Index

Show enters and exits. Hide enters and exits.

00:11:25roo leaves the room.
00:11:38rue enters the room.
00:17:47benburkert leaves the room.
00:19:57imajes_ enters the room.
00:21:04imajes leaves the room.
00:27:16benburkert enters the room.
00:28:54brixenok, seems the kill spec is responsible for the hangup I see on freebsd
00:29:44brixenyay! 0/0 on gentoo 64
00:29:46evani thought we turned those off
00:30:17brixenhowever, 308 sec for the run is insane
00:30:31brixenevan: kamal_fariz removed some fails tags for stuff that was passing
00:30:36brixenthey may have gotten included
00:30:51brixenI'm going to look at it in more detail as soon as I push the 64bit fix
00:31:17brixenI think the specs sat in Singleton.instantiate for 30 sec
00:32:02evanpikes.
00:32:03evanyixes.
00:32:06evangah.
00:32:09evanyikes
00:32:18brixenall those :)
00:32:48evanthose were marked as failing because they're unstable
00:32:50evanas I recall
00:33:38brixenyeah, makes sense
00:33:49brixenI'm going to try to fix them
00:34:24brixenor tag them: fails(This causes solar flares. Do not use near your computer):...
00:34:27brixenor something :)
00:34:36evanhehe
00:34:42evanunstable seems like a good tag
00:35:03brixenyeah
00:36:44boyscout1 commit by Brian Ford
00:36:44boyscout * Fix libgdtoa for 64bit.; bf3caae
00:37:07brixendjwhitt: could you force a pull/build on the 64bit bot?
00:38:46imajes enters the room.
00:39:35imajes_ leaves the room.
00:41:02djwhittbrixen: just checked, looks like it's running specs right now
00:41:36KirinDave enters the room.
00:42:22rhubarbSolar flares could be a source of wireless power!
00:43:21brother_rspec enters the room.
00:43:47rhubarbI was thinking about setting up a combined wireless power source/cardio exercise machine
00:43:55rhubarbJust need to slightly modify a tesla coil
00:45:22djwhittwow, some of the specs chew up a lot of memory
00:46:01chop3 leaves the room.
00:46:02djwhittgets up to 533mb resident
00:47:29rubuildius_amd64Brian Ford: bf3caae22; 1756 files, 6075 examples, 20326 expectations, 0 failures, 0 errors; http://rafb.net/p/lnMyfx32.html
00:47:45djwhittw00t
00:48:06djwhittlooks like it's fixed
00:50:00brixendjwhitt: cool
00:50:31brixendjwhitt: yeah, I see mem use got way up when running the Generator specs after core, but haven't had a chance to look at it yet
00:50:49benburkert leaves the room.
00:51:13brixenrhubarb: we'll run all the unstable specs on your computer, to power your wifi :)
00:52:01djwhittit would be really fun to record performance data from every ci run
00:52:23djwhittmaybe after I get a better build system going I'll work on that
00:53:30ezmobius leaves the room.
00:54:08rhubarbNeed some power, the adapter is pretty hungry
00:55:34imajes leaves the room.
00:57:38KirinDave leaves the room.
01:02:05KirinDave enters the room.
01:02:14brother_rspec_ enters the room.
01:02:27lachie leaves the room.
01:05:22brother_rspec_ leaves the room.
01:05:48lachie enters the room.
01:14:19brother_rspec leaves the room.
01:15:08boyscout3 commits by Brian Ford
01:15:09boyscout * Tagged unstable Process.kill specs that cause hangup on linux.; 1334092
01:15:10boyscout * Added tag 'unstable' to exclusions for mspec-ci.; 0161ed1
01:15:11boyscout * Added ruby -v for mspec -tr. Silenced MRI warnings for MSpec.; 81d14f4
01:17:56brixenzenspider: added -v, I think that's what you wanted, holla if not
01:18:23brixendinner, bbl...
01:25:16rubuildius_amd64Brian Ford: 133409245; 1756 files, 6071 examples, 20318 expectations, 1 failure, 0 errors; http://rafb.net/p/4Noclj63.html
01:25:21lachie leaves the room.
01:31:17ezmobius enters the room.
01:31:22jrun enters the room.
01:37:47imajes enters the room.
01:38:19slaney enters the room.
01:42:29GMFlash enters the room.
01:44:41eventualbuddha leaves the room.
01:44:46imajes leaves the room.
01:53:49KirinDave leaves the room.
01:55:21rhubarbIs there an agenda for next week, anything to prep?
01:57:04jinjing enters the room.
01:58:18lachie enters the room.
01:58:43evanrhubarb: whats going on next week?
01:59:35lachie leaves the room.
01:59:44bitbang leaves the room.
02:01:11chop3 enters the room.
02:02:51dysinger leaves the room.
02:03:34cremes_ enters the room.
02:05:37evanwhat output formatter do I use to make unit_diff happy?
02:06:36jinjing leaves the room.
02:06:47jinjing enters the room.
02:06:58rhubarbevan: iCal?
02:07:20evanhuh?
02:09:04Defilerevan: 'u' is the format you want
02:09:35evanbin/mspec -fu ... | unit_diff -u
02:09:36evan?
02:09:39rhubarbponders how long before he gets swatted
02:09:46rhubarbevan: Call, 19th
02:10:02evanoh, rue, is that you?
02:10:08rhubarbYes..
02:10:17rhubarbI really need to stop with the phonetic alternatives, huh?
02:10:23evanprobably.
02:11:14evancall on the 19th is a status update
02:11:35Defilerevan: Yeah, that is the correct command, as I recall
02:11:44evanhrm.
02:11:47evandoesn't seem to work.
02:12:03Defilerbrian changed the format recently I think, and broke it
02:12:06evanthe sexp aren't split across lines
02:12:25DefilerI believe I heard Ryan threatening blood vengeance :)
02:12:39evanhrm. this was discussed, i didn't realize it was broken
02:13:04rue_is_takenDefiler, not to pry but are you in the rainy place permanently now?
02:13:41DefilerNope. Just here for a 10-day change of venue
02:13:43evanwhere is pretty_inspect coming from in mspec?
02:13:45evanpp?
02:14:38jtoy enters the room.
02:14:41DefilerI believe it is just an alias for inspect right now
02:14:47evanbut... where?
02:14:50DefilerI don't see any code that contradics that
02:14:58Defilermspec/spec_helper.rb
02:15:03Defiler(not spec/)
02:15:13evani don't have any aliases in there
02:15:18evani haven't updated in a little bit
02:15:23evanbecause i'm in the middle of a big change
02:15:27DefilerIt reopens Kernel
02:15:33DefilerFor some reason not an alias
02:16:13evanoh
02:16:14evanthats in head
02:16:17evanbut i don't have that...
02:16:22evanI wonder where it's coming from...
02:16:24rubuildius_ppc leaves the room.
02:16:33rubuildius_ppc enters the room.
02:16:40DefilerYou should just be able to do control-]
02:16:44Defilerin vim, and hop to it
02:16:54rue_is_takenstdlib/pp.rb
02:17:08evanonly if pp is required
02:17:11evanwhich it doesn't seem to be
02:17:51rue_is_takenCould it be an old stable?
02:17:53Defilerspec/ruby/1.8/spec_helper.rb requires pp
02:17:59evanrue_is_taken: no
02:18:00Defileron head
02:18:46rue_is_takenAh, mspec/spec_helper.rb
02:19:13rue_is_takenHrm, too slow.
02:19:25ezmobius leaves the room.
02:19:25rue_is_takenAlso too long. Feel like a spammer here.
02:20:28evanhm, where was it in 49d48c3
02:21:16rue_XIVLooks like brixen just added it
02:21:16Defilerspec/ruby/1.8/spec_helper.rb line 6
02:21:25agile enters the room.
02:21:27Defileron 49d48c3
02:22:40MenTaLguY enters the room.
02:25:02evani'm not pulling that file in though
02:25:10evanbecause i'm just running spec/compiler
02:25:50DefilerRight. Should get moved up, IMO
02:26:03evanok.
02:26:11DefilerAny implemenation that can't run pp can easily just make their own pretty_inspect alias that does 'inspect'
02:26:16evanbut that doesn't explain where it is when i run just the compiler specs
02:27:03evanforget it
02:27:06MenTaLguYhello
02:27:09DefilerNo idea, actually
02:27:09evani've given up on it.
02:27:21DefilerI just checked out that rev, and I don't see it
02:27:45rue_XIVIt is inevitable that a codebase this size becomes either a sentient being or a repository of wild magic
02:30:12boyscout1 commit by MenTaLguY
02:30:13boyscout * proper Mutex and ConditionVariable implementation using per-wait Channels; fe7d5e5
02:30:30MenTaLguYokay guys, that's Mutex and ConditionVariable for you
02:30:44jinjing leaves the room.
02:30:54MenTaLguYin JRuby I've also redone much of the rest of stdlib thread stuff in terms of Mutex+ConditionVariable; should I port?
02:32:16evanwhat has been changed?
02:32:37MenTaLguYin thread.rb, or in stdlib generally?
02:32:42evanstdlib generally
02:33:15MenTaLguYWell, generally it's been part of a project to eliminate the use of Thread.critical in an effort to improve peformance/throughput
02:33:22rue_XIVDefiler: The lappy has been really nice, by the way. Couple kinky spots but nothing that an hour's research has not unequivocally solved
02:33:33MenTaLguYalso correctness in some cases
02:33:51Defilerrue_XIV: Cool
02:34:38evanMenTaLguY: i'd say sure, go ahead and port them into rubinius
02:34:41rue_XIVDefiler: Battery life is not that great until they get the ACPI side and SpeedStep working better, sound is still external only and unless you select the Atheros WiFi mPCIe card, usb devices for network
02:35:02rue_XIVMenTaLguY: Nice! How much have you gotten rid of?
02:35:10MenTaLguYall the obvious stuff
02:35:33MenTaLguYthe major remaining bits are debug.rb, trace.rb, and the shell.rb, shell/*.rb stuff
02:35:43MenTaLguYand finalizer.rb
02:36:38MenTaLguYoh, and drb
02:37:47srbaker enters the room.
02:38:41rue_XIVCoolness
02:40:11MenTaLguYevan: hm. what about stuff that isn't currently in Rubinius stdlib, like Monitor?
02:40:14rubuildius_amd64MenTaLguY: fe7d5e5cf; 1756 files, 6071 examples, 20318 expectations, 1 failure, 0 errors; http://rafb.net/p/UfAKxz97.html
02:40:28MenTaLguYwhoops, let me see to that failure
02:40:57benburkert enters the room.
02:41:01MenTaLguYah, Process.kill failure
02:41:02MenTaLguYnever mind
02:41:08djwhittMenTaLguY: I don't think that failure was your fault
02:41:25djwhittbuild before this one had that same failure
02:41:43MenTaLguYwell, yeah
02:41:48MenTaLguYbut maybe I MAKE MISTAKES BACK IN TMIE
02:41:53djwhittheh, wow
02:42:11rue_XIVHeh
02:42:21MenTaLguYLook, just now I even made my past self mis-spell "TIME"
02:42:46MenTaLguY(you'll notice it was spelled right a moment before)
02:43:04rue_XIVControlling time is a big responsibility
02:44:16MenTaLguYIt's also a thankless job.
02:44:43MenTaLguYYou come back from a mission and say "Guess what guys? I just prevented President McKinley's assassination!"
02:44:54MenTaLguYThen everyone just looks at you like, "..what?"
02:45:24rue_XIVNow that you mention it, can you slow it down a bit? It seems that you have been trying to get out of the office early recently, the hours just fly by
02:45:50dewd leaves the room.
02:45:51rue_XIVHm. Apparently about:config may "void my warranty" in FF3
02:46:36cremesi have 4 hung rubuildius_ppc jobs on my box; was there a hanging problem with recent commits?
