Index

Show enters and exits. Hide enters and exits.

00:00:42cremes leaves the room.
00:00:42wmoxam leaves the room.
00:06:01evanbrixen: no, working on EBADF still
00:06:17imajes enters the room.
00:06:28brixenevan: ok
00:06:45krisps leaves the room.
00:06:57evanit's related to TCPSocket getting killed while running accept
00:07:06brixendoing a last clean build ci run and then I'll push the MT change
00:07:09krisps enters the room.
00:07:09evansome data isn't being cleaned up when the Thread gets terminated
00:07:12evanbrixen: yay!
00:07:12brixenahh
00:07:30brixenstill a ways to go to remove e.g. const stuff from using Hash
00:08:29dbussinkbrixen: well, other people can help with that then too
00:09:04wifelette enters the room.
00:09:21brixendbussink: as long as we're not all standing on one another's toes ;)
00:09:31brixenkind of a small space in terms of loc
00:09:38ttmrichter enters the room.
00:09:39ttmrichter leaves the room.
00:10:03dbussinkhehe, that's true
00:10:42brixenmust say, git absolutely rocks
00:11:53brixenhere it comes :)
00:12:37wifelette_ enters the room.
00:13:01boyscout4 commits by Brian Ford
00:13:02boyscout * MethodTable now is a kind of LookupTable.; 9542a82
00:13:03boyscout * Give LookupTable back to stables.; 297c3fd
00:13:04boyscout * Fixed LookupTable#delete. Added LookupTable#entries, #dup.; eb937c8
00:13:05boyscout * Fix #inspect, #to_s for LookupTable to use instance.class for name.; 5bab28e
00:13:24wifelette leaves the room.
00:13:29benburkert leaves the room.
00:22:30benburkert enters the room.
00:23:32headius_ enters the room.
00:23:33headius leaves the room.
00:24:54rubuildius_ppcBrian Ford: 9542a821b; 1741 files, 5873 examples, 19940 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163102
00:26:40perdiy leaves the room.
00:28:52wifelette_ leaves the room.
00:33:05wifelette enters the room.
00:44:04wmoxam enters the room.
00:47:25cfoo enters the room.
00:49:27xmlhacker_ leaves the room.
00:52:03xmlhacker enters the room.
00:55:20crossblaim leaves the room.
01:00:51crafterm enters the room.
01:01:43headiusso the sprint is going right now, yes?
01:02:09craftermheadius: yes mate! its all on right now
01:02:13headiushuzzah!
01:02:28headiuswho's working on what?
01:02:51craftermthere's a bunch of people working on new specs and fixing specs at the moment
01:03:18headiusahh neato...write more IO specs
01:03:22headiusIO/Socket
01:04:09craftermok :)
01:05:16brixenperfect, rain here just in time for hiking this weekend
01:05:23boyscout1 commit by Brian Ford
01:05:24boyscout * Tweaks to LookupTable. Converted Errno::Mapping to use LT.; 407e1a4
01:07:13craftermtrying to upload some photos to flickr now but net connection is a bit flakey...
01:10:04chop3 enters the room.
01:12:20benburkert leaves the room.
01:13:11AndrewO enters the room.
01:14:55benburkert enters the room.
01:16:58chop3 leaves the room.
01:19:33radarek leaves the room.
01:19:49imajes leaves the room.
01:20:13aotearoa enters the room.
01:21:11chop3 enters the room.
01:21:25rubuildius_ppcBrian Ford: 407e1a419; 1741 files, 5866 examples, 19933 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163128
01:24:48headius leaves the room.
01:36:21benburkert leaves the room.
01:39:38djwhittlooks like ci is about 1 sec faster for me with the LookupTable changes
01:40:11brixenheh, I regularly see a 3-6 sec variance
01:40:22djwhittthat was over multiple runs
01:40:32headius enters the room.
01:40:34geekounet leaves the room.
01:40:44djwhittit's actually pretty consisten on the system I'm running it on too
01:40:45brixenthat's cool, what's the 1 sec relative to?
01:40:54djwhitt44 vs 43
01:41:04brixenahh ok
01:41:10djwhittso that's not huge, but still
01:41:12djwhittnot bad
01:41:24brixenyeah
01:41:35brixenworking on replacing constants with LT now
01:41:56headius leaves the room.
01:41:58roo enters the room.
01:42:07brixenwb roo
01:42:10headius enters the room.
01:43:24rue leaves the room.
01:43:40rueBetter
01:45:45macournoyer enters the room.
01:49:08cfoo leaves the room.
01:56:10warren_s enters the room.
01:59:35benburkert enters the room.
01:59:37quik_ enters the room.
02:00:13krisps leaves the room.
02:01:12benburkert leaves the room.
02:01:40benburkert enters the room.
02:06:43headiushmmm
02:08:38vintrepid leaves the room.
02:09:25benburkert leaves the room.
02:09:53KirinDave leaves the room.
02:11:08krisps enters the room.
02:11:08macournoyer leaves the room.
02:11:37macournoyer enters the room.
02:23:20quikcrafterm: hello?
02:23:22brixenhm, consts are resisting my mallet, might have to use a ball-peen hammer on them
02:23:31boyscout1 commit by Brian Ford
02:23:32boyscout * Added lookuptable_find that returns Qundef.; 4b7a935
02:23:46brixenbut first, dinner with the gf, bbl...
02:30:41aotearoa_ enters the room.
02:35:49rubuildius_ppcBrian Ford: 4b7a93575; 1741 files, 5866 examples, 19933 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163150
02:42:17aotearoa__ enters the room.
02:45:49VVSiz_ enters the room.
02:48:48wmoxam leaves the room.
02:50:36macournoyer leaves the room.
02:55:11aotearoa leaves the room.
03:00:26krisps leaves the room.
03:04:06VVSiz leaves the room.
03:06:49antares leaves the room.
03:08:39aotearoa_ leaves the room.
03:10:08cremes enters the room.
03:10:37cremesstrange crash while running bin/mspec ci from the command line on the box that runs rubuildius_ppc
03:10:42cremessee here: http://pastie.caboo.se/163154
03:11:06cremesthis only occurs when running from the cli; runs from a subprocess (or cron) work okay
03:11:18tmornini enters the room.
03:11:35cremesi can't tell which test is causing the problem; runs with 'bin/mspec ci -f s' do not err out
03:16:59cremeshitting the sack for the night... i can help troubleshoot this in about 10 hours (yes, i'm *very* sleepy)
03:21:25jrun leaves the room.
03:24:29craftermquik_: hi mate!
03:26:08quikcrafterm: Sorry I'm unable to get up there!
