DireFog has quit [Read error: Connection reset by peer]
heftig has quit [Ping timeout: 260 seconds]
DireFog has joined #rubinius
heftig has joined #rubinius
josh-k has quit [Remote host closed the connection]
josh-k has joined #rubinius
josh-k has quit [Ping timeout: 250 seconds]
|jemc| has quit [Ping timeout: 244 seconds]
amsi has quit [Quit: Leaving]
havenwood has quit []
dimday has joined #rubinius
josh-k has joined #rubinius
dimday has quit [Quit: Leaving.]
dimday has joined #rubinius
houhoulis has joined #rubinius
|jemc| has joined #rubinius
brixen has quit [Ping timeout: 245 seconds]
nirix has quit [Ping timeout: 272 seconds]
nirix has joined #rubinius
brixen has joined #rubinius
justinmcp has quit [Remote host closed the connection]
justinmcp has joined #rubinius
houhoulis has quit []
postmodern has quit [Quit: Leaving]
meh` has quit [Ping timeout: 272 seconds]
craigp has joined #rubinius
craigp has quit [Remote host closed the connection]
josh-k has quit [Remote host closed the connection]
josh-k has joined #rubinius
josh-k has quit [Read error: Connection reset by peer]
rtl has quit [Remote host closed the connection]
JohnBat26 has joined #rubinius
havenwood has joined #rubinius
havenwood has quit [Client Quit]
houhoulis has joined #rubinius
craigp has joined #rubinius
postmodern has joined #rubinius
houhoulis has quit [Remote host closed the connection]
dimday has quit [Ping timeout: 272 seconds]
craigp has quit [Remote host closed the connection]
havenwood has joined #rubinius
flavio has joined #rubinius
benlovell has joined #rubinius
noop has joined #rubinius
flavio has quit [Quit: WeeChat 0.4.1]
elia has joined #rubinius
flavio has joined #rubinius
flavio has quit [Changing host]
flavio has joined #rubinius
havenwood has quit [Remote host closed the connection]
havenwood has joined #rubinius
havenwood has quit [Ping timeout: 264 seconds]
noop has quit [Ping timeout: 260 seconds]
<yorickpeterse>
morning
<yorickpeterse>
|jemc|: you can also roll a "reverse visitor pattern" or w/e you'd call it. That is, you have some Evaluator class that has on_XXX methods for generating bytecode
<yorickpeterse>
That way you don't have to create dozens of AST classes
noop has joined #rubinius
josh-k has joined #rubinius
imajes has quit [Ping timeout: 244 seconds]
Mugatu has quit [Ping timeout: 245 seconds]
headius has quit [Ping timeout: 260 seconds]
headius has joined #rubinius
Mugatu has joined #rubinius
imajes has joined #rubinius
postmodern has quit [Quit: Leaving]
benlovell has quit [Ping timeout: 250 seconds]
cremes has quit [Ping timeout: 245 seconds]
cremes has joined #rubinius
benlovell has joined #rubinius
<yorickpeterse>
oh hey, today was apparently the last day of RGSoC
<cremes>
i think i understand his cryptic in-channel comments from yesterday but i want the promise of “real soon now” to be on a public branch.
<yorickpeterse>
They're uploading the videos in a bit of a weird order
<yorickpeterse>
matz was the first one on day 1, brixen the first one on day 2
<cremes>
yorickpeterse: i want to see the ruby buddha one too :)
<yorickpeterse>
but Pat was like the whatever-nth on day 1
<yorickpeterse>
or day 2, don't remember
<yorickpeterse>
also can confirm interesting things coming up
<yorickpeterse>
but I'll not spoil the fun
<cremes>
i found the part about “fun” in addition to “method” to be interesting
<cremes>
yorickpeterse: looks like a pun with “fun” in your comment, heh
<cremes>
do you have a rubinius-x branch of oga using fun? ;)
<yorickpeterse>
nah, first gotta make it not leak memory on Rbx
<yorickpeterse>
but for that we need to fix some CAPI handles stuff
<yorickpeterse>
and for that I need to be able to actually compile master :?
