Parrot 2.5.0 Released! | parrot.org | Log: irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | Priorities: review experimental features for promotion or removal, fix 'make html' (talk to Coke), update tutorial (talk to tcurtis), pre-release testing.
Set by moderator on 13 July 2010.
dalek rrot: r48106 | Chandon++ | branches/gsoc_threads (17 files):
[gsoc_threads] Merge from trunk.
 00:08
00:24 bacek joined 00:31 kid51 joined 00:36 nbrown joined
kid51 gsoc_threads branch: Getting test failures in t/pmc/pmc.t and t/pmc/scheduler.t 00:39
00:41 aloha joined
kid51 smolder.plusthree.com/app/projects/...ails/34937 00:42
00:48 bacek joined
Chandon kid51: If we're doing automated testing I should probably set those to ignore or something. 00:53
kid51 But aren't we *always* doing automated testing? 00:54
Chandon Didn't know that. Probably should have assumed.
kid51 Any branch in the repository is fair game for testing. 00:55
And for those of us who do Smolder tests, the difference between testing trunk and testing a branch is just one word.
I don't claim any insight into the subject of that branch ... 00:56
... but I do know that tests that are passing in trunk but failing in branch are a red flag. 00:57
Chandon That sounds like a reasonable assumption. 00:59
kid51 I'm glad to see you've got a number of TODO items identified in your branch. 01:01
trunk at r 48106: make fulltest PASS linux/i386 01:03
01:26 bacek joined 02:14 bacek joined
kid51 make fulltest PASS also on darwin/ppc 02:14
02:15 aloha joined 02:17 hercynium joined
dalek rrot: r48107 | Chandon++ | branches/gsoc_threads (9 files):
[gsoc_threads] Task queue only has tasks in it. Also, skip scheduler tests.
 02:36
02:54 janus joined, potato joined 03:13 jsut_ joined 03:41 LoganLK joined 04:09 yobert joined 04:18 TiMBuS joined 04:27 theory joined 05:08 TiMBuS joined 05:09 atrodo joined 05:29 snarkyboojum joined 06:04 he_ joined 06:08 uniejo joined
dalek rrot: r48108 | khairul++ | branches/gsoc_instrument (4 files):
Added test for gc event class.
 06:57
cotto yay for tests 07:01
07:26 baest joined 07:59 AzureStone joined
dalek rrot: r48109 | moritz++ | trunk/CREDITS:
[CREDIT] update Audrey Tang's nick name, by her request
 08:03
08:19 dalek left, dalek joined 08:36 AndyA joined 09:35 kj joined
dalek p-rx: 7b1280f | pmichaud++ | src/ (3 files):
Add backtrack indicators to Cursor, Regex "pass" nodes.
 09:38
p-rx: eb16d23 | pmichaud++ | src/ (2 files):
Initial version of backtracking. Adds Cursor.next and helper methods to restart
p-rx: b796e3d | pmichaud++ | src/Regex/Cursor.pir:
Be more robust about backtracking into failed/nonbacktracking cursors.
p-rx: a1c5ce3 | pmichaud++ | src/ (2 files):
Add backtracking into subrule code. Split !cursor_next from Cursor.next .
p-rx: c2c2fbe | pmichaud++ | src/PAST/Compiler-Regex.pir:
Eliminate (unsaved) uniquereg used for markers in greedy/ratchet quantifier.
p-rx: 7b0bf98 | pmichaud++ | src/PAST/Compiler-Regex.pir:
Make sure uniquereg for frugal quantifiers is reset properly if re-entering
p-rx: cf03cdb | pmichaud++ | build/PARROT_REVISION:
Bump PARROT_REVISION.
p-rx: 6aedfc2 | pmichaud++ | src/stage0/ (4 files):
Update bootstrap.
kudo: 9005478 | pmichaud++ | build/PARROT_REVISION:
Bump PARROT_REVISION to get regex backtracking fixes from nqp-rx. Fixes RT
 09:40
09:42 jsut joined
dalek rrot: r48110 | pmichaud++ | trunk/ext/nqp-rx/src/stage0 (4 files):
[nqp-rx]: Update bootstrap with regex backtracking fixes.
