Index

Show enters and exits. Hide enters and exits.

00:01:09nicksieger leaves the room.
00:03:12Arjen_ leaves the room.
00:04:57robin_dewd leaves the room.
00:06:35benburkert enters the room.
00:06:36benburkert_ leaves the room.
00:06:57nicksieger enters the room.
00:07:20therealadam leaves the room.
00:19:01olabini leaves the room.
00:22:01RyanTM leaves the room.
00:23:51pauldix leaves the room.
00:28:07wmoxam enters the room.
00:30:00perplexes enters the room.
00:37:22nexcastellan leaves the room.
00:38:10RyanTM enters the room.
00:39:54enebo leaves the room.
00:48:33anonuserCan someone recommend a decent rails profiling tool?
00:49:19drbrainaction_profiler
00:49:30zenspideranonuser: what do you want to profile?
00:49:43anonuserthe entire application end to end.
00:50:02anonuserit could be the host but I'm not so sure.
00:52:24timoruby-prof ?
00:52:34timoit has a couple patches to do object/allocation profiling as well
00:53:02timoI also find valgrind/callgrind to be useful in certain circumstances
00:58:38pauldix enters the room.
00:58:45pauldix leaves the room.
01:01:08nullstyle leaves the room.
01:06:29hemulen leaves the room.
01:07:22anteaya leaves the room.
01:09:43benburkert leaves the room.
01:10:08benburkert enters the room.
01:10:31benburkert leaves the room.
01:11:09benburkert enters the room.
01:12:49blakewatters leaves the room.
01:16:16zenspideranonuser: yes, but WHAT do you want to profile?
01:16:46anonuserA request through the application.
01:17:21anonuserI guess it's a bit premature but I just want to keep something in mind because once I'm finished I'll really need to figure out what's happening
01:18:08hemulen enters the room.
01:19:01jero5 enters the room.
01:29:40anteaya enters the room.
01:29:53benburkert leaves the room.
01:30:19benburkert enters the room.
01:30:47benburkert leaves the room.
01:31:24benburkert enters the room.
01:32:31benstiglitz leaves the room.
01:32:54benburkert leaves the room.
01:32:57yugui enters the room.
01:33:29benburkert enters the room.
01:36:03hemulen leaves the room.
01:36:10benburkert_ enters the room.
01:37:32benburkert leaves the room.
01:39:56ShayArnett enters the room.
01:44:56AndrewO enters the room.
01:44:57jbarnette leaves the room.
01:47:31cored enters the room.
01:48:29ShayArnett leaves the room.
01:52:39perplexes leaves the room.
01:53:00anteaya_ enters the room.
01:53:32perplexes enters the room.
01:54:11michalw leaves the room.
01:57:34wmoxam leaves the room.
01:59:44benburkert enters the room.
01:59:44benburkert_ leaves the room.
02:00:05jtoy enters the room.
02:01:06anteaya leaves the room.
02:15:57ShayArnett enters the room.
02:16:44mernen enters the room.
02:19:36perplexes leaves the room.
02:19:46benburkert leaves the room.
02:23:02ezmobius leaves the room.
02:24:39VVSiz_ enters the room.
02:27:41nullstyle enters the room.
02:28:09benburkert enters the room.
02:39:21hemulen enters the room.
02:40:11xhanjian leaves the room.
02:42:30VVSiz leaves the room.
02:57:48xhanjian enters the room.
03:08:24cored leaves the room.
03:08:24AndrewO leaves the room.
03:13:40sambo82 enters the room.
03:14:58lstoll enters the room.
03:17:52perplexes enters the room.
03:18:13crafterm enters the room.
03:20:10perplexes leaves the room.
03:21:57ShayArnett leaves the room.
03:22:57jbarnette enters the room.
03:26:09trythil enters the room.
03:26:18hemulen leaves the room.
03:29:16lstoll_ enters the room.
03:31:14jeremydurham enters the room.
03:32:13crafterm leaves the room.
03:33:49jeremydurham leaves the room.
03:35:49benburkert leaves the room.
03:37:17lstoll leaves the room.
03:40:23jbarnetteevan: first use of pastejour in the wild ftw
03:42:38trythil_ enters the room.
03:42:38trythil leaves the room.
03:51:19shame leaves the room.
03:53:08gnufied leaves the room.
03:53:22evanjbarnette: wooo!
03:53:49jbarnetteevan: we could not be more awesome if we were wearing Flavor Flav's pants
03:54:05evanFLAVOR FLAAAAV!
03:55:31shame enters the room.
04:00:01couthi evan
04:12:14vertiginou2 enters the room.
04:23:03rueYou mac geeks *shake*
04:23:18jbarnetteheh
04:28:22trythil_ leaves the room.
04:31:18Yurik enters the room.
04:35:46random8r enters the room.
04:38:50masswhenever I hear flavor flav, I think of this: http://www.jelsert.com/products_flavor_ice.asp
04:39:01rueReally?
04:39:11mass"yes kids, eat our delicous snack and you too can have teeth like flavor flav!"
04:39:24rueI just think of the fall of a promising musician to a weird charicature appearing apparently on VH1
04:41:07anonuserrue, Yeah Bridget Nielson is great. ;)
04:42:30random8r leaves the room.
04:45:44RyanTM leaves the room.
04:47:31dlee leaves the room.
04:57:55mernen leaves the room.
05:05:34benny leaves the room.
05:13:06jbarnette leaves the room.
05:14:31therealadam enters the room.
05:33:27anonuser leaves the room.
05:43:41anteaya_ leaves the room.
05:44:45anonuser enters the room.
05:50:45yugui leaves the room.
05:51:46anonusergit.rubini.us still the "central" repository to pull from?
05:52:45evanyep
05:55:03anonuserthank ya
05:55:24anonuserI'm going to guess it's not okay to post a link to a full copy of the blue book
05:56:25anonuserthat inline work William Morgan is exciting.
05:56:42rueNot unless you have the rights for it. Chapters 26-29 are available online legally
05:56:53rueI dunno about licencing for the rest of it
05:58:21anonuserI searched the title of the book and a full copy was in the search results
05:59:46anonuser740 pages of poorly scanned pages.
06:05:54ezmobius enters the room.
06:10:58vincentlubrixen: hi, I just got an error for rake spec:update. it says "permission to brixen/rubyspec denied"
06:11:29evanvincentlu: oh, brixen was working on it.
06:11:30evando
06:11:34evanrake spec:committer
06:11:37evanthen try it.
06:12:46vincentluworks now :D
06:12:51vincentluevan: thanks!