03:26:21craftermquik_: no worries mate :)
03:26:39quikAfter being robbed and trying to replace things, I'm horribly broke
03:26:51quikthat, and I couldn't get a laptop sorted quick enough
03:27:37craftermquik_: thats ok - we'll run another one in melbourne :)
03:27:49quikI hope so
03:28:01quikI really wanted to come, I feel pretty torn this weekend
03:30:15jinjing leaves the room.
03:33:43GMFlash leaves the room.
03:34:15craftermall cool mate, no worries
03:37:32boyscout3 commits by Eric Hodel
03:37:33boyscout * Add Zlib::GzipReader #eof?, #pos, #read w/length; 8551da4
03:37:34boyscout * Ensure #pos clears internal eof flag; a4dba83
03:37:35boyscout * rake spec should build; 03d0af8
03:39:17craftermsprint pics: http://www.flickr.com/photos/crafterm/2317816680/
03:39:30crafterm... and http://www.flickr.com/photos/crafterm/2317006693/
03:40:17headiusgetting a lot done?
03:42:12crafterma few new tickets are in lighthouse atm
03:42:29craftermlots of discussion and people getting into things which is great
03:45:37headiushow long is the sprint supposed to go? through tomorrow?
03:45:42headiusI'm looking forward to a flood of new specs
03:47:54craftermwill last till end of tomorrow
03:49:48rubuildius_ppcEric Hodel: 8551da47a; 1741 files, 5871 examples, 19942 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163167
03:49:54smtlaissezfaire enters the room.
03:51:30smtlaissezfairewhat is the top level constant in MRI for the patchlevel version?
03:51:50smtlaissezfaireVERSION gives "1.8.6" - what top level constant will give the patchlevel?
03:53:20tarcieri<pedant> I believe VERSION is deprecated in favor of RUBY_VERSION </pedant>
03:53:46aotearoa enters the room.
03:59:12boyscout2 commits by Evan Phoenix
03:59:13boyscout * Fix looking for constants in modules included in Object; d21b475
03:59:14boyscout * Save and restore errno on task changes; 5696b2b
03:59:31evanok, EBADF is fixed.
03:59:55smtlaissezfaire leaves the room.
04:00:07ubiquitous enters the room.
04:02:51boyscout1 commit by Eric Hodel
04:02:52boyscout * Fix Zlib::GzipWriter#finish. Pair: Lincoln.; 1e039fb
04:05:09hassoxcrafterm: those photo's make it look like you guys are enjoying yourselves ;)
04:10:38AndrewO leaves the room.
04:10:56pd enters the room.
04:13:29tarciericy: heh
04:13:37tarcierierr, misfire :/
04:16:06quik_ leaves the room.
04:16:35chop3 leaves the room.
04:16:44rubuildius_ppcEric Hodel: 1e039fb5c; 1741 files, 5872 examples, 19943 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163171
04:16:45rubuildius_ppcEvan Phoenix: d21b475f7; 1741 files, 5871 examples, 19942 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163170
04:18:06aotearoa__ leaves the room.
04:33:25craftermhassox: yes mate, lots of fun :)
04:36:58boyscout1 commit by Eric Hodel
04:36:59boyscout * Fix IO#pos EOF spec. Pair: Lincoln, Evan.; 41b26c4
04:49:40rubuildius_ppcEric Hodel: 41b26c49f; 1741 files, 5872 examples, 19943 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163181
04:55:58mutle enters the room.
05:12:20notahat enters the room.
05:12:28evanbrixen: you around?
05:14:22brixenyep
05:14:37brixenlayered with kitten, but I can kinda type :P
05:14:55evanhehe
05:14:55evanso
05:15:05evanbin/mspec seems to be running ci whether I want it to or not
05:15:20brixenwhat's the command?
05:15:33evanbin/mspec run spec/ruby/1.8/core/io
05:15:37evanonly gives me
05:15:44evan59 files, 137 examples, 208 expectations, 0 failures, 0 errors
05:16:39brixenwhat about just: bin/mspec spec/ruby/1.8/core/io ?
05:17:05evanyep
05:17:08brixenI get the same for either of those actually
05:17:21brixenlet me try rspec
05:17:22evani'm doing
05:17:22evanbin/mspec spec/ruby/1.8/core/io/ungetc_spec.rb
05:17:24evanand getting
05:17:44evan1 file, 1 example, 1 expectation, 0 failures, 0 errors
05:18:26brixenany local changes? I'm 1/9/8/0/9
05:18:44brixenlet me git update
05:18:58evanI added a new mspec formatter
05:19:01evanlet me retry
05:20:32evanwtf.
05:20:33evanok
05:20:34evanone sec
05:20:37evanseems to be my changes.
05:20:48brixenpastie? maybe I can spot it
05:21:58evani got it
05:22:04evani made a change to tally that seems to blow something up
05:22:13brixenok
05:22:13evanlet me rework it
05:22:16brixensure
05:23:04robinhoode enters the room.
05:28:02evanok, fixed it
05:28:09brixencool
05:28:18krisps enters the room.
05:30:10evanpushing the new formatter now
05:30:10evanit was our fun fix
05:30:10evanadded a spinner formatter
05:30:29lstoll enters the room.
05:31:15brixennice
05:31:30brixenyeah, we're getting a lot of dots :)
05:35:52boyscout1 commit by Eric Hodel
05:35:53boyscout * Fix RDoc for 1.8, no String#ord; dd1bd4c
05:35:58ttmrichter enters the room.
05:41:01boyscout1 commit by Evan Phoenix
05:41:02boyscout * Add spinner output formatter; 8974f7b
05:41:23d2dchat enters the room.
05:41:30evanyay spinner!
05:42:42lstollthe spinner is FTW. and the a is for awesome.
05:46:52brixenhah, takes me back to the early days of boot linux
05:47:04brixenbooting
05:48:46jrun enters the room.
05:56:04tizianobis_ enters the room.
05:56:20rubuildius_ppcEvan Phoenix: 8974f7bae; 1741 files, 5872 examples, 19943 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163195
05:56:22rubuildius_ppcEric Hodel: dd1bd4c48; 1741 files, 5872 examples, 19943 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163194
05:57:00headiusyou guys do work on some weird stuff when you sprint
05:57:38evanit's important to have fun stuff
05:57:49headiusI tend to find it all fun myself
05:57:56headiusbut that's just me
05:58:08evanwell, silly
05:58:14headiuswhat's not fun about compiler optz and such
05:58:16evanstuff
05:58:17evanlike a spinner mspec output formatter
05:58:24evanit's not a group activity
05:58:27evanthere are 22 people here
05:58:55_matta_ enters the room.
06:03:39headiusyeah, it's good to see the commits flowing
06:03:55headiusVVSiz or I will probably poke around any new specs this weekend
06:04:06headiusI hope someone's looking at IO specs
06:12:41RyanTM leaves the room.