 09:46
11:05 lucian joined 11:36 snarkyboojum joined 11:38 clinton joined
dalek kudo: 8d8114f | moritz++ | src/Perl6/Compiler.pir:
print parrot version in --version output, and athe revision only if != 0
 11:43
12:15 khairul joined 12:22 whiteknight joined
whiteknight good morning, #parrot 12:28
12:49 jsut_ joined
Coke whiteknight: zzzzzzzzzzzzz 12:51
whiteknight Coke: aaaaaaaaaaaaaaaaaaaaaaaa 12:54
12:55 ambs joined 12:57 bluescreen joined 13:08 bluescreen joined 13:23 khairul_ joined 13:31 bluescreen joined
Coke khairul: fixed your blog in planet parrot. 13:32
13:52 JimmyZ joined 13:53 macroron joined 13:54 bluescreen joined
whiteknight purl msg ash_: Have you seen github.com/graydon/rust ? it's an interesting language with an interesting interface to C function calls. Looks like it creates a separate stack to call them on. Don't think it effects your project, but definitely worth a look at. 13:54
purl Message for ash_ stored.
13:56 tcurtis joined 14:01 bubaflub joined 14:07 Andy joined
dalek kudo: 40c6ec7 | (Š’ŃŃ‡ŠµŃŠ»Š°Š² ŠœŠ°Ń‚ŃŽŃ…ŠøŠ½)++ | src/Perl6/Actions.pm:
Implement nested Whatever-currying

much (suggested by jnthn++, conducted by moritz).		 14:14
rakudo: cf5f19a | moritz++ | CREDITS:
14:15 dalek joined
whiteknight sounds like something suspicious you could get at an indian restaurant: "I'll have the whatever curry" 14:18
moritz is that the curry with last week's left-overs? :-) 14:20
atrodo that's not something I'd order. I'm hit or miss on the curry anyways
14:22 theory joined 14:24 bubaflub joined
Coke one more day for bugfixes and platform reports. 14:34
I will probably cut 2.6 in the evening. 14:35
(tomorrow)
dalek uaak-tutorial: 6133e26 | tcurtis++ | doc/tutorial_episode_6.pod:
Remove superfluous {*} in Ep 6
 14:44
uaak-tutorial: abe0ce4 | tcurtis++ | doc/tutorial_episode_6.pod:
$(...) -> .ast in Ep 6.
nxed: r561 | NotFound++ | trunk/examples/fly.winxed:
improve fly example
 14:58
15:04 snarkyboojum joined 15:06 particle left 15:07 particle joined 15:10 [1]Casan joined, theory joined 15:16 particle left 15:19 kthakore joined
kthakore hi folks 15:19
bacek_at_work: around? 15:20
15:20 particle joined
Coke just had 3 smolder failures come through. 15:20
These are the ones that fail on a regular basis. can someone todo these tests?
15:26 davidfetter joined
dalek kudo: 2f4733e | moritz++ | src/Perl6/Grammar.pm:
implement NYI messages for six different phasers
 16:05
16:08 eternaleye joined 16:12 ruoso joined 16:21 whiteknight joined
dalek uaak-tutorial: cde4870 | tcurtis++ | doc/tutorial_episode_3.pod:
Refactor to be more correct to the grammar in Ep 3.
 16:30
uaak-tutorial: 7827432 | tcurtis++ | src/Squaak/ (2 files):
Make the appropriate changes to the appropriate grammar in accordance with
purl hmmm... previous commit is the 'fix'
Coke What is the point of including the name of the file in the NAME section of POD? 16:37
dalek rrot: r48111 | khairul++ | branches/gsoc_instrument (4 files):
Add NULL check when destroying InstrumentVtable.
