Parrot 3.8.0 "Magrathea" | parrot.org | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 29 September 2011.
00:35 lateau joined 00:36 eternaleye_ joined 01:24 benabik joined
lateau is this not allowed in pir? I've got error:imcc:syntax error when I tried this => unless config['MAJOR'] >= 3 goto hooray 01:31
01:31 jsut joined
lateau error:imcc:syntax error, unexpected '[' ('[') 01:31
dalek rrot: 6026393 | jkeenan++ | lib/Parrot/Harness/DefaultTests.pm:
Use correct, long-since-current name of subroutine in documentation.
01:41
01:48 GodFather joined 02:51 lateau joined 04:12 preflex_ joined 04:39 fperrad joined 05:12 davidfetter joined 05:22 Khisanth joined
dalek rdinal: ca8f0bd | (Daehyub Kim)++ | setup.pir:
patch for issue #13
05:43
rdinal: 57004ee | dukeleto++ | setup.pir:
Merge pull request #17 from lateau/lateau

patch for issue #13 (detect minimum parrot version)
06:40 mj41 joined 07:32 he joined
plobsing lateau: PIR is not a full language, but a sugar layer on top of an assembly language. there is no op implementing the complex operation you are trying to do (at least none that IMCC knows about for that syntax). I would suggest breaking up your line slightly, possibly moving the keyed lookup out and using a temporary variable. 07:41
07:43 mj41 joined 07:55 jsut_ joined 08:16 lucian joined
cotto ubuntu++ 08:36
definitely dislike 11.10 less than 11.04 08:39
or like more
lucian cotto: did you previously dislike unity?
cotto lucian, some bits were annoying. I'm still not sure why a hard limit of 4 virtual desktops is a good idea. 08:47
power management is at least sane now though.
lucian cotto: there's a hard limit? somehow i never noticed 08:48
cotto I love that everything's in the title bar.
lucian i'm just in love with the vertical screen space
yeah, me too
it adds 1/5 to my netbook's useful vertical resolution 08:49
cotto It's usually a huge waste of vertical space, which is scace.
:D
tadzik that's why I like sort of windowsy very small bar at the bottom
my eyes lie at the top most of the time anyway, it seems least annoying at the bottom 08:50
lucian tadzik: but then you still have title and menu bars
osx fixes it a little with the menu bar. but unity goes even further, and at least i like it
tadzik well, I have those reduced to minimum too, but true
generally, I consider window borders quite useless, but KDE does not really allow me to remove them completely 08:55
09:52 mj41 joined 11:37 Psyche^ joined 11:50 whiteknight joined
whiteknight good morning, #parrot 12:00
nine Good morning whiteknight :) 12:06
whiteknight hello nine
I wanted to play with green_threads last night, but we needed to go grocery shopping and the rest of my time was spent updating Ubuntu 12:07
12:08 lateau joined
nine whiteknight: well, today's just another day :) 12:08
tadzik: KDE does allow to remove window borders completely 12:09
tadzik: System settings/Workspace appearance/Window decorations/Configure decoration/General/Border size/No Border 12:11
12:21 bluescreen joined
tadzik nine: that's borders, but not title bars 12:29
atrodo =~ 12:43
13:25 mtk joined 13:41 bluescreen joined 13:52 kthakore joined
nine whiteknight: do you happen to have a quite fast computer? 14:05
whiteknight define "quite fast? 14:10
also, probably not 14:11
nine Just noticed that on a very loaded system t/pmc/alert.t runs quite reliably 14:15
14:17 lucian joined 14:25 jsut joined 14:31 lucian joined
nine Thou shalt not disable the scheduler with alarms still pending! 14:50
tadzik (: 14:52
nine Now it would just be great if I could come up with a test that can reliably reproduce this bug... Chances are that the scheduler will see some epic refactoring in the future 14:53
whiteknight it better 14:55
nine How difficult can it be to write a test for a bug you supposedly understand? 15:22
moritz 0..Inf
tadzik hmm. The bug is alarms pending, but scheduler disabled? 15:28
nine yep 15:29
moritz what disables the scheduler? 15:30
tadzik could you perhaps fire scheduler disabling as an alarm, the last of them all?
oh, that doesn't help
moritz oh, another ticket migration 15:32
to another system where you can't open bug reports by email
*sigh*
tadzik maybe it's easier to fix github than trac
nine Well it probably just demonstrates why I'm reluctant to commit this code. It works, but I hate it and I don't understand it completely.
tadzik you could possibly set up an daemon that'd read emails from a certain mailbox and file bugs via GH api 15:33
dukeleto tadzik: seems feasible 15:38
tadzik: or we can just ask github to enable an email-to-issue feature :) 15:39
tadzik I vote for the latter
whiteknight is email-to-ticket really such an important feature? 15:41
dukeleto whiteknight: for the folks who like RT, it is 15:42
moritz whiteknight: it makes submitting bugs much simpler for me
15:45 awwaiid joined
whiteknight moritz: okay. I've never used the feature so I don't know how useful it is 16:01
dukeleto whiteknight: the perl world usually has an email for perl core internals that you can send a bug report and it opens a ticket 16:04
i am starting to lean towards the git philosophy of ticket systems
whoa, this is cool: graphs-new.mozilla.org/ 16:11
tadzik argh. 500 what? 16:12
16:17 lucian joined 16:25 fperrad joined 16:34 lucian joined
dukeleto blargpoop 17:09
17:18 mj41 joined 17:47 dmalcolm joined
Tene dukeleto: git philosophy of ticket systems? 17:52
dukeleto Tene: there is no ticket system for git.git 17:53
Tene: only the main development mailing list
17:55 benabik joined
moritz so error reports from people who don't keep nagging are ignored, exactly as it is now. Correct? 18:11
benabik o/ 18:16
nine Ok, I think, I will just commit this now. I don't like it and I still was unable to write a good test. But the whole alarm handling and scheduler interaction is so screwed I'll have to change things anyway. 18:24
benabik Commit early, commit often. 18:25
nine yep 18:28
whiteknight nine: yeah, get more eyes on it
nine pushed 18:34
benabik nine++ 18:36
cotto ~~ 18:37
18:38 Coke joined
Coke cotto: there's an ubuntu update? maybe my netbook is slightly more useful than a brick now! 19:07
whiteknight Coke: Yeah, 11.10 is pretty awesome 19:08
19:08 soh_cah_toa joined
whiteknight It's much more stable than 11.04, from my testing 19:09
Coke will give it a shot tonight, thanks. I liked some of the visual updates from 10.x to 11.04, but it basically broke hardware sleep mode.
whiteknight oh, I turned off my laptop's sleep mode. I haven't tried that again 19:12
that's something I should try
cotto 11.10 restored the ability to do nothing on laptop lid close, which makes me happy.
whiteknight it's surprising how often "do nothing" is an essential feature in software 19:17
dalek rrot/green_threads: cbd81fd | (Stefan Seifert)++ | / (2 files):
Enable preemption only when more than one task is active.

We enable the scheduler alarm and enable_scheduling only in situations where more than one task is in the queue. This way, there is no preemption overhead in the common single-task case.
19:20
rrot/green_threads: a074845 | Whiteknight++ | / (2 files):
Merge branch 'green_threads' of git://github.com/niner/parrot into green_threads
benabik Hm. Looks like someone needs to be added to AUTHORS.
cotto lulz: johnkary.net/git-1-7-7-changes-affe...developer/ "Adding cache plugins, back up in a sec :)" 19:21
whiteknight msg nine when you get a chance, add an entry for yourself in CREDITS
aloha OK. I'll deliver the message.
benabik CREDITS, AUTHORS… Whichever.
19:29 zpmorgan joined 19:36 zpmorgan joined 19:44 zpmorgan joined
dukeleto this looks pretty hilarious: learnyousomeerlang.com/ 21:15
tadzik the haskell one is second or so on my to-read list 21:16
I should finish SICP first
21:44 perlite_ joined 22:09 rfw joined 22:18 silug joined 22:47 schmooster joined