02:46:38MenTaLguYactually I've been sick all week
02:46:43MenTaLguYhence my crap commits earlier
02:47:34MenTaLguYfeeling better now
02:47:42MenTaLguYhence I'm committing code that works
02:47:48rue_XIVcremes_: There was an issue with kill, could be that
02:47:52rue_XIVWas/is
02:48:06cremesrue_XIV:
02:48:18cremesrue_XIV: hmmm... okay, will kill the ones in flight
02:48:44rue_XIVcremes_: Can you see where it is hanging? Are you running with -fs?
02:49:03djwhittcremes_: there was a build earlier that was hanging
02:49:07cremesrue_XIV: no, can't see; i'll check the rubuildius logs
02:49:13djwhittcremes_: they should be working now
02:49:28rue_XIVAlso, if someone wants a little-used kingdom, I want ff3 cookie dialogs stop stealing focus
02:50:23cremesdjwhitt: building one in my private directory to confirm...
02:50:48jrun leaves the room.
02:51:16boyscout1 commit by MenTaLguY
02:51:17boyscout * bring in Sync and Monitor from JRuby; f1be211
02:54:04boyscout1 commit by MenTaLguY
02:54:05boyscout * threadsafe (and secure, when possible) tempfile rewrite from JRuby; 2e9b9a8
02:58:36boyscout1 commit by MenTaLguY
02:58:37boyscout * rewritten mutex_m from JRuby; b840ee8
02:59:09DefilerNice tempfile change
02:59:18MenTaLguYthanks
02:59:27MenTaLguYdespite its claims, the original isn't particularly threadsafe at all
02:59:52benburkert leaves the room.
03:01:10MenTaLguYfigured I'd throw in a security fix and some minor cleanups while I was at it
03:02:56wifelette leaves the room.
03:04:30cored enters the room.
03:05:04rubuildius_amd64MenTaLguY: b840ee825; 1756 files, 6071 examples, 20318 expectations, 1 failure, 0 errors; http://rafb.net/p/RI4Chc34.html
03:05:05rubuildius_amd64MenTaLguY: f1be211a8; 1756 files, 6071 examples, 20318 expectations, 1 failure, 0 errors; http://rafb.net/p/nEXEzN57.html
03:08:13wycats leaves the room.
03:10:40cavalle_ enters the room.
03:11:39chop3 leaves the room.
03:12:10cremesrubuildius_ppc hangs after the Float#ceil test : - returns the smallest Integer greater than or equal to self
03:12:17cremesit's stuck there
03:14:41cremesi see that is hasn't compiled the float/coerce_spec.rb yet
03:16:24cremesif i try to manually run "bin/mspec spec/ruby/1.8/core/float/coerce_spec.rb" it hangs too;
03:16:29cremeshow should i troubleshoot?
03:16:53srbaker_ enters the room.
03:20:32djwhittcremes_: does it do that when you run it by hand?
03:20:43cremesdjwhitt: yep
03:20:48benburkert enters the room.
03:20:53djwhittweird
03:21:05rubuildius_ppcMenTaLguY: b840ee825; bin/ci failed! http://pastie.caboo.se/paste/165954
03:21:06rubuildius_ppcMenTaLguY: 2e9b9a8d9; bin/ci failed! http://pastie.caboo.se/paste/165951
03:21:07rubuildius_ppcMenTaLguY: fe7d5e5cf; bin/ci failed! http://pastie.caboo.se/paste/165950
03:21:08rubuildius_ppcBrian Ford: 133409245; bin/ci failed! http://pastie.caboo.se/paste/165942
03:21:21djwhittI know brixen made a change earlier that was causing it to hang on my system, but he fixed it and it's been fine ever since
03:21:33benburkert leaves the room.
03:21:41cremes??
03:22:06djwhittbrixen: you around?
03:25:18xif leaves the room.
03:26:08cavalle leaves the room.
03:26:32cremestrying 'rake distclean; rake build' to see if something odd stuck around between builds
03:31:25srbaker leaves the room.
03:38:27mediogre enters the room.
03:38:27rubuildius_ppc leaves the room.
03:40:01geekounet leaves the room.
03:40:03geekounet enters the room.
03:40:15cremestoo many manhattans... look again in the morning
03:40:18cremes_ leaves the room.
03:40:44yipstar leaves the room.
03:43:29cored leaves the room.
03:45:50rubuildius_amd64 leaves the room.
03:46:54mgreenly enters the room.
03:47:26rubuildius_ppc enters the room.
03:47:54rubuildius_amd64 enters the room.
03:52:06MenTaLguY leaves the room.
04:02:32mediogre leaves the room.
04:05:41srbaker_ leaves the room.
04:05:55srbaker enters the room.
04:06:47mgreenly leaves the room.
04:07:22webmat leaves the room.
04:11:15AndrewO leaves the room.
04:45:40rue_XIVDamnit, this cookie stuff is getting tiresome
04:52:14bitbang enters the room.
04:58:43dkubb leaves the room.
05:02:11rubuildius_amd64 leaves the room.
05:07:57seydar enters the room.
05:08:12seydaryou all are going to hate me for being so brief, but on the latest git pull i get this: b
05:08:14seydarwhoops
05:08:53seydaroh shoot. copied wrong thing. http://pastie.caboo.se/pastes/165967
05:09:00seydaron my Tiger 10.4.11/ PPC
05:09:05seydari must sleep!
05:09:06seydar leaves the room.
05:09:09RyanTM leaves the room.
05:09:29evanhrm
05:09:35evanwhy did brixen link it as a .dylib
05:09:36evanthats wrong.
05:09:39evanit should be a .a
05:18:08yipstar enters the room.
05:20:09daesan enters the room.
05:20:13daesan leaves the room.
05:22:18aotearoa enters the room.
05:34:46zuwiki enters the room.
05:38:31enebo enters the room.
05:40:23enebo_ enters the room.
05:41:08enebo_ leaves the room.
05:57:30lachie enters the room.
05:58:50enebo leaves the room.
06:25:53mediogre enters the room.
06:34:34zuwiki leaves the room.
06:38:56jtoy leaves the room.
07:08:52perdix enters the room.