 16:40
nxed: r562 | NotFound++ | trunk/examples/fly.winxed:
add automatic mode to fly example
arnsholt Hmm. Writing tests for backtracking is hard 16:45
Coke also wonders what the diff between NAME and TITLE is,. 16:51
tcurtis Coke: Don't forget about Name. 16:53
16:55 Mythmon joined
Mythmon Coke: ping 16:56
hmm. since he doesn't seem to be around, has anyone noticed an issue where when you edit a page on the trac.parrot.org wiki, you get logged out? 16:59
I am a sysadmin for the server its on, I am trying to fix it, but I can't reproduce the bug myself.
NotFound Mythmon: tell people to check his system clock. 17:00
cotto_work ~~ 17:01
Mythmon NotFound: ?
NotFound Mythmon: In my experience, that is the cause of that kind of problems 99% of times. 17:02
Mythmon Ah. I was expecting that it had to do with our caching service being a bit to aggressive and caching the page that commits the edit.
NotFound Mythmon: don't know, but checking clocks as first step doesn't harm. 17:03
Mythmon true. i will have him check that. 17:05
dalek uaak-tutorial: d185c17 | tcurtis++ | doc/tutorial_episode_6.pod:
Some further changes in Ep 6.
NotFound japhb: ping 17:07
japhb NotFound, pong
(Here by the miracle of Telepresence!"
NotFound japhb: check last revision of fly.winxed if you want a nice opengl demo
japhb NotFound, ah, cool, thanks! 17:08
kthakore hi NotFound
japhb Needs an update of parrot, too?
kthakore hi japhb
japhb Hi, kthakore!
NotFound japhb: no, last you tried will be fine
japhb OK, cool
particle Coke: Mythmon: you should talk
kthakore japhb: hi
particle Mythmon: thanks for joining. coke asked me to submit the request for trac.parrot.org
Mythmon ah. 17:09
17:09 [1]Casan joined
kthakore messages? 17:09
purl To access purl's messages, msg me with the word "messages".
dalek rrot: r48112 | coke++ | branches/html_cleanup/lib/Parrot/Docs/PodToHtml.pm:
Default title to what POD::Simple::HTML thinks the title is.
 17:13
rrot: r48113 | coke++ | branches/html_cleanup (12 files):
Make POD conform to 'make html' expectations a bit more.
uaak-tutorial: a12ad27 | tcurtis++ | doc/tutorial_episode_6.pod:
Make sub_call a statement.
 17:17
uaak-tutorial: 1f22fac | tcurtis++ | src/Squaak/ (2 files):
Make sub_definitions work the way they should in the example compiler.
tracwiki: v6 | coke++ | CleanupMakeHtml 17:18
tracwiki: track latest changes WRT title.
tracwiki: trac.parrot.org/parrot/wiki/Cleanup...ction=diff
Coke Mythmon: I had that problem days ago. 17:22
I can no longer reproduce it myself.
in the past, it was due to the caching, yes.
There is a reason I did not open a ticket, though. I blame particle for your wasted time, sorry. =-)
dalek uaak-tutorial: 18859f4 | tcurtis++ | src/Squaak/ (2 files):
Add sub calls to example compiler.
 17:23
Coke Alrightee, diffs in 'make html' now down to about 3.9KLOD
Mythmon Coke: no problem. We got the ticket about the time that we had changed something with the cacher, so I blamed that at first. I'm glad to hear it is working. 17:24
kthakore howdy folks .... I am attempting to crack TT#1639
Coke Mythmon: prolly a blip. I'll definitely open a ticket if it recurs. 17:25
Mythmon++ # danke
kthakore but I am unsure what StringBuilder (src/pmc/stringbuilder.pmc) replaces for (src/pmc/string.pmc)
and where does the overlap needs to be changed in StringHandle
anyone?
purl Somewhere, someplace, in some universe, somebody uses whatever you just asked about.
kthakore oh well
I will wait for bacek 17:26
Coke kthakore: Stringhandle uses a STRING as an internal buffer. 17:27
that buffer should be a StringBuilder instead of a String.
the main reason is the guts of push_string(), which stringbuilder does better.
kthakore so STRINGHANDLE ?
Coke whoops. moment. 17:28
whiteknight kthakore: what's the question?