06:13:39ttmrichter leaves the room.
06:15:14TheVoiceI hear you guys are causing some trouble tonight, good stuff.
06:16:17ezmobius leaves the room.
06:18:03ezmobius enters the room.
06:18:07TheVoice leaves the room.
06:19:30TheVoice enters the room.
06:19:41krisps leaves the room.
06:19:53TheVoice leaves the room.
06:20:12tizianobis leaves the room.
06:21:11TheVoice enters the room.
06:21:36TheVoice leaves the room.
06:22:09ttmrichter enters the room.
06:22:11ttmrichter leaves the room.
06:23:32rubymaverick enters the room.
06:25:29brixenwow, I switched external_ivars from Hash to LT in one shot I think
06:28:37wifelette leaves the room.
06:28:51wycats leaves the room.
06:31:58ezmobius leaves the room.
06:31:59_matta_ leaves the room.
06:32:03crafterm leaves the room.
06:41:55_matta_ enters the room.
06:41:59crafterm enters the room.
06:43:44masswaves hello
06:43:51boyscout1 commit by Brian Ford
06:43:51brixenhey mass!
06:43:52boyscout * Converted external instance vars to use LookupTable.; fb05dae
06:44:12massthis openid code is kicking my butt
06:44:37brixenwhat code is that?
06:44:49massjust trying to clean up some flow
06:44:53srbaker leaves the room.
06:44:57massyou wouldn't like it, its struts 2 :)
06:45:14brixenheh, you're always programming crazy stuff :)
06:45:32massto summarize, this app at work has some really badly designed parts
06:45:37massand I volunteered to fix it
06:45:44brixenheh, hmm, this sounds familiar
06:45:46massor rather, it finally bothered me enough that I'm fixing it on my own
06:46:02massand it wound up being harder than I suspected :)
06:46:21brixenhey, did you read the steps paper the vpri folks posted?
06:46:35massand now the one part I didn't want to touch is not behaving because it was a lil fragile
06:46:40masssteps paper?
06:46:42brixensome very cool stuff in there
06:46:44brixensec..
06:46:46massnope, I did read about shoes though
06:47:01masswhy is sorta crazy
06:47:02brixenheh, almost bought the shoes book at lulu the other day
06:47:24brixenhttp://vpri.org/pdf/steps_TR-2007-008.pdf
06:47:36massI want to ask him if he assembles lots of model airplanes or something :)
06:48:18brixenI tried to assemble a model once, used super glue, took a week to unstick my fingers :P
06:48:30massI've done some rather cool things cleaning this code though, learning-wise
06:48:45massbut once I do this I've touched every function :P
06:49:02massis this a 12 step program?
06:49:08brixenclose
06:49:24brixenstep 1: admit that your C++/java code base is unmanageable :P
06:49:47brixenstep 4: make a personal code inventory and make amends to the generations of coders you've damaged with said code
06:49:47masspfft, I don't have a problem, I can quit C++/java anytime I like
06:49:52brixenhah!
06:50:09massI just.. need to finish debugging this smart pointer :P
06:50:16mass:( oh god, I have a problem!
06:50:22brixentcp/ip in < 200 loc gives hope to the human race
06:50:39massanyone can do tcp/ip in < 200 loc
06:50:44masswith wide-enough paper
06:50:50brixenheh
06:51:16brixenC has those arbitrary 80 char limits to a line :P
06:51:40massreally?
06:51:57brixenyeah, that's why it takes 10-20k loc
06:52:01massC++ is way better than C
06:52:12massC is for people who haven't truly embraced evil
06:52:24brixenthis looks pretty cool too: http://www.nabble.com/ANN:-New-Objective-C-Runtime-td13648692.html
06:52:58massever since I started leaving those C/C++ User Journals on my front door the Jehovah's Witnesses and Mormons have stopped coming by
06:53:12brixenhah, that's hilarious
06:53:49brixenI should try that, I've been getting the little tracks folded up and tucked by my door knob
06:53:51massprobably because they realize they don't want to get in discussions with me
06:54:10mass"have you accepted jesus christ as your personal savior?"
06:54:20brixensatan
06:54:27mass"have you accepted that associating logic with data is a better way?"
06:54:33jrun leaves the room.
06:54:35eugenebrechko enters the room.
06:55:00rubuildius_ppcBrian Ford: fb05daeda; 1741 files, 5872 examples, 19943 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163199
06:55:01mass"have you let object oriented design into your heart?"
06:55:14mass"have you ever asked yourself, 'What Would Stroustroup Do?' "
06:56:05massyou've been getting satanic literature? thats a sign you may be in the wrong neighborhood
06:56:40brixenit's all the same to me :P
06:57:01massobjc is a neat language
06:57:05masssorta insane, but neat :)
06:57:29massit seems like an interesting search term when it comes to research papers though
06:57:46jero5 leaves the room.
06:58:48massC and Smalltalk's love child
06:58:50wifelette enters the room.
06:59:44massman I don't want to work on this openid code anymore
06:59:52massbut I can't move onto my own projects until I finish it :-/
07:00:13evan_s leaves the room.
07:00:31brixenI could try to find you a semi-difficult 2-week project in rbx :)
07:00:49brixenlast time you had painful code, we got 2-3 new libraries
07:03:46wycats enters the room.
07:04:20digitalfallout enters the room.
07:04:45masshehe, each about one file :)
07:06:02brixenthis was interesting too, haven't tried it out yet: http://tanjent.livejournal.com/756623.html
07:08:22massneat
07:09:30massI have a new , highly efficient hashing algorithm
07:09:41masstake the first four characters of a string, and treat them like a 32 bit integer
07:09:44mass:)
07:10:27brixennow why didn't I think of that?!
07:10:53MenTaLguYmaybe because it works badly for URLs
07:11:06massit also has the benefit of not requiring you to cache the hashed value, as long as you pad with zeroes
07:11:09mass:P
07:11:11masshttp://www.google.com
07:11:18massftp://ftp.idsoftware.com
07:11:25massmailto:david@wherever.com
07:11:28massall looks fine to me
07:13:36massif you have a problem with collisions, you obviously don't have enough buckets
07:16:18brixenheh, mass we gotta get you away from that code, noxious fumes obviously
07:17:00massok, how about the LAST four characters then?
07:17:01d2dchat leaves the room.
07:17:03mass:D
07:17:13context leaves the room.
07:19:25brixendunno why these constants are giving me such a hard time
07:20:09brixenheh, well because it's trying to do a hash operation on a lookuptable
07:20:17brixencheers for machine_handle_fire
07:22:55massI dislike password complexity rules, they are almost always stupid
07:23:24massgoogle wouldn't let me reset my password to J4n1#~x because it was too short
07:23:35massbut it was fine with "Password1"
07:23:39brixenheh
07:24:18quik_ enters the room.