07:10:47aotearoa leaves the room.
07:12:44rueDoobie doo
07:13:42scooprjionks
07:16:04antares enters the room.
07:24:07mediogre leaves the room.
07:27:56GMFlash enters the room.
07:29:21rueScoopstenator
07:31:04dbussinkmorning!
07:31:22ruezenspider: Completely OOB but I tried out Things.app, I like it a lot better than OmniFocus
07:31:24ruedbussink: *wave
07:31:25scooprroo =)
07:31:55ruedbuss..eroo..tor?
07:33:17dbussinkwtf?
07:34:05brixenDefiler: mspec/spec_helper.rb has nothing to do with spec/** specs
07:34:17brixenDefiler: it's only for running mspec specs like so: spec mspec/spec
07:34:29brixenwith rspec
07:35:41GMFlash enters the room.
07:37:14mae enters the room.
07:39:23ruebrixen: I think they were saying that the #pretty_inspect was getting picked up
07:39:26rueMaybe a path issue?
07:39:46maehey guys
07:40:03ruemaeprnator
07:40:37maewow
07:40:42maewhat ?
07:40:44maeheh
07:41:14maeso is this like a vm for ruby, the pure and lazy function language, with type inferencing and strong types?
07:42:16brixenrue: there's no way, nothing includes mspec/spec_helper.rb
07:42:32brixenspec/ruby/1.8/spec_helper.rb requires pp.rb
07:42:35dkubb enters the room.
07:43:07rueDoes it? I only found the one
07:43:10rueBut I was a bit hasty
07:43:15brixenone what?
07:43:22rueThe one in mspec/
07:43:34ruemae: Well, it used to be
07:44:15ruemae: Currently we are actually implementing a pretty synergestic multi-tiered business solutions architecture framework integrator
07:44:15antares leaves the room.
07:44:22antares enters the room.
07:55:15GMFlash leaves the room.
07:55:20GMFlash enters the room.
07:56:08GMFlash leaves the room.
07:56:09octopod enters the room.
07:56:16GMFlash enters the room.
08:00:35antares_ enters the room.
08:00:36antares leaves the room.
08:12:23qwert666 enters the room.
08:13:04qwert666hi
08:13:23antares_ leaves the room.
08:13:40dbussinkbrixen: still there?
08:16:32octopod leaves the room.
08:16:52brixendbussink: yeah, briefly
08:17:11boyscout2 commits by Brian Ford
08:17:13boyscout * Exclude failing UNIXServer.new spec.; 004662e
08:17:14boyscout * Remove dylib from libgdtoa.; f317583
08:17:24dbussinkbrixen: ah ok, well, was looking at openbsd and the 0.0 / -0.0 is still there, i don't think that even can be fixed
08:17:31dbussinkbrixen: mri exposes the same behavior too
08:17:42brixenhrm, ok
08:17:48dbussinkbrixen: problem is that the system doesn't differentiate between it
08:17:49brixenfixed it on gutsy
08:18:06brixenat what level?
08:18:08dbussinkprintf("f: %f\n", -0.0); => 0.0 on openbsd
08:19:21brixendo you have a parallels image for openbsd?
08:19:26dbussinkyeah
08:19:44brixenis there a way I could ftp it from you or something?
08:20:06dbussinki have an ftp
08:20:19dbussinkbut i've limited the upload speed, so it's gonna take quite some time
08:20:53brixenok, perhaps later
08:21:07brixenso, ppc needs some tweaks on the libgdtoa
08:21:20brixen#define NO_LONG_LONG I believe
08:21:33brixenbut even with that, float/constants_spec.rb still hangs
08:21:34dbussinkopenbsd also needs them
08:22:05dbussinkNO_FENV_H
08:22:06brixenconstants_spec has the float literals with a very large number of digits
08:23:01brixenI can log in again to cremes's machine
08:23:07brixenbut I'll have to look at it tomorrow
08:23:12brixensleep time :)
08:23:47brixenI hate hw manufacturers and their stupid fp quibbles
08:23:49dbussinknight!
08:23:53brixenjust have a standard already :P
08:24:00brixennight! :)
08:24:01dbussinkhehe, yeah
08:28:09rue_XIV leaves the room.
08:37:37maerue: wow, you've got to be kidding me ; )
08:46:50mediogre enters the room.
08:54:46Fullmoon enters the room.
09:06:16mediogre leaves the room.
09:12:03thehcdreamer enters the room.
09:13:14wycats enters the room.
09:13:45wifelette enters the room.
09:16:33perplexes leaves the room.
09:25:18Fullmoon leaves the room.
09:27:41GMFlash leaves the room.
10:10:04zimbatm enters the room.
10:15:54tmornini enters the room.
10:16:20thehcdreamer leaves the room.
11:02:45yipstar leaves the room.
11:05:49chris2 enters the room.
11:34:38hassox enters the room.
11:49:10Fullmoon enters the room.
11:49:47Fullmoon leaves the room.
11:53:37tmornini leaves the room.
12:27:35qwert666_ enters the room.
12:30:03Arjen_ enters the room.
12:32:56wireless_ enters the room.
12:33:06w1rele55 leaves the room.
12:37:20webmat enters the room.
12:45:21qwert666 leaves the room.
12:49:47dewd enters the room.
12:54:35lopex enters the room.
12:59:56motion leaves the room.
13:03:52nemerle enters the room.
13:14:04radarek enters the room.
13:15:02rubuildius_ppcBrian Ford: 004662e54; bin/ci failed! http://pastie.caboo.se/paste/166030
13:15:03rubuildius_ppcMenTaLguY: b840ee825; bin/ci failed! http://pastie.caboo.se/paste/166029
13:15:20cremes_ enters the room.
13:20:32ctennis enters the room.
13:31:21dewd leaves the room.
13:33:58nicksieger enters the room.
13:38:41imajes enters the room.
13:46:22imajes_office leaves the room.
13:57:31dewd enters the room.
14:10:13VVSiz enters the room.
14:18:12GMFlash enters the room.
14:35:29lopex leaves the room.
14:51:38RyanTM enters the room.
15:11:02pauldix enters the room.
15:11:45cremes_ leaves the room.
15:12:08radarek leaves the room.
15:12:10jacen_ enters the room.
15:14:20cavalle_ leaves the room.
15:21:24pauldix leaves the room.
15:40:23bitbckt enters the room.
15:41:29VVSizmorning, folks! does anybody have 64 bit MRI? I'd like to check one simple command output.
15:42:08VVSizTime.utc(-1) - what this would return in 64bit MRI?
15:42:11ctennisVVSiz: sure
15:42:16VVSizcool
15:42:28ctennisTiirb(main):001:0> Time.utc(-1)
15:42:29ctennis=> Fri Jan 01 00:00:00 UTC -1
15:42:44VVSizvery good! and this: Time.utc(39)
15:42:56ctennisirb(main):002:0> Time.utc(39)
15:42:56ctennis=> Sat Jan 01 00:00:00 UTC 39
15:43:02dbussinksame in the year 39
15:43:21dbussinkVVSiz: testing what jruby should do?
15:43:42VVSizgood. So indeed 64bit MRI does not raise exception on these values (but 32 bit MRI does)
15:43:57VVSizdbussink: yep, we have some new Time rubyspec failures, so I'm trying to figure them out
15:44:14VVSizdbussink, ctennis: how about this one: Time.utc(2**30) ? :)
15:44:15dbussinkVVSiz: yeah, well, this is because time_t is 32 bit or 64 bit
15:44:31ctennisirb(main):001:0> Time.utc(2**30)
15:44:31dbussinkand Time is a direct translation from that
15:44:31ctennis=> Thu Jan 01 00:00:00 UTC 1073741824
15:44:43VVSizdbussink: yep, just wanted to make sure
15:44:46VVSizthanks!!
15:44:54ctennissure
15:45:15dbussinkwith a 64 bit int, time can be modelled from before the big bang until really far far in the future :P
15:45:54VVSizhopefully, we'll never switch to 128 bit! :)
15:46:46dbussinkwell, for timestamps it wouldn't be necessary
15:47:14dbussinkbut the time between between each increase in bit size is growing and growing
15:47:22dbussinkso i think it's gonna take a while
15:47:32VVSiz:)
15:49:25dbussinkis java's time stuff 64 bit based?
15:52:33VVSizdbussink: we use 3rd party JodaTime library, and it keeps the offset in 64 bit value
15:52:43VVSizthe same is for Java built-in Time value
15:53:02dbussinkah ok, well, we want to replace ours eventually too
15:53:20dbussinkbut there are some nasty things there with respect to timezone discovery etc
15:57:05nicksieger leaves the room.
16:05:08rueMoo
16:05:16VVSiz:)
16:05:42nicksieger enters the room.
16:08:10webmatHey guys, do you know why the installation instructions say "CURRENTLY PLEASE DO NOT INSTALL"?
16:08:19webmatObviously I've tried and yes it fails...
16:08:29webmatI'm just curious as to why :-)
16:08:41dbussinki don't know, i never install it :P
16:08:54dbussinki think there are still issues with setting up the correct paths etc.
16:09:04dbussinkand making sure that the correct versions are picked up
16:09:46webmatAh, ok, thanks :-)
16:10:55dbussinkbut if you want to look into it, please do
16:12:15webmatI just might. I have a couple of things in the queue on other subjects, but I'm just starting to really look into Rubinius and god I'm excited about it :-)
16:16:44dbussinkthat's nice to hear
16:28:20VVSizhmmm, the updated udpsocket specs are problematic for me
16:28:34VVSizthey hang hard on Linux, with MRI and JRuby
16:29:44evanwebmat: i need to change the instructions
16:29:48evanit's ok to install now
16:29:53evanwe've ironed out the problems
16:30:19webmatevan: nice!
16:37:47olabini enters the room.
16:37:48webmatEarlier I reported that it didn't work for me. Actually I didn't fiddle with it at all, since the web page said it didn't work. Turns out I just needed to sudo :-) I had assumed the rake task would call sudo by itself like Merb's does :-) Moral of the story, sudo rake install works like a charm on OSX Leopard.
16:38:07enebo enters the room.
16:40:13jayWHY enters the room.
16:43:51dbussinkevan: ah, the collision problems with existing installs when compiling a new one are also fixed?
16:44:18dbussinkevan: how's the work on the order evaluation going?
16:48:37headius enters the room.
16:49:14KirinDave enters the room.
16:50:08VVSizis Matt Palmer here? I'd like to discuss the udpsocket rubyspec changes that lead to spec run hangups on Linux
16:56:09headius leaves the room.
16:56:18crossblaim enters the room.
16:56:25crossblaimhello
16:57:18enebo leaves the room.
16:57:38enebo enters the room.
17:00:10enebo leaves the room.
17:00:38enebo enters the room.
17:03:10wdperson enters the room.
17:03:14enebo leaves the room.
17:03:14enebo_ enters the room.
17:05:31enebo enters the room.
17:06:16enebo leaves the room.
17:09:15enebo enters the room.
17:10:25enebo_ leaves the room.
17:14:17nicksieger leaves the room.
17:15:16nicksieger enters the room.
17:24:21jayWHY leaves the room.
17:25:44rubuildius_amd64 enters the room.
17:26:11rubuildius_amd64 leaves the room.
17:27:37rubuildius_amd64 enters the room.
17:30:27nemerle leaves the room.
17:34:07imajes_ enters the room.
17:35:03rubuildius_amd64Brian Ford: 004662e54; 1756 files, 6070 examples, 20315 expectations, 1 failure, 0 errors; http://rafb.net/p/kTWi7V81.html
17:36:26ctennis leaves the room.
17:36:39jayWHY enters the room.
17:39:54bigl enters the room.
17:40:09djwhittodd... when I checked on the build bot this morning it looked like it was hung trying to run the specs
17:40:15djwhittI restarted it and everything seems ok
17:41:30enebo leaves the room.
17:43:40mgreenly enters the room.
17:44:40krisps enters the room.
17:44:43brixendgtized: just ran then on a gentoo 64 and got 0/0
17:44:53brixennot sure what's up with that one failure reported
17:44:57brixenis the bot on HEAD?
17:45:22djwhittyep, should be
17:45:23brixenalso, I think I've got the ppc issue with gdtoa figured out
17:45:25brixenok
17:45:50djwhittI'll see if I can figure out what's going on with that failure
17:45:57brixencool, thanks
17:46:12tizianobis enters the room.
17:46:22ctennis enters the room.
17:49:54mae leaves the room.
17:50:23mae enters the room.
17:54:48rubuildius_amd64 leaves the room.
17:55:04rubuildius_amd64 enters the room.
18:01:03tizianobis_ enters the room.
18:05:00rubuildius_amd64Brian Ford: 004662e54; 1756 files, 6070 examples, 20315 expectations, 0 failures, 0 errors; http://rafb.net/p/db30SU69.html
18:05:26djwhittthere we go
18:05:48brixenyay
18:05:56djwhittthe zsh script was catching HUP before the Ruby process could get it
18:06:24djwhittit wasn't doing anything useful with it so I just removed the signal handlers from the zsh script
18:06:51nicksieger leaves the room.
18:07:07thehcdreamer enters the room.
18:07:07thehcdreamer leaves the room.
18:07:11djwhitthopefully that won't have any nasty side affects
18:07:25brixenahh
18:07:44brixenI'm using zsh on freebsd, so that probably explains the hangup there
18:07:54brixenbut I'm seeing it on gutsy too, and I use bash on that
18:10:10djwhittyou're seeing the same failure?
18:10:28brixenwell, I was seeing hangup, I'll have to look at it again
18:10:33brixenI don't have that vm booted
18:10:39djwhittmight not be the same thing
18:10:43brixenyeah
18:10:47djwhittin my case the problem was I wasn't receiving a hangup
18:10:51djwhittthe spec was expecting it
18:10:59brixenahh ok
18:11:13imajes leaves the room.
18:13:06imajes_ enters the room.
18:19:05tizianobis leaves the room.
18:21:56brixenfinally! 0/0 spec run on ppc
18:22:08brixencremes: thanks for making that ppc box available, you rock :D
18:30:07dbussinkbrixen: if you commit it i can test freebsd / openbsd and fix up some stuff there
18:30:13imajes leaves the room.
18:30:30brixendbussink: super
18:30:40brixendbussink: I've got 5 different systems building right now :)
18:30:55dbussinkah, what types?
18:31:02brixenwait 4
18:31:12brixenos x intel, os x ppc, freebsd, gentoo 64
18:32:00brixenI've got a partially installed opensuse I should finish
18:32:16brixenand I've got a xsde (?) solaris image that I hate
18:32:31brixenbut I should probably get it running
18:36:44VVSizwhat the hell, completely bogus spec added: http://git.rubini.us/?p=code;a=commit;h=9b3988436a21f61c86168a7566d472c4dfa22162
18:37:18VVSizfirst of all, it uses '=' instead of '==', and second of all it verifies something that is not true for MRI, JRuby
18:38:02benburkert enters the room.
18:38:02brixenhah! wowsers
18:38:32djwhittheh, you can blame zenspider for that one
18:38:37djwhitthe comitted it
18:38:46VVSiz:)
18:38:58djwhittI'm guess that was a patch from the sprint
18:39:01djwhitt*guessing
18:39:06VVSizheheheheh
18:39:14RyanTM leaves the room.
18:39:52brixensmack 'im!
18:40:21VVSizI'll just revert the commit since it obviously wrong for now
18:40:41brixenyeah
18:40:51RyanTM enters the room.
18:41:03brixendrat, 0/0 spec runs on os x intel/ppc and now gentoo 64 hangs
18:41:57VVSizbrixen: test runs for me with JRuby/MRI also hanging due to #421 (udpsocket test hangups), and the tests seems to be as well obviously broken
18:42:19brixenVVSiz: ok, running with -fs now to try to locate it
18:42:54hoopy leaves the room.
18:43:56boyscout1 commit by Vladimir Sizikov
18:43:57boyscout * Revert "Added simple spec for range splatting".; ff5e9d3
18:44:04KirinDave leaves the room.
18:45:09VVSizbrixen: just in case: http://rubinius.lighthouseapp.com/projects/5089-rubinius/tickets/421
18:45:29brixenVVSiz: thanks
18:46:18brixenhmm, 64bit seems hung on Time.xmlschema
18:51:22brixenugh! but just ran it with -T --gdb and no hang
18:52:09djwhittbrixen: is that happening on 64bit linux?
18:52:12nemerle enters the room.
18:52:31brixenyeah
18:52:38brixendjwhitt: seems nondeterministic
18:52:53djwhittI wonder if that's what was hanging rubuildius earlier today
18:54:24brixenwell, I've got 2 clean runs on 64, clean on os x intel/ppc, and seems ok on freebsd, except that the process gets killed in Generator specs for using up all my swap
18:54:41brixenI'll push this and we can see what future fun awaits :)
18:56:11boyscout2 commits by Brian Ford
18:56:12boyscout * Added .gitignore for libgdtoa config generated files.; d4ada26
18:56:13boyscout * Remove libgdtoa config generated headers from version control.; 271edf6
18:56:47rubuildius_amd64 leaves the room.
18:57:04rubuildius_amd64 enters the room.
18:57:30hoopy enters the room.
19:02:40enebo enters the room.
19:04:22nicksieger enters the room.
19:09:41tizianobis_ leaves the room.
19:15:35xif enters the room.
19:16:21xif leaves the room.
19:17:10mgreenly leaves the room.
19:18:52djwhittI think it's hung
19:19:00xif enters the room.
19:19:38KirinDave enters the room.
19:20:04rubuildius_ppcBrian Ford: d4ada2679; 1756 files, 6073 examples, 20342 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/166124
19:20:13djwhittoh, never mind
19:20:13djwhittheh
19:20:24djwhittoh wait ppc...
19:20:55brixenyeah, just started another run on 64
19:20:59djwhittok yeah, looks like it's hung trying to run this command 'sh -c find /var -type s 2> /dev/null'
19:21:02brixendunno what's up with that
19:21:15brixenhmm, wth
19:21:18djwhittthat shows up below mspec-ci in the process tree
19:21:27djwhittyeah, I have no idea where it's getting that from
19:21:40brixenwhat command did you use to find that?
19:21:45djwhitthtop
19:21:52djwhittlooked at the process tree
19:21:57brixenok
19:22:21roo enters the room.
19:22:46djwhittI did a grep on the code for 'find /var' and didn't get any results so I'm not sure where it's coming from
19:22:55thehcdreamer enters the room.
19:23:55brixenI see something similar in file_types.rb in core/file/fixtures
19:24:16brixenwith this comment above:
19:24:16brixen # TODO: This is probably too volatile
19:24:20brixenhehe
19:24:32djwhittnice
19:24:33djwhittheh
19:29:18nicksieger leaves the room.
19:29:28brixendjwhitt: every time I run it like so bin/mspec ci -T --gdb
19:29:32benburkert enters the room.
19:29:32brixenI cannot get a hang
19:31:53rueThose are my proud achievements
19:32:23KirinDave leaves the room.
19:32:27rueThey were hardcoded before, I tried to make it slightly more permissive of different devfs' and directory layouts
19:33:09brixenrue: yeah, I'm not seeing any hangs there
19:33:19nicksieger enters the room.
19:33:22brixenI cannot get this thing to hang with -fs either
19:33:38brixenmaddening
19:34:10brixenperhaps a race somewhere
19:35:08djwhittrue: I think it's perhaps a bit too permissive now
19:35:40djwhittrue: perhaps we could have a list of likely socket locations instead of doing a find
19:36:44brixenperhaps we should make the spec only work if you have nc or netcat?
19:36:52brixenand make a socket to test, would that work?
19:37:05GMFlash leaves the room.
19:37:32djwhittwhat would it do if you didn't have netcat installed?
19:37:36djwhittjust not run the spec?
19:37:43brixendjwhitt: dunno what's up, I ran mspec ci without hang: plain, -T --gdb, -fs, -f*
19:37:52brixenyeah, just not run the spec
19:37:57brixenwe'd need yet another guard
19:37:59brixen:P
19:39:00brixenperhaps a committee to name the committee to recommend the guard name to the committee to name the committee to poll the community for the panel to approve the working group to study and recommend the best name
19:39:11brixenfor the guard
19:39:15enebo leaves the room.
19:39:17djwhittheh
19:39:32brixenand not to forget the paint color :P
19:40:56ezmobius enters the room.
19:40:59dbussinki object your honor!
19:42:45webmat leaves the room.
19:43:37rubuildius_amd64 leaves the room.
19:43:40webmat enters the room.
19:44:54webmat_ enters the room.
19:46:24yipstar enters the room.
19:46:41AndrewO enters the room.
19:49:33enebo enters the room.
19:50:31chris2 leaves the room.
19:55:45wifelette_afk leaves the room.
19:55:46wifelette enters the room.
19:55:49wycats leaves the room.
19:56:10wycats enters the room.
19:58:47dbussinkbrixen: i also have a hang with mri on 64 bit
20:00:03dbussinkUDPSocket.send
20:00:23webmat leaves the room.
20:00:56xmlhacker leaves the room.
20:01:29ruedjwhitt, brixen: The problem is circular speccing
20:02:14rueI chose the file types as the point where we have to be able to validate against the OS so that other specs can work with that assumption
20:02:54thehcdreamer leaves the room.
20:03:54webmat_ leaves the room.
20:03:57yaroslav enters the room.
20:04:50djwhittyeah, I get that, but doing a find against all of /var and /tmp is a little scary
20:05:03yaroslav leaves the room.
20:05:25dbussinkexpecially /var could take some time on certain systems
20:06:16djwhittyeah, I think doing a find against /var could even do nasty stuff like hit proc mounts on some systems
20:07:53ezmobius leaves the room.
20:08:44djwhittso, does Mac OS 'find' have the '-mount' option?
20:15:49djwhittlooks like it doesn't
20:16:17djwhittdoes seem to have -x though...
20:16:26djwhitt-xdev seems to be deprecated
20:17:34tizianobis_ enters the room.
20:22:57tizianobis leaves the room.
20:24:19djwhittrue, brixen, dbussink: do you think it would be safe to change that find command to hit /var/run instead of /var?
20:24:56dbussinkdjwhitt: that should work for the systems i know
20:25:21brixendjwhitt: sure
20:25:46brixenI think I'd still prefer making the spec depend on the system having nc or netcat
20:27:09djwhittbrixen: yeah, that would be better I think. I'm just not sure what's involved in creating that guard
20:27:57brixenok, le'me think about it a bit
20:28:10brixengotta run some errands, bbl..
20:28:16djwhittbrixen: can I just go ahead and make the /var/run change for now?
20:28:22brixenyep, please do
20:28:24djwhittk
20:36:05BluR enters the room.
20:36:18imajes_ leaves the room.
20:46:01wdperson leaves the room.
20:46:22boyscout1 commit by David Whittington
20:46:23boyscout * Modified file type specs to search for sockets in /var/run instead of /var; ad64c0e
20:47:08rubuildius_amd64 enters the room.
20:49:33boyscout1 commit by Dirkjan Bussink
20:49:33boyscout * Fix build on OpenBSD; df86fe6
20:51:03ruedbussink: Nitpicking again, can you make sure to make a little note about what specifically was broken with the build etc.?
20:51:18rueI rely on RSS occasionally and I just see the message, not the commitdiff
20:51:32rue(Which I should fix anyway but that notwithstanding)
20:51:39dbussinkah ok, well, try to think of it next time then
20:52:41jayWHY leaves the room.
20:56:25menator enters the room.
20:57:45jayWHY enters the room.
21:00:06rubuildius_amd64Dirkjan Bussink: df86fe61e; 1756 files, 6070 examples, 20315 expectations, 0 failures, 0 errors; http://rafb.net/p/r78Qjt76.html
21:00:07rubuildius_amd64David Whittington: ad64c0ea7; 1756 files, 6070 examples, 20315 expectations, 0 failures, 0 errors; http://rafb.net/p/FijgYb66.html
21:01:49nicksieger leaves the room.
21:02:28rubuildius_ppcDirkjan Bussink: df86fe61e; 1756 files, 6073 examples, 20344 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/166151
21:16:21MenTaLguY enters the room.
21:30:56nicksieger enters the room.
21:31:24MenTaLguYhowdy nick
21:31:53nicksiegerhi MenTaLguY
21:32:40dbussinkhey nicksieger, i think i had some beers with you back at railsconf europe
21:32:58nicksiegerhi dbussink. dirkjan, right?
21:33:09nicksiegerdid i spell that correctly
21:33:11dbussinkyeah, that's right
21:33:17dbussinkeven that :)
21:33:40dbussinkforeigners spell it correct more often than other dutch people :0
21:33:42dbussink:)
21:34:10nicksiegerat pycon trying to absorb the tasty bits of pythin
21:34:13nicksiegerpython rather
21:35:11dbussinkhaving fun there?
21:35:26dbussinkin chicago i see
21:36:17nicksiegeryeah, it's pretty good. don't know too many folks here, but...
21:36:26nicksiegertalks are very mixed
21:37:03nicksiegerthere are definitely some people new to presenting, but in general everyone is received well at least
21:37:57dbussinkah, they even have dedicated development sprints
21:37:58wycats leaves the room.
21:41:31djwhitthmm... just had an idea regarding build bots
21:42:26AlanNorbauerI did a rake git:update to pull the most recent code and can no longer build. anyone have an idea?: http://pastie.caboo.se/166162
21:43:02dbussinkAlanNorbauer: you tried rake distclean build ?
21:43:34dbussinkbrixen: perhaps an idea to do some sprinting at railsconf?
21:43:38AlanNorbauerI'll give that a shot, thanks.
21:43:42AlanNorbauer(I'm new, sorry)
21:44:04djwhittI'd like to centralize buildbot reporting a bit more, but setting up a central server could cause problems too since someone has to maintain the server etc.
21:44:30djwhittso I'm thinking maybe I could just use a separate IRC channel for build bots
21:44:49djwhittand then have a bot that pulls output from that channel and summarizes it here
21:45:00djwhittthat way noise could be kept down in this channel
21:45:06dbussinkdjwhitt: i think i've heard that idea somewhere before, but it's definitely useful
21:45:19djwhittyeah, I think maybe benny was talking about it
21:45:35djwhittI like it because it keeps it easy for people to setup bots
21:45:45djwhittbut still keeps the noise down
21:50:06wycats enters the room.
21:58:18antares enters the room.
22:11:08jayWHY leaves the room.
22:15:24chop3 enters the room.
22:22:03MenTaLguYhow's Rubinius doing with rake and gems, OOC?
22:22:16headius enters the room.
22:22:29MenTaLguYcharlie!
22:22:53headiusI'm here!
22:22:55headiuspycon wifi has been pretty goofy, but better today
22:23:14MenTaLguYcool
22:23:30MenTaLguYI've got a question
22:23:38MenTaLguYdo we care about supporting trace.rb and debug.rb in JRuby?
22:23:48headiuswhy, they don't work?
22:24:01headiusI don't particularly care, but others might
22:24:29MenTaLguYwell, they rely on Thread.critical at the moment, in ways that may not be entirely satisfied by the "loose" Thread.critical implementation we can even do
22:24:39djwhittMenTaLguY: gems work now I think
22:25:23djwhittthere's a rubygems directory in lib
22:26:18MenTaLguYheadius: anyway, I guess it's not really something I'm asking for an immediate answer on
22:26:24MenTaLguYheadius: more something to mull over
22:27:25headiusMenTaLguY: I'm not particularly concerned about making them work if we can just have something equivalent
22:27:52headiusand really, if we want debugging support, getting compiled execution cleaner and using or wrapping jdb is almost enough
22:28:19headiusI've played with jdb and compiled ruby code, and you can set breakpoints in compiled ruby just fine
22:28:27headiusanyway, more a #jruby discussion too :)
22:37:50jayWHY enters the room.
22:40:40ezmobius enters the room.
22:43:09antares leaves the room.
22:46:06AlanNorbauerShould I expect /bin/mspec ci to always pass unless I'm doing something wrong? 1 expectation is failing
22:46:18djwhittAlanNorbauer: which one?
22:46:28AlanNorbauerSubtendMethodVis should have global method smv_test_global_function FAILED
22:46:30DefilerIt should pass, but might not if you have a platform we don't normally try it with
22:46:39AlanNorbauerDefiler: I'm on OSX
22:47:16Defilerand 'rake' thinks it doesn't need to build anything?
22:48:23AlanNorbauerI just ran a rake distclean build and it exited without any errors, as far as I can tell
22:49:35AlanNorbauerand running rake build again does nothing
22:50:08cjheath leaves the room.
22:51:35imajes enters the room.
22:52:54cjheath enters the room.
23:00:14jayWHY leaves the room.
23:09:48zimbatm_ enters the room.
23:10:12zimbatm leaves the room.
23:16:40aotearoa enters the room.
23:21:17jayWHY enters the room.
23:22:42nicksieger leaves the room.
23:28:45geekounet leaves the room.
23:28:56geekounet enters the room.
23:29:54jayWHY leaves the room.
23:30:15AndrewO leaves the room.
23:31:30enebo leaves the room.
23:33:20headius_ enters the room.
23:33:24headius leaves the room.
23:34:20headius_ leaves the room.
23:42:12qwert666_ leaves the room.
23:47:15imajes leaves the room.
23:49:59imajes enters the room.
23:52:48wireless_ leaves the room.
23:54:22RyanTM leaves the room.
23:55:14UncleD leaves the room.
23:56:30AlanNorbauer leaves the room.
23:56:41RyanTM enters the room.
23:59:39womble leaves the room.