Index

Show enters and exits. Hide enters and exits.

00:01:32obvio leaves the room.
00:02:25mkrauskopf leaves the room.
00:03:38headius leaves the room.
00:04:02enebo leaves the room.
00:10:58trythil leaves the room.
00:11:42srbaker enters the room.
00:15:15lstoll enters the room.
00:16:56wireless_ leaves the room.
00:17:35thudson_ leaves the room.
00:25:09zenspidergah... the specs take a long time now
00:25:18evanyeah
00:25:23evanit's the running of the stdlib specs
00:25:24zenspiderand wow... the linear growth has really slowed down
00:25:30evani think we need to stop running them as part of ci
00:25:32zenspider(linear growth of the count)
00:26:08boyscout1 commit by Ryan Davis
00:26:09boyscout * String#unpack overhaul. NO extra methods littered through Fixnum/Integer/String. ...; 6a13357
00:26:18mernen enters the room.
00:26:26zenspiderrawr
00:26:37zenspidergod that sucked more than Array#pack
00:27:44brixendamn you zenspider
00:27:53brixenyou just had to do that while I'm in the middle of this :P
00:28:06evanack.
00:28:37zenspiderbrixen: I have git:pull radar. :P
00:28:49brixenmaintaining the airplane while it flies, fun
00:29:29zenspiderif buck rogers can do it IN SPACE... you can too. :P
00:29:35brixenheh
00:30:04zenspiderisn't git great? I can blindside you DISTRIBUTED
00:30:17brixenwhew, rebase origin worked ok
00:30:52zenspiderfuck. my brain hurts
00:31:59srbaker leaves the room.
00:37:52rubuildius_amd64Ryan Davis: 6a1335746; 2105 files, 6854 examples, 24677 expectations, 0 failures, 0 errors; http://rafb.net/p/Bdd6WG11.html
00:39:03zenspiderrawr
00:40:44brixenok folks, *please* do not push any commits for a minute
00:41:29anteaya enters the room.
00:42:19rubuildius_ppcRyan Davis: 6a1335746; 2105 files, 6856 examples, 24703 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/193988
00:43:52zenspiderpushes a commit just for fun. :P
00:45:36Defilerzenspider: Tasty amount of code deleted in 6a13357. Nice.
00:46:58zenspiderDefiler: I actually increased the flog of unpack by about 15%. :/
00:47:08zenspiderbut considering pack shrunk by 33%, I'm ok with that
00:47:27Defiler15%?? Project status critical!
00:47:45zenspiderI've got a change to uuencode in pack that halves the time again
00:47:52zenspideroriginal:mine:new :: 3:2:1
00:48:41zenspiderDefiler: well... I got rid of all the procs within, and there were a TON of stupid public methods littering throughout Fixnum, Integer, and String. They're all removed
00:48:50zenspidermost were only called once at that... so freakin' lame
00:48:57zenspiderI gotta run.
00:49:05DefilerYeah, I saw that in the diff. Cool
00:49:26wmoxam leaves the room.
00:50:07boyscout3 commits by Brian Ford
00:50:08boyscout * Made rake build depend on smarter mspec:init task.; 1960197
00:50:09boyscout * Put MSpec on a submodule tether.; 849d9fd
00:50:10boyscout * The emancipation of MSpec.; d6abe18
00:50:23brixenok, so please try rake git:update and rake build
00:50:35brixenI'm dubious that git will handle the mspec dir correctly
00:51:08brixenif this breaks, I suppose we can make git:update task patch everything up
00:53:59jtoy enters the room.
00:54:38thoughtfiz leaves the room.
00:56:07brixenevan: have you run git gc on the main repo recently? indexing seems to take a fair amount of time
00:56:17benstiglitz leaves the room.
00:57:24Fullmoon leaves the room.
00:57:53rubuildius_amd64Brian Ford: 19601972b; 2105 files, 6854 examples, 24677 expectations, 0 failures, 0 errors; http://rafb.net/p/P8Axm569.html
00:58:39benburkert leaves the room.
00:58:53thoughtfiz enters the room.
01:00:26brixendjwhitt: ping
01:01:20thoughtfiz leaves the room.
01:03:47Defilerwycats_: ping
01:05:30rubuildius_ppcBrian Ford: 19601972b; 2105 files, 6856 examples, 24703 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194000
01:07:20Fullmoon enters the room.
01:09:07boyscout1 commit by Wilson Bilkovich
01:09:08boyscout * Handle classes with no subclasses in Class#__subclasses__; ed76fd2
01:14:07jtoy leaves the room.
01:16:34evanbrixen: i'll run it now.
01:16:56brixenevan: cool
01:17:20dhvsfj enters the room.
01:17:55rubuildius_amd64Wilson Bilkovich: ed76fd25c; 2105 files, 6854 examples, 24677 expectations, 0 failures, 0 errors; http://rafb.net/p/xuZwVz37.html
01:18:47evanbrixen: crap.
01:18:52evanrake gives me an error
01:19:10evanabout mspec existing but not being empty or a git repo
01:19:16brixenevan: yeah
01:19:24brixenlike this: http://ci.rubini.us/ci/1210291786.M366149P12107.hoshi.fallingsnow.net.html
01:19:38evanyeah
01:19:45evanit didn't nuke the directories
01:19:52evanyou need to do that in the rake task probably
01:20:05evanrm -rf mspec
01:20:06brixenperhaps rake git:update ?
01:20:07evanfixes it
01:20:09brixenyeah
01:20:11evanno
01:20:15rueI would just have people remove it manually this time
01:20:20evanjust have mspec:init detect this case and remove the mspec directory
01:20:23evanno
01:20:23evanno
01:20:28evanthe task needs to do it
01:20:35evanotherwise we'll send the next 2 weeks telling people this
01:20:38evanover and over and over and over
01:20:40brixenevan: not sure what the criteria is
01:20:43rueNo, the task needs to do it /only once/
01:20:49brixenevan: I first had a check for .gitmodule contents
01:20:59brixenthen changed that to checking mspec/bin/mspec exists
01:21:00evanbrixen: easy
01:21:05evanthere is an mspec/bin directory
01:21:07evanand it's empty
01:21:08evanif so
01:21:12evannuke mspec
01:21:14evanthen continue
01:21:17rueUse the commit hash
01:21:22evanwe'll remove it later
01:21:50evanif you don't add it, i will.
01:22:01brixenyeah, I can
01:22:07evanno please.
01:22:11evaner.
01:22:12evannow please.
01:23:32brixenevan: the existing test is: unless File.exist? "mspec/bin/mspec"
01:23:40evanwell, that didn't work for some reason
01:23:46evanbecause it barfed on me.
01:23:51brixenevan: seems I just need rm -r mspec before the git submodule init?