06:12:54evanno prob
06:19:52boyscout1 commit by Vincent Lu
06:19:53boyscout * Throws OutOfRange when some iconv() returns out-of-range buffer size.; 10b3fc3
06:33:27rubuildius_ppcVincent Lu: 10b3fc39e; 2188 files, 7304 examples, 26090 expectations, 0 failures, 0 errors
06:33:28rubuildius_amd64Vincent Lu: 10b3fc39e; 2188 files, 7245 examples, 25942 expectations, 1 failure, 0 errors; http://rafb.net/p/L1kwde99.html
06:47:36wycats leaves the room.
06:48:16thehcdreamer leaves the room.
07:06:01wycats enters the room.
07:16:15jacen_ leaves the room.
07:17:05trythil enters the room.
07:22:47random8r enters the room.
07:29:32nullstyle leaves the room.
07:30:02blakewatters enters the room.
07:35:26jeremydurham enters the room.
07:35:30jbarnette enters the room.
07:35:57blakewatters leaves the room.
07:38:22trythil_ enters the room.
07:41:02trythil leaves the room.
07:43:54jeremydurham leaves the room.
07:52:15sambo82 leaves the room.
07:52:34sambo82 enters the room.
07:57:35Skip enters the room.
08:03:58jbarnette leaves the room.
08:14:19thehcdreamer enters the room.
08:18:27sambo82 leaves the room.
08:19:11qwert666 enters the room.
08:23:35Arjen_ enters the room.
08:35:36krisps enters the room.
08:35:42edwardam leaves the room.
08:37:20krisps enters the room.
08:39:07trythil_ leaves the room.
08:41:04TheVoic1 enters the room.
08:45:12octopod enters the room.
08:49:04TheVoice leaves the room.
08:52:16ezmobius leaves the room.
08:54:36atduskgr_ enters the room.
08:54:47mutle enters the room.
08:55:34atduskgreg leaves the room.
08:58:04jtoy leaves the room.
09:04:56jtoy enters the room.
09:08:12db-keen enters the room.
09:16:26Maledictus enters the room.
09:21:55massM is for model.. V is for view...
09:24:17imajes_office leaves the room.
09:24:53olabini enters the room.
09:24:58wycatsC is for Cuntroller?
09:25:20YurikCumtroller
09:26:36imajes enters the room.
09:27:02lstollcoffee?
09:33:03boyscout2 commits by Eero Saynatkari
09:33:04boyscout * Better alternative extension check for FFI on OS X. Closes #583.; 6f53b0c
09:33:05boyscout * Whitespace cleanup in ffi.c before fix.; b5c1315
09:33:05random8r leaves the room.
09:35:00atduskgr_ leaves the room.
09:45:52michalw enters the room.
09:46:25rubuildius_ppcEero Saynatkari: 6f53b0c7f; 2188 files, 7304 examples, 26090 expectations, 0 failures, 0 errors
09:48:21rubuildius_amd64Eero Saynatkari: 6f53b0c7f; 2188 files, 7245 examples, 25942 expectations, 1 failure, 0 errors; http://rafb.net/p/TY1wdq38.html
10:10:48yugui enters the room.
10:12:35chris2 enters the room.
10:32:53lstoll leaves the room.
10:39:22gnufied enters the room.
10:52:52dctanner enters the room.
11:07:00Arjen_ leaves the room.
11:09:26blakewatters enters the room.
11:10:14radarek enters the room.
11:16:43jtoy leaves the room.
11:22:56xhanjian leaves the room.
11:23:13krisps leaves the room.
11:27:12blakewatters leaves the room.
11:32:41maninred enters the room.
11:43:08chenkai enters the room.
11:45:02lstoll enters the room.
11:45:54maninred leaves the room.
11:46:03chenkai leaves the room.
11:46:24chenkai enters the room.
12:05:33chenkai leaves the room.
12:05:48chenkai enters the room.
12:23:54blakewatters enters the room.
12:25:56_mutle enters the room.
12:27:07mutle leaves the room.
12:27:51naeu enters the room.
12:33:26benny enters the room.
12:35:04RyanTM enters the room.
12:41:20blakewatters leaves the room.
12:43:41chenkai leaves the room.
12:48:11chris2 leaves the room.
12:50:13rue leaves the room.
13:01:17qwert666 leaves the room.
13:02:17anteaya enters the room.
13:09:53kAworu leaves the room.
13:14:57kAworu enters the room.
13:14:58moofbong enters the room.
13:32:06rue enters the room.
13:44:14pauldix enters the room.
14:01:11blakewatters enters the room.
14:03:16ShayArnett enters the room.
14:08:00AndrewO enters the room.
14:29:35moofbong leaves the room.
14:29:52moofbong enters the room.
14:37:04wmoxam enters the room.
14:46:45nullstyle enters the room.
15:16:11wmoxam leaves the room.
15:18:55wmoxam enters the room.
15:31:12dschn leaves the room.
15:35:59gdagley enters the room.
15:36:50smparkes leaves the room.
15:38:24kevwil enters the room.
15:39:06rue leaves the room.
15:39:21rue enters the room.
15:39:23gnufied leaves the room.
15:44:44nexcastellan enters the room.
15:48:33dbussinkwell, back in the netherlands :)
15:51:23dbussinkprobably still too early for westcoast ppl :P
15:51:29rueProbably
15:51:40rueFun trip overall, then?
15:52:46benburkert enters the room.
15:52:53shame leaves the room.
15:54:18dbussinkrue: yeah, had a lot of fun
15:55:29kevwil leaves the room.
15:57:06ch0wda enters the room.
16:14:14nullstyle leaves the room.
16:24:02rubuildius_ppc leaves the room.
16:24:08lopex enters the room.
16:24:43rubuildius_ppc enters the room.
16:26:38nullstyle enters the room.
16:29:35hemulen enters the room.
16:32:17nexcastellan enters the room.
16:37:53wvdschel enters the room.
16:38:33chenkai enters the room.
16:41:33ruedbussink: Finally on the last page of unread e-mail :/
16:42:32edwardam enters the room.
16:44:52VVSiz_ enters the room.
16:58:35evanmorning.
17:01:59rueGuten morgeroo
17:02:28wmoxam leaves the room.
17:03:14VVSiz leaves the room.
17:03:16shame enters the room.
17:13:12anteaya leaves the room.
17:23:52chenkai leaves the room.
17:24:08chenkai enters the room.
17:26:05qwert666 enters the room.
17:30:24anteaya enters the room.
17:30:59rueIs the bot using an old version of the specs or am I imagining that failure?
17:35:40DefilerIt uses the frozen specs
17:35:51DefilerWhich probably haven't been updated recently enough
17:36:17rue5/24 is the commit so probably not
17:36:54DefilerI started to do it at the conference, but it was tricky, and brixen volunteered
17:36:58DefilerI now recall
17:37:59perplexes enters the room.
17:38:43Defilerdrbrain: Did the new rubygems stuff get pulled into the rbx repo yet?