07:25:22lachie enters the room.
07:25:56MenTaLguYrainbow tables are an issue these days
07:26:15MenTaLguYthough the minimum password lengths really need to be longer than just 9 chars
07:26:21MenTaLguYto deal with that
07:26:54brixenwe might as well just dispense with passwords now, once quantum computing hits mainstream it won't matter anyway
07:27:11brixengot back to an X on the dotted line
07:27:17brixenor go
07:27:28notahat leaves the room.
07:28:39context enters the room.
07:31:55massrainbow tables are only an issue if they get your hashed passwords though, no?
07:35:00MenTaLguY yes
07:35:17rueOh, yeah, quantum computing
07:35:27rueLogging in may cause your machine to have been hacked
07:35:36rueWhat are you geeks up to today?
07:35:48masskilling cats by opening the box
07:35:49massyou?
07:36:30massmy theory on quantum computers is kinda like my theory on time travel
07:36:43massi.e. "if time travel was possible, don't you think we'd know about it by now?"
07:37:29MenTaLguYquantum computers are less likely to drop by from the future though
07:38:07rueOr maybe they are?
07:38:13rueWe do not know until we see one
07:38:25brixenwell, quantum computing is not sci-fi, it's quite real
07:38:32brixenwhen it will be practical is another matter
07:38:33rueAt which point it may or may not have had wont been future sent
07:39:04quikcrafterm: hows it going?
07:39:07brixenI wish shotgun could be in a quantum superposition of states where LookupTable and Hash overlapped
07:39:31brixenbecause I think I'm going to have to code it to handle constants in either a LT or hash
07:39:58brixensince stables have to be in sync with shotgun
07:40:22MenTaLguYisn't it just a matter of making LT and Hash duck-type the same?
07:40:32ruebrixen: Well, there are downsides to superpositions
07:40:48rueFor example, you might just end up with a year-old repo
07:41:09brixenMenTaLguY: I suppose I could have made the tuple entries in LT the same as hash
07:41:17rueOr possibly locked in some type of deadly trap capsule orbiting the earth and whatnot
07:41:20brixenthis is probably easier though, once I figure out exactly what's going on
07:41:53brixenrue: heh, you wouldn't worry long if you were trapped in a capsule orbiting the earth
07:42:03brixenthe dead cat certainly has the easier lot of the two ;)
07:42:25rueThat depends on how many dimensions you are banking on
07:42:56ruebrixen: The MethodTable instantiation is somewhat convoluted
07:43:07brixenrue: ahh, MT is done :D
07:43:13brixenthis is the constants table
07:43:41brixenMT is yesterday's agony
07:43:58craftermquik_: really good mate, there's 27 new patches in lighthouse :)
07:44:19brixenholy moly
07:44:22brixenpatches or tickets?
07:44:25quikgood stuff. I've been painting to keep my mind off it
07:44:31brixenwhere's those committers? :P
07:45:52rueMan, I do not think I have even pulled since last week
07:46:22quikpulled the repo or pulled chicks?
07:46:41craftermbrixen: tickets - but the guys have put 1 patch per ticket in
07:46:49brixencrafterm: awesome!
07:46:58craftermlots of admin work for us tonight then :)
07:47:06brixenquik_: he's married, let's hope he means pulled the repo
07:47:11quikwhere is the lighthouse account at?
07:47:17brixenquik_: although, what does pulled chicks mean? :)
07:47:31headiuscrafterm: nice, are those fixes or spec updates?
07:47:32brixenhttp://rubinius.lighthouseapp.com/projects/5089-rubinius/tickets
07:47:38craftermquik_: http://rubinius.lighthouseapp.com/projects/5089-rubinius/overview
07:47:41craftermheadius: a mix mate
07:48:53headiusexcellent, looking forward to running the new specs
07:50:15mediogre enters the room.
07:51:08benburkert enters the room.
07:56:17rueWhen is your birthday, headius? I am going to buy you a bullwhip
07:56:40headiusa bullwhip, eh?
07:57:32ttmrichter enters the room.
07:57:48rueMaybe some chewing tobacco
08:06:13rueInteresting, I have more battery power left after I reboot :(
08:06:52jacenthe nsa-trojan doesn't boot up right at start
08:06:56jacenit waits a bit.
08:07:39rueN-uh, I have a Mac now, I am impervious to any and all security threats past and yet imagined
08:08:32jacenThat's what they tell you.
08:10:43headius leaves the room.
08:13:37ezmobius enters the room.
08:16:11rueHey hey
08:17:08rueIf I have one complaint about OS X, it is that display sleep is implemented completely retardedly
08:20:20quik_ leaves the room.
08:28:04imajes enters the room.
08:31:17jinjing enters the room.
08:34:55crafterm leaves the room.
08:37:23_matta_ leaves the room.
08:39:10Fullmoon enters the room.
08:39:29quik_ enters the room.
08:41:29imajes leaves the room.
08:41:51perplexes leaves the room.
08:44:59kamal_farizwoah .. were all those tickets from the sprint
08:45:31ezmobius leaves the room.
08:47:00crafterm enters the room.
08:47:03_matta_ enters the room.
08:49:32Fullmoon leaves the room.
08:51:40jinjing leaves the room.
08:56:41quik_ leaves the room.
09:03:23crafterm leaves the room.
09:03:43dbussinkmorning!
09:15:42eugenebrechko leaves the room.
09:16:08_matta_ leaves the room.
09:16:42lachie leaves the room.
09:25:20lachie enters the room.
09:27:34_matta_ enters the room.
09:32:33dbussinkhow's the sprint going?
09:36:48rueMorning
09:40:07dbussinkquite some tickets from the sprint
09:40:19dbussinkwhat's the policy on handling them?
09:43:48dbussinkbrixen: you still there?
09:45:18benburkert leaves the room.
09:46:07radarek enters the room.
09:49:38rueI think you can pretty much apply if able
09:49:59rueLessee what the Wiki says
09:52:37rosejn enters the room.
09:55:22dbussinkwell, gotta go now anyway
09:55:37dbussinkbut i'll probably have some time tomorrow to handle some
10:14:40crossblaim enters the room.
10:19:50quik_ enters the room.
10:57:50chris2 enters the room.
10:58:26octopod enters the room.
10:59:06lachie leaves the room.
11:00:50_matta_ leaves the room.
11:06:26dewd enters the room.
11:14:35rosejn leaves the room.
11:39:05quik_ leaves the room.
11:50:31tizianobis_ enters the room.
11:52:46rue"As you see people arriving from another flight, you observe that some of them have burn marks. One of them comments that 'merging with the stars doesn't work.'"