01:23:52evani can check
01:23:56evanoh
01:23:56evanyeah
01:23:58brixenok
01:24:01evando it as the 1st thing
01:24:04brixensure
01:24:46brixenevan: well, since I can't repro here, can you add: rm_f "mspec" at Rakefile:348 and test it?
01:24:56evanit's easy to repro
01:25:10rubuildius_ppcWilson Bilkovich: ed76fd25c; 2105 files, 6856 examples, 24703 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194008
01:25:19evanjust check out in the past
01:25:48brixenyeah, ok
01:27:06evanor just check that git submodule returns not 0
01:27:10evanand nuke the directory
01:27:11evanand start over
01:27:56evani have to run.
01:28:00brixenk
01:30:11brixenhrm, I cannot repro it. I'll just add the rm
01:33:55boyscout1 commit by Brian Ford
01:33:56boyscout * Remove mspec dir before attempting git submodule init.; 6834933
01:34:55ruebrixen: Should send an e-mail to the list just in case
01:36:08brixenrue: ok, will do
01:42:56rubuildius_amd64Brian Ford: 6834933e9; 2105 files, 6854 examples, 24677 expectations, 0 failures, 0 errors; http://rafb.net/p/X23CE862.html
01:42:58brixenrue: posted, please correct/add to if necessary
01:44:07thoughtfiz enters the room.
01:47:32brixenwell that fixed it for ci.rubini.us
01:49:22thoughtfiz leaves the room.
01:49:39rubuildius_ppcBrian Ford: 6834933e9; 2105 files, 6856 examples, 24703 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194015
01:51:23blakewatters enters the room.
01:52:27shame enters the room.
01:55:21thoughtfiz enters the room.
01:56:57dgtizedevan: definitely having major problems with the cpp branch
01:57:19thoughtfiz leaves the room.
01:57:51dgtizedI tried wiping it and recreating, and now it asks for iseq_instructions_names.hpp everywhere
01:59:05dgtizedruby instructions.rb still isn't in the rake file for recreating as part of the rake task
01:59:21dgtizedbut running that still doesn't fix everything
01:59:23ruebrixen: Should be fine. If lots of folks seem confused, can send another one tagged [LISTEN UP NOW]
01:59:44ruedgtized: Should that be instruction_names ?
02:00:14dgtizedrue: the hpp file or the rb file?
02:01:36rueAugh, they are still talking about the stabby operator on -core
02:02:04ruedgtized: Either one; I mean could it be a problem with the filename?
02:02:06yugui enters the room.
02:02:57rueI really wish they could be as enthusiastic about actual issues
02:03:04ttmrichter enters the room.
02:03:25dgtizedrue: no when I run instructions.rb it fixes the hpp problem, but it still won't compile
02:03:37sambo82 enters the room.
02:04:04dgtizedsecondly how do we get the new mspec stuff working, git stash keeps complaining about copying over a dirty state
02:04:47rueIs your branch clean?
02:06:34dgtizedrue: seems to be, show-branch shows same as HEAD, and I don't see why it's doing git stash apply anyway since I don't have anything stashed
02:06:51jtoy enters the room.
02:06:51dgtizedrue: I mean I have 4 or 5 untracked files
02:06:59dgtizedbut other then that
02:07:45thoughtfiz enters the room.
02:07:57dlee leaves the room.
02:08:26kw enters the room.
02:10:23rueDoes git-status give you a branch? I would associate that with not being on a branch (at all, not even master)
02:10:34dgtizednm, mspec:init fixed it
02:10:40dgtizedand yes I was on a branch
02:11:02dgtizedso what do we have to do special to commit/update the mspec branch?
02:11:29rueHm, weird. Might want to clear the nonexisting stash
02:12:27dgtizedrue: did so, I think it was more like I wiped out mspec and then didn't init it, and then git forgot to tell me it was missing, but thought stuff was dirty
02:12:57dgtizedagain though, how do we commit/update for the mspec branch now or is that documented somewhere?
02:13:05ruedgtized: I think brixen has or will put in rake tasks but generally, you would make your change in the mspec branch, commit it and then push it which updates the *mspec* repo. Then you would drop down to Rubinius proper and commit the submodule update and then push that to the Rubinius repo
02:13:19rueNote that this is mspec/, not yet spec/
02:13:58dgtizedk
02:14:08rueLet me send my submodule link on the list for background information
02:14:49dgtizedare updating the git browser on git.rubini.us to be fancier so it knows about submodules or does any version of that browser do that?
02:15:07dgtizedsorry, I meant are we?
02:15:57wmoxam enters the room.
02:16:11thoughtfiz leaves the room.
02:17:43brixenrue: this is quite a nice thread, linus is amusing
02:17:53brixensome parallels between e.g. rbx dev and jruby dev
02:17:55ruedgtized: I think there is a config variable for that
02:18:00brixenrue: http://kerneltrap.org/mailarchive/linux-kernel/2008/4/30/1663344
02:18:03ruebrixen: Which?
02:18:05rueTa
02:18:56ruedgtized: I am pretty sure GitHub automatically handles that if the repos are both on there
02:18:57brixenrue: nice http://kerneltrap.org/mailarchive/linux-kernel/2008/4/30/1667934
02:19:19brixendgtized: you need to have a github account
02:19:48brixendgtized: docs for this will come out. if you have a burning commit to mspec, post it to: rubinius-dev ml or rubyspec ml
02:19:53sambo82_ enters the room.
02:20:27lopex leaves the room.
02:20:37sambo82 leaves the room.
02:21:39wmoxam leaves the room.
02:23:44brixenrue: ack, how did you sneak in 37d4825
02:24:14brixenrue: that belongs in spec_helper.rb for whatever is using mspec, not in mspec itself
02:25:23dgtizedbrixen: I have a github account, are we tracking everything through github now?
02:25:34brixendgtized: no, just mspec and rubyspecs
02:25:42brixenI'll add people to those accounts
02:25:54brixendgtized: le'em see if I can figure out how ;)
02:26:04dgtizedk
02:26:38brixendgtized: what's your github user?
02:29:19ezmobius leaves the room.
02:29:33dysinger leaves the room.
02:29:35dhvsfj leaves the room.
02:29:56dysinger enters the room.
02:31:41dysinger leaves the room.
02:31:46VVSiz_ enters the room.
02:32:37dgtizedbrixen: dgtized
02:33:13brixenheh
02:33:34ShayArnett enters the room.
02:33:35brixenohh, look at that fancy autocomplete
02:33:54brixendgtized: ok, added as a collaborator
02:34:19brixendgtized: http://github.com/brixen/mspec/tree/master
02:34:46brixenhmm, need to figure out the easiest way to do this with existing folks
02:34:56brixenohh, a LH page
02:35:04dhvsfj enters the room.