17:40:57thehcdreamer leaves the room.
17:44:04jbarnette enters the room.
17:47:47yukito enters the room.
17:48:26benstiglitz enters the room.
17:48:32wvdschel leaves the room.
17:50:39benstiglitz leaves the room.
17:51:56yukito leaves the room.
17:52:07wvdschel enters the room.
17:53:19enebo enters the room.
17:56:41benstiglitz enters the room.
18:00:57evanDefiler: HA!
18:01:03evanI just had a mind flash
18:01:17evanand realized why File::Stat always shows up very high in the profiling output
18:01:42eneboevan: because it does a stat()? :)
18:01:44DefilerOh, because it doesn't show the IO time
18:01:44Defilerright
18:02:00evanwell, sure, stat() is slow
18:02:00Defiler(separately)
18:02:05evanbut also, because of the darwin bug
18:02:16evanswitching INTO the kernel likely causes the itimer hook to fire
18:02:22evanso it doesn't fire properly normally
18:02:29evanbut switching into kernel land 'fixes' it
18:02:43Defileraha!
18:02:45evansince stat() is a syscall(), itimer fires
18:02:48DefilerThat has to be exactly it
18:03:18evanbecause the rest of running code does few syscalls
18:03:30evanand thus itimer fails to fire properly
18:03:39evanwhile the pthread spins in userland
18:03:58evangiven the arch of the mach kernel, that makes sense.
18:04:36evanit's possible that all other slices are recorded when the kernel decided to suspend the thread
18:04:53evanthus itimer doesn't work at all while the kernel lets the process spin
18:05:22DefilerSo, I need to add -S support to our command-line
18:05:31evan-S?
18:05:36Defilerdoes anyone know if that does $: << *ENV["PATH"] ?
18:05:44Defileror does it just do it temporarily to find the script arg?
18:05:45evanah
18:06:05evanit probably doesn't add it to $:
18:06:16evanjust includes it when looking for the script to run
18:06:18evanbecause remember
18:06:23evanloading the script doesn't use $:
18:06:32evanruby blah
18:06:39evan1.8 doesn't look for blah in $:
18:06:42dgtized enters the room.
18:06:47evanit looks in the current directory
18:06:51evanunless you put -S on there
18:06:57evanthen it also looks in $PATH
18:07:22DefilerOK, yeah, I just wrote a test that satisfies me of that as well
18:09:30JonathanT_ enters the room.
18:10:46dc_ leaves the room.
18:15:05evanheadius / enebo: what hotel are you staying at in Tokyo again?
18:15:06nexcastellanWhat's the status of RubyGems in Rubinius? Last I heard, mkmf.rb doesn't work quite yet and I am wondering if that has changed recently?
18:15:22evannexcastellan: it's actively being worked on
18:15:26eneboOh god...in Tsukuba?
18:15:36nexcastellanAnything I can do to help?
18:15:36evanwe recently fixed rbconfig.rb, so mkmf.rb should be happier.
18:15:46evannexcastellan: send in patches!
18:15:49Defilernexcastellan: We support RubyGems, but ones that have native extensions don't install properly yet
18:15:53evanenebo: um. is that in Tokyo?
18:15:58eneboWe have the most convoluted travel arrangements...let me check
18:16:09eneboRuby Kaigi is outside of Tokyo in Tsukuba
18:16:15evanenebo: I was going to stay in the same hotel in Tokyo
18:16:28evansince headius was saying it's pretty nice and in a nice place.
18:16:43evanenebo: right. I'm staying in the conf hotel out there
18:16:50evanbut i'm staying one night in Tokyo
18:16:56enebooh ok...Cerulean tower in Tokyo
18:17:04evanah yes.
18:17:05evanthats the one.
18:17:06evanthanks!
18:17:11atmos_ enters the room.
18:17:19eneboA tip when using the reservation system...uncheck breakfast
18:17:39eneboThe breakfast is worth it, but if you don't it will say no rooms match your request
18:17:54Arjen_ enters the room.
18:18:09evank.
18:20:28wvdschel leaves the room.
18:20:44twbray enters the room.
18:20:51octopod leaves the room.
18:21:05wycats leaves the room.
18:21:59naeu leaves the room.
18:24:43mutle leaves the room.
18:26:24NoKarma enters the room.
18:28:32chenkai leaves the room.
18:29:42atduskgreg enters the room.
18:34:53olabini leaves the room.
18:41:11Defilerevan: Is it cool if I move 'class Backtrace' from kernel/core/context to kernel/core/backtrace?
18:41:21evanit's very cool.
18:43:43anonuserway cool.
18:44:24twbray leaves the room.
18:45:09anonuserwoo building.
18:45:13Defilerawesome, I have the rspec spec suite running
18:45:35Defilersh: /usr/local/lib/rubinius/bin/rbx: No such file or directory
18:45:48DefilerI'm still getting that crap though, even though I am setting the bindir by hand to the correct path
18:45:52evanyeah
18:45:55evani'm going to sort that out today
18:46:36naeu enters the room.
18:46:39Defilerawesome
18:46:40boyscout1 commit by Wilson Bilkovich
18:46:41boyscout * Move Backtrace into its own file so I can remember where it is; e90b46e
18:47:05Defilerrad. the slowest spec still only takes 1.4 seconds
18:47:24Defilerhttp://pastie.org/208790
18:48:06evanrspec shows you the slowest specs?
18:49:00thehcdreamer enters the room.
18:50:05anonuserwants that damn garbage collector book
18:50:39twbray enters the room.
18:58:35rubuildius_amd64Wilson Bilkovich: e90b46ed2; 2188 files, 7245 examples, 25942 expectations, 1 failure, 0 errors; http://rafb.net/p/SKdV1c86.html
18:58:45vincentlua question... is o.extend(M) equivalent to class << o; include M; end; ?
18:59:41vincentluI mean the language. The implementation might not agree.
18:59:49rubuildius_ppcWilson Bilkovich: e90b46ed2; 2188 files, 7304 examples, 26090 expectations, 0 failures, 0 errors
18:59:53Defilerevan: Yeah, it runs with the profiler on
19:00:11Defilerevan: (at least for its own specs.. you have to ask for it in the spec.opts file in your own projects)
19:00:36Defilervincentlu: I can't think of a scenario where those two would produce different behavior off-hand..
19:01:19wycats enters the room.
19:02:01rubuildius_ppc leaves the room.
19:02:41rubuildius_ppc enters the room.
19:06:01jbarnette_ enters the room.
19:06:33carllerche enters the room.
19:09:48DefilerOh man rspec is a superb source of bugs for us
19:09:51DefilerTicket-tastic
19:10:24anonuser?
19:11:19evanwoo!
