Index

Show enters and exits. Hide enters and exits.

00:00:20evancoachella
00:00:24evanbig festival
00:00:32evansaw like 20 bands
00:00:39agardinerawesome!
00:00:47evanincluding Portishead and Prince
00:01:12agardinerwow! Don't seem to hear much about Prince these days...
00:01:33agardinersounds like a great gig
00:01:34brixenportishead, sweet
00:01:47evanyeah
00:01:51evani'm not really into his music
00:01:56evanbut he puts on one hell of a show
00:02:09agardinerhehe, yeah
00:02:16agardinerhe's a showman alright
00:07:12fbuilesv enters the room.
00:08:38brixenDefiler: this is a bit simpler: http://pastie.org/188393
00:08:44brixenDefiler: put that in a file in the root of rbx
00:09:08brixenthen compare the output of: bin/mspec <file>; bin/mspec -tr <file>; and spec <file>
00:12:15mentz_ leaves the room.
00:13:01Defilerwith or without the other patches?
00:13:19brixenwith
00:13:33brixenwell, just the IOStub patch
00:13:44brixenthe 2nd patch is just modifying specs
00:16:24Defilerso.. why is there even anything waiting on stdin in this case?
00:17:01brixenwaiting? do you mean the string you get in mri with mspec?
00:19:59DefilerOK, so..
00:20:06Defilerthe io helper code is not being invoked under MRI
00:21:10brixenwell, with mspec, but it is with rspec
00:21:18AndrewO enters the room.
00:23:01DefilerOK, so.. what you are saying is that there is some problem in mspec under MRI
00:23:10Defilerthat does not occur in rspec or in mspec under rbx
00:23:51brixenyeah
00:26:48brixenDefiler: no worries, just wanted some confirmation that I wasn't totally seeing things
00:27:01brixengotta run to class...
00:27:47Defilerhrm
00:28:55DefilerThis is a wacky one
00:29:31brixenyeah, it is
00:29:46brixenok, back later...
00:30:09rueSo in summary, I think you all suck ass
00:30:15rueOh, and good morning
00:32:21DefilerWell put
00:42:03ctennis leaves the room.
00:42:45wycats_ enters the room.
00:43:01ctennis enters the room.
00:43:03ctennis leaves the room.
00:44:29nicksieger leaves the room.
00:54:37agile enters the room.
00:58:53wycats leaves the room.
01:00:55nicksieger enters the room.
01:01:19jtoy enters the room.
01:06:20benstiglitz leaves the room.
01:07:01kw enters the room.
01:12:59marnen enters the room.
01:20:10jtoy leaves the room.
01:24:31mkrauskopf leaves the room.
01:39:17benburkert leaves the room.
01:40:37jennyw enters the room.
01:42:56twbray leaves the room.
01:45:45pd enters the room.
01:47:30imajes leaves the room.
01:49:54binary42 enters the room.
01:54:22crafterm enters the room.
01:56:57dlee leaves the room.
01:57:26twbray enters the room.
01:59:52xhanjian enters the room.
02:00:49jtoy enters the room.
02:13:30marnen_ enters the room.
02:15:59ixx leaves the room.
02:16:46benburkert enters the room.
02:19:15ixx enters the room.
02:21:43MenTaLguY enters the room.
02:23:17AndrewO leaves the room.
02:24:14VVSiz_ enters the room.
02:28:12benny leaves the room.
02:30:29marnen leaves the room.
02:31:32VVSiz leaves the room.
02:31:57dfg59 enters the room.
02:33:15evanmuhah
02:33:19MenTaLguYaha?
02:33:37evantaught Compiler::ExecuteContext how to handle blocks
02:33:59yugui enters the room.
02:34:14dfg59i've been working on getting File#join to work with recursive arrays and it seems like it makes sense to add a private helper method on Array
02:34:19MenTaLguYyay!
02:34:20dfg59similiar to recursively_flatten
02:34:24dfg59what's the procedure for that?
02:34:29tarcieriMenTaLguY: so I'll probably try out that MVM Mongrel load balancer this weekend
02:34:34MenTaLguYtarcieri: cool
02:34:36tarcierisorry I didn't respond to your e-mail
02:34:41MenTaLguYs'ok
02:34:44evantarcieri: i'm working on a new patch for ragel
02:34:50tarcierievan: nice
02:34:51evanbeen email with Adrian
02:34:53MenTaLguYtarcieri: I was mainly CCing you in case I got some details wrong
02:34:57tarcierievan: that's awesome
02:35:05tarcieriMenTaLguY: yeah that's pretty much what I was thinking, so...
02:35:44tarcierievan: I'm wondering if I should move http11.rb down a directory to be more compatible with the present Mongrel, which looks for http11 in the toplevel
02:35:53evanplease do
02:35:56tarcieriok
02:35:57MenTaLguYtarcieri: btw, I think I've figured out inter-VM actors
02:35:57evanthe layout should be the same
02:36:04tarcieriyeah
02:36:13tarcieriMenTaLguY: what about them?
02:36:21tarcierimessage serialization?
02:36:25MenTaLguYtarcieri: after talking with Tim Bray last night I started thinking about the missing bits real hard, and I got the marshaling portions sorted out
02:36:27MenTaLguYyeah
02:36:32tarcierinice
02:36:56MenTaLguYone of the more interesting developments is that I think we should add a per-actor flag whether to trap messages that fail to un-serialize
02:37:08MenTaLguY(by default they would get discarded)
02:37:14MenTaLguYor something along those lines
02:37:21tarcieriMenTaLguY: if we could get something like Ports going it'd be a lot easier to factor the "VM Manager" into its own class
02:37:35MenTaLguYyeah
02:37:40MenTaLguYbut I think I've worked out a really minimal design
02:37:44tarcierinice
02:38:14MenTaLguYit helps that Rubinius is finally far along enough with Marshal
02:38:19MenTaLguYdon't think it was for a while
02:39:20MenTaLguYanyway, hopefully I'll be able to have at least a proof-of-concept in by tonight
02:39:29boyscout1 commit by Tony Arcieri
02:39:30boyscout * Move http11.rb to where the Mongrel gem expects it; 515c7af
02:39:38MenTaLguYevan: is there a way to get the VM id of the current VM?
02:39:45MenTaLguYif not I think I'll need to add that ability
02:39:48tarcieriRubinius::VM_ID
02:39:55MenTaLguYoh, nice
02:39:58tarcierithe current VMActor stuff relies on that
02:40:04MenTaLguYit would need to yeah
02:40:27evanyep
02:41:05MenTaLguYI've also been sketching out an implementation of the Rubinius MVM API for JRuby
02:41:29evand0pe.
02:41:30evan:)
02:41:32tarcieriso what's the status of the whole standard MVM interface thingy that everyone seems to want
02:41:45MenTaLguYI think we need to try some stuff and see what sticks
02:41:56anteaya enters the room.
02:42:04MenTaLguYRubinius' API is a good starting place
02:42:24tarcieriI mean, there's this, supposedly:
02:42:25tarcierihttp://www.sun.com/aboutsun/pr/2008-02/sunflash.20080227.5.xml
02:42:57MenTaLguYyep
02:43:15fbuilesv leaves the room.
02:43:23MenTaLguYbut as far as stock Ruby goes, the API is more or less the smallest issue
02:43:30tarcieriheh
02:43:32tarcieriindeed
02:43:57MenTaLguYI don't think we'll be stepping on any toes by trying out the Rubinius API in JRuby to see how well it carries over
02:44:06tarcieriyeah, definitely
02:45:13headius enters the room.
02:45:21MenTaLguYhey charlie
02:46:02benny enters the room.
02:47:40rubuildius_amd64Tony Arcieri: 515c7afc6; 2091 files, 6722 examples, 23657 expectations, 0 failures, 0 errors; http://rafb.net/p/BNatkk25.html
02:47:53boyscout1 commit by Evan Phoenix
02:47:54boyscout * Taught ExecuteContext about blocks; d88906a
02:48:54dfg59so if i add a new private helper function to Array, where should i put it's spec?
02:49:56dfg59i ask because i don't see a spec for recursively_flatten (also a private method on Array) anywhere
02:51:44brixendfg59: you can put it in spec/core/array
02:51:58brixensince it's for us and not part of standard ruby
02:52:03radarek leaves the room.
02:52:08brixenjust use the same naming convention
02:52:19brixene.g. Array#fooby => fooby_spec.rb
02:52:23dfg59brixen: alright, i think it's really done this time :) the helper function will use RecursionGuard to remove any extra outer arrays.
02:52:29brixensweet
02:52:35brixenthat was one hell of a thorny problem
02:52:38boyscout1 commit by MenTaLguY
02:52:39boyscout * split out Actor::Filter to a separate file; 0896973
02:52:52dfg59do you know Charles L irc name? want to thank him for his help
02:53:01brixenactually, I don't
02:53:05brixenI wanted to thank him too :)
02:53:09tarcieriMenTaLguY: so if I wanted to take a crack at Port
02:53:17tarcieriwhat's the best way to interface with the current implementation?
02:53:19tarcieri@ready?
02:53:30dfg59i felt like i was ping-pong pairing...but distributed :)
02:53:30brixen@get_set, @go?
02:53:50brixendfg59: yeah, it was cool to follow along :)
02:54:19tarcieriMenTaLguY: as in, Port would send the fileno of the "active mode" IO object to... some channel
02:55:02jennyw leaves the room.
02:55:14headiushiya MenTaLguY
02:55:23dfg59brixen: upon adding a new spec, is it necessary to update it's meta information anywhere?
02:56:40bitbang_ enters the room.
02:58:09MenTaLguYtarcieri: hm, not sure really
02:58:30evantarcieri: want the new ragel patch?
02:58:32MenTaLguYtarcieri: I think you'd want to modify Scheduler.send_on_writable so that it could send a specific object rather than nil
02:58:33evani sent it to Adrian
02:58:34tarcierievan: Sure
02:58:37MenTaLguYtarcieri: to begin with
02:58:40evantarcieri: email?
02:58:49tarcierievan: tony@medioh.com
02:59:03tarcieriMenTaLguY: well, this is for multiplexing IO reads with incoming Actor messages
02:59:24evanok, sent.
02:59:28tarciericool
02:59:39tarcieriI'll let you know if it fixes the parse error in the output
02:59:48evanshould
02:59:52tarciericool
03:00:05brixendfg59: should not be necessary unless it fails (i.e. add a tag), but I expect it will pass :)
03:00:23dfg59brixen: let's hope :\
03:00:39MenTaLguYtarcieri: something I've also been wanting for a while is a Scheduler primitive that reads an object from a channel and writes it to another channel
03:00:52MenTaLguYas a schedulable/cancellable action
03:01:19tarcierinice
03:01:35tarcierithat'd be great for selectively multiplexing different types of messages
03:01:38headius leaves the room.
03:01:39MenTaLguYexactly.
03:01:52dfg59oh man...Array#== doesn't work with recursive arrays...this should make writing a spec interesting
03:02:07MenTaLguYit would make timeoutable read on a channel possible too
03:02:20MenTaLguYwhich I've come to think is pretty important
03:02:32tarcierisome sort of tool for multiplexing multiple channels in general would be nice, regardless of how it's implemented
03:02:33MenTaLguYbut isn't something that should be rolled into Channel directly
03:02:39headius enters the room.
03:02:48dfg59guess i'll open a ticket and tackle that first
03:03:02MenTaLguYwell, if you want a general tool you probably also would want to have some mechanism for transforming the message before it got written to the second channel
03:03:22tarcieriyeah pretty much
03:03:29rubuildius_amd64MenTaLguY: 089697357; 2091 files, 6722 examples, 23657 expectations, 0 failures, 0 errors; http://rafb.net/p/nPkoLa30.html
03:03:30rubuildius_amd64Evan Phoenix: d88906a56; 2091 files, 6722 examples, 23657 expectations, 0 failures, 0 errors; http://rafb.net/p/PNEl6V11.html
03:03:35tarcierisomething which encapsulates the message with where it came from somehow
03:03:40MenTaLguYyes
03:06:01tarcieriyep, that'd definitely help with the Actor code, at the very least
03:06:57bitbang leaves the room.
03:07:22obvio171 leaves the room.
03:10:27obvio enters the room.
03:11:37rubuildius_ppcMenTaLguY: 089697357; 2091 files, 6724 examples, 23683 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188465
03:11:38rubuildius_ppcEvan Phoenix: d88906a56; 2091 files, 6724 examples, 23683 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188464
03:11:39rubuildius_ppcTony Arcieri: 515c7afc6; 2091 files, 6724 examples, 23683 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188461
03:12:57Fullmoon leaves the room.
03:14:12MenTaLguYhm, I think I'm going to head off early tonight
03:14:14MenTaLguY'night
03:14:33MenTaLguY leaves the room.
03:16:21lstoll leaves the room.
03:19:38headius leaves the room.
03:20:34nicksieger leaves the room.
03:20:48loincloth enters the room.
03:23:25Fullmoon enters the room.
03:31:50boyscout4 commits by Drew Olson
03:31:51boyscout * Updated File#join to use Array#remove_outer_arrays; ac64eaa
03:31:52boyscout * Added private method Array#remove_outer_arrays.; f02b62e
03:31:53boyscout * Added spec for Array#remove_outer_arrays; df94214
03:31:54boyscout * Added more edge cases for recursive arrays to spec for File#join; ec4ece9
03:40:52yugui leaves the room.
03:41:08yugui enters the room.
03:42:17rubuildius_amd64Drew Olson: ac64eaa9a; 2092 files, 6737 examples, 23673 expectations, 0 failures, 0 errors; http://rafb.net/p/mKPvRV46.html
03:45:05rubuildius_ppcDrew Olson: ac64eaa9a; 2092 files, 6739 examples, 23699 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188480
03:46:30benburkert leaves the room.
03:47:20tarcierihey evan... just in case you're wondering, your patch fixes the parse errors
03:50:46wmoxam enters the room.
03:51:09crafterm leaves the room.
03:51:26twbray leaves the room.
03:52:59boyscout1 commit by Tony Arcieri
03:53:00boyscout * Regenerate Mongrel parser using Evan's latest Ragel patch; b28e634
03:59:43bitbang_ leaves the room.
04:01:46tarcieriPerformance is unchanged, but I assume JIT will fix that eventually :)
04:02:16rubuildius_amd64Tony Arcieri: b28e63414; 2092 files, 6737 examples, 23673 expectations, 0 failures, 0 errors; http://rafb.net/p/WiYqxm71.html
04:05:44femtowin enters the room.
04:06:02rubuildius_ppcTony Arcieri: b28e63414; 2092 files, 6739 examples, 23699 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188488
04:07:32femtowinerror: Failed connect to git.rubini.us:8080; No such file or directory
04:07:56femtowin(curl_result = 7, http_code = 0, sha1 = fb6fb420828518c0995df12be823e85a838d3da5)
04:08:26tarcieriheh curl
04:13:48femtowin leaves the room.
04:15:23meanphil_ enters the room.
04:16:17Rich_MorinI'm trying to understand the way that Ruby 1.9 handles method parameters. Apparently, a list of parameters with defaults can be placed before, in the middle of, or after the list of "ordinary" parameters. Is this simply a "convenience" in writing parameter lists or does it affect usage, as well?
04:18:52femtowin enters the room.
04:18:58boyscout2 commits by Marnen Laibow-Koser
04:18:59boyscout * Fix regression in BigDecimal#add.; 2349742
04:19:00boyscout * Committing so we can bisect.; a11a107
04:21:00meanphil leaves the room.
04:21:54d2dchat enters the room.
04:27:17rubuildius_amd64Marnen Laibow-Koser: 234974229; 2092 files, 6746 examples, 23976 expectations, 0 failures, 0 errors; http://rafb.net/p/xq47c960.html
04:27:25lstoll enters the room.
04:32:05rubuildius_ppcMarnen Laibow-Koser: 234974229; 2092 files, 6748 examples, 24002 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188497
04:36:35GMFlash leaves the room.
04:36:43GMFlash enters the room.
04:38:49agardinerDefiler, evan: did you resolve those issues related to class << self you were working on last Friday?
04:39:17agardinersounded like you'd worked through most/all of it, but I don't see a commit...
04:40:57benburkert enters the room.
04:45:32ubiquitous enters the room.
05:00:45anteaya leaves the room.
05:05:44ubiquitous leaves the room.
05:06:36dfg59 leaves the room.
05:10:06loincloth leaves the room.
05:14:26loincloth enters the room.
05:15:29wmoxam leaves the room.
05:21:31cyndis leaves the room.
06:02:56ttmrichter leaves the room.
06:03:34ttmrichter enters the room.
06:13:33rueagardiner: I do not recall that having been fully fixed
06:14:06agardineri thought evan had cracked it...
06:15:11ezmobius enters the room.
06:30:54joachimm enters the room.
06:45:33benburkert leaves the room.
06:45:40benburkert enters the room.
06:57:42loincloth leaves the room.
07:00:10mentz_ enters the room.
07:00:29nicksieger enters the room.
07:05:10joachimm leaves the room.
07:14:09Maledictus enters the room.
07:17:49d2dchat leaves the room.
07:28:52femtowin enters the room.
07:31:35headius enters the room.
07:34:10mkrauskopf enters the room.
07:37:28wycats enters the room.
07:49:11jtoy leaves the room.
07:57:18cout leaves the room.
07:57:41headius leaves the room.
08:00:09femtowin leaves the room.
08:00:14dlee enters the room.
08:06:17dlee leaves the room.
08:07:05agardiner leaves the room.
08:15:05benburkert leaves the room.
08:18:18dysinger leaves the room.
08:20:33cout enters the room.
08:22:30thehcdreamer enters the room.
08:23:50dysinger enters the room.
08:34:28mentz__ enters the room.
08:36:44mentz enters the room.
08:38:10jtoy enters the room.
08:41:08trythil leaves the room.
08:44:17rby_ enters the room.
08:44:24yugui enters the room.
08:46:25w1rele55 enters the room.
08:52:28mentz_ leaves the room.
08:53:19imajes enters the room.
08:53:47mentz__ leaves the room.
08:54:46octopod enters the room.
08:56:23mentz leaves the room.
08:56:34mutle enters the room.
09:04:04lstoll leaves the room.
09:08:25olabini leaves the room.
09:08:46VVSiz_ enters the room.
09:17:53danlucraft enters the room.
09:26:38VVSiz leaves the room.
09:26:49Arjen_ enters the room.
09:27:05yugui leaves the room.
09:30:23ezmobius leaves the room.
09:37:10olabini enters the room.
09:39:56kw leaves the room.
09:40:16kw enters the room.
09:57:42_mutle enters the room.
09:58:59mutle leaves the room.
09:59:56mutle enters the room.
10:05:15hornbeck leaves the room.
10:14:02_mutle leaves the room.
10:14:08octopod_ enters the room.
10:17:38octopod leaves the room.
10:25:41kw leaves the room.
10:41:04mentz enters the room.
10:45:32aquasync enters the room.
10:50:03JimM2 enters the room.
10:51:36kw enters the room.
10:51:51kw leaves the room.
10:56:16olabini leaves the room.
10:57:22olabini enters the room.
11:05:23qwert666 enters the room.
11:11:42GMFlash leaves the room.
11:35:22be9 enters the room.
11:55:43Fullmoon leaves the room.
12:00:40yaroslav enters the room.
12:06:42riffraff enters the room.
12:07:18riffraffhi
12:12:39dbussinkhowdy
12:23:46mentz leaves the room.
12:25:52jtoy leaves the room.
12:27:18rby_ leaves the room.
12:39:52dbussinkanother quiet day in #rubinius :P
12:40:48rue'S cause you are never here ;)
12:42:54arjen__ enters the room.
12:42:55Arjen_ leaves the room.
12:48:41ctennis enters the room.
12:56:48dbussinkrue: i'm here, but talking to myself seems so useless :)
12:57:26wdperson enters the room.
13:04:19lstoll enters the room.
13:12:56w1rele55 leaves the room.
13:16:31mentz enters the room.
13:30:37w1rele55 enters the room.
13:33:55pd leaves the room.
13:36:33JimMc leaves the room.
13:39:05mentz leaves the room.
13:45:16pd enters the room.
13:45:31radarek enters the room.
13:46:42radarek_ enters the room.
13:47:47radarek leaves the room.
13:48:35ruedbussink: Well, true enough. You are a salaryman now
13:50:36JimMc enters the room.
14:02:00pd leaves the room.
14:06:10mkrauskopf leaves the room.
14:06:25mkrauskopf enters the room.
14:08:03AndrewO enters the room.
14:10:15radarek enters the room.
14:11:10radarek_ leaves the room.
14:11:47moofbong enters the room.
14:12:19moofbong leaves the room.
14:12:20mentz enters the room.
14:18:18d2dchat enters the room.
14:19:26moofbong enters the room.
14:22:15smparke1 leaves the room.
14:27:23imajes leaves the room.
14:27:36fbuilesv enters the room.
14:27:36aquasync leaves the room.
14:29:58agile leaves the room.
14:33:06srbaker leaves the room.
14:38:30srbaker enters the room.
14:38:40probablycorey enters the room.
14:43:23headius enters the room.
14:44:46wmoxam enters the room.
14:46:01GMFlash enters the room.
15:05:19jtoy enters the room.
15:05:35jtoy leaves the room.
15:11:02pd enters the room.
15:14:19yaroslav leaves the room.
15:17:29srbaker leaves the room.
15:20:26fbuilesvevan: did you ever upload the doxygen config file?
15:22:57djwhittfbuilesv: http://rubini.us/static/doxygen/
15:23:15Rich_Morin_ enters the room.
15:23:15djwhittcorundum: doxygen is http://rubini.us/static/doxygen/
15:23:16corundumokay
15:23:24djwhittcorundum: doxygen?
15:23:25corundumdoxygen is http://rubini.us/static/doxygen/
15:23:27djwhittnice
15:23:34fbuilesvdjwhitt: do you know where the config file for that is?=
15:24:01djwhittfbuilesv: oh sorry, didn't fully read your message
15:24:02djwhittno I don
15:24:05djwhitt*don't
15:24:16Rich_MorinIs there an irb (irbx?) for Rubinius? If not, where does it sit on the timeline?
15:24:32fbuilesvRich_Morin_: rbx works as an irb
15:24:36fbuilesvworks as irb*
15:24:42djwhittRich_Morin_: yeah, should fire up if you run rubinius without telling it what script to run
15:25:01Rich_MorinThanks!
15:30:42Rich_MorinIt doesn't seem happy with me: http://pastie.caboo.se/188652
15:31:01smparke1 enters the room.
15:31:30danlucraftThose docs are nice.
15:31:49danlucraftThough you might say, tab-happy. :)
15:32:42TheProkrammerStill would like to get rubinius compiling under cygwin...
15:32:56fbuilesvRich_Morin_: does it happen too when you run $ ./shotgun/rubinius.bin ?
15:33:52TheProkrammerActually let me rephrase, I'd like to get it running properly with FFI. I have it compiling.
15:33:55chris2 enters the room.
15:34:43Rich_Morinfbuilesv: yes FYI, this is on a Power Mac G5, running OSX 10.4.11
15:35:11therealadam enters the room.
15:35:36fbuilesvRich_Morin_: I wonder if it has something to do with it being PPC. Another guy kept complaining about half of the specs didn't pass when using his PPC Mac.
15:35:50fbuilesvI wouldn't think it could matter sine rubuildus_ppc runs just fine tho
15:36:14djwhittyeah, but he's running 10.4.11 ...
15:36:16dbussinkthe difference between 10.4 and 10.5 might be a bigger problem than the arch difference
15:36:44fbuilesvgood point
15:37:22rueTheProkrammer: What was the issue with FFI?
15:38:21TheProkrammerrue: It can't find the exported functions. So it bombs out.. even when I compiled it into one executable.
15:38:27Rich_MorinI have both Intel and PPC machines, running both Tiger and Leopard. If it would be useful, I could run some tests.
15:39:38dysinger leaves the room.
15:40:15cyndis enters the room.
15:43:25Rich_MorinAside from trying rbx, what test(s) should I run?
15:47:05fbuilesvRich_Morin_: run the specs and see how they're failing? Maybe someone with more experience knows why 10.4 seems to be so picky
15:47:36agile enters the room.
15:47:55Rich_Morinfbuilesv: IIRC, there's more than one way to run them. Can you give me the particular commandyou have in mind?
15:48:17qwert666_ enters the room.
15:48:31fbuilesvRich_Morin_: rake spec:ci just tto make sure that the ones that are ok are indeed passing
15:49:15Rich_MorinShoud I run that under script(1) or will it capture the results somewhere?
15:49:52fbuilesvRich_Morin_: that prints the results, you can output them to a file if you want but you could just read them. I'm guessing that you'll get _a lot_ of failures
15:50:00dlee enters the room.
15:51:29Rich_Morinfbuilesv: Is there an email addres where I should send the results?
15:51:59fbuilesvRich_Morin_: I'd post them here on a pastie or fill a bug in LH so more people can see them.
15:52:19Rich_MorinGonna be a BIG pastie...
15:53:00Rich_MorinAnyway, I gott go. I'll see what I have when I get home...
15:54:57dblack leaves the room.
16:00:11rby_ enters the room.
16:05:12qwert666 leaves the room.
16:10:07rueTheProkrammer: Both libffi and ltdl should work on Cygwin. Do you happen to have a paste?
16:10:21obiejuan enters the room.
16:12:52TheProkrammerrue: give me a few minutes..
16:14:45benburkert enters the room.
16:17:40TheProkrammerrue: http://pastie.caboo.se/188678
16:20:22yugui enters the room.
16:27:30rueTheProkrammer: Ah. It might be *specifically* that you are trying to open the same process
16:27:32dodecaphonic enters the room.
16:28:55TheProkrammerrue: Well right now I have all the lib stuff compiled as a seperate dll linked to rubinius.bin ...
16:31:36mkrauskopf leaves the room.
16:31:52mkrauskopf enters the room.
16:32:39rueDid you try -DLIBLTDL_DLL_IMPORT ?
16:36:52TheProkrammerHmm.. nope :)
16:38:37binary42 leaves the room.
16:38:50TheProkrammerlooks like I'll have to def __WINDOWS__ as well...
16:39:05TheProkrammerthat may help though, thanks!
16:40:59MaledictusHow do I test the cpp branch? I currently get a segfault in rubinius.local.bin
16:43:56rueMaledictus: `make test` unless there is a Rake task now. It will not run like shotgun does
16:44:15rueHm, make that it /does/ not run like shotgun does
16:46:25benburkert leaves the room.
16:46:33marnen enters the room.
16:47:02Maledictusgmake: Nothing to be done for `test'
16:47:20Maledictuseh, I mean how do I compile it ;)
16:47:31benburkert enters the room.
16:52:04evanMaledictus: cd vm/test; rake
16:52:12EugZol enters the room.
16:52:50Maledictus../builtin_iseq.hpp:18:42: error: gen/iseq_instruction_names.hpp: No such file or directory
16:53:21Maledictus`rake` in the root of that branch didn't work
16:53:39MaledictusDo I need the rubinius.local.bin from the master branch to bootstrap?
16:53:41evanyeah, rake at the root doesn't work, i know.
16:53:41evanno
16:53:47evango into vm
16:53:47rueTheProkrammer: Also, in shotgun/lib/subtend/ffi.h there are two macro sets.. see if using the FreeBSD version works. Something says symbols may not be lazy loaded. You need -no-undefined or --export-all-symbols, probably --enable-auto-import and if the archive something options are available
16:53:49evanand do
16:53:50evanmkdir gen
16:54:10Maledictusdone
16:54:19NoKarma enters the room.
16:55:01marnen_ leaves the room.
16:55:39rueTheProkrammer: But I think the import might be the main part
16:57:30Maledictusand now?
16:58:16headiusevan: you thought about putting in any peephole optz at compile time?
16:58:29benburkert leaves the room.
16:58:33evanheadius: yeah, considered it
16:58:36headiusI was trying to benchmark some stuff yesterday and got absurd numbers from 1.9 because it did such optz
16:58:40evanjust haven't gotten around to it
16:58:46evanabsurd good?
16:58:58lopex enters the room.
16:58:59headiusyeah, it reduced a long string of 100 ivar accesses to a single access
16:59:00evan1.9's peephole optz are silly
16:59:08headiusso the perf numbers were...unexpected ;)
16:59:11evanyeah
16:59:23evanit does 'access' folding
16:59:47evanremoves the classic 'set; pop; push' sequence to 'set'
16:59:52evanthats about it
16:59:58brixenwhen would you realistically have a long line of accesses like that though?
17:00:07evanin terms of real programs, i'm not sure the benefits are there
17:00:09headiusbrixen: I haven't run into any optz that were actually useful
17:00:12brixenpipes in and runs to class
17:00:17brixenheadius: ahh
17:00:19evanheadius: exactly
17:01:11headiusthere are such optz that probably would be useful, like eliminating a conditional with a literal false
17:01:17headiusor loops that won't run
17:01:34RyanTM enters the room.
17:01:41RyanTM_ enters the room.
17:02:09evanyeah
17:02:14evanwe just haven't bothered
17:02:32evanthe benefits from peephole optz are almost impossible to quantify
17:02:35RyanTM_ leaves the room.
17:05:19mutle leaves the room.
17:05:38trythil enters the room.
17:06:26mutle enters the room.
17:06:33yaroslav enters the room.
17:06:34joachimm enters the room.
17:07:38EugZol leaves the room.
17:11:35djwhittGoRuCo talks are up: http://goruco2008.confreaks.com/
17:11:48headiusnice
17:12:09headiusI really need to get to goruco someday
17:12:19djwhittit was great
17:12:22djwhitteven better this year
17:12:45headiusI think it's time for another inflammatory blog post
17:12:54headiusput on your flame-retardant jumpsuits
17:13:03djwhittegad, what now?
17:14:28brixencan't we all just get along?
17:14:58brixenlearned of and installed nu last night, quite interesting
17:15:03enebo enters the room.
17:15:19brixensee y'all in a couple..
17:24:07benburkert enters the room.
17:25:08evanmorning.
17:25:30evanoh, we talked before.
17:25:37evani need coffee.
17:25:59Maledictusit did compile. needed to remove -ldl. I'm on FreeBSD.
17:26:12dgtized leaves the room.
17:26:32djwhitthehe, you guys need a "donate coffee to rubinius" link somewhere
17:27:45benburkert leaves the room.
17:29:11marnen_ enters the room.
17:29:43yugui leaves the room.
17:31:58dgtized enters the room.
17:36:11Rich_Morin enters the room.
17:37:46Rich_MorinI'm thinking about using Capistrano to automate building rbx and running specs on several machines. Has anyone already done this? If not, any caveats or suggestions?
17:39:11evanRich_Morin: what kind of machines?
17:39:15evanwhat do you want to get out of it?
17:39:46mapar enters the room.
17:40:46Rich_MorinThey're all Mac, running OSX of various flavors on various CPUs. I ran into an rbx bug (irb mode) this morning and thought it might be useful to find out which machines exhibit it.
17:41:03thehcdreamer leaves the room.
17:41:22evanRich_Morin: that bug just needs you to run 'rake distclean' then 'rake build'
17:41:24evanit will go away
17:41:31evanit's from stale .rbc files
17:41:56marnen leaves the room.
17:43:12evanthat being said
17:43:18evani'm not aware of anyone using cap
17:43:20evanto do that
17:43:21Rich_Morintrying that now (ssh is coool(
17:43:34evanit's not very useful if you can't easily see the results though.
17:44:52danlucraft leaves the room.
17:46:02danlucraft enters the room.
17:46:07Rich_MorinWell, my plan was to record the output in files for examination.
17:46:49benburkert enters the room.
17:47:03NoKarma leaves the room.
17:47:47Rich_MorinI have G3's, G4's. G5's. and Intel boxes, running (variously) 10.4 and 10.5 I was thinking it might be useful to run the specs on them all and compare the results. But, if you don't think so, I won't bother.
17:48:06dblack enters the room.
17:48:58evanthe only difference will be possibly between 10.4 and 10.5
17:50:57Rich_MorinWell, that reduces the scope of the effort :-)
17:51:30mkrauskopf leaves the room.
17:51:48mkrauskopf enters the room.
17:54:45KirinDave enters the room.
18:00:08mutle leaves the room.
18:00:33moofbong leaves the room.
18:01:52nicksieger leaves the room.
18:05:55mentz leaves the room.
18:09:04anteaya enters the room.
18:09:23headius_ enters the room.
18:13:19dblack leaves the room.
18:14:58mapar leaves the room.
18:17:43marnen enters the room.
18:18:13moofbong enters the room.
18:21:45octopod leaves the room.
18:21:48RyanTM leaves the room.
18:23:51headiusit is done: http://headius.blogspot.com/2008/04/rubyists-are-wrong.html
18:24:39Rich_MorinAfter running "rake distclean", "rake build", "sudo rake install", and "rehash", interactive rbx is still getting errors.
18:25:14dbussinkRich_Morin: what if you don't install and run ./shotgun/rubinius from the rubinius root dir?
18:25:43evanheadius_: another hard hitting morning!
18:25:59headius leaves the room.
18:26:15Rich_MorinWell, I already installed, but ./shotgun/rubinius also barfs
18:26:20nicksieger enters the room.
18:26:44Rich_MorinWanna pastie?
18:26:51joachimm leaves the room.
18:27:31evanheadius: http://www.gemaffair.com/content/July_Birthstone_Jewelry_Guide.htm
18:27:33evanRich_Morin: sure
18:28:00headiusevan: I'm just brutal, I know
18:29:05evani like the "Fancy Cut" or the one on the bottom left of the 2 lines of them
18:29:07Rich_Morinhttp://pastie.caboo.se/188771
18:29:18evanthe one that is heavily tapered into points
18:29:40evanarg.
18:29:48Maledictusheadius_: http://forum.ruby-portal.de/search.php?search_id=newposts another one for you ;)
18:29:51evanRich_Morin: you've got the latest code?
18:30:20headiusthat brings me to a login
18:30:38Maledictushttp://forum.ruby-portal.de that should work
18:30:53Rich_Morinrecent, certainly, but I can certainly do a pull and try again.
18:31:12Rich_Morin(rdm@cerberus) 17: rbx -v
18:31:13Rich_Morinrubinius 0.8.0 (ruby 1.8.6 compatible) (b46817fa2) (04/29/2008) [powerpc-apple-darwin8.11.0]
18:32:25headiusMaledictus: ahh I see, the ruby
18:32:29headiusyeah, that one's weird
18:32:33evanRich_Morin: what about not installing it
18:32:41evanjust run it as ./shotgun/rubinius in the build directory
18:33:10headiusevan: yeah, that's a nice cut
18:33:29Rich_Morinerm, the pastie I sent you was a run of ./shotgun/rubinius in the build directory - does installing make that invalid?
18:34:37marnen_ leaves the room.
18:34:54evanRich_Morin: no, just curious
18:35:01evansome people had this problem a while back
18:35:03evanbut it went away
18:35:11evanyou should do 'rake uninstall'
18:35:21evanand build a totally fresh checkout
18:35:49Rich_Morinevan: Could you send me a pastie giving a complete set of commands you want me to run, in order? I'll run them and report back.
18:36:19evanmkdir ~/git/rbx-clean; cd ~/git/rbx-clean
18:36:23evanack.
18:36:38evancd ~/git; git clone git://git.rubini.us/code rbx-clean
18:36:46evancd rbx-clean; rake; ./shotgun/rubinius
18:36:50evanthe last 2 lines
18:37:22djwhittmight want to make sure RUBYOPT isn't set too
18:38:17Rich_Morinno; buy MACOSX_DEPLOYMENT_TARGET=10.4
18:50:00marnen leaves the room.
18:52:34olabini leaves the room.
18:56:43Rich_Morinevan - ./shotgun/rubinius works now. What should I try next?
18:56:59evanyou could install it and try
18:57:02evanshould be fine now
18:57:08evansomething stale wasn't getting cleaned out.
18:57:29Rich_Morin"sudo rake install; rehash; rbx" ?
18:57:50trythil leaves the room.
18:59:34evansure
19:01:41mentz enters the room.
19:02:35benburkertevan: are you still presenting at Dallas Tech Fest?
19:02:53evandoesn't look like it's going to work out, no.
19:02:58evansorry :/
19:03:10benburkertboo hiss
19:03:21benburkertohh well
19:04:32dambalah enters the room.
19:06:20Rich_Morinevan - that cleared it up for rbx. Thanks!
19:06:25evanno prob
19:11:52sholden enters the room.
19:16:54fbuilesvdoes this way of closing attributes in REXML make sense for anyone: http://pastie.caboo.se/188813 ?
19:18:32fbuilesvclosing tags*
19:20:50rueIf by "way of closing" you mean "NOT closing" then yes, it does :)
19:20:59fbuilesvrue: well, that :P
19:21:04wycats leaves the room.
19:21:05kw enters the room.
19:21:35fbuilesvI'd love to see some documentation on the method to see what I'm supposed to spec but oh well :P
19:22:11rueSER has pushed quite a bit of test/unit code for REXML recently, that would be a good start
19:22:27fbuilesvrue: SER?
19:22:53rueSean Russell
19:23:40fbuilesvrue: which branch?
19:24:52fbuilesvI don't see it on 1.8 or 1.9
19:27:10olabini enters the room.
19:28:10Arjen_ leaves the room.
19:31:30rueHum, me either. I could have sworn I saw it on CIA
19:32:05fbuilesvrue: Many fixes were indeed included for 1.9. I thought the patches I sent were never applie but they just weren't closed on the original tracker
19:32:56fbuilesvor attributed for all I see :S
19:36:16wycats enters the room.
19:37:51rby_ leaves the room.
19:37:58joachimm enters the room.
19:39:38rueheadius: Do you recall seeing SER fixing stuff in REXML?
19:39:55rueOr committing tests to be particular
19:41:23TheVoice enters the room.
19:42:13rby_ enters the room.
19:45:58hornbeck leaves the room.
19:46:13hornbeck enters the room.
19:49:57antares_ enters the room.
19:58:20yaroslav leaves the room.
20:07:58dysinger enters the room.
20:09:48benny enters the room.
20:14:03joachimm_ enters the room.
20:16:38enebo leaves the room.
20:19:35be9 leaves the room.
20:20:38dodecaphonic leaves the room.
20:24:29TheVoic1 enters the room.
20:25:40joachimm_ leaves the room.
20:26:03joachimm_ enters the room.
20:29:14joachimm leaves the room.
20:35:26KirinDave leaves the room.
20:41:05TheVoice leaves the room.
20:48:03Rich_MorinI'm interested in being able to grab the sexp just as it comes out of the parser, modify it, and then run the resulting code. Can anyone give me a clue about how to do this?
20:50:09rueTrace from lib/bin/compile.rb
20:50:36DefilerRich_Morin: check out lib/compiler/compiler.rb 'def self.compile_file'
20:50:55rueI have a bit of staging done for sexp manipulation but might be a bit before it goes in
20:51:00DefilerThe easiest way is to insert your own code into the compilation process, so you don't have to reinvent everything
20:51:18DefilerYou could also hook into File.to_sexp
20:51:29mkrauskopf leaves the room.
20:51:46mkrauskopf enters the room.
20:53:30Rich_MorinI guess part of what I'm missing is the mechanics of modifying Rubinius, then having it run on the script. I could, for example, imagine reaching into the code and modifying it, then doing a require of the "real" file. But there may be a better WTDI
20:55:17Defilerwhat now?
20:55:25DefilerIt's just ruby code.
20:55:58DefilerJust implement your own, say, File.to_sexp that calls some code you want to run that tweaks up the parse tree however you want
20:56:12boyscout1 commit by Federico Builes
20:56:13boyscout * Adds more specs for REXML.; 823683a
20:56:20Defiler"run on the script" is the part of that question I don't follow, specifically
20:57:32joachimm_ leaves the room.
20:57:46enebo enters the room.
20:59:19Rich_MorinThis may be too simple a question, but. Let's say that I have a script "foo" and want to run it using a dynamically-modified version of rbx. If "foo" modifies rbx, it'll be too late, because foo has already been parsed, etc. So, I have to modify rbx (somehow), then cause it to parse and run foo.
20:59:30anteaya leaves the room.
21:00:27Defilerrequire 'thing_that_modifies_rubinius' as the first line, or -r on the command-line, etc
21:00:37anteaya enters the room.
21:01:20rueThat still leaves the stables
21:01:27Rich_MorinBut isn't require an executable statement? So, the script containing the require has already been compiled?
21:02:11Defilerrue: Not really.. you can monkeypatch the stables after the system loads, which it would have done already at this point
21:02:40DefilerOh, I see what you are saying. Yeah. You would either want to put it in the first file that you load, or put it on the command-line
21:02:53DefilerIt's like require 'rubygems'. You have to do it before anything else if you want to require gems
21:03:02Rich_Moringot it
21:03:20DefilerWhat are you planning to do?
21:03:27DefilerIf it isn't secret sauce
21:07:18rubuildius_amd64Federico Builes: 823683a86; 2097 files, 6765 examples, 23999 expectations, 0 failures, 0 errors; http://rafb.net/p/p7CP1e97.html
21:11:41rueLooks like it is
21:11:47rubuildius_ppcFederico Builes: 823683a86; 2097 files, 6767 examples, 24025 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188875
21:13:59ruivaldo enters the room.
21:15:27VVSizfbuilesv: looks like your new XML specs fail on JRuby :) since the hash order is diffrent on JRuby/Ruby1.9
21:15:54fbuilesvVVSiz: let me correct that then
21:15:59fbuilesvthanks for the advice!
21:16:15VVSizcool! :)
21:19:38fbuilesvVVSiz: could you pastie a log of the failing ones while I finish downloading JRuby?
21:22:20tarcieriugh... I need two different versions of Ragel installed :(
21:23:33VVSizpastie: for fbuilesv
21:23:46pastiefbuilesv: http://pastie.org/188884 by VVSiz.
21:23:48Rich_MorinDefiler: sorry, I got called off to a standup meeting. I mostly want to play, but one possibility is adding a check for implicit coercion to Boolean in (say) an if or while statement.
21:23:55fbuilesvVVSiz: ty
21:24:21VVSizfbuilesv: actually, if you'd like to follow JRuby, the best way is to use subversion (or git svn) directly, and have up-to-date version
21:24:32fbuilesvVVSiz: yes, finishing the co right now
21:24:47fbuilesvVVSiz: doing a .sort on those keys and values to make sure they'll be the same, any objection to that?
21:25:00VVSizsounds good to me
21:25:54headius leaves the room.
21:25:56VVSizalso, if you have some jruby-related questions, feel free to ask on #jruby , somebody is always there
21:26:12fbuilesvwill do, thanks
21:31:10boyscout1 commit by Federico Builes
21:31:11boyscout * Fixes REXML::Element#namespaces specs; 4a846f8
21:32:56imajes_ enters the room.
21:35:20VVSizfbuilesv: now is "all clear" :)
21:36:21fbuilesvVVSiz: Good, now I have to find out: 1) Why my XChat window went totally clear, and 2) How to get ruby1.9 to run the specs
21:36:45VVSizjut supply -t path/to/ruby1.9
21:37:14VVSizthat's what I do: bin/mspec -t /opt/ruby1.9/bin/ruby spec/ruby/1.8/
21:37:36fbuilesvVVSiz: I tried that but I get a lot of errors about character encoding and stuff
21:37:39VVSizbut be warned, MRI 1.9 fails A LOT of specs
21:37:42VVSizyes
21:37:56VVSizI also get SIGSEGVs and stuff, crashes
21:38:12VVSizbasically, the specs currently are "no-go" with ruby 1.9
21:38:21fbuilesvoh cool, then I'll just keep testing with -t ruby and -t jruby while it gets stable
21:38:31drbrainthe specs aren't supposed to run on 1.9
21:38:46VVSizbut it useful to run some targeted specs against 1.9 (those that you develop) just to get the feel what's going on (or what will happen in MRI 1.9 ongoing)
21:38:54drbrainthere isn't a spec/ruby/1.9 directory
21:39:05fbuilesvoff to class, see ya!
21:39:24evandrbrain: picked up GTA4 just now.
21:39:42drbrainevan: mine arrived on my doorstep at 10AM
21:39:44danlucraft leaves the room.
21:39:50drbrainthen it crashed at the first save point :(
21:39:51macournoyer enters the room.
21:40:04drbrainI watched some TV though
21:42:08evancrap.
21:42:18rubuildius_amd64Federico Builes: 4a846f807; 2097 files, 6765 examples, 23999 expectations, 0 failures, 0 errors; http://rafb.net/p/h9Ku2h38.html
21:42:33dbussinkone more expectation needed :)
21:43:31pd leaves the room.
21:44:49rubuildius_ppcFederico Builes: 4a846f807; 2097 files, 6767 examples, 24025 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188915
21:45:10Maledictusis that 32bit ppc?
21:45:50imajes leaves the room.
21:46:43drbrainMaledictus: it says which platform on the first line of the paste
21:46:53drbrainIIRC, 64 bit PPC is powerpc64
21:47:04drbrainthis one is [powerpc-apple-darwin9.2.2]
21:47:17Maledictusah, ok
21:50:52radarek leaves the room.
21:52:00ctennis leaves the room.
21:52:48Maledictus leaves the room.
21:52:53dgtizedmethinks we need another round of optimizations, these specs are getting slow to run again
21:54:37dbussinkhey, they released java 6 for leopard
21:54:56sholden leaves the room.
21:55:33dbussinkVVSiz: guess that makes you happy ;)
21:56:25ezmobius enters the room.
21:57:15ruivaldo leaves the room.
21:57:39atmos leaves the room.
21:57:46atmos enters the room.
22:10:31imajes_ leaves the room.
22:16:31atmos leaves the room.
22:17:01atmos enters the room.
22:17:50obvio leaves the room.
22:20:44NoKarma enters the room.
22:20:57NoKarmafbuilesv: Hi
22:22:26atmos leaves the room.
22:22:33atmos enters the room.
22:24:58nicksieger leaves the room.
22:25:12ctennis enters the room.
22:26:03nicksieger enters the room.
22:26:47atmos leaves the room.
22:26:51atmos enters the room.
22:28:52AndrewO leaves the room.
22:30:04qwert666_ leaves the room.
22:30:09santana_ enters the room.
22:31:43santanahi
22:33:24rueHello
22:33:57santanahey rue, how can I get commit access
22:34:35dleesantana_: you have to submit a patch that get accepted into rubinius
22:34:55drbraina) submit a patch to the tracker b) get it reviewed and committed c) send evan your SSH key
22:34:57santanaI have dlee
22:36:39santanafour patches applied
22:36:43santanaand four pending
22:36:55drbrainok, you can move to step C
22:37:07santanaok
22:37:53santanawhat's his e-mail address
22:38:07drbrainhe is in-channel
22:38:13drbrainping him
22:38:20drbrainyou can just pastie your key to him
22:38:29santanaoh, ok
22:40:23santanaRSA?
22:40:51drbraindoes not matter
22:41:16drbrainjust make sure it is your public key
22:41:16santana_ leaves the room.
22:41:45santana_ enters the room.
22:45:55santanadbussink: could you tell evan about commit b5cd669?
22:46:14benburkert leaves the room.
22:50:14moofbong leaves the room.
22:55:59macournoyer leaves the room.
22:56:04bitbang enters the room.
22:58:16santana__ enters the room.
22:58:16santana_ leaves the room.
23:03:10headius enters the room.
23:05:38benny leaves the room.
23:11:26nkpart enters the room.
23:16:02boyscout1 commit by Gerardo Santana
23:16:03boyscout * Merge branch 'master' of git@git.rubini.us:code; b9b95b8
23:16:26evanone commit...
23:16:53evanoh
23:16:57evanboyscout got confused.
23:18:01santana:-S
23:19:35santanais there something wrong?
23:20:37evannope
23:20:42evanyou did the commits a few days ago
23:20:47santanayes
23:20:48evanso the commit history looks funny
23:20:50evanbut it's correct
23:20:51santana6 days ago
23:20:55santanaaah
23:20:55evanbecause it's sorted by date
23:21:00santanaheh
23:22:57marnen enters the room.
23:26:29antares_ leaves the room.
23:27:20rubuildius_amd64Gerardo Santana: b9b95b8e2; 2097 files, 6765 examples, 23999 expectations, 0 failures, 0 errors; http://rafb.net/p/iW9Ayc36.html
23:28:12agile leaves the room.
23:29:04rubuildius_ppcGerardo Santana: b9b95b8e2; 2097 files, 6767 examples, 24025 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/188985
23:30:01benny enters the room.
23:30:35rby_ leaves the room.
23:38:53kw leaves the room.
23:43:30marnen leaves the room.
23:44:24agardiner enters the room.
23:45:19agardinerhowdy
23:45:26evanhey there.
23:45:55mentz leaves the room.
23:46:11agardinerso, making progress getting ruby-debug running on rubinius...
23:46:19evanyay!
23:46:23NoKarma leaves the room.
23:46:24evanwonderful news!
23:46:32tarcieri?
23:46:39agardinerhehe, yeah it should be useful
23:46:45tarcieriC++ branch running code?
23:46:58evanlook up.
23:47:11tarcierioh
23:47:12tarcieriheh
23:47:19tarcieriwell on a completely unrelated note then...
23:50:24evanyes?
23:50:51agardinerso evan, do you recall we talked about a while back was having a way to access the root CM for each loaded file?
23:51:00evansure.
23:51:16agardinerits something i need to be able to implement breakpoints when given a file and line number
23:51:20tarcierievan: heh, just wondering if the C++ branch was running code yet
23:51:29evanoh. getting there.
23:51:41evani'm being slow because i'm testing everything as I go along.
23:51:50tarcieriaah
23:52:26santanaevan: is the code public?
23:52:30evanyeah
23:52:33evanit's in the cpp branch
23:52:38evanof git.
23:52:45evangit branch cpp origin/cpp
23:52:52evanwill make a local cpp branch off the remote one
23:52:55santanaaha, I'll take a look!
23:54:44headius enters the room.
23:54:51headius enters the room.
23:54:54headiushmm
23:54:55evanthe headius bounce!
23:55:00headiusgoofy colloquy
23:57:09agardinerevan: so how do you feel about me adding to CompiledMethod a hash of the $LOADED_FEATURE filename => root CompiledMethod?
23:57:30evannot sure i follow.
23:57:33agardinerit would allow you to access the root CMs using CompiledMethod[<filename>]
23:57:38evanoh.
23:57:40evanum.
23:57:41evanno.
23:57:48evanplease don't.
23:57:53agardinerok
23:58:00agardinersuggestions?
23:58:10evanhm.
23:58:14rueWhy not?
23:58:50rue(I am sure there is a reason, I have just been too out of it to know)
23:58:56evanactually, i guess that might be ok
23:59:19santana_ enters the room.
23:59:20evanbut it should be something like CompiledMethod.scripts[<something thats in $">]
23:59:42drbrainyes, that will be more clear
23:59:47agardinersure, that would be clearer
23:59:59evanyou should be able to wire it up inside require only