Parrot 2.0.0 "Inevitable" released! | parrot.org | Priorities: merge tt_389_fix and one_make branches | Roadmap: icanhaz.com/parrotroadmap | Latest modified TT's: icanhaz.com/parrotbugs
Set by moderator on 20 January 2010.
Tene Soon, but not quite yet. Some of PGE is under-tested. I haven't, for example, tried running an HLL with the poison pill in method fetches. 00:12
But yes, soon.
If rakudo is running fine on 2.0, I should work on tt389 tonight?
Tene afk
pmichaud Tene: for tonight, I'd suggest tt389, yes. 00:13
(and thank you for taking this up)
dalek rrot: r43522 | jkeenan++ | trunk/docs/submissions.pod:
Applying, with small modifications, patches to documentation re submission of
 00:15
chromatic Tene, mostly we need to merge that PGE branch into the tt389_fix branch, except for the modification to src/global.c. 00:17
Tene chromatic: have you tried running anything else on that PGE branch yet? I haven't even run 'make test', as I ran out of time last night. I only checked PGE's 'make test'.
chromatic I haven't.
00:18 cconstantine joined
Tene I guess that falls into part of "need to merge", though, so that's fine. 00:18
chromatic Yeah. I can't imagine anything else would fail that I didn't already fix on the other branch. 00:19
If so... well, we can fix that too.
00:23 mikehh joined
Tene drives home and errands and be back later. 00:24
00:25 cconstantine joined 00:28 mikehh joined
cotto_work seen kid51 00:49
purl kid51 was last seen on #parrot 20 hours, 55 minutes and 36 seconds ago, saying: must sleep
00:50 cognominal joined 00:52 jsut_ joined 01:02 cconstantine joined, hercynium joined 01:10 bacek joined 01:19 cognominal joined 01:20 davidfetter joined 01:44 abqar joined
dalek tracwiki: v147 | chromatic++ | WikiStart 01:45
tracwiki: trac.parrot.org/parrot/wiki/WikiSta...ction=diff
TT #451 closed by chromatic++: Anticipated changes to bytecode before 2.0 01:50
TT #602 closed by chromatic++: testing sprint (especially opcodes)
TT #603 closed by chromatic++: documentation sprint
TT #606 closed by chromatic++: prune c data structures
01:52 mariano joined 02:19 mtk joined, mtk left 02:40 kid51 joined 02:45 cconstantine joined 02:56 japhb joined
kid51 cotto_work: ping 02:57
03:08 davidfetter joined 03:19 plobsing joined
kid51 must sleep 03:27
purl $kid51->sleep(8 * 3600);
03:35 cotto joined 04:05 TonyC joined 04:08 nopaste joined 04:18 nopaste joined 04:25 bacek joined 04:27 JimmyZ joined 05:10 abqar joined 06:08 bacek joined
dalek rrot: r43523 | tene++ | branches/pge_no_namespace_methods (2 files):
PGE/TGE method/ns fixes
 07:00
07:01 cotto joined 07:05 theory joined
Tene okay, testing this merge before committing. 07:06
I hope we don't need an SVN merge on tt389_fix from pge_no_namespace_methods. 07:07
07:09 uniejo joined
dalek rrot: r43524 | tene++ | branches/tt389_fix (11 files):
Merge the changes from pge_no_namespace_methods up through r43523
 07:16
07:26 brooksbp joined
chromatic I don't think we need an SVN merge. We should be okay. 07:26
As far as I know, everything should pass on that branch. 07:27
07:32 cognominal joined
Tene Nope! 07:33
I've fixed one failure.
There are a few more.
chromatic PGE/TGE or something else?
nopaste "tene" at 76.27.121.193 pasted "failures on the branch" (26 lines) at nopaste.snit.ch/19316 07:34
chromatic Same sort of thing, I'm sure.
They're probably easier to diagnose on the other branch though. 07:35
Tene t/compilers/pge/06-grammar.t is kind of tricky. I'm not sure if it's misusing PGE, or if PGE is broken, or both.
I'm pretty sure the last.