11:52:51rueAnd on that note, good night.
12:13:48lstoll leaves the room.
12:26:17mediogre leaves the room.
12:28:00aotearoa leaves the room.
12:48:46thehcdreamer enters the room.
12:49:38imajes enters the room.
12:56:53jinjing enters the room.
12:58:19ubiquitous leaves the room.
12:58:32tizianobis_ enters the room.
12:59:20tizianobis_ leaves the room.
13:07:55thehcdreamer leaves the room.
13:08:47notahat enters the room.
13:13:11notahat leaves the room.
13:14:33tizianobis leaves the room.
13:19:02RyanTM enters the room.
13:23:55lachie enters the room.
13:29:49digitalfallout leaves the room.
13:32:07lachie leaves the room.
13:43:27lstoll enters the room.
13:43:44lstollSPRINT
13:48:22retnuH_ enters the room.
13:54:51retnuH leaves the room.
13:58:31benburkert enters the room.
13:58:33digitalfallout enters the room.
13:59:26antares enters the room.
14:00:29crossblaim leaves the room.
14:01:21crossblaim enters the room.
14:01:31chop3 enters the room.
14:04:10nicksieger leaves the room.
14:04:27enebo enters the room.
14:06:51enebo leaves the room.
14:14:12dewd leaves the room.
14:16:49crossblaim leaves the room.
14:17:59crossblaim enters the room.
14:18:49enebo enters the room.
14:25:33GMFlash enters the room.
14:29:45boyscout2 commits by Dirkjan Bussink
14:29:46boyscout * Fix Sprintf for positive non decimal notation; a720bba
14:29:47boyscout * Switch FFI::TypeDefs to LookupTable; 21ec5c8
14:30:16obvio enters the room.
14:31:33dbussinkfor who's in, i committed some patches from the sprint
14:31:48dbussinkused them for fixing some ipaddr specs
14:40:25rubuildius_ppcDirkjan Bussink: a720bba16; 1741 files, 5881 examples, 19974 expectations, 0 failures, 2 errors; http://pastie.caboo.se/paste/163251
14:42:10chop3 leaves the room.
14:42:27dbussinkhmm, something is still not ok on big endian
14:43:39dbussinkcremes: you there?
14:44:10chop3 enters the room.
14:44:39obvio171 enters the room.
14:45:17VVSizdbussink: it seems that the new Array#pack specs are incorrecty
14:45:26dbussinkVVSiz: yeah, i noticed
14:45:41dbussinki thought they were, there were explicit comments in it about it :P
14:45:50VVSiz'n' directive is always for network-order
14:45:52dbussinkbut i don't have a big endian system i can test on
14:45:54dbussinkyeah
14:46:03VVSizJRuby now fails them, and MRI on Solaris (big-endian) also fails
14:46:18VVSizruby 1.8.4 (2005-12-24) [sparc-solaris2.9]
14:46:21dbussinkthat was my guess too
14:46:23VVSiz[1234].pack('n')
14:46:23VVSiz=> "\004\322"
14:50:08dbussinkVVSiz: fix incoming
14:50:16boyscout1 commit by Dirkjan Bussink
14:50:17boyscout * Fix the Array#pack specs, network order is the same everywhere; 4612812
14:51:42cremesdbussink: here now; what's up?
14:52:16dbussinkcremes: i was trying to connect to your system to test with the current failures, but i couldn't connect
14:52:28VVSizdbussink: excellent, the failures are gone
14:52:35VVSizthanks
14:52:53dbussinkVVSiz: should have looked better before applying the patch
14:53:25cremesdbussink: ip addr changed; try 75.3.137.196
14:54:45dbussinkcremes: ah, thnx :)
14:59:45dbussinkcremes: testing the fix in your system now for those two build failures
15:00:08cremesdbussink: ok; preparing for the extra system load... NOW!
15:00:15cremes:)
15:00:22dbussinkalways nice to be able to test it first then just praying it will work :P
15:00:28obvio leaves the room.
15:01:17cremesdbussink: what kind of programmer does that? i just code & squeeze my eyes shut to run ./a.out
15:02:14dbussinkhell has been unleashed on me too often, so i'm a convert ;)
15:02:56benburkert leaves the room.
15:03:54dbussinkpfff, a build after a distclean takes looooong
15:04:41cremesdbussink: after letting the local git repos fall that far behind, it's the only sane way to try a build
15:04:57dbussinkcremes: yeah, i know, but it's still annoying :)
15:05:33headius enters the room.
15:10:06digitalfallout leaves the room.
15:10:24benburkert enters the room.
15:12:44rubuildius_ppcDirkjan Bussink: 4612812bd; 1741 files, 5881 examples, 19974 expectations, 0 failures, 4 errors; http://pastie.caboo.se/paste/163258
15:13:20dbussinkah, more failures is actually good here :)
15:19:25headiusmm, time to pull specs
15:19:43cremesdbussink: the bug is at line 1129 of kernel/core/array.rb
15:19:52dbussinkcremes: yeah, i know
15:20:01dbussinkcremes: somehow the build failed in my checkout
15:20:06dbussinkso doing a clean checkout now
15:20:18cremesdbussink: buzz me if you need some help
15:20:25headiusaw, rats, no new ruby specs yet
15:20:50dbussinkheadius: some for Array#pack
15:20:59dbussinkheadius: added some minor cases because of a bug we had
15:21:13headiushmm, don't see those in my update
15:21:25headiusare they in a ticket?
15:21:41dbussinkno, just pushed it in
15:21:46dbussinkhttp://git.rubini.us/?p=code;a=commitdiff;h=a720bba1619deb4358b453f58913d30a1a311b07
15:22:10dbussinki'm telling bogus, not array pack, but string modulo :)
15:22:11headiusoh, I see them
15:22:35cremesare the sprinters asleep right now?
15:22:47headiusprobably
15:23:03dbussinkit's day here, so night there :P
15:23:12cremesokay, day one of the sprint was probably for familiarization
15:23:23cremesday two will knock our socks off
15:23:35headiusday one didn't knock your socks off?
15:23:48dbussinkcremes: there are quite some pending tickets on lh
15:23:52cremesmy socks are attached via those little garter thingies...
15:24:05cremeshard to knock off
15:24:13headiushmm, well array still passes, that's good
15:24:29cremesdbussink: do y'all need some help committing tickets?
15:24:51dbussinkcremes: dunno what the policy is during the sprint
15:25:07dbussinkcremes: i grabbed a ticket because i wanted to fix some other specs too
15:25:34cremesdbussink: i see; i don't have a lh acct anyway...
15:25:54dbussinkcremes: ah, well, you can apply some patches if you like
15:25:57cremeswell, i have a read-only acct; i can't close tickets
15:26:14dbussinkcremes: someone can close it if it's in, just post a comment
15:26:17obvio171_ enters the room.
15:26:36cremesk; i'll have a look at a few this morning then
15:27:47headiushmm, only one new failure
15:28:14cremesdbussink: btw, fix is: if (kind =~ /n/i) && (little_endian)
15:28:29dbussinkcremes: yeah, i have it here
15:28:36benburkert leaves the room.
15:28:41cremesawesome; off to look at some tickets
15:29:04antares leaves the room.
15:29:11srbaker enters the room.
15:31:08jinjing leaves the room.
15:31:39headiuswhere does the platform stuff determine :wordsize
15:32:23boyscout1 commit by Dirkjan Bussink
15:32:24boyscout * Fix Array#pack for network byte order; 8860852
15:34:00dbussinkheadius: you mean in shotgun/mkconfig.sh ?
15:34:37cremesheadius: look at shotgun/lib/machine.c, line 576 where it gets set
15:35:41cremesheadius: dbussink is more correct
15:35:59dbussinkcremes: all looks ok again on ppc now
15:36:31cremesyay! right now this is the *only* rubuildius machine so we have to keep it error-free!
15:36:44headiussorry, I just meant in mspec, trying to figure why there's one failure in string for jruby
15:37:02headiusfound it
15:37:07obvio171 leaves the room.
15:38:31dbussinkcremes: it's pretty useful, because most devs have a intel machine, so it picks up problems faster :)
15:38:47cremesagreed!
15:50:44rubuildius_ppcDirkjan Bussink: 886085200; 1741 files, 5881 examples, 19976 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163264
15:53:22quik_ enters the room.
15:53:23d2dchat enters the room.
15:53:26imajes leaves the room.
15:55:53quik_ leaves the room.
15:59:01cremesgit question... after my latest pull why does it tell me this? : http://pastie.caboo.se/163266
15:59:28cremesi'm specifically curious about the stuff in shotgun/external_libs/onig that "needs update"
15:59:43squeegycremes: that looks fine to me
16:00:06cremessqueegy: but if i try to commit someting, won't those "changes" be included?
16:00:18dbussinkwhat does git status say?
16:00:56cremesgit status says: http://pastie.caboo.se/163268
16:01:34cremesi didn't modify those shotgun/* files; the build system did
16:01:48cremesshouldn't they be ignored or something?
16:01:49dbussinkcremes: strange, you can do git checkout shotgun/external_libs/onig/ so it reverts
16:02:32dbussinkwell, i've seen it before that they changed
16:02:38dbussinkand people accidently committed them
16:02:45dbussinkdunno what exactly triggers it
16:03:24cremeshmmm... okay, the checkout did revert it; my next build will likely bring those changes back
16:04:05cremesbtw, anybody see this ruby fork? http://www.sapphire-lang.org/projects
16:06:03headiusweird
16:06:13headiusdan berger
16:06:34cremesheadius: he's pretty active with ruby-core but i didn't even see an announcement there
16:06:49cremeshe must be getting upset with matz's leadership on mri
16:08:54edward_g enters the room.
16:10:01headiuswell, sometimes a fork is a good thing
16:10:28headiusnothing runs ruby as well as MRI right now, so if this helps push it forward, it would be a good thing
16:10:35binary42 enters the room.
16:11:38srbaker leaves the room.
16:11:50cremesagreed; getting the source to compile cleanly will probably expose (& fix) lots of bugs too
16:17:13srbaker enters the room.
16:22:23sholden enters the room.
16:25:29digitalfallout enters the room.
16:28:25mae enters the room.
16:32:41AndrewO enters the room.
16:36:32sholden leaves the room.
16:37:07sholden enters the room.
16:38:49jrun enters the room.
16:41:00digitalfallout leaves the room.
16:46:11sholden leaves the room.
16:47:22sholden enters the room.
17:00:35sholden leaves the room.
17:01:32sholden enters the room.
17:06:24sholden leaves the room.
17:19:33benburkert enters the room.
17:22:31benburkert leaves the room.
17:23:03binary42 leaves the room.
17:30:22benburkert enters the room.
17:44:25thehcdreamer enters the room.
17:46:15tizianobis enters the room.
17:48:03dewd enters the room.
17:49:26benburkert leaves the room.
17:56:05tizianobis_ enters the room.
18:00:41benburkert enters the room.
18:05:48brixenhmm, wonder why berger didn't just throw in with rbx
18:06:13brixendbussink: you rang? (a few hrs ago)
18:13:37thehcdreamer leaves the room.
18:14:48tizianobis leaves the room.
18:27:18GMFlash leaves the room.
18:27:26GMFlash enters the room.
18:28:22rmm5t enters the room.
18:31:06tizianobis_ leaves the room.
18:34:02binary42 enters the room.
18:35:18enebo leaves the room.
18:41:02dkubb leaves the room.
18:44:42benburkert leaves the room.
18:49:17thehcdreamer enters the room.
18:50:09VVSizbrixen: you there?
18:51:18brixenyep
18:51:30VVSizgood :) quick question
18:52:05VVSizfor mocks in mspec, do I need to do anything else to make sure the methods called appropriate number of times
18:52:17VVSiz(obj = mock("blah")).should_receive(:ping).twice.and_return(5)
18:52:23brixenhmm, not that I know of
18:52:26brixennot working?
18:52:36VVSizwell, that what I was afraid of
18:52:38VVSizyep, don't work
18:52:42VVSizpastie: for brixen
18:53:02pastiebrixen: http://pastie.org/163309 by VVSiz.
18:53:15brixendoes it pass with rspec?
18:53:21VVSizas you see I define a mock and explicitly say I need the method called twice
18:53:25VVSizwell, it doesn't fail :)
18:53:40brixenwith rspec though?
18:53:48VVSizI can add Mock.verify_count , and then it fails
18:53:49VVSizlet me try
18:54:20VVSizwith spec it correctly fails
18:54:24brixenk
18:54:31VVSizso mspec has a problem it seems
18:54:44brixenis verify_count not getting called?
18:54:58VVSizthat's what rspec says: Mock 'blah' expected :ping with (any args) twice, but received it 3 times
18:55:08VVSizyep, verify_count is not being called in mspec
18:56:07brixenyeah, looking at mspec/runner/state.rb, RunState#process
18:56:54brixenif you add: protect "Mock.verify_count", lambda { Mock.verify_count }
18:57:06brixenbefore the protect "Mock.cleanup"...
18:57:08brixendoes that fix it?
18:57:12GMFlash leaves the room.
18:57:15VVSizlet me try
18:57:20GMFlash enters the room.
18:58:25brixenhmm, or you could add the call to verify_count in Mock.cleanup
18:58:30VVSiznope, not working
18:58:39brixenprobably better to have it separate
18:59:09brixenlet me try
19:00:04VVSizbrixen: oh, sorry, it does work after all :)
19:00:23VVSizonce I added (and SAVED!) that protect "Mock.verify_count" ... :)
19:00:48brixenheh
19:00:53brixenI was going to say... :P
19:01:14brixencan you fix up the specs for RunState#process and commit all that?
19:01:24VVSizwill do
19:01:28brixensuper, thanks!
19:01:55crossblaim leaves the room.
19:02:11brixenthis is one reason I really like being compatible with rspec, nice to have an independent check
19:02:22VVSizyep
19:02:26GMFlash leaves the room.
19:02:27brixenthere's some failures now with rspec that I need to fix up
19:02:34GMFlash enters the room.
19:03:40GMFlash leaves the room.
19:03:44rmm5t leaves the room.
19:03:48GMFlash enters the room.
19:04:45boyscout1 commit by Brian Ford
19:04:46boyscout * Converted Selector and primitives method table to use LookupTable.; 7ac2e51
19:05:31nicksieger enters the room.
19:06:48boyscout1 commit by Vladimir Sizikov
19:06:49boyscout * Mspec should invoke Mock.verify_count for every test.; ccc82d7
19:07:56VVSizbrixen: btw, I see one spec failure for mspec:
19:07:57VVSiz1)
19:07:57VVSiz'MSpec provides .register_exit to record the exit code' FAILED
19:07:57VVSizexpected: 0,
19:07:57VVSiz got: 1 (using ==)
19:07:57VVSiz/opt/work/rubinius/mspec/spec/runner/mspec_spec.rb:22:
19:08:31VVSizit was there before my change, and remained there after it
19:08:45brixeninteresting, I just ran it after pulling your changes and it's clean
19:08:58brixenyou ran with rspec, right?
19:16:13VVSizyeah
19:17:07brixenwhat version?
19:20:16jrun leaves the room.
19:24:22boyscout1 commit by Vladimir Sizikov
19:24:23boyscout * Adedd a couple of GzipReader#rewind specs.; 9b9d821
19:30:50jrun enters the room.
19:34:13dkubb enters the room.
19:35:15thehcdreamer leaves the room.
19:36:04chris2 leaves the room.
19:36:54rubuildius_ppcVladimir Sizikov: 9b9d82160; 1741 files, 5881 examples, 19976 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163318
19:36:55rubuildius_ppcVladimir Sizikov: ccc82d78a; 1741 files, 5881 examples, 19976 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163315
19:36:56rubuildius_ppcBrian Ford: 7ac2e511b; 1741 files, 5881 examples, 19976 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163314
19:43:54benburkert enters the room.
19:46:04d2dchat leaves the room.
19:57:41maduyb__ enters the room.
20:05:39Defilercremes: Where did you hear about Sapphire? I don't see anything on ruby-talk or ruby-core
20:05:42brixendrat, I put this dependency on Array#each in LookupTable#each
20:06:07brixenstables are nice, but maybe we need a facility to run some patch code when using stables
20:06:37brixenDefiler: we need to lasso berger over here
20:07:18DefilerYeah. I understand not wanting to bet on an unproven system (us)
20:07:46Defiler..but surely rewriting our core in C would be easier than making MRI be C99 and otherwise meet his stated goals
20:07:51d2dchat enters the room.
20:08:14brixenno doubt
20:08:20DefilerWe should experiment with that, actually.. write a couple of the 'hottest' kernel methods in C and see what happens
20:08:37DefilerThen throw them away, of course
20:08:41brixenheh
20:08:55brixendid you read through the steps paper from vpri yet?
20:10:45DefilerNope. Got a link?
20:10:47benburkert leaves the room.
20:11:01brixenyep, sec..
20:11:32brixenhttp://vpri.org/pdf/steps_TR-2007-008.pdf
20:12:55benburkert enters the room.
20:14:48maduyb__ leaves the room.
20:16:35rmm5t enters the room.
20:17:42rmm5t leaves the room.
20:21:15aotearoa enters the room.
20:22:34tizianobis enters the room.
20:31:01tizianobis leaves the room.
20:32:49wmoxam enters the room.
20:37:02ezmobius enters the room.
20:48:14Defilerdrbrain: Where are we at in terms of having rubygems included in lib/?
20:52:10benburkert leaves the room.
20:57:19ezmobius leaves the room.
20:59:21ruebrixen: imperator is an opinionated bastard :) He sent an e-mail to the Google Group a while back with some questions
21:00:17Defilerbrixen: I am of the opinion that: bin/mspec run -t r spec should pass
21:00:21Defilerbrixen: What do you think?
21:06:09lachie enters the room.
21:08:36brixenrue: ahh, imperator is dan berger?
21:08:52brixenDefiler: bin/mspec -tr spec/ruby yes
21:08:55brixenspec, no
21:09:16brixensince spec/* includes rbx specific stuff
21:09:26DefilerOK, so you disagree
21:09:28brixenalmost got consts using lookuptable
21:09:38DefilerI was thinking that we should wrap anything rbx-specific with the appropriate guard
21:09:44brixennope
21:09:52brixentoo much junk for no reason
21:10:21brixenmiddle of lunch, bbiab..
21:10:56boyscout1 commit by Wilson Bilkovich
21:10:57boyscout * Fix Module#method_defined? and friends for accessors; b748efa
21:12:26ubiquitous enters the room.
21:17:29MenTaLguY enters the room.
21:18:43Defilerbrixen: When you get back, could you tell me where we are on class variable compatibility with MRI?
21:21:39brixenback.. dunno exactly, there's #306 and whatever specs we have
21:21:56Defilerbrixen: Can you think of anything specific that would make this not work? http://pastie.org/163367
21:22:11brixenbtw, obj.__kind_of__ is probably an OOM faster than #is_a?
21:22:24brixenwe should probably use it when appropriate in core lib code
21:22:44brixenhmm, that pastie looks like #306 or something else ez reported
21:22:47brixenle'me check
21:23:25brixenDefiler: is that code in rails?
21:23:32brixenI can't find ez's paste
21:23:42DefilerYeah, that is in activesupport
21:23:59Defilerbrixen: I thought is_a? had a compiler plugin
21:24:16brixenlast time I did describe on those two, they were different
21:24:22brixenI had a bench somewhere
21:24:35DefilerWe should fix that, if possible
21:24:47Defiler(by which I mean, have the code code make use of __kind_of__)
21:25:05brixenI'll check right quick
21:25:28brixenDefiler: can you repro the cvar stuff outside of eval?
21:26:34nemerle leaves the room.
