yes, and yes it can, but we've had this problem in the past so many times where RVM tries to upgrade RubyGems and fail at it for whatever reason
<yorickpeterse>
it's why that option was added in the first place
<mpapis>
yorickpeterse, running setup.rb is still supported method of installing rubygems ... can we find why it fails?
<yorickpeterse>
well, dreinull would have to provide the full rubygems log as a start
<mpapis>
dreinull, can you show us /home/user/.rvm/log/1415874961_rbx-2.3.0/rubygems.install.log
<Rotonen>
wth MRI, buliding still silently breaks if there is a C_INCLUDE_PATH set in the system (even if it is set to sensible things) and building only works if that shell variable is unset? O.o
<Rotonen>
by silently breaking i mean completing and producing a subtly non-functional interpreter (one fine example is not finding symbols when referencing SHA1 code)
<yorickpeterse>
dreinull: problem starts at line 5, apparently it's trying to load corrupted bytecode
<mpapis>
yorickpeterse, check line 9 - checking if I can disable installation of docs
<yorickpeterse>
That shouldn't be causing it
<mpapis>
there it is --no-document - it's the first thing I set in my gemrc - that's why I never had this problem
<yorickpeterse>
I suspect the LLVM bug is caused due to the invalid bytecode
<mpapis>
yorickpeterse, it worked in line 8, docs fail in line 9
<yorickpeterse>
or is at least, somehow related
<yorickpeterse>
mpapis: I'm suspicious though
<yorickpeterse>
either way, if docs crash the JIT then that's a bug we need to fix
<yorickpeterse>
(in rbx)
<mpapis>
yorickpeterse, you can investigate the bug I will disable the docs so it works on older versions
<dreinull>
meanwhile I'm compiling…
<dreinull>
might as well replace the central heating unit with my laptop
<mpapis>
dreinull, rvm get head should fix the documentation crash issue
<dreinull>
mpapis next run
<dreinull>
with rubygems ignoer it worked well
josh-k_ has quit [Remote host closed the connection]
josh-k has joined #rubinius
kagaro has joined #rubinius
josh-k has quit [Ping timeout: 250 seconds]
jnh has joined #rubinius
jnh has quit [Ping timeout: 244 seconds]
noop has quit [Ping timeout: 240 seconds]
<cremes>
brixen: :P
noop has joined #rubinius
benlovell has quit [Ping timeout: 256 seconds]
<mpapis>
dreinull, as always commited and for got to push changes, the update is now online - if you had it running break it and run again after `rvm get head`