Coke whiteknight: the ticket is vague.
ok. stringHnale has an attr "*stringhandle" - that's the one that needs replacing. 17:29
kthakore whiteknight: yeah ...
purl doo wop doo wop // shang a langa shanga langaa
dalek uaak-tutorial: 2ac5a71 | tcurtis++ | doc/tutorial_episode_6.pod:
Superfluous {*}s in Ep 6.
uaak-tutorial: 9996dfc | tcurtis++ | doc/tutorial_episode_6.pod:
for_statement -> statement:sym<for>
kthakore Coke: ok
whiteknight: Coke can that ticket be clarified?
Coke it's the "puts" method that makes it want the update. 17:30
kthakore ok 17:31
Coke: that narrows done code to look like?
Coke kthakore: you could start by changing the open() method to initialize the stringhandle attr with a StringBuilder instead of a string and seeing what breaks.
japhb NotFound, Finally got a chance to run the fly.winxed demo. Very nice! :-) 17:32
NotFound japhb: thanks 17:33
japhb Did you get a chance to try the multisampling on a system that had HW OpenGL?
NotFound japhb: I looked at some docs and failed to find a way to use it without complicating a lot the code.
japhb Huh. Surprising, that. 17:34
NotFound Most probably my fault. 17:35
Coke I will hack on the html_cleanup branch again this evening. eyes welcome in the meantime. 17:38
japhb NotFound, I have it working here. 17:41
Would you like a diff? 17:42
NotFound japhb: Sure!
nopaste "japhb" at 192.168.1.3 pasted "fly.winxed diff for MULTISAMPLE" (39 lines) at nopaste.snit.ch/22122 17:43
japhb There you go. :-)
particle Mythmon++ # sorry about the snipe hunt 17:44
NotFound japhb: Just that? I'll have confused something in my attempts, then. 17:45
japhb Yup. :-)
(Mis-pasted a constant, maybe?)
NotFound Probably.
kthakore Coke: sory $work came in 17:46
Coke: um ... would be posibble .. kind sir ... to update that ticket with your stuff? 17:47
17:56 bubaflub joined
Coke I would appreciate it if someone else with wiki rights could do that. otherwise I'll do it when I get off $DAYJOB 17:57
kthakore Coke: can you see TT#1639
NotFound japhb: I don't appreciate any difference.
Coke kthakore: yes.
tcurtis kthakore: link and what you want updated?
Coke I have admin rights. just a little busy atm.
tcurtis I think I can edit it. 17:58
japhb NotFound, didn't you tell me you were running with software GL on your usual box?
Coke wonders who carl williams is and why he is spamming me.
NotFound japhb: I've tested on my laptop.
japhb NotFound, what's OS and GPU on the laptop? 17:59
Coke ugh. 18:00
NotFound Ubuntu, Intel Mobile 4 Series 18:01
Coke Ok. I'll update the ticket. :P
kthakore Coke: ok
Coke: tcurtis I updated the ticket
Coke: I ask for clarification on puts() method there 18:02
Coke I do feel that if I bother clarifying too much, I might as well just write the patch. if you have specific questions, those would be good to note.
kthakore Coke: ok
Coke: np
Coke: can atleast confirm the files there?
src/pmc/stringbuilder.pmc string.pmc and stringhandle.pmc? 18:03
I am sorry I am total newb
japhb NotFound, Intel graphics? There's your problem right there, sadly. REALLY poor support of not-absolutely-core features.
Which Ubuntu, OOC? 18:04
NotFound japhb: I'll commit it anyway, then.
japhb NotFound, nod, thx
NotFound japhb: 10.04
japhb has a laptop with 10.04 and nvidia graphics. 18:05
I'll test there as well, just for fun.
Already committed?