02:37:21ruebrixen: Sneakily
02:37:50brixenhmm, or I could just guess I suppose
02:38:48brixenrue: what's yours?
02:38:54dgtizedbrixen: I care more about access to the spec set once that is published, but thanks
02:40:00VVSiz leaves the room.
02:40:17brixendgtized: yeah, I'll add ya
02:40:58brixengrabbing some food, bbiab..
02:41:10ruebrixen: Guess
02:41:47rueYou are not duck-adding people ;)
02:42:42brixenrue: duck-adding?
02:42:53brixenrue: oh, autocomplete didn't show you the first time
02:45:47MenTaLguY enters the room.
02:46:08benburkert enters the room.
02:47:31thoughtfiz enters the room.
02:49:16thoughtfiz leaves the room.
02:51:58enebo enters the room.
02:52:52lstoll leaves the room.
02:53:36headius enters the room.
03:06:24rubuildius_ppc leaves the room.
03:06:27rubuildius_ppc enters the room.
03:16:46cored enters the room.
03:19:34ezmobius enters the room.
03:20:07headius leaves the room.
03:24:05trythil_ enters the room.
03:29:49rubuildius_ppc leaves the room.
03:29:51rubuildius_ppc enters the room.
03:31:31dhvsfj leaves the room.
03:33:54anteaya leaves the room.
03:34:12binary42_ enters the room.
03:34:15binary42 leaves the room.
03:34:38enebo leaves the room.
03:37:54Fullmoon leaves the room.
03:40:19sambo82_ leaves the room.
03:43:54dysinger enters the room.
03:46:03rubuildius_ppcBrian Ford: 6834933e9; 2105 files, 6856 examples, 24703 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194081
03:47:42nicksieger enters the room.
03:47:57MenTaLguY leaves the room.
03:48:25Fullmoon enters the room.
03:54:16dhvsfj enters the room.
03:56:11lstoll enters the room.
03:59:36cored leaves the room.
04:00:17dhvsfj_ enters the room.
04:01:02stepheneb enters the room.
04:05:10dhvsfj__ enters the room.
04:05:10dhvsfj_ leaves the room.
04:08:58djwhittbrixen: you pinged?
04:09:39brixendjwhitt: ah, just wondering how hard it would be to change the bot to use bin/mspec ci -B full.mspec (or somethin gsimilar)
04:10:13djwhittnot hard
04:10:20djwhittlet me try that right now
04:10:26brixenok, cool
04:10:53brixenwell, it's not implemented yet (i.e there's only spec/default.mspec)
04:11:06djwhittah, ok
04:11:22brixenI'll ping you if this passes muster with evan and I commit that
04:11:26djwhittk
04:11:34brixenthanks!
04:11:37djwhittsure, no prob
04:12:53rueMeep, I am glad I am not developing IronRuby. They have more annotation and whatever /*!*/ does than code
04:13:29dhvsfj leaves the room.
04:15:09stepheneb leaves the room.
04:15:10nicksieger leaves the room.
04:19:31evanno fix for the mspec problem?
04:19:46brixenit's fixed according to ci.rubini.us
04:19:53evanok
04:20:00evanah yeah
04:20:05evani missed it in the back trace
04:20:15evanand i'm full of gnochi, beer, and coffee
04:20:20brixenheh
04:21:31brixenevan: so, if we want to not run spec/ruby/1.8/library by default with bin/mspec ci, I can add a file like full.mspec (similar to spec/default.mspec) and have the bots run bin/mspec ci -B full.mspec
04:21:43brixenthen the default.mspec file will not include the library specs
04:21:55evanthat would be great
04:21:56brixensound reasonable
04:21:58brixenok
04:23:06evanhow about
04:23:08evanbin/mspec ci full
04:23:16evanjust to make it similar
04:23:19brixenwell, that's a lot more complicated
04:23:24evanok.
04:23:25brixenthere's already the -B option
04:23:30evanalrighty
04:24:54brixenin fact, folks can already have their own ~/.mspecrc and set :ci_files to whatever they want
04:25:08brixenbut by default, it loads spec/default.mspec
04:25:23evanoh, is there a write up about using that?
04:25:35brixenwill be in the new docs :)
04:25:39evanok.
04:25:58dhvsfj__ leaves the room.
04:26:56srbaker enters the room.
04:34:29wycatsDefiler: pong
04:34:53atmos leaves the room.
04:35:24lstoll_ enters the room.
04:36:01lstoll leaves the room.
04:36:56dhvsfj enters the room.
04:44:02benny leaves the room.
04:44:55crafterm enters the room.
04:45:00brixenevan: is the plan to have ci.rubini.us report something in channel?
04:45:23evanit will take over the functions of the rubuildius bots eventually
04:45:28brixenok
04:45:55thoughtfiz enters the room.
04:49:00evanawesome
04:49:14evanI'm going to eliminate perform_hook from the VM
04:49:15evanit's a hack.
04:49:25evanand anything that requires hooks is going to have to go first class
04:49:30rueRename it to perform_hack()
04:49:53evan%s/perform_hook/back_the_fuck_up/g
04:50:07evanit's only used a couple places now
04:50:15evan{add, attach}method
04:50:23evanand open_class
04:50:38evan{add, attach}method are already first class
04:50:46evanand we've talked about doing the same for class opening
04:50:48evanso it's a go.
04:53:19ttmrichter leaves the room.
04:53:56ttmrichter enters the room.
04:55:24thoughtfiz leaves the room.
04:59:47dhvsfj leaves the room.
05:01:30rueevan: `cat #rubinius.topic | mail -t "rubinius-dev@groups.google.com"`
05:03:03brixenhm, it would have been cool to keep all the history for mspec
05:03:14brixenI wonder how I might do that with the rubyspec files
05:03:40brixenI could just push a clone of rbx, then rm everything and move rubyspec files up to the toplevel dir
05:04:03ShayArnett leaves the room.
05:05:41rueYou have all the history of it?
05:06:04brixenrue: ?
05:06:06rueIt even shows on gitweb, actually
05:06:35brixenthere's no history here: http://github.com/brixen/mspec/tree/master
05:07:53thoughtfiz enters the room.
05:08:37rueAh, yes. That repository does in fact have no history since it was just created :)
05:09:34brixenright, but it'd be cool to migrate the history, but since there's no partial clones, what about a full clone then rm what you don't want and move stuff around?
05:09:50thoughtfiz leaves the room.
05:14:18rueYou could do that, sure. I would just dump the history in a file though
05:15:08rueYou are thinking in terms of 'project history' rather than 'repository history.' Dunno if it would cause any issues either way
05:15:34brixenwell, I'm thinking in terms of git blame
05:15:51brixenit's very helpful to understand the context of a change to a piece of code
05:18:06rueTrue
05:18:33rueAlthough in this case, if it is not in the new repo and if it is not you, you can blame it on me :)
05:19:18brixenheh
05:21:11yipstar leaves the room.
05:23:34ezmobius leaves the room.
05:24:59rueI do not really know of any other way to do it
05:27:52wycats_ leaves the room.
05:27:53brixencrap, this was all sorts of broken
05:31:52lstoll_ leaves the room.
05:36:52Blinchik leaves the room.
05:37:41Blinchik enters the room.
05:44:32thoughtfiz enters the room.
05:45:02Defilerwycats: I was wondering if you had had a chance to look at Merb + webrick yet. But ezmobius is all over it
05:46:41thoughtfiz leaves the room.
05:47:03brixenevan: what do you think of having the mspec and rubyspec github repos report through our LH?
06:00:15wycats_ enters the room.
06:02:50benburkert leaves the room.
06:19:39lstoll enters the room.
06:20:23thoughtfiz enters the room.
06:29:27lstoll leaves the room.
06:30:13Erlang00t enters the room.
06:30:33Erlang00tyou guys running compiler.rb etc, thru ruby or thru rubinius?
06:31:15rueRubinius
06:31:41rueCurrently MatzRuby is only needed for Rake
06:31:52rueWell, actually that might not even be true anymore
06:32:03Erlang00tso how do I debug it, if I want to learn its behavior?
06:32:09Erlang00tI mean the compiler.rb etc stuff
06:34:19brixenrue: sweet, rake mspec:sync worked like a charm :) thanks again for those docs
06:35:01brixenthis has been less painful than I feared
06:35:51brixenErlang00t: have you read through: http://rubini.us/doc/introduction-to-the-compiler/
06:36:33Erlang00tyes, I know how it worked
06:36:48Erlang00tbut I want to debug it how it work
06:36:59jtoy leaves the room.
06:37:36jtoy enters the room.
06:37:56Erlang00tlike in a netbeans environment, and see how the compiler.rb works
06:38:36brixenhmm, I don't know anything about netbeans
06:38:40boyscout4 commits by Brian Ford
06:38:41boyscout * Updated MSpec submodule to 849d9fd4; 199ca08
06:38:43boyscout * Made rake build depend on spec:update.; 216bca1
06:38:43boyscout * Added spec:full task that includes stdlib specs in CI.; c0070e6
06:38:44boyscout * Excluded stdlib specs from default CI run. Added spec/full.mspec.; e4d844b
06:40:32Erlang00tnetbeans uses jruby
06:40:39Erlang00tbut I guess it can also use ruby
06:45:36brixenErlang00t: along with the new vm, evan has got the compiler working under ruby again, so in the near future you should be able to use ruby with netbeans and explore how it works
06:45:45rubuildius_amd64Brian Ford: 199ca08bc; 1280 files, 5649 examples, 20924 expectations, 0 failures, 0 errors; http://rafb.net/p/XkFyjC72.html
06:48:45rueErlang00t: If you want to step through the compiler in the debugger, you can do just that
06:49:06rueSetting breakpoints is somewhat tricky but that should not be much of an issue
06:51:37ruebrixen: Cool, that should be a good help. Put a conditional to sync?
06:52:15rubuildius_ppcBrian Ford: 199ca08bc; 1280 files, 5651 examples, 20948 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194137
06:54:54ruebrixen: You think it would be easiest to just have a task for `rake submodule:mspec:commit_push_and_sync` or whatever?
06:55:09blakewatters leaves the room.
06:58:15brixenrue: well, mspec:sync just does the update and commit. I wouldn't include a push in that
06:58:41brixenrue: the spec:sync will update, make sure ci passes, and if it does, commit
06:58:44brixenor something
06:59:22brixenI mean, it does a fetch, rebase, commit
07:00:01dfg59 enters the room.
07:05:22rueWell, that is what I mean :) Do you think it should just be left so that first one does the semi-normal commit and push in the submodule and then uses sync or should there be a task to combine the whole thing
07:05:29rue+?
07:07:39rueHm. Is there a way in LH to get to the changesets directly? I see none
07:08:00rueIf not, they might as well change the link to go to GitHub instead
07:08:07mkrauskopf enters the room.
07:11:14mkrauskopfErlang00t: NetBeans are completely interpreter agnostic. They support MRI, JRuby and recently Rubinius as well.
07:11:59brixenrue: well, the issue with making changes in rbx_root/mspec is that the submodule uses the public url (e.g. git://) and not the committer url
07:12:25mkrauskopfErlang00t: Today I'll add support for debugging with JRuby trunk and also today or tomorrow I'll play with Adam Gardiner's ruby-debug Rubinius port
07:12:25brixenrue: since changes to mspec should be quite rare, I think expecting you to handle that manually is not too unreasonable
07:13:05brixenrue: however, for spec/ruby, I was thinking it would clone with the committer url
07:13:16dfg59 leaves the room.
07:13:21mkrauskopfErlang00t: so everything which is possible from CLI wrt. debugging should be possible from IDE as well.
07:13:22brixenwhile spec/frozen would be a submodule with the public url
07:13:31evanwe shouldn't (and I don't believe you can) commit to a submodule
07:13:37evanwe shouldn't try in any way
07:13:37evanimho.
07:13:46brixenevan: yeah, I agree
07:13:55evanhaving revision lock is going to help
07:13:56brixenit's just confusing (although it is possible it appears)
07:13:56evannot hurt.
07:14:31brixenspec/frozen will be the submodule, and you shouldn't need to do anything unless you know what you're doing
07:14:39evanyep.
07:14:40evanexactly.
07:14:41brixen(i.e. rake spec:sync)
07:15:17brixenevan: what do you think of having mspec and rubyspec report commits on our LH?
07:15:31evansure.
07:15:35evanit's easy enough
07:15:37evanyou can do it
07:15:44brixenyeah, do I need special info?
07:15:45evanjust put a LH token into github
07:15:47evannope
07:15:59evanyou should be able to generate tokens for rubinius
07:16:18evanon the "My Profile" page
07:16:21evanon the right hand side
07:16:30evangenerate a token for rubinius
07:16:31brixenso, subdomain: rubinius.lighthouseapp.com
07:16:47evanyep.
07:16:50evanoh
07:16:51evanon github?
07:16:55brixenyeah
07:16:56evansubdomain is just rubinius
07:17:01brixenahh, yah
07:17:01evanproject is the number
07:17:16evan5089
07:17:17brixenoh, ok here's the token
07:17:23brixenhow do I get that?
07:17:24evanthen the token
07:17:32evanget what?
07:18:00brixenahh, got it
07:18:07brixenthe token
07:18:15evank.
07:20:22brixenevan: what do you think of removing the R: in CONTRIBUTORS? I tried to merge AUTHORS but that makes hard
07:20:28brixeni.e. just have email, name, irc
07:21:31evanplease
07:21:36Skip enters the room.
07:21:44evanit's very stale
07:21:53evanredo any way you like.
07:21:56brixenok
07:22:40rueUnless you have a separate repository for specs, you must commit in the submodule
07:23:03evanno
07:23:09evanmspec is the only submodule right now
07:23:15evanif you want to change mspec
07:23:19rueI dunno which way is more desirable but there is no way around that
07:23:20evanyou need to do it to a clone of mspec
07:23:31evannot to the submodule of mspec included into rubinius
07:23:43evanthere is no way around what?
07:23:47rueThat is fine, but it needs to be documented somewhere.
07:23:50evanthe way I understand it
07:23:53rueWhat about specs then?
07:23:55evanyou can't commit to a submodule directly
07:24:01evanrue: another ball of wax
07:24:11evanthey're being moved to a submodule too
07:24:11rueYou can. It is just another repository if you attach to it
07:24:17rueI understand all this.
07:24:28evanbut the specs are going to be include into rubinius twice
07:24:31rueHence me asking what we are going to do with it
07:24:32evanonce as a frozen submodule
07:24:40evanand once as a live, tracked version
07:24:44evanpeople can commit to the 2nd
07:25:46brixenrue: spec/frozen is a submodule of https://github.com/brixen/rubyspec/tree
07:25:52evanthat gets us the fixed version ci thats been talked about many times
07:25:54brixenrue: spec/ruby will be a clone of that repo
07:26:05rueWait, the edge spec is going to remain in the Rubinius repository?
07:26:06evanbut still lets people work on the specs within the context of rubinius easily
07:26:09evanrue: no.
07:26:12brixenrue: bin/mspec ci will run what is in spec/frozen
07:26:24evanit will be a clone to under the rubinius clone
07:27:00evanbrixen: don't forget to put the proper .gitignore's in place
07:27:04brixenrue: git submodule add https://github.com/brixen/rubyspec/tree spec/frozen
07:27:14evanso that git, run in the context of rubinius, ignores spec/ruby
07:27:33brixenrue: git clone git@github.com:brixen/rubyspec.git spec/ruby
07:27:36brixenevan: will do
07:27:42evanbrixen: be sure the submodule is the anon too
07:27:46evannot the https one
07:27:47brixenyep
07:28:03evaner. whatever the anon is.
07:28:05brixencat .gitmodules ;)
07:28:07evani might be https, i forget.
07:28:20brixenurl = git://github.com/brixen/mspec.git
07:28:26evanyep
07:28:28evancool.
07:28:45evanok, i have to go.
07:28:55brixenok
07:28:56evanbrixen: good job
07:29:04brixenevan: thanks!
07:29:51imajes enters the room.
07:30:35rueWe are manually going to track the repository. Colour me unhappy
07:31:15brixenrue: manually now?
07:31:20brixener how?
07:31:26evanrue: we've gone over and over this.
07:31:40evanif you'd like to revise how brixen is looking to do it
07:31:45evandiscussion over IRC is not going to work
07:31:52evanlets setup a phone call tomorrow to discuss
07:32:19evanok?
07:33:09rueI do not think I was there
07:33:11brixenrue: hah, you just posted to rubinius-dev that you like to manually use the git commands
07:33:28rueI do.
07:33:33brixenrue: we can tweak this, nothing is set in concrete
07:33:49evanwhat time is good for you guys?
07:33:49rueevan: Yeah, call tomorrow unless we can sort this out.
07:34:08brixenany time is fine for me
07:34:09evanlets schedule the call
07:34:13evanrue: what time is good for you?
07:34:27ruebrixen: Can you describe me the process of adding a new spec? Just the commands is fine
07:34:42brixensure
07:34:44rueevan: Any time
07:34:47evanok
07:34:50evan11am PST
07:34:55brixenk
07:34:58evan1am RTZ
07:35:00evan(rue time zone)
07:35:07evaner
07:35:09evan1PM
07:35:13evanRTZ always confuses me.
07:35:15evan:)
07:35:15rueCheck
07:35:29rueI have my own personal time zone
07:35:52brixenrue: rake spec:init => ensure spec/frozen is init'd and updated (submodule) and spec/ruby is cloned
07:36:07brixenrue: edit a spec in spec/ruby/... and bin/mspec spec/ruby/.. to test it
07:36:29brixengit commit; git push to github/brixen/rubyspec.git as normal
07:36:39brixennone of that touches spec/frozen
07:36:47brixenwhich is what bin/mspec ci uses
07:36:55brixenbecause it's sync'd to spec/tags
07:37:02evanok, talk to you guys tomorrow.
07:37:06brixenevan: ok
07:37:07evannite.
07:37:10brixennite
07:37:39brixenrue: if you want to sync spec/frozen to github/brixen/rubyspec.git, you use rake spec:sync
07:38:00brixenand ensure that the tags are sync'd so that bin/mspec ci is clean
07:38:08brixenthen you push to rbx
07:38:17brixenand everyone's spec/frozen is updated
07:38:20brixenrue: follow?
07:38:30rueSec, cat trying to eat my food
07:39:01brixenheh
07:39:43ruebrixen: OK, but that spec change is not making it to other people unless they update their spec/ruby then?
07:41:16rueAlso, if the infrastructure is there, we could just use the mechanism in all three cases rather than have two be submodules
07:44:59brixenrue: not sure I understand that first one. no change makes it to people until they fetch (or pull)
07:45:03brixennothing different here
07:45:51brixenthe submodules are useful in that they are a frozen revision automatically
07:46:06brixenif we did everything with clones, where do we store the revision?
07:46:19brixenI think this way is easiest overall
07:46:39rueIf I do `git pull`, it will not update the specs
07:46:55brixencorrect
07:47:00brixenbut you have to do rake build
07:47:11brixenthat's why build => spec:update
07:47:32brixen=> in the rake task sense
07:48:00mkrauskopf leaves the room.
07:48:15mkrauskopf enters the room.
07:48:22brixengit submodules are not ideal imo because git pull doesn't also include git submodule update
07:48:31brixenI suppose we could submit a git patch
07:51:03rueIf the frozen tree is already a submodule, then there is no need to keep track of the revision
07:51:39rueOr, more precisely, if it is already a repository in its own right
07:55:03brixenhmm, no idea what you mean there
07:55:15brixenthe point is, people will be changing rubyspecs a lot
07:55:24brixenbut spec/frozen will not change unless we change it
07:55:32brixenand it will be sync'd to our spec/tags dir
07:57:53rueI am saying that if frozen exists as a repository, we can just clone and update it like we do the edge specs. There is no need to use a submodule to designate a revision since it would be frozen's HEAD
08:01:17rueIf you are already going to have the infra in place to replicate HEAD-tracking submodule behaviour, then there is no reason to bring submodules in at all and have to support two systems
08:01:51brixenthere is no infra to track rubspec in a manner that is frozen to a particular revision
08:01:57brixenthat's what the submodule is for
08:02:18brixenrubyspec repo will be changing independent of rbx
08:02:56brixenunless we use a submodule, we have to store a particular revision of rubyspec somewhere so that folks updating rbx will only have that revision of rubyspec, which is sync'd to *our* spec/tags dir
08:03:11brixenI think this will make more sense once the tasks are in
08:03:29rueYou said spec/frozen is a submodule of rubyspec
08:03:34brixenit is
08:03:37brixenhence frozen
08:03:44brixen*because* it's a submodule
08:03:52brixenuntil we update it to a new revision
08:03:55rueYes, *in* rubyspec.
08:04:01brixenhuh?
08:05:50rueBy "submodule of rubyspec," do you mean "submodule of the rubinius repository, tracking a revision of the rubyspec repository" ?
08:08:25brixenI mean submodule in whatever sense mspec is now a submodule
08:08:40brixenspec/frozen will be exactly like mspec is now
08:09:51qwert666 enters the room.
08:11:57rueOK. I read that as frozen being a submodule inside the spec repository
08:13:11rueNeed to document this pretty carefully.
08:13:27imajes leaves the room.
08:13:45brixenyeah, much to document
08:15:31qwert666 leaves the room.
08:20:09thehcdreamer enters the room.
08:20:40GMFlash enters the room.
08:20:46kw leaves the room.
08:29:10Yurik leaves the room.
08:52:57mutle enters the room.
08:54:22trythil_ leaves the room.
08:57:06crafterm leaves the room.
09:02:56hosiawak enters the room.
09:04:41benny enters the room.
09:08:08imajes enters the room.
09:13:28naeu enters the room.
09:23:12octopod enters the room.
09:29:29riffraff enters the room.
09:31:38riffraffhi
09:33:52TheVoice leaves the room.
09:37:10Arjen_ enters the room.
09:45:59rueHello
09:53:52jzj enters the room.
09:54:53jzjquestion: im playing with the bytecode.rb, can i debug it to understand how it works?
09:55:18jzjor if there's other appoach to understand it
09:55:52jzjand also ghe generator.rb
09:55:59jzj*the
09:57:15Fullmoon leaves the room.
09:58:16gnufied enters the room.
10:03:43Maledictus enters the room.
10:05:21Form enters the room.
10:06:15FormI just read Ezra's article on mod_rubinius
10:06:44Formwill it only have a Rack interface or will it also support deeper hooks into Apache?
10:07:59Yurik enters the room.
10:08:06rueForm: Just Rack for starters. I am not certain anything more is worthwhile
10:08:28Formrue: okay
10:08:38ruejzj: Yes, you can run it in the debugger. Also, from http://rubini.us/doc you can read the compiler intro
10:08:49rueForm: Did you have a particular use case?
10:09:14Formrue: well, I have used AuthenHandlers with both mod_perl and mod_python
10:09:39Formand would love to be able to write those in Ruby
10:14:18jzj leaves the room.
10:15:02rueForm: Mm. It is certainly possible to do
10:15:16jzj enters the room.
10:15:36rueForm: Rack has some type of auth middleware
10:15:48ruejzj: --> Yes, you can run it in the debugger. Also, from http://rubini.us/doc you can read the compiler intro
10:16:06jazen leaves the room.
10:16:40Formrue: cool, I didn't know that
10:17:02Formrue: still, in my mod_perl days I used more than just Authen
10:18:03brainopia enters the room.
10:18:59jzjrue: how to run the debugger?
10:19:25jzji read the compiler intro before
10:21:03rue-debug
10:22:20rueForm: Sure, there are definitely cases where it would be nice to be able to use Ruby. Personally I hope Apache dies a quick death though :/
10:22:51Formrue: unlikely.
10:22:53Formwhat would you prefer?
10:23:05rueAt this point, anything :)
10:23:16Formhehe
10:23:32Skip leaves the room.
10:23:40rueIt is really horrible internally. Nginx is sort of like a sane Apache, but I would honestly go with Ebb or something
10:27:06Skip enters the room.
10:31:15rubuildius_ppc leaves the room.
10:31:20rubuildius_ppc enters the room.
10:47:53Erlang00thow does rubinius subcommand work?
10:48:13Erlang00tor I mean, where the related code does rubinus handle subcommand?
10:48:22rueIn loader.rb
10:48:59rueIf it looks like a subcommand, loader checks lib/bin/ for a file with the same name
10:49:03jzj leaves the room.
10:51:04jzj enters the room.
10:51:48Yurik leaves the room.
10:55:50xhanjian_ leaves the room.
10:56:38xhanjian enters the room.
10:58:51Yurik enters the room.
11:00:14xhanjian leaves the room.
11:00:35jtoy leaves the room.
11:01:39xhanjian enters the room.
11:07:18xhanjian leaves the room.
11:07:34rubuildius_ppcBrian Ford: 199ca08bc; 1280 files, 5651 examples, 20948 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194208
11:09:07xhanjian enters the room.
11:11:05Blinchik_ enters the room.
11:13:41Blinchik leaves the room.
11:20:45yugui_ enters the room.
11:21:07yugui leaves the room.
11:34:32jzj leaves the room.
11:35:39Formrue: when you get to the point where it makes sense to start thinking about "internal" Apache hooks, I could help you out
11:35:53FormI don't have time to work on more projects right now
11:36:09Formbut that's definitely a feature I want, so I should help with the implementation
11:36:13Formopen source at it's best
11:46:36benny leaves the room.
11:47:56benny enters the room.
11:55:10Erlang00t leaves the room.
11:57:32rueForm: Sure, have to see what can be done. The model is to basically just be a shim, so it may not fit in with few enough compromises.
11:58:14rueForm: Mind you, mod_ruby should be fine for administrative stuff even if it pretty much sucks for actual application serving
12:06:35obvio enters the room.
12:15:19GMFlash leaves the room.
12:16:23jacen_ enters the room.
12:18:35qwert666 enters the room.
12:21:47chris2 enters the room.
12:27:41imajes leaves the room.
12:28:39imajes enters the room.
12:29:31Yurik leaves the room.
12:59:49Fullmoon enters the room.
13:00:35benny leaves the room.
13:00:51benny enters the room.
13:18:23gnufied leaves the room.
13:23:34gnufied enters the room.
13:31:07blakewatters enters the room.
13:33:05Skip leaves the room.
13:34:07Ski1 enters the room.
13:43:56mutle leaves the room.
13:44:37mutle enters the room.
13:47:39blakewatters leaves the room.