<yorickpeterse>
* :/
<yorickpeterse>
well, I can do that by disabling the console
<cremes>
what? that’s bullshit, man!
<cremes>
time to deprecate C API and make a C API 2.0
<yorickpeterse>
oh yeah uhh....I'll keep quiet about that :P
<yorickpeterse>
But no, the CAPI won't go away any time soon
<yorickpeterse>
Considering we've discussed this here anyway: basically we talked about adding some kind of flag in "well behaved extensions"
<yorickpeterse>
so that we can do better cleanups in the CAPI (something like that from what I remember)
<cremes>
yeah, the current C API has the start of that with some flags regarding handling of RSTRING if i recall
<yorickpeterse>
so if I remember correctly we keep CAPI handles around forever, or at least for a long time
<yorickpeterse>
which is because we don't know if extensions keep data around forever or not, or stuff it somewhere in a global variable
<yorickpeterse>
In this case the flag would say "Hey, I play nice and don't do that", allowing us to remove handles nicely
<cremes>
so that resolves a memory leak then
<yorickpeterse>
Yup
havenwood has joined #rubinius
<brixen>
yorickpeterse: for your case, we should be able to avoid the C-API completely
<yorickpeterse>
oh?
<brixen>
by rewriting your parser not to use racc
<brixen>
brb, need to reboot to install os x command line tools lol
<yorickpeterse>
racc isn't the problem actually
<yorickpeterse>
well, not just
<yorickpeterse>
The Ragel lexer is also in C
<yorickpeterse>
Rewriting the parser in something else isn't something I'm going to do any time soon either, unless I really have to
benlovell has quit [Ping timeout: 245 seconds]
benlovell has joined #rubinius
caiooliveira has joined #rubinius
<caiooliveira>
Anyone failing to check PKCS5_PBKDF2_HMAC()
<yorickpeterse>
eh?
<caiooliveira>
Rake is aborting whenever trying to check this function during rubinius installation.
<caiooliveira>
It actually fails to com[pile a c code which calls this function
<caiooliveira>
Any clue? I think I`m being a little vague
|jemc| has quit [Ping timeout: 260 seconds]
<yorickpeterse>
Do you mean when installing a Gem, or Rubinius itself?
<caiooliveira>
Rubinius itself
<caiooliveira>
It might be something in my system
<caiooliveira>
I think
|jemc| has joined #rubinius
havenwood has quit [Remote host closed the connection]
havenwood has joined #rubinius
havenwood has quit [Remote host closed the connection]
<brixen>
yorickpeterse: pegs don't have a separate lexing step
<brixen>
also, I wouldn't expect you to rewrite it
<yorickpeterse>
brixen: I have strong doubts about kpeg and the likes being more performant than what I have now
<yorickpeterse>
but that's based on fairy dust and FUD
<brixen>
I'm not thinking kpeg
<|jemc|>
yorickpeterse: kpeg is pretty slow for my grammar - but I'm moving to a compiled bytecode parser
<yorickpeterse>
|jemc|: parsing time is one thing
<yorickpeterse>
memory usage is another beast
<|jemc|>
yorickpeterse: haven't measured memory usage yet - but compared to kpeg, this compiled parser constructs the AST in a second step rather than as it goes
<|jemc|>
so it doesn't waste memory/time constructing AST nodes on an ultimately failed path
<yorickpeterse>
The AST nodes aren't the biggest problem actually, at least not in my case
<yorickpeterse>
it's String allocations that mess it up
<|jemc|>
my parser also doesn't slice up any strings until the second step as well - it only notes the indices of captures
havenwood has joined #rubinius
<yorickpeterse>
Dynamic slicing is something I have planned for the future, but it's pretty tricky to get right
<yorickpeterse>
"Those are probably using an LLVM-enabled rubinius, and our rubinius port
<yorickpeterse>
isn't LLVM-enabled as it doesn't support the version of LLVM in ports."
<yorickpeterse>
what the fuck is that
<yorickpeterse>
"oh it's slower because we're not enabling core features of it"
heroux__ has joined #rubinius
<cremes>
IIRC openbsd is on like llvm 3.0 or something ancient
thegcat has quit [Ping timeout: 244 seconds]
gaustin has quit [Ping timeout: 244 seconds]
heroux_ has quit [Ping timeout: 244 seconds]
gaustin_ is now known as gaustin
the_g_cat is now known as thegcat
<cremes>
i think there were also issues with how llvm was compiled, RE: rubinius wants it without some flags enabled (RTTI?)
|Blaze| has joined #rubinius
<cremes>
but those flags are on by default in openbsd. i don’t remember the details though i bet they’re in an issue somewhere.
<yorickpeterse>
meh, their cup of tea. heading home, bbl
|Blaze| has quit [Remote host closed the connection]
|Blaze| has joined #rubinius
elia has joined #rubinius
elia has quit [Ping timeout: 260 seconds]
<jeremyevans>
cremes: Actually, our LLVM was too new, not too old :)
<cremes>
jeremyevans: ah! i couldn’t remember. how about now… still too new?
<jeremyevans>
Not sure. I haven't seen a mention in the release notes that newer versions of LLVM are supported
<jeremyevans>
So I haven't tried it
<|jemc|>
LLVM 3.5 support is on master now I think, but not yet released
<cremes>
^^
<jeremyevans>
Well, when it is released, I'll hook LLVM support back up :)
<cremes>
cool, that would be great
havenwood has quit [Remote host closed the connection]
<cremes>
at baruco brixen mentioned a new release within a few weeks so hopefully we’ll see one shortly here.
havenwood has joined #rubinius
<jeremyevans>
The OpenBSD LLVM port maintainer is very aggressive about updating the LLVM version, which is the reason it's often out of sync with what our rubinius port supports
havenwood has quit [Remote host closed the connection]
benlovell has joined #rubinius
benlovell has quit [Ping timeout: 272 seconds]
elia has joined #rubinius
elia has quit [Quit: Computer has gone to sleep.]
elia has joined #rubinius
rtl has joined #rubinius
noop has joined #rubinius
<johnmuhl>
Any ideas why Dir.glob(...).size on the same directory might return a different number than MRI or JRuby (which return the same size)?
<|jemc|>
johnmuhl: do you see any pattern to what is being included/excluded in/from in rubinius?
<|jemc|>
(in/from the glob in rubinius)
<johnmuhl>
ah ha. Dir.glob("A*") includes stuff that begins with, for example, Ä
<johnmuhl>
in rubinius that is
guilleiguaran___ has quit [Ping timeout: 260 seconds]
mjc_ has quit [Ping timeout: 272 seconds]
guilleiguaran___ has joined #rubinius
mjc_ has joined #rubinius
johnmuhl has quit [Ping timeout: 260 seconds]
johnmuhl has joined #rubinius
guilleiguaran___ has quit [Ping timeout: 260 seconds]
alexsuraci has quit [Ping timeout: 260 seconds]
alexsuraci has joined #rubinius
guilleiguaran___ has joined #rubinius
<cremes>
johnmuhl: sounds like an encodings bug
<johnmuhl>
cremes: do you mean in my script or in rubinius?
<GitHub134>
rubinius/1.8.7-README-updates-bison 031146e jc00ke: Adding notes on building w/different versions of bison...
<GitHub134>
[rubinius] jc00ke created 1.8.7-README-updates-bison (+1 new commit): http://git.io/C9lxmw
GitHub48 has joined #rubinius
<GitHub48>
[rubinius] jc00ke opened pull request #3150: Adding notes on building w/different versions of bison (1.8.7...1.8.7-README-updates-bison) http://git.io/lK6glw