21:27:26rubuildius_ppcWilson Bilkovich: b748efa99; 1741 files, 5881 examples, 19977 expectations, 0 failures, 0 errors; http://pastie.caboo.se/paste/163369
21:27:53Defilerbrixen: OK, found the minimal thingy, I think
21:29:18Defilerhttp://pastie.org/163370
21:29:34Defilercheck_foo prints the correct result, the next line throws an error
21:29:36cremesDefiler: found the sapphire page via reddit
21:29:57cremesi immediately looked around on ruby-talk & core to see if i had missed the announcement but found nothing
21:30:03jrun leaves the room.
21:30:36jrun enters the room.
21:31:41brixenDefiler: they are different, but the bench I just ran is not showing the same behavior that I observed when working on the Stat stuff: http://pastie.org/163372
21:31:46jrun leaves the room.
21:32:09jrun enters the room.
21:32:25jrun leaves the room.
21:33:12brixenDefiler: so, if I understand the result of running your pastie, it's related to eval?
21:34:10jrun enters the room.
21:37:31Defilerbrixen: Let me check. There is actually another possibility
21:39:38aotearoa_ enters the room.
21:39:38jrun leaves the room.
21:39:51jrun enters the room.
21:39:58Defileroof. We have another problem I just found. Heh
21:40:23Defilerwait, no. false alarm on that part
21:40:29radarek leaves the room.
21:42:33Defilerbrixen: It is weirder than that: http://pastie.org/163374
21:42:50Defiler@@foo is fine inside the class_eval
21:43:13d2dchat leaves the room.
21:43:42brixenhmm, ok, then probably something wrong with the lookup chain
21:44:16MenTaLguY leaves the room.
22:03:55headiusDefiler: I agree with you, but I don't think rubinius specs should be guarded, I think they should be in a separate tree
22:04:02headiusor rubyspecs should be moved off-project
22:04:10headiusI know it's on its way
22:04:19headiustaking a remarkably long time though :)
22:04:41DefilerOh crap headius agreed with me. I retract my opinion :)
22:04:59headiusI agree with you retracting your opinion
22:05:36brixenregardless of specs moving off project, it doesn't make sense to have them in a separate tree *within* this project
22:05:38DefilerToo bad I can't guard my IRC conversations with Timeout
22:06:14headiusok, rubyspecs should be in a separate tree :)
22:06:15brixencurious, why do we need bin/mspec -tr spec to run?
22:06:21eugenebrechko enters the room.
22:06:26brixenheadius: why?
22:06:29brixenfor this project?
22:06:40DefilerIt just seems to me that we have gotten into a place where it is not obvious, to new users of the specs, what should and shouldn't work
22:06:50headiusbecause it's not just rubinius devs contributing specs
22:07:14aotearoa leaves the room.
22:07:26brixenheadius: I'm not talking about ruby specs being separate, I'm asking why they should be in a separate dir in this project?
22:08:21brixenanyway, gotta run some errands, back later..
22:08:23headiusreally I guess they are separate enough for me in-project
22:08:26headiusspec/ruby
22:08:30headiusis simple enough
22:11:47eugenebrechko leaves the room.
22:14:13headiushmmm
22:14:25headiuswould a spec for define_method(:foo, method(
22:14:29headiuser
22:14:44headiusdefine_method(:foo, method(:bar).to_proc) go under method or define_method
22:14:47headiusdecisions
22:15:48MenTaLguY enters the room.
22:16:23DefilerI would say define_method
22:16:40DefilerIf you were just testing what method(:bar).to_proc did, that would be one thing
22:16:57headiuswell, it's a bit of both
22:17:25headiustesting that define_method can accept procs seems separate to me from testing that the proc produced by to_proc is usable everywhere other procs are
22:17:30headiusI think it should go under to_proc
22:17:36DefilerWhere are the existing specs for define_method(:foo, Proc.new ...)
22:17:38Defilerassuming they exist
22:18:28headiusthere are define_method specs
22:19:52evanmorning
22:19:56headiushowdy
22:20:03lachie leaves the room.
22:20:42evanman the internet here sucks
22:21:32DefilerFreaking southern hemisphere, am I right?
22:21:52headiusIP over wallaby
22:23:46boyscout1 commit by Charles Nutter
22:23:47boyscout * Adding a spec for Method#to_proc proc used in define_method.; 142222e
22:24:14ezmobius enters the room.
22:25:43DefilerAnybody have technomancy's IM info?
22:29:31eugenebrechko enters the room.
22:32:46ezmobiusyou in seattle Defiler ?
22:33:08DefilerYep
22:33:17ezmobiuscool
22:33:34DefilerSitting in the holy sanctum now (Vivace)
22:33:41d2dchat enters the room.
22:35:57rubuildius_ppcCharles Nutter: 142222e41; 1741 files, 5882 examples, 19977 expectations, 0 failures, 1 error; http://pastie.caboo.se/paste/163393
22:37:04d2dchat leaves the room.
22:37:21headiushmmm
22:38:22headius_ enters the room.
22:38:22headius leaves the room.
22:39:24headiusboth jruby and rubinius choke on a 100kloc source file
22:39:29headiusjruby blows up in JVM's verifier, and rubinius just seemed to run away (had to kill it around 600MB memory
22:40:29ezmobiusdont think ive ever seen a ruby file with 100kloc
22:41:07headiusrdoc's parse_rb.rb is around 65k
22:41:12headiusbut it's got classes and method bodies
22:41:16headiusso that breaks it up
22:41:19ezmobiuscrazy
22:41:25Defiler65kloc in one file?
22:41:26Defilerwtf
22:41:28headiusyeah, we use it for parser benchmarking
22:41:45headius_ enters the room.
22:41:46headius leaves the room.
22:41:51evan100kloc of what?
22:42:00evanhow big is it in bytes?
22:42:52headiusa += 1
22:42:54headiussome test case vvsiz came up with, no idea why
22:42:58headiuslook like 700kb
22:43:39headiushttp://jira.codehaus.org/browse/JRUBY-2245
22:44:00headius leaves the room.
22:44:08headius enters the room.
22:44:38headius leaves the room.
22:45:40headius enters the room.
22:45:46evaninteresting
22:45:52evani wonder what blows up
22:45:58evanprobably inside the compiler
22:46:43headiuswell, we'll see if a wired connection is better
22:46:45headiusevan: it might just take a lot of memory, but 600 is about as high as I can go without dogging the rest of the system
22:46:48headiusand obviously that's a big much
22:46:50headiusbit
22:47:00evanyeah
22:49:25aotearoa_ leaves the room.
22:53:58headius leaves the room.
22:54:32crafterm enters the room.
23:03:54eugenebrechko leaves the room.
23:11:50GMFlash leaves the room.
23:11:58GMFlash enters the room.
23:14:38GMFlash leaves the room.