19:11:55evanhow about for a rubinius shirt, one of the 50s style guys giving a thumbs up, with banner text above that says "Bugtastic!"
19:12:18headiusI'm not sure that sends the right message
19:12:24evanbah.
19:12:25evan:)
19:12:33headiusmight as well have him say "It's craplicious!"
19:12:37anonuserKind of like the fallout buddy?
19:12:42evananonuser: yeah!
19:13:16TheProkrammer"More Bugs Today, But Still Less Then Yesterday!"
19:13:39DefilerI like bugs because then we get to fix them
19:14:04headiusa found bug is a fixed bug has been our motto on JRuby
19:14:19DefilerRubinius - It'll Ruby your face off!
19:14:38TheProkrammerRubinuius: The Ruby in all of us.
19:14:46loincloth enters the room.
19:14:48headiusRubinius: 100% Open Source
19:15:04anonuserevan, kind of like the faded cartoon in the background? http://computershopper.com/shoptalk/fallout_screen001355_screen.jpg
19:15:12evanI think Wilson should apply one of his patent sayings to it
19:15:27evansomething like "Rubinius - Like pull the blood sword from the rivers of hell."
19:15:46anonuserhaha
19:15:52headiusRubinius: MagLev without the OODB and Lock-in
19:15:57evanhah
19:15:57anonuser:(
19:16:12anonuserheadius, I absolutely loved your write up on MagLev.
19:16:19headiusRubinius: 1986 Was Better Anyway
19:16:23anonuserHAHA
19:16:33evanhehe
19:16:36DefilerNext time we talk about contributors I demand to be allowed to use this on a slide: http://www.piratesk12site.net/dune2.jpg
19:16:43nullstyle_ enters the room.
19:16:45Defilerexcept a way better screen-cap from the HD-DVD release
19:16:54headiusnice
19:16:56dysinger enters the room.
19:17:03anonuserMuadib sees into rubinius.
19:17:04evanRubinius: More '80s technology than limo full of coked up strippers.
19:17:08Defilerhahaha
19:17:09evanthan a limo.
19:17:11anonuserRubinius - Fear is the mind killer.
19:17:15DefilerI have to have that shirt
19:17:35anonuserRubinius - The Betamax of Ruby implementations
19:17:36DefilerRubinius - His name is a killing word
19:17:42TheProkrammerRubinius - even Muadib was really just a moon-rat.
19:17:48dbussinkwants it too
19:17:48headiusRubinius: Shai Hulud for the Ruby Fremen
19:18:02anonuserheadius, haha.
19:18:09TheProkrammerRubinius - He who controls the VM controls the universe.
19:18:17anonuserRubinius - Not just his blood, but all a mans water.
19:18:21headiusok, back to work
19:18:39tarcierihaha
19:18:47TheProkrammerShould put something on the rubinius homepage that cycles through user submitted slogans...
19:18:54evanoh
19:18:57evannot a bad idea
19:19:52TheProkrammercould be fun, could be dangerous :) maybe review them first.
19:20:18anonusermy god, there's going to be a remake of Dune.
19:20:20dbussinknah, the motto is not ask permission, but ask forgiveness :P
19:21:14jbarnette__ enters the room.
19:23:21jbarnette leaves the room.
19:23:37nullstyle leaves the room.
19:23:39nullstyle_ leaves the room.
19:23:43nullstyle enters the room.
19:27:22jbarnette_ leaves the room.
19:29:44kevwil enters the room.
19:30:20olabini enters the room.
19:32:13atmos_ leaves the room.
19:32:40antares_ enters the room.
19:33:55DefilerThis is a fun one http://rubinius.lighthouseapp.com/projects/5089-rubinius/tickets/587-regexperror-target-of-repeat- operator-is-invalid
19:34:02DefilerThanks, rspec!
19:36:22kevwil leaves the room.
19:36:23dctanner enters the room.
19:39:28trythil enters the room.
19:40:58anonuserThis is peculiar, URI#+ spec fails when I run rake spec:check
19:41:14anonuserbut if I run mspec directly against the file all seems well
19:42:13MenTaLguY enters the room.
19:43:24evanDefiler: eh gads.
19:43:31evanDefiler: thats onig reporting the error.
19:44:12dbussinkis it broken in 1.9 then too?
19:44:23evanvery likel
19:44:25evany
19:46:31DefilerWhat's the best way to install 1.9 these days without making it the default
19:46:45evancheck out how multiruby does it
19:46:48drbrain--program-suffix=19
19:46:51evanyeah
19:46:53evanthat.
19:47:06drbrainor put it in its own directory with --prefix=
19:47:10drbrain(to configure)
19:47:14Defilerk
19:47:18drbrainmultiruby does the latter
19:47:28DefilerIs that what you use?
19:47:31brixenDefiler: I'm using --program-suffix= because it makes it easy to use -t with mspec
19:47:41drbrainI use --program-suffix
19:47:43Defileraah
19:47:44brixenI've got 19, 187, 186, etc
19:47:53brixen186 head*
19:48:08drbrainwith --program-suffix, 186 will use 187 std lib
19:48:17drbrainthat's the only downside
19:48:19brixenack
19:48:23brixendidn't realize that
19:48:36brixenruby-lang has 1.8.7 docs up standard now btw
19:48:47brixenjust so you don't get confused writing specs ;)
19:48:49drbrainthey changed that in 1.9, so 1.9.0 and 1.9.1 will use different directories
19:48:54brixenawful awful project management
19:49:41Defilerwe should dns hijack them to roll it back to 1.8.6
19:50:15Defilerhow can you make it worse without.. *hand gesture* a website
19:50:27evandrbrain: they use the tiny version in the lib/ path now?
19:50:36drbrainevan: yes
19:50:50evanso tiny upgrades will require full reinstallation, right?
19:50:57Defilera phoenix of fail rising from the ashes of the old world
19:50:59evangems, extensions, etc.
19:51:05evanhey, i'm a phoenix!
19:51:09drbrainalso, sadly, 27% of RubyGems hits on gems.rubyforge.org are for 0.9.4 or less :(
19:51:14brixenevan: what's your opinion on this: http://pastie.org/208842
19:51:15Defileryou are a phoenix of succeed though
19:51:27drbrainI think RubyGems still uses only major and minor
19:51:29brixenevan: mkmf uses 'archdir' to find ruby.h (ugh)
19:51:30evanbrixen: fine.
19:51:33evanbetter
19:51:33drbrainI'd have to double check
19:51:48drbrainbut things installed into site ruby will need to be reinstalled
19:51:53brixenevan: ok, just concerned because it means we maintain a different version of mkmf
19:51:55evandrbrain: i could see rubygems using major.minor
19:52:07evansince the idea is that the installed lib directory is 'frozen' on installation
19:52:16evanthat you don't put crap in /usr/local/lib/ruby/1.9.0/
19:52:21evancourse, people do.
19:52:46drbrainI have a /usr/local/lib/ruby/gems/1.9.0 directory, so I think gems also uses teeny now
19:53:10evank.
19:53:31evanmajor.minor.teeny.kitten :)
19:56:08evanlunch time.
19:57:03dctanner leaves the room.
19:58:19edwardam_ enters the room.
20:04:38evanlight enters the room.
20:06:55edwarda__ enters the room.
20:07:14edwardam_ leaves the room.
20:07:24Cosmos95 leaves the room.
20:07:27Defiler example_group = Class.new(ExampleGroup) do
20:07:30Defilerwhoa.. this exists?
20:07:30edwardam leaves the room.
20:07:38DefilerThat's pretty cool
20:09:44boyscout1 commit by Vincent Lu
20:09:45boyscout * Readline::HISTORY includes (#extend) Enumerable.; cb94da0
20:10:52anteaya_ enters the room.
20:11:09edwarda__ leaves the room.
20:11:20MenTaLguYyeah, it's a godsend for specs
20:11:22edwardam enters the room.
20:11:39MenTaLguYmetaprogramming specs particularly
20:11:50TheVoice enters the room.
20:13:39Defileroh, weird.
20:13:55Defilerthat's a version of initialize that doesn't return self
20:14:38carllerche leaves the room.
20:15:05carllerche enters the room.
20:15:19anteaya leaves the room.
20:15:52brixenthe return value of #initialize is always ignored, isn't it?
20:16:05jicksta leaves the room.
20:16:13elight leaves the room.
20:16:30jicksta enters the room.
20:17:41rueYes, it is
20:18:23brixenDefiler: which part of example_group = Class.new(ExampleGroup) do was surprising? (just so I understand)
20:19:12brixenwe've got class/superclass specs that use Class.new(Fixnum), but no class/new specs for that behavior that I see
20:20:37rueBlock-form Class and Module ar the best things ever.
20:20:49rueIn fact, I shall never use `class` or `module` keywords again
20:21:46brixentakes away rue's keyboard
20:22:01brixenrue: you'll have to do that with the mouse if you want :P
20:22:30rueGood thing I installed MS Paint
20:22:44rueNow to see how I can export a .bmp into a source file..
20:22:54dctanner enters the room.
20:23:00nullstyle_ enters the room.
20:23:03rubuildius_ppcVincent Lu: cb94da01c; 2188 files, 7304 examples, 26090 expectations, 0 failures, 0 errors
20:23:17jbarnette enters the room.
20:24:14perplexes leaves the room.
20:24:31DefilerFor those playing the home game, this is what the bug turned out to really be:
20:24:33Defilerhttp://rubinius.lighthouseapp.com/projects/5089-rubinius/tickets/588-class-dup-fails-to-copy-metho ds
20:24:42perplexes enters the room.
20:25:12brixenahh
20:25:28jbarnette__ leaves the room.
20:26:03rueDoes it fail without the block form?
20:26:04cezarsa leaves the room.
20:26:30cezarsa enters the room.
20:26:45rueI probably need to sort out this spec mess so I can pull and run again
20:26:49DefilerIt doesn't seem to
20:26:59Defileraah, wait.. not the same scenario..
20:27:28DefilerYes, it does fail
20:27:30Defilerwithout a block
20:27:35Defilerbut only for class methods, not instance methods
20:28:33michalw leaves the room.
20:29:15nullstyle leaves the room.
20:31:30anonuserDefiler, why are those tickets now marked as spam?
20:32:16Defilerwhat the hell
20:32:38DefilerHow do I un-spam a ticket?
20:32:54evanlight leaves the room.
20:33:18kevinclarkhere too I think: http://rubinius.lighthouseapp.com/projects/5089/tickets/583-subtend-ffi-c-compiler-warning#ticket- 583-1
20:33:25kevinclarksee rue's comment as spam?
20:33:50kevinclarkyou can click to see it, but it shouldn't be spam in the first place
20:34:50rueIt should be under /spams
20:36:26kevinclarkI've got my rubinius days back (yay not prepping for launch). What needs doing? When I saw brixen at the conf he said he was working on Config::CONFIG stuff, should I look there?
20:36:55brixenkevinclark: working on a small fix, will push in a sec
20:37:31brixenDefiler: are you hamming all the spam tickets already?
20:37:35brixenif not, I will
20:37:59brixenheh, looks like that's a yes
20:38:18rubuildius_amd64 leaves the room.
20:38:23rueI just did
20:38:27brixenahh
20:38:30brixenthanks rue
20:38:33mjwhitt leaves the room.
20:39:49Defilerbrixen: hamming?
20:40:12rueSpamming, hamming.
20:40:14Defilerkevinclark: We need to work through the rspec failures as fast as possible
20:40:22Defilerkevinclark: want to see the list?
20:40:27kevinclarkplease
20:40:37rueI really do think it would be faster to craft some type of cyborg
20:42:07Defilerhttp://pastie.org/208878.txt
20:42:19brixenDefiler: the way to mark tickets as "not spam" is to mark them as "ham"
20:42:21DefilerI have the 'no method 'examples'' one figured out (it is Class#dup failing)
20:42:34Defilerand the one about a regexp pattern fails in 1.9 as well
20:42:48DefilerOther than that, fair game.. if you want to replicate my environment, pull from my copy of rspec here:...
20:43:04Defilerhttp://github.com/wilson/rspec/tree/master
20:43:10DefilerMy changes haven't been merged back to the main repo yet
20:43:41Arjen__ enters the room.
20:43:41DefilerI'm headed to the gym now. Let me know if you need to commit an rspec change
20:43:54ezmobius enters the room.
20:45:08kevinclarkwhere's rakebinius from?
20:45:50rubuildius_amd64 enters the room.
20:45:55mjwhitt enters the room.
20:46:36Defilerjust a copy of the rake stub with the shebang line changed to /usr/local/bin/rbx
20:46:45kevinclarkgotcha
20:46:57DefilerOh, and I hacked my rbconfig to have /usr/local/bin as the bindir
20:47:07Defilerevan is going to fix that problem today apparently
20:47:18evanyep
20:47:33DefilerBecause he is the man.. that problem has been so irritating
20:50:37kevinclarkthe file for the shebang only needs to be +x, yeah?
20:50:40kevinclarkgrumbles
20:50:44kevinclarkit doesn't like my rubinius
20:50:50evanit can't be a shell script
20:50:52kevinclarktrying to execute as bash
20:50:53evanbash doesn't allow that.
20:50:55kevinclarkahhh
20:51:04evanyou can't have a shebang point to another shell script
20:51:09evanin zsh, that works fine.
20:51:12evanbash doesn't like it though
20:52:03elight enters the room.
20:52:06lstoll_ enters the room.
20:52:16Defilerso we reeaally need to make "rbx rake" work, then
20:52:47evani shall today
20:52:48rubuildius_amd64 leaves the room.
20:52:58evangets the book of mystical unix hacks off the shelf
20:53:05mjwhitt leaves the room.
20:54:16mjwhitt enters the room.
20:55:03IngmarDon't forget your unix beard :)
20:56:09carllerche leaves the room.
20:56:30carllerche enters the room.
20:56:36chris2 enters the room.
20:56:41trythil leaves the room.
20:59:52headiuserp
20:59:55headiusyou need to use env to call another shell script in shebang
21:01:06lstoll leaves the room.
21:01:43evancan env take a full path?
21:02:05headiusyes
21:02:13evanok.
21:02:15headiusI believe so
21:02:25Arjen_ leaves the room.
21:02:27headiuscan't pass arguments to the script though
21:02:33evanright.
21:02:35headiusor at least...gnu env you can't, bsd env you can
21:03:53kevinclarkwhere does rubinius's gems look for installed gems?
21:04:15evanlike /usr/local/lib/rubinius/gems
21:04:16brixenkevinclark: lib/rbx/gems/
21:04:18evansomething like that.
21:04:58kevinclarkand I should just be using gem -I to put them there?
21:05:16brixenkevinclark: installed rbx or dev dir?
21:05:22kevinclarkde dir
21:05:23kevinclarkdev
21:05:32brixenshotgun/rubinius gem install does the right thing
21:05:35evanyeah
21:06:01kevinclarksweet
21:06:12evanand the speed is nice now
21:06:14evanthanks to drbrain
21:06:18brixenheh, yes it is :)
21:06:34brixenit's damn snappy for sure
21:08:11ShayArnett leaves the room.
21:08:13ezmobius leaves the room.
21:08:35naeu leaves the room.
21:09:29trythil enters the room.
21:09:30elight leaves the room.
21:09:39naeu enters the room.
21:10:12naeu leaves the room.
21:12:41naeu enters the room.
21:14:06seydar enters the room.
21:15:17elight enters the room.
21:16:05vertiginous enters the room.
21:20:31headiusnicksieger: http://wiki.jruby.org/wiki/Warbler seems to be a little out of date...is there a better wiki article?
21:20:33headiusoops
21:20:36headiussorry
21:22:11evanyou're forgiven.
21:29:01anteaya__ enters the room.
21:32:48atduskgreg leaves the room.
21:33:34masswhats new in rubinius-land?
21:33:54drbrainfast RubyGems!
21:34:33masshas anyone deployed a rails app on rubinius yet? mephisto blog or the like? :)
21:34:58massrubygems head has some nice optimizations. nice considering they deoptimized the crap out of it in 1.1.1
21:34:59brixenmass: soon
21:35:07drbrainto the real-world? I don't think so
21:35:14drbrainmass: that was me!
21:35:15elight leaves the room.
21:35:21masshehe
21:35:34drbrainmass: I had to be done to figure out how to make it awesome
21:35:35massI had to grab head on my vps to get rails 2.1 on :)
21:35:37drbrainI/it
21:35:40brixenevan: I think we may want to consider having Config::CONFIG['wordsize'] be a string ('32', '64') rather than a Fixnum
21:35:57evani found a couple bugs in rbconfig.rb
21:35:58massits really fast now
21:36:01massso congrats :)
21:36:01evanlooking at it now.
21:36:16dctanner leaves the room.
21:36:18drbrainmass: thanks
21:36:20massI decided against telling people on the forum to check it out of svn to fix their probs ;-)
21:36:29brixenevan: just worried about code in the wild that expects values to be strings and does regexp matching
21:36:37evanthen we shouldn't change it.
21:36:49brixen"it" ?
21:36:54evani'm fixing stuff so that the headers get installed also
21:37:03evanso that extensions can be built when we're installed.
21:37:15brixenk
21:37:16evanit == the value
21:37:22evanalso, i had
21:37:31evanCONFIG["install_prefix"] = /usr/local
21:37:35evanie, no quotes
21:37:37evanthats a bug, right?
21:37:43evanyou didn't mean a crazy regex
21:37:43brixenack, yeah
21:37:46evank
21:37:46brixennope
21:37:50evani'm doing a few other fixes
21:37:51evanpush shortly.
21:37:57brixenk, I'll hold off then
21:38:06massthe interesting thing about the whole register window stuff, at least for me
21:38:35massI had considered a while ago trying to make a CPU that used address windows rather than named registers
21:38:54evanmass: thats basically what squirellfish uses
21:39:05evanthe addresses are just numbered from an offset.
21:39:10evanand the offset is slide forward.
21:39:16massyep :)
21:39:31massnow I know what I'll be doing if I ever go back for a graduate degree
21:39:39evanhehe
21:39:44evanthats how they get zero copy arguments too
21:39:53evanwe did that in rubinius at one point too
21:39:56evanusing zero copy stack args
21:40:03massI'll add an add fixnum opcode for you too if I make one
21:40:11massalthough I don't think I'll have access to the best fabs
21:40:24evanjavascript can do that because it has fixed argument counts.
21:40:27evanno splats or anything.
21:40:32mass*nod*
21:41:43anonuserQuick question does the Dir.glob spec fail for anyone else?
21:41:43masssilly splats
21:42:32drbrainanonuser: it does not
21:42:40anonuserI must be special then.
21:42:41brixenanonuser: it's failing on linux
21:42:53drbrainhow is it failing?
21:42:54seydar enters the room.
21:43:02seydardrbrain: how are objects represented in rubinius?
21:43:03anonuserThe ordering in the glob is different.
21:43:11drbrainI may be able to point someone in the right direction for fixing it
21:43:13brixendrbrain: like so
21:43:15brixenhttp://ci.rubini.us/ci/1212606731.M470571P13187.hoshi.fallingsnow.net.html
21:43:38nexcastellanevan: CONFIG["install_prefix"] = /usr/local I think I just ran into that problem right now. You end up with (?-mix:\A) or some such, right?
21:43:46brixennexcastellan: yeah
21:44:08drbrainseydar: I haven't looked
21:44:11anonuserdrbrain, files and directories are in reverse ordering than expected.
21:44:17nexcastellanbrixen, evan: You guys working on mkmf.rb right at the moment?
21:44:17anonuserwhen doing a glob
21:44:22brixennexcastellan: yes
21:44:36brixenanonuser: you get the same failures as that link I pasted, yes?
21:44:42anonuserbrixen, correct.
21:44:42seydardrbrain: oh wait! i have the blue book as a pdf. i'll just check there
21:44:46nexcastellanbrixen: Hahaha, okay well you are making me happy.