07:36 fperrad joined
Tene Ah, no, it's just misusing PGE. 07:37
chromatic Excellent. 07:38
dalek rrot: r43525 | tene++ | branches/tt389_fix/t/compilers/pge/04-compile.t:
Fix one more test
 07:48
rrot: r43526 | tene++ | branches/tt389_fix/t/compilers/pge/06-grammar.t:
[pge] Fix another test
rrot: r43527 | tene++ | branches/tt389_fix/t/compilers/tge/parser.t:
[tge] Fix another test
07:55 TiMBuS joined
Tene P6object is still failing. :( 07:55
chromatic Try that test on the other branch. That should narrow it down substantially. 07:56
07:59 iblechbot joined
Tene oh... this is the add_method one. 08:01
08:02 mikehh joined
dalek rrot: r43528 | tene++ | branches/tt389_fix/compilers/json/JSON.pir:
[json] Fix PGE usage
 08:05
Tene I think it's down to like 2 or 3 failures left. Testing now 08:10
dalek rrot: r43529 | tene++ | branches/tt389_fix/t/library/p6object.t:
[P6object] Fix one test
 08:21
Tene I think I'm probably done for the night. 08:29
chromatic If I finish my writing, I may be able to poke at it in a bit. 08:31
Tene I've looked at the three remaining test failures, and I don't really get what's happening at all, but I might just be sleepy. 08:32
chromatic Which time zone is yours now? 08:33
Tene Mountain. 08:34
purl mountain is just to the east of the continental divide
Tene I haven't been traveling for a while now.
I'm planning on moving to Pacific in the not-too-distant future. Still trying to work out the details, though. 08:35
chromatic One important detail is to head west, then you'll be here.
Tene Oh man, you just solved like 90% of this for me. :)
chromatic Story of my life. 08:36
Tene mostly looking at the bay area so far. 08:37
nopaste "tene" at 76.27.121.193 pasted "remaining tt389 test failures" (14 lines) at nopaste.snit.ch/19317
chromatic I'll take a look. 08:38
Tene P6object.t *does* pass on pge_no_namespace_methods after you apply that patch on tt389_fix. 08:40
lemme commit that patch there, actually.
chromatic Which patch?
Tene r43529, the latest, the one that changes t/library/p6object.t 08:41
chromatic That looks sane.
Tene or just update pge_no_ns_methods to r43530 08:42
It passes in pnnm, but not in tt389
chromatic I might not have caught all of the culprits in src/globals.c. 08:43
Tene Ahh, that would do it, yes. :) 08:44
And... my headphones finally broke the rest of the way. I think that means it's time for bed. 08:45
08:47 riffraff joined
dalek rrot: r43530 | tene++ | branches/pge_no_namespace_methods (5 files):
Cherry-pick some patches from tt389_fix (r43525-r43529)
 08:54
09:01 bacek joined 09:37 cognominal joined 11:07 lucian joined 11:34 payload joined 11:41 bacek joined 11:52 cconstantine joined 12:06 bacek joined 12:24 cconstantine joined 12:27 ruoso joined 12:37 clinton joined 12:43 bluescreen joined 12:52 bluescreen joined 13:49 payload joined 13:51 smash joined
smash hello everyone 13:51
Coke another good reasons for a 2.0.1 release would be to make sure we can actually cut a point release. =-) 14:03
pmichaud: Glad to hear you have ng working. 14:04
was it all context related?
pmichaud Yes. 14:05
Coke woot. 14:06
lists.perl.org/list/cvs-parrot.html (found via latest NOC post). whee. 14:09
14:10 particle joined
smash pmichaud: testing ng against parrot-2.0 or HEAD ? 14:13
Coke smash: I'm guessing 2.0 14:14
PerlJam Coke: you think making 2.0.1 will be significantly different from making 2.0.0 or 2.1.0 ? 14:15
Coke PerlJam: in that it will involve a branch, yes. 14:19
14:32 cognominal joined 14:35 Essobi joined 14:43 plobsing joined 14:59 mj41 joined
smash Coke: thks, running tests now 15:02
dalek nxed: r367 | julian.notfound++ | trunk/winxedst1.winxed:
for in string in stage 1
 15:18
nxed: r368 | julian.notfound++ | trunk/examples/misc.winxed:
fix example
 15:23
15:26 iblechbot joined
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#31819), fulltest) at r43530 - Ubuntu 9.10 amd64 (gcc with --optimize) 15:40
15:44 theory joined 15:47 Psyche^ joined, davidfetter joined 15:55 payload joined 16:01 brooksbp joined 16:26 Andy joined 16:34 Andy joined 17:07 jsut joined
cotto_work good morning bacek 17:09
darbelo clock?
purl darbelo: LAX: Thu 9:09am PST / CHI: Thu 11:09am CST / NYC: Thu 12:09pm EST / LON: Thu 5:09pm GMT / BER: Thu 6:09pm CET / IND: Thu 10:39pm IST / TOK: Fri 2:09am JST / SYD: Fri 4:09am EST /
17:19 cotto_work joined 18:03 payload joined 18:23 ruoso joined, gaz joined 18:48 cognominal joined 18:51 chromatic joined 19:24 whiteknight joined
whiteknight hello #parrot 19:24
Coke ~~ 19:26
19:29 joeri joined
whiteknight dukeleto: ping 19:30
dukeleto whiteknight: pong!
cotto_work It looks like svn del t/tools/0[12345]*.t && perl tools/dev/mk_manifest_and_skip.pl is enough to get Configure.pl --test passing in one_make. I don't think further effort is justified. 19:31
whiteknight dukeleto: how do we compare floats using the test::more library?