13:54:52_mutle enters the room.
13:55:47mutle leaves the room.
14:07:05AndrewO enters the room.
14:13:14brainopia leaves the room.
14:17:17webmat enters the room.
14:26:58stepheneb enters the room.
14:27:20moofbong enters the room.
14:29:08jazen enters the room.
14:29:39jacen_ leaves the room.
14:32:38rff enters the room.
14:35:16riffraff leaves the room.
14:36:41jpease leaves the room.
14:36:45foysavas leaves the room.
14:37:32ShayArnett enters the room.
14:37:56jpease enters the room.
14:46:00benny leaves the room.
14:46:36trythil enters the room.
14:47:27trythil leaves the room.
14:51:50mutle leaves the room.
14:52:40mutle enters the room.
14:55:26GMFlash enters the room.
14:58:14foysavas enters the room.
14:59:42dgtizedevan: ping
15:01:59rueProbably out for an hour or two still
15:02:45rueVm still broken?
15:02:49dgtizedyea
15:03:20dgtizedhe keeps checking things in in huge batches so I just presume it's something that was forgotten
15:04:04dgtizedright now the problem is it's unhappy linking with the regex library somehow
15:09:41gdagley enters the room.
15:11:17rueWhat is the repro?
15:18:11blakewatters enters the room.
15:18:14mutle leaves the room.
15:20:40dgtizedrepro?
15:20:41rff leaves the room.
15:24:29dgtizedit can't find ffi.h or onigurama.h
15:27:33rueWhat is the command line?
15:27:43mutle enters the room.
15:27:52stepheneb leaves the room.
15:33:46benburkert enters the room.
15:36:47dgtizedrake?
15:37:08dgtizedafter doing a ruby instructions.rb in vm
15:37:17dgtizedsince it's still not in the makefile
15:37:35dgtizedoh and you might need to do a top level rake just to make sure the external libraries are built
15:38:32yugui_ leaves the room.
15:40:22riffraff enters the room.
15:41:57rueHm, might need a distclean, getting duplicate symbol errors
15:46:38dgtizedI don't know if distclean works
15:49:38rff enters the room.
15:55:05benburkert leaves the room.
15:57:52riffraff leaves the room.
16:04:17benstiglitz enters the room.
16:14:22trythil enters the room.
16:16:36macournoyer enters the room.
16:19:51lopex enters the room.
16:20:29rueWth with this build
16:22:27stepheneb enters the room.
16:22:42DefilerOK, I have a retarded question
16:23:27DefilerThe 'empty' SHA1 hash value is always the same, right?
16:23:33DefilerIt's not implementation specific, presumably
16:24:32jpease leaves the room.
16:31:38rueFor an empty string?
16:31:48rueNo special rules
16:32:13octopod_ enters the room.
16:32:39mapar enters the room.
16:35:12benburkert enters the room.
16:42:06naeu leaves the room.
16:42:52hosiawak leaves the room.
16:45:00thehcdreamer_ enters the room.
16:45:05brainopia enters the room.
16:46:19benny enters the room.
16:47:15octopod leaves the room.
16:48:19shame leaves the room.
16:51:48boyscout1 commit by Vladimir Sizikov
16:51:49boyscout * New rubyspecs for BigDecimal's #floor and #ceil.; aba022a
16:51:51srbaker leaves the room.
16:53:39thehcdreamer leaves the room.
16:56:14therealadam enters the room.
16:56:26dlee enters the room.
17:00:47rubuildius_amd64Vladimir Sizikov: aba022a66; 1280 files, 5649 examples, 20924 expectations, 0 failures, 0 errors; http://rafb.net/p/WHG9MT39.html
17:04:58rubuildius_ppcVladimir Sizikov: aba022a66; 1280 files, 5651 examples, 20948 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194366
17:06:33brainopia leaves the room.
17:10:13yipstar enters the room.
17:10:51srbaker enters the room.
17:12:57joachimmwhat do I need to do to generate typechecks.gen.cpp?
17:19:01dgtizedthe essential problems with the new vm is that the paths are relative in the rake file and I think that's messing some things up
17:21:19evandgtized: you should be running just 'rake' in test.
17:21:20imajes leaves the room.
17:21:22evanthen everything works.
17:21:31evandoing something else is outside the scope of the current work.
17:21:42evanjoachimm: it should be autogenerate by running field_extract
17:21:50evanjoachimm: run 'rake field_extract' in test/
17:22:15evandgtized: which paths are you talking about?
17:22:30joachimmevan: thanks, I was reading the rakefile just now.
17:23:56gnufied leaves the room.
17:26:11lchin enters the room.
17:27:12mutle leaves the room.
17:30:21DefilerIf I want to return a subclass instance based on the args to 'initialize', I need to implement my own SomeClass.new, right?
17:30:59DefilerOpenSSL::Digest::Digest.new("SHA1") needs to return an OpenSSL::Digest::SHA1 instance, to be specific
17:32:41srbaker leaves the room.
17:33:08shame enters the room.
17:33:46evanyeah
17:33:54evanyou have to write your own new
17:33:59evanor your own allocate
17:34:12evanyou need .new for this though
17:34:23evansince you need to look at the args to figure out what subclass you need.
17:44:17srbaker enters the room.
17:48:51chris2_ enters the room.
17:50:57chris2 leaves the room.
17:51:03benny leaves the room.
17:53:39rueevan: It does not quite work automatically, something is not being generated
17:53:48evanhm.
17:53:57evanrue: you mean the new vM?
17:54:52rueYeah. External libs were causing a problem earlier, futzing with it now
17:55:11evanyeah
17:55:19evanexternal libs aren't built automatically now
17:55:31evanit's just using them via them being built from shotgun
17:55:36evanthat logic needs to be moved over.
17:56:25rueYeah. Could start pulling stuff to the main Rakefile?
17:56:53mapar leaves the room.
17:57:01rubuildius_ppc leaves the room.
17:57:04rubuildius_ppc enters the room.
17:57:42thehcdreamer_ leaves the room.
17:57:47srbaker leaves the room.
18:03:45evanhey!
18:03:55evanslaps ChanServ on the wrist
18:03:55DefilerYeah, freenode is like that
18:04:17evanoh, mode lock
18:04:28srbaker enters the room.
18:05:06elight enters the room.
18:05:08DefilerI think you can tell chanserv to change it and it will stick, though
18:05:14evanyep
18:05:16evandoing it now
18:05:36evanbrixen: set the channel topic
18:05:42brixenevan: k
18:05:50brixenwill do when this is more complete
18:05:55evank
18:06:27evanhah
18:07:12mapar enters the room.
18:07:20imajes enters the room.
18:07:40srbaker leaves the room.
18:08:15obiejuan enters the room.
18:08:59joachimmmake coverage fails with duplicate symbol _main, is there something I need to clean up before?
18:09:48brixendjwhitt: ping (or can you see about build bot running rake spec:full or bin/mspec ci -B full.mspec)
18:09:54evanjoachimm: oh ack.
18:09:55evanyeah
18:09:57evancrap.
18:10:14evanvm/main.cpp needs to be excluded from being built in coverage
18:10:42evanjoachimm: i'll fix that shortly
18:10:52evanfor now, just mv vm/main.cpp vm/main.cpp.blah
18:10:55evanthat should fix i.t
18:12:06rubuildius_ppcVladimir Sizikov: aba022a66; 1280 files, 5651 examples, 20948 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194405
18:12:55kw enters the room.
18:13:33joachimmevan is there some particular reason for using class instead of typename in template declarations?
18:13:44evanno
18:13:46evanthey're the same.
18:14:27yipstar leaves the room.
18:15:22wycats_ leaves the room.
18:16:27Arjen_ leaves the room.
18:17:34wycats_ enters the room.
18:18:49technomancy enters the room.
18:20:38ruejoachimm: The convention is to use "class" when it must be a user-defined type and "typename" when it need not. Personally, I just use "typename" everywhere
18:22:59technomancywhat's the convention in specs for putting constants in their own files?
18:23:13technomancydo we want to keep the spec files themselves clear of noise?
18:23:29technomancyis looking at digest/sha1/digest_spec.rb
18:23:31evantechnomancy: what kind of constant?
18:23:51evanah
18:23:56evanyeah, typically
18:24:22technomancyevan: the SHA1Constants module contains a Contents that's just lorem stuff and then values for blank digests, hex digests, etc.
18:24:37evanI think thats ok
18:24:37technomancyit looks like there's something behind the Contents value though; what's the deal with that?
18:24:56evandrys things up
18:24:57evani think
18:25:44imajes leaves the room.
18:26:49technomancyoh, i see where it's being used; never mind
18:35:59mapar leaves the room.
18:40:38enebo enters the room.
18:41:09rueWell, this GCC compile will probably take forever
18:42:30ruetechnomancy: Shared stuff should go in fixtures, usually. I personally prefer to keep any of that close to the actual spec so long as it is not overwhelming
18:43:02technomancyrue: "close to" being in the file or just in the same dir?
18:43:53rueIn the same file. If you are moving it out of the file, fixture dir would be best
18:44:09rueOr if it already exists outside the file, naturally
18:44:15headius enters the room.
18:46:34mkrauskopf_ enters the room.
18:47:25mkrauskopf leaves the room.
18:50:40brixenrue: we should open a discussion of fixtures on the rubyspec list. the challenge is 1. clashes, 2. duplication, 3. obfuscation when a set of specs should depend on a single bigger structure
18:51:00brixenI'm all for locality, but we've never been able to adequately address those 3 points
18:51:00rueSure
18:51:07brixenhence, the fixtures dir
18:51:50evanthe mailling list is a good place to have that discussion
18:54:21yaroslav enters the room.
18:59:21rubuildius_amd64 leaves the room.
18:59:57octopod_ leaves the room.
19:01:37rubuildius_amd64 enters the room.
19:02:19djwhittbrixen: just updated the bot. worked fine in test. you should see a repeat run here in just a sec
19:02:36brixendjwhitt: awesome, thanks!
19:02:49djwhittbrixen: yep, no prob
19:03:09djwhittbrixen: assuming everything looks ok I'll ping cremes and he can update his as well
19:03:22brixencool, that'd be great
19:06:53wycats_ leaves the room.
19:07:33wycats_ enters the room.
19:12:54rubuildius_amd64Vladimir Sizikov: aba022a66; 2105 files, 6854 examples, 24744 expectations, 0 failures, 0 errors;
19:13:28brixensweet, all on one line
19:20:23boyscout1 commit by Vladimir Sizikov
19:20:24boyscout * Some more rubyspecs for BigDecimal's #floor and #ceil.; 686c284
19:21:23therealadam leaves the room.
19:23:21blakewatters leaves the room.
19:23:45wycats_ leaves the room.
19:25:13technomancyI get "no such file or directory - mspec/bin/mspec" when I try to run bin/mspec... what's changed with this?
19:25:36VVSiz:)
19:26:30VVSiztechnomancy: brixen can tell
19:26:45blakewatters enters the room.
19:28:35atmos enters the room.
19:29:21wycats_ enters the room.
19:29:58mkrauskopf_ leaves the room.
19:30:05headiushey, you guys might be interested in this: http://blogs.sun.com/bangle/entry/maxine_vm_to_be_announced
19:30:08technomancydid it get spun out into a gem?
19:30:20headiusit's a self-hosting metacircular VM written in Java with a native compiler
19:30:32headiuscool stuff from Sun Labs
19:32:58rubuildius_amd64Vladimir Sizikov: 686c28493; 2105 files, 6854 examples, 24764 expectations, 0 failures, 0 errors;
19:33:12VVSiztechnomancy: http://groups.google.com/group/rubinius-dev/browse_thread/thread/da19a433520afad5
19:33:24technomancythanks
19:33:32rubuildius_ppcVladimir Sizikov: 686c28493; 1280 files, 5651 examples, 20948 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/194461
19:34:23mapar enters the room.
19:36:11mkrauskopf enters the room.
19:37:49rueheadius: Nah, that would never work. Java is so slow
19:38:00VVSizheheheh
19:38:16VVSizrue: you've been here for the last 24 hours! never sleep? :)
19:38:17evanheadius: yeah, i saw that
19:38:19evanheadius: looks cool.
19:38:29rueWait, GPLv2? Is that actually possible?
19:38:36rueVVSiz: Tuesdays
19:39:16technomancyif I've got a bunch of constants in a module for a spec, can I include the Module somewhere in the spec so I don't have to dereference it all the time?
19:40:03brixentechnomancy: follow the examples in the specs that require 'fixtures/classes'
19:40:11brixentechnomancy: what spec are you working on?
19:41:06technomancybrixen: Defiler threw some OpenSSL HMAC specs my way
19:41:11technomancysaid they're among the last merb blockers
19:41:22ezmobius enters the room.
19:41:59yaroslav leaves the room.
19:42:30brixentechnomancy: ok, what's dir are you in? spec/ruby/1.8/library/openssl ?
19:42:34technomancyyeah
19:42:43brixendid you use mkspec to create the dirs?
19:43:10technomancyglances around nervously.
19:43:11technomancyno
19:43:15brixentechnomancy: http://rubinius.lighthouseapp.com/projects/5089/specs-creating-files-with-mkspec
19:43:19technomancyaha; thanks
19:43:26technomancyman, you guys don't sit still long. =)
19:43:57benny enters the r