21:44:49anteaya_ leaves the room.
21:44:52anonuserbrixen, irb gives me the same ordering though
21:45:02brixennexcastellan: :) almost had sqlite3-ruby compiling last night
21:45:13anonuserSo it's not so much a bug as it is an architectural difference.
21:45:23brixenanonuser: "same ordering" as what?
21:45:35drbrainanonuser: I'm not sure why it does that, as in glob_helper it checks the match_dir flag first
21:45:40anonuserbrixen, I looked at the spec and ran the test in MRI
21:45:58anonuserMRI gives me the same order that the failing spec gives me.
21:46:04brixenahh, ok
21:46:22drbrainanonuser: I'm not sure how it could be building up the pattern differently
21:46:25anonuserI get "dir_filename_ordering", "dir/filename_ordering"
21:46:30drbrainto have match_dir patterns come before match_all patterns
21:46:44drbrain... it shouldn't be platform dependent
21:47:40nexcastellanbrixen: I almost have an extension building as well. :)
21:47:58atduskgreg enters the room.
21:49:00drbrainanonuser: you you add: require 'pp'; p path; pp patterns to the top of Dir::glob_helper and run just that spec?
21:49:09brixennexcastellan: sweet, which one are you working on?
21:49:24anonuserdrbrain, will do
21:49:37anonuserI was looking at dir.rb, it's complicated o.O
21:49:42anonuserbut yes give me a moment.
21:49:57nexcastellanbrixen: Well, one of ours, but later I'm going to try getting the following working: json, mechanize, bbcode, rmagick, hpricot, recaptcha.
21:50:00drbrainit's a port of the MRI code
21:50:15brixennexcastellan: ambitious! hehe, great to hear though
21:50:40drbrainhaha, rmagick
21:50:42brixenanonuser: if you think dir.rb is complicated, you really have to enjoy the original C code :)
21:50:49nexcastellanPurely selfish, I assure you. Once we get those working and assuming all our other gems are working, our framework will run on Rubinius. That's my current goal.
21:50:58anonuserbrixen, :(
21:50:59brixennexcastellan: awesome
21:51:06nexcastellanBut of course we'll be contributing patches.
21:51:08drbrainnexcastellan: that's the best way to do it
21:52:26dgtizedDefiler: shouldn't that backtrace.rb split have a # depends: class.rb?
21:52:30anonuserdrbrain, hrrm are you sure that's where you want the printing?
21:53:03dgtizedDefiler: or at least # depends on: context.rb
21:53:08evandgtized: no
21:53:09drbrainanonuser: yes
21:53:19evandgtized: the depends lines are only for what it needs to load
21:53:23evannot what it needs to run.
21:53:43dgtizedevan: yes but it used to be in context.rb which has a # depends on: class.rb
21:54:01evanthats actually not needed
21:54:03evansomeone added those
21:54:22evanyou don't need class.rb to create a class.
21:54:38dgtizedok, so that could be stripped as a dependency for context?
21:54:40kevinclarkwhen I'm running rubinius from a dev dir, what does this indicate? Everything seems to be built and running ok: sh: /Users/kev/code/rbx/bin/rbx: No such file or directory
21:54:41evanthats why it works.
21:54:45evandgtized: yeah, but don't bother.
21:54:49ezmobius enters the room.
21:54:55evankevinclark: yeah, i know.
21:54:59evani'm working on that right now.
21:55:04kevinclarkevan: ahhh
21:55:15evanhow did you get that?
21:55:33kevinclarktrying to run defiler's rspec stuff. Also, trying to install mongrel.
21:55:42anonuserdrbrain, I get no output from Dir::glob_helper
21:55:49evankevinclark: install via Rakefile?
21:55:57evanit's because Config is stupid as shit.
21:56:09evani'm going to have to make it aware that it's in the dev directory.
21:56:30drbrainanonuser: did you remember to rake after editing?
21:56:37drbrainfor fun
21:56:42anonuserdrbrain, oops
21:56:52drbrainsometimes you need to kick it in the pants
21:57:15bitsweat enters the room.
21:57:44antaresevan: gems that install binaries (like merb) put them into wrong place as well, I had to symlink all of them to run merb
21:57:57evanyes
21:57:59evanall related.
21:58:03evani'm working on it now
21:58:04drbrainantares_: yeah, we're working on that too
21:58:14evancontinued complaining about it wont make me work any faster.
21:58:21anonuserThere we go!
21:58:35wvdschel enters the room.
21:58:43wvdschel leaves the room.
21:59:23antaresevan: no complains, mind you :) just seen someone having similar problem I did
21:59:41kevinclarkevan: With gem. I wanted to see how far it got.
21:59:50evankevinclark: I don't know what "with gem" means.
22:00:01kevinclark./shotgun/rubinius gem install mongrel
22:00:07kevinclarkwith gem
22:01:25ezmobiuskevinclark: talk to tarcieri, he has a rbx compatible mongrel building
22:01:35anonuserdrbrain, you want me to send that somewhere?
22:01:47kevinclarkezmobius: I've got it building (outside of using gem), but it segfaults
22:01:51kevinclarkdidn't used to :\
22:01:54kevinclarkneed to track it down
22:01:56drbrainanonuser: just use rafb.net/paste or similar
22:02:09ezmobiuskevinclark: using the Rubinius.asm http11 or the subtend one?
22:02:18kevinclarkezmobius: subtend
22:02:22ezmobiusahh ok
22:02:33kevinclarkwhat's Rubinius.asm http11?
22:02:45ezmobiusragel can output a goto machine that uses Rubinius.asm
22:02:51ezmobiuspure ruby that runs on rbx
22:03:08ezmobiustarcieri has a mongrel building based on that
22:04:26anonuserIs there a way to run the single part of the spec rather than the whole file?
22:05:57anonusernevermind got it.
22:06:52anonuserdrbrain, http://rafb.net/p/DgCO9L66.html
22:07:26anonusertried to only get it to run that particular spec because it's way way too long if you run them all
22:07:28evankevinclark: oh! did mkmf work trying to compile the extension then? thats rad if so.
22:08:21kevinclarkevan: it bailed with the error I gave earlier trying to run extconf
22:09:14evankevinclark: it bailed with the sh command error trying to find rbx?
22:09:26evani'll have that figured out shortly.
22:09:42drbrainanonuser: and, hard to figure out which output matches which glob pattern
22:10:06brixenevan: I've pulled in mkmf.rb from 1.8.6 head and I only have a 1-line change to rbconfig, perhaps I should push so you can test everything with what you're doing?
22:11:06evango ahead.
22:11:06anonuserdrbrain, hold on I'll make it triggered print.
22:11:08evani'll merge on this side.
22:11:15brixenevan: k
22:11:27drbrainanonuser: it's ok, I see what's happening
22:11:34drbrainlooking at the code to see why
22:11:44kevinclarkevan: correct
22:11:49anonuserdrbrain, okay
22:11:54anonuserI'll stop then
22:12:10boyscout2 commits by Brian Ford
22:12:11boyscout * Imported 1.8.6 HEAD mkmf.rb and moved to lib dir.; c10f8c9
22:12:12boyscout * Added CROSS_COMPILING flag to rbconfig.; 9ba6c89
22:12:32anonuserdrbrain, is it that it's adding match_all before match_dir?
22:13:28drbrainlinux walks Dir.pwd like %w[deeply special dir_filename_ordering subdir_two dir]
22:13:33drbrain(I took out a bunch
22:13:35brixenevan: if we require Config values to be strings, this would be the diff: http://pastie.org/208945
22:13:57brixenevan: and the 'wordsize' key is a rbx addition (doesn't exist in MRI)
22:13:59drbrainOS X walks Dir.pwd in alphabetical order
22:14:04drbrainthat's the difference
22:14:15anonuseryeah.
22:14:34drbrainthe spec should be renamed
22:14:52mahargis that a fix for looking for rbx in lib when doing a compiled gem?
22:14:58maharg(the mkmf thing)
22:15:12brixenmaharg: which mkmf thing?
22:15:14anonuserdrbrain, should it even matter? or do we need specs per platform?
22:15:23trythil leaves the room.
22:15:39drbrainanonuser: rubinius Dir::glob should walk in alphabetical order
22:16:09anonuserMRI Dir::glob respects the platform though.
22:16:19brixenyeah, I don't think Dir.glob ordering is specified
22:16:30brixenwe should be checking contents, unless it's the dir before file thing
22:16:32drbrainit's whatever readdir() returns
22:16:44brixenthe spec should proably sort then?
22:16:54drbrainreaddir(3)
22:16:57evanif it depends on the order, yes.
22:17:27dgtizedit's checking to see what is contained by checking for array equality so it's unecessarily dependent on order
22:18:11brixenyeah, add a sort then
22:18:54drbrainwe use Dir.foreach which seems to match directly
22:19:21drbrainso, the spec must be broken, since linux returns entries from readdir(3) randomly
22:19:32drbrainwhile OS X returns them in order
22:20:00jicksta_ enters the room.
22:20:09dgtizedwell I'm getting the same type of spec failure on Hash#select
22:20:14dgtizedthe array order is reversed
22:20:53anonuserSo just get rid of the spec.
22:21:19dgtized?
22:21:24drbrainanonuser: yes, I think you're right
22:21:39anonuserdrbrain, I'll happily submit a patch to delete the spec ;)
22:21:42dgtizedis that spec only for ordering?
22:21:47anonuserdgtized, correct
22:21:51drbrainanonuser: can you check the history first to see why the Dir::glob ordering spec was added in the first place
22:21:56drbrainthe Dir::glob one is
22:22:05anonuserdrbrain, sure.
22:22:18drbrainI don't see any code here in 1.8 that ensures something happens before something else
22:22:24drbraindir_filename_ordering vs dir/filename_ordering
22:22:50edwardam leaves the room.
22:22:59anonuserdrbrain, I don't know why the spec was written.
22:23:00blakewatters leaves the room.
22:23:11brixenlet's ask VVSiz about this
22:23:25brixenhistory will be hard to recreate based on the rubyspecs move
22:23:27moofbong leaves the room.
22:23:33brixenVVSiz: you around?
22:23:39w1rele55 leaves the room.
22:24:27anonuserbrixen, correct.
22:24:29evanmkmf.rb is shear madness.
22:24:36rubuildius_amd64 enters the room.
22:24:42drbrainevan: 10 years old!
22:24:47evanmore!
22:24:56evanin another few months will be 11!
22:24:59drbrain10 years 4 months :/
22:25:05dgtizedbrixen: how come you didn't export the rubyspec history when you moved it over to github?
22:25:07anonuserdrbrain, I have no idea how to track down the commit,
22:25:09evandrbrain: oh, i thought it was 8.
22:25:25nexcastellanevan: Hahaha, but lots of fun, no? I'm just one more fix away from having mkmf.rb hacked up enough to compile our local extension.
22:25:25dgtizedbrixen: is there no way to extract file level history?
22:25:27brixendgtized: impossible to. the commits were mixed with rubinius code and git lost almost all history no matter what
22:25:43anonuser:(
22:25:45drbrainat least, that's when it was checked into CVS for the first time
22:25:49evannexcastellan: how much have you changed it?
22:26:11dgtizedbrixen: hmm, that's rather unfortunate -- I'm suprised it couldn't just go through each commit and keep the messages and changes specific to the set of directories you are dealing with
22:26:13anonuserwell, for now I'll get rid of the thing and it will plague us no more. if that's okay with you
22:26:19drbrainI guess for now we should just comment it out and put a big message on it
22:26:26anonuserkk
22:26:29rubuildius_ppcBrian Ford: c10f8c9ac; 2188 files, 7304 examples, 26090 expectations, 0 failures, 0 errors
22:26:30nexcastellanevan: Only a few lines, really, but I've been focusing on getting it to work rather than getting a nice patch. How far have you got?
22:26:34brixenanonuser: usually, you can do e.g. git log 1.8/core/dir/glob_spec.rb
22:26:41drbrainlike "This spec doesn't seem to exhibit a behavior of MRI"
22:26:42anonuserbrixen, already did
22:26:45anonuseronly four entries.
22:26:47evannexcastellan: i'm looking at how it uses Config
22:26:51evanthe problem is that is does
22:26:53dgtizedanonuser: right because of my question above
22:26:57brixenanonuser: yes, because the code got moved recently
22:26:58evanFile.join(Config::CONFIG["bindir"], CONFIG["ruby_install_name"])
22:27:02anonuserbrixen, correct.
22:27:04nexcastellanHahaha.
22:27:04evanwhich is why it doesn't work in dev mode.
22:27:06anonuserthat's the first entry.
22:27:27evanbrixen: did you push your mkmf change?
22:27:31brixenevan: yeah
22:28:00brixenevan: see also http://pastie.org/208945 RE requiring Config values to be String only
22:28:33evank.
22:28:48brixenanonuser: does that spec fail under MRI 1.8.6p114?
22:29:04anonuseryes.
22:29:07evannexcastellan: what are your other changes?
22:29:15anonuserIt seems to be platform dependent
22:29:21brixenanonuser: k, can you please add a ticket on rubyspec.org so I can ask VVSiz about it
22:29:28anonuserkk
22:29:31brixenthanks
22:30:37drbrainanonuser: and note that ordering seems to be dependent upon readdir(3)
22:30:48anonuser