I'm going to start converting t/op/trans.t from perl to PIR, and I want to know best practices
cotto_work The other test doesn't seem to do much either. 19:32
Coke perl:exists::javascript:?? 19:33
(nevermind) 19:36
whiteknight dukeleto: I think we're definitely going to be comparing the floats with a delta 19:40
dukeleto backlogging... 19:41
whiteknight: is takes 4 args 19:42
whiteknight: is(x,y,msg,delta)
whiteknight ah, okay. Awesome
cotto_work Coke, do you see any reason not to delete the pmc2c tests outright? 19:52
chromatic I see no good reason.
cotto_work There's probably some other supporting code that can go too. 19:53
I'll have enough time tonight to clobber it if nobody beats me to it. I've had it in for those tests for a long time. 19:54
Coke cotto_work: I would just ask that you reply to the email i cc'd you on earlier today if you're going to do that. 20:00
to keep jim in teh loop.
cotto_work sure
in the loop is where he will be 20:01
Coke checking those 5 tests...
cotto_work they're the 5 that fail
Coke cotto_work: ENOFILES.
cotto_work sorry
t/tools/pmc2cutils/0*.t
darbelo ITYM "svn rm t/tools/pmc2cutils/0*.t" 20:02
20:02 cghene joined
cotto_work all my codes are perfect 20:03
both of them 20:04
purl both of them are good
dukeleto purl, drink some fire 20:05
purl dukeleto: huh?
Coke 08 isn't failing.
cotto_work right, but it doesn't do very much 20:06
darbelo Kill it anyway.
Coke those tests seem to be doing more than testing PCCMETHOD_BITS
cotto_work yes, but they rely on its presence
s/rely/depend/
Coke At this point, I think removing them from whatever Configure.pl --test calls and moving them to one of the non-standard test targets is sufficient.
the question of whether or not we need a test for our tools that is not "run parrot" is a different one. 20:07
cotto_work I'd strongly prefer to make them go away.
Coke yes, but that's not what this branch is about. =-) 20:08
20:12 kurahaupo joined
cotto_work My thinking is that deleting the tests is the easiest way to move forward with the branch and will mean one less roadblock when work resumes on nqp-based pmc2c and ops2c tools. 20:13
dukeleto this Shelby Moore person seems to be a quack 20:15
cotto_work Was that the first person who responded to your Neko message? 20:16
nm. It was,
.
20:20 jan joined
kurahaupo Good morning from LCA! 20:21
(With apologies to those who aren't being rained on)
dukeleto kurahaupo: mornin' 20:28
whiteknight yay! my first commit in a long time 20:30
dukeleto whiteknight: that always feels good
whiteknight I'm working on open solaris right now, and it refuses to permanently accept parrot's certificate 20:32
the more I use this OS, the less I like it 20:33
for git it refuses to remember my ssh password, even with ssh-agent running
chromatic Could be permissions. 20:34
dalek rrot: r43531 | whiteknight++ | trunk (3 files):
move trans.t to trans_old.t for old perl test. Create new trans.t which is pure-pir to hold the new versions of tests. Convert tests for sin, cos, tan, adding some new tests for specific interesting cases
whiteknight chromatic: I'm sure it could be a lot of things, but the documentation is lousy and I can't really be bothered to dig
I don't use it enough to warrant too much fighting 20:35
Coke I thought we got a valid ssh cert for parrot. 20:39
from a reputable authoritah. 20:40
whiteknight it is valid.
Coke ok. then it's your OS's problem? =-)
whiteknight yes
Coke just making sure.
WOOO
particle++
cotto_work: one could argue that if the tests are testing the right thing, they will actually ENABLE switching to a different implementation. 20:41
cotto_work they aren't
chromatic QED 20:42
whiteknight these trigonometric ops annoy me to no end 20:43
particle if only we had the complete set... 20:44
whiteknight we don't even have a complete set. We're missing csc and cot for starters
I would like to see them all moved out of core, maybe into a dynop
Coke there's a ticket for that, IIRC.
particle yes, there is, somewhere
whiteknight hell, I would like to see the library of them moved and maintained in a completely separate project
get all these things and their godforsaken tests out of the repo 20:45
Coke are those ops in math.ops?
whiteknight yeah
Coke trac.parrot.org/parrot/ticket/449
20:45 davidfetter joined
whiteknight about 80% of the way down the file 20:45
Coke If you want to add more to the list, comment on that ticket.
chromatic They're on the DEP list. 20:46
cotto_work you know what that means
cotto_work hands whiteknight a chainsaw
Coke chromatic: where?
whiteknight a separate mathematical operations project would not be a bad idea. We could add a lot more related ops without worrying about bloat 20:47
particle sounds gsoc-ish
Coke sounds like an afternoon's worth of work.
whiteknight I think it's more rote than that
Coke and moving it to a dynoplib gets us 90% of the way there.
cotto_work +1 to Coke
chromatic The deprecation plan has listed them as "move to dynops" for a while, and it's time.
particle <3 spring cleaning 20:48
cotto_work spring cleaning?
purl spring cleaning is probably www.ibiblio.org/Dave/Dr-Fun/df20020...020402.jpg
Coke chromatic: I don't see a ticket or mention for it other than "math.ops" in the TT I mentioned.
cotto_work good. That's as it should be.
chromatic =item moved to dynop [eligible in 1.1] 20:49
Parts or all of: bit.ops, debug.ops, io.ops, math.ops, set.ops
(the obscure and rarely used parts), sys.ops.
Coke that's the TT I mentioned, yes.
chromatic That's in DEPRECATED.pod too. 20:50
20:50 mikehh joined
Coke ... my point is that while the trig functions are covered by the super-vague ticket, they were never mentioned as something specifically to move, even when we talked about this in #ps a few weeks ago. 20:51
so, add a comment, give folks a few days, /THEN/ rip them out.
(and in the meantime, move over all the ones that we already agreed on.)
particle they were talked about at parrot workshop at yapc::na 2009
Coke particle: if it's not in the ticket, it doesn't count.
particle but a comment period won't make this task any harder 20:52
whiteknight Coke: they did specificaly mention the trascendental functions 20:53
and the trig functions are a subset of that
Coke ... you have me there.
apparently I have been away from math for too long ^W^W long enough. 20:54
in that case, have fun. =-)
(The best part is, I wrote that comment that mentions them. =-) 20:55
particle do you feel transcendent? 21:00
particle notices 4 coverity defects 21:01
cotto_work only 4? 21:02
particle Query: In run 149, defects, classification in {Bug, Pending, Uninspected}; sort: CID(Desc)
4 *new* coverity defects, anyway
oh, no, that's 4 outstanding defects, they're not new 21:03
21:04 mikehh joined
Coke I was just thinking today we haven't seen particle in a while, I should text him. 21:04
and here he is.
particle i recently returned from a 23-day vacation in south america
Coke ... I cannot fathom having 23 days in a row to scrape together that didn't involve $DAYJOB. 21:05
I hope you had a good time. =-)
particle i did, indeed.
summer rocks. 21:06
GeJ Good morning everyone
particle 'course, now i'm thinking of heading to chile this summer, so i can go skiing there...
21:09 bluescreen joined 21:10 mikehh joined 21:18 plobsing joined
Coke particle: 2 vacations in the same year? Madness. 21:22
what are you, european?
NotFound Are you a prince of Monaco? 21:23
particle i work for myself.
and i have no kids.
Coke yah, just jealous. :P 21:25
particle i know :D 21:26
21:27 mikehh joined 21:33 patspam joined 21:38 cotto joined 21:50 mikehh joined 21:54 mikehh joined 21:59 mikehh joined 22:07 Casan joined
dukeleto particle: welcome back 22:07
purl my mind is.. not going.
davidfetter open the pod bay doors, purl 22:09
purl I'm sorry, davidfetter, I'm afraid I can't do that. I still have the utmost confidence in the mission.
22:10 plobsing joined 22:11 brooksbp joined
Coke mmmheheh. 22:11
kotter?
dukeleto Coke: do you need the Heimlich?
22:15 mikehh joined, perlpilot joined 22:16 perlpilot left 22:19 cconstantine joined 22:20 mikehh joined 22:28 mikehh joined, patspam1 joined 22:33 mariano joined 22:37 mikehh joined 22:38 patspam joined 22:43 mikehh joined 22:53 mikehh joined 22:57 cognominal joined 22:59 mikehh joined 23:14 mikehh joined 23:23 mikehh joined 23:28 payload joined 23:38 kid51 joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!