Coke ticket updated. 18:06
and I'm just busy today, sorry I'm grouchy. =) 18:07
18:07 joeri joined
dalek nxed: r563 | NotFound++ | trunk/examples/fly.winxed:
enable multisample in fly example, japhb++
 18:07
japhb Ah the joy of a fresh OS reinstall: Want to get something done? First, install all of the packages you had before the reinstall! Three hours later, you're ready to go! :-/ 18:08
cotto_work I love that game. 18:12
Coke seen szabgab? 18:29
purl szabgab was last seen on #p5p 19 days, 8 hours, 36 minutes and 41 seconds ago, saying: ribasushi, need commit bit? [Jun 30 09:52:43 2010]
18:38 mmcleric joined
dalek p-rx: 5725781 | pmichaud++ | (2 files):
Fix protoregex bug with classnames beginning with Q.
 18:42
p-rx: 11cd811 | pmichaud++ | src/stage0/ (4 files):
Update bootstrap.
pmichaud japhb: I maintain a script that installs all of my packages and settings for me. 18:44
japhb pmichaud, that would probably work ... if not for the fact that I changed distros. :-)
pmichaud japhb: ah, that does make it more difficult. :) 18:45
18:45 Mythmon left
pmichaud locally: 18:46
particle sheesh, why doesn't everyone just run things in virtual machines? 18:49
dalek kudo: cd2d123 | pmichaud++ | build/PARROT_REVISION:
Bump PARROT_REVISION to get fix for Q-named classes. Fixes RT #74276.
Coke particle: there's an edge case problem.
particle who cares what your base distro is as long as the vm hypervisor runs there? 18:50
japhb particle, because GPU performance sucks in a VM. 18:51
particle export DISPLAY ;) 18:52
you're right, gpu performance is still one of the weak links in vm infra 18:53
dalek rrot: r48114 | pmichaud++ | trunk/ext/nqp-rx/src/stage0 (4 files):
[nqp-rx] Update nqp-rx with fix for protoregexes causing Q-classes to fail.
Coke is reminded that he has yet to find a valid key/disk combination to run XP on his linux box. 18:54
19:09 whiteknight joined
particle coke: are you running xp to test parrot? 19:10
Coke oh hell no
particle wrong answer
if it's to test parrot, i can give you a key :)
Coke "I would be happy to run xp to test parrot if people bought me a machine for it."
particle i can give you any OS and a key, but the hw you'll need another sponsor for. 19:11
any *windows* OS
Coke no, this is decidedly not for parrot use, I won't abuse your codes. 19:12
dalek kudo: 21d67d5 | jonathan++ | src/Perl6/BacktracePrinter.pm:
Instead of starting to make a list of 'ignore this frame' for the backtrace

its purposes. Then it's just one thing to check.		 19:19
kudo: b62db16 | jonathan++ | src/builtins/control.pir:
Mark die and warn as being invisible to the backtrace printer.
kudo: c0df20c | jonathan++ | src/metamodel/Attribute.nqp:
Mark attribute accessors as invisible frames so the errors give the line number
rrot: r48115 | gerd++ | trunk/lib/Parrot/Install.pm:
to take care about symlinks; according to tt #509; also tested with a FAT32 destination, there the library will be copied
 19:27
tcurtis pmichaud: ping 19:31
pmichaud tcurtis: pong
tcurtis pmichaud: How can I make "PAST::Op.new( $iter, $stepsize, :pirop('add'), :node($/) )" turn into an add_p_p instead of add_p_p_p? More accurately, what's the incantation I really want? 19:32
pmichaud tcurtis: you're wanting to add $stepsize to $iter? 19:33
tcurtis pmichaud: right.
pmichaud PAST::Op.new( $iter, $stepsize, :pirop('add__0PP'), :node($/) ) 19:34
tcurtis Thanks. pmichaud++
pmichaud note that the result of the node will be $iter after the add, too.
the '0' says to use the first (0th) argument as the return value for this node.
as opposed to add__PP which would allocate a new register for the result 19:35
(and not really give what you want anyway :-)
slightly better might be 19:36
add__0P+
which avoids coercing the second argument to a PMC
but that really depends a bit more on HLL knowledge
tcurtis pmichaud: thanks. Now I just need to figure out what else is broken about Squaak's for. :) 19:38
pmichaud afk # heading towards airport
tcurtis is done with Episode 6 of the tutorial. :) 19:50
whiteknight tcurtis++ 19:51
I would love to help with that, but I've been spending every ounce of my available energy getting Kakapo running again 19:52
dalek uaak-tutorial: d640b72 | tcurtis++ | doc/tutorial_episode_6.pod:
$(...) -> .ast in Ep 6.
uaak-tutorial: e372aba | tcurtis++ | doc/tutorial_episode_6.pod:
Fix Ep 6 exercise solution.
uaak-tutorial: 21bbd5c | tcurtis++ | src/Squaak/ (2 files):
Bring example compiler in line with last commit.
19:56 hanekomu_9 joined
Coke if folks can test gerd's patch on various OSen, i'd appreciate it (TT#509) 20:02
20:25 theory joined
dukeleto tcurtis++ # nice work on the squaak tutorial 20:25
tcurtis dukeleto: thanks. Episode 7 is going to be...fun. Time to reverse engineer NQP's optable. 20:26
20:36 jan joined
cotto_work tcurtis++ 20:43
Coke how many eps are there?
tcurtis++
tcurtis 9.
Coke (no one has tackled this in aeons. =-)
cotto_work If I'm lucky, I'll have tuits to help tonight. 20:44
It's nice to see that GitHub displays POD well. 20:46
tcurtis Although 9 doesn't really require any work since it's just an example Squaak program and some exercises without solutions.
cotto_work++: great. :) 20:49
cotto_work 8 doesn't look too bad 20:57
tcurtis, if I ever see you at a YAPC or something, remind me to buy you a beer or something. Awesomeness isn't worth much if nobody knows how to use it. 21:02
or something
purl hmmm... something is really wrong out there :)
tcurtis cotto_work: The "or something" would definitely have to be the one anytime in the next 3 years(unless it's in a country with a lower drinking age). :) 21:04
dalek nxed: r564 | NotFound++ | trunk/examples/fly.winxed:
typo
cotto_work That's why I put it there.
dukeleto tcurtis: Canada's drinking age is 18 :) 21:05
cotto_work: i asked the github peeps to render POD a while ago, glad to hear that it deploys and works 21:06
cotto_work github.com/ekiru/squaak-tutorial/bl...sode_6.pod <-there it is
dukeleto++
Coke contributing to the deliquency of a minor-- !
cotto_work I'd only do that as far as is legal. 21:07
There's plenty of delinquency in Parrot already. 21:08
tcurtis cotto_work: there what is? 21:09
cotto_work github's POD rendering
dalek kudo: 981e44e | moritz++ | t/spectest.data:
run two more test files
 21:10
21:25 particle joined 21:29 hercynium joined 21:44 ash_ joined
dalek rrot: r48116 | mikehh++ | branches/html_cleanup/lib/Parrot/Docs/PodToHtml.pm:
fix codetest failure - trailing whitespace
 21:56
kudo: 0b12127 | masak++ | src/core/operators.pm:
[operators.pm] changed parameter name
 22:03
22:07 payload joined 22:11 bubaflub joined
dalek uaak-tutorial: 012459d | tcurtis++ | (2 files):
Properly panic on redeclaration of variables.
 22:14
22:16 lucian_ joined 22:20 theory joined 22:32 theory joined
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#34957), fulltest) at r48116 - Ubuntu 10.04 amd64 (g++ with --optimize) 22:40
22:42 snarkyboojum joined 22:43 theory joined 23:19 AndyA joined
dalek uaak-tutorial: 2e2f878 | tcurtis++ | (2 files):
Bring back operator stuff.
 23:19
23:20 japhb joined 23:23 whiteknight joined
mikehh rakudo (0b12127) builds on parrot r48116 - make test PASS, spectest_smolder -> #34959 (pugs r31770) FAIL - Ubuntu 10.04 amd64 (g++ with --optimize) 23:39
t/spec/S32-temporal/DateTime-strftime.t - Non-zero exit status: 1 - No subtests run
14 TODO PASSes in 2 files
23:43 davidfetter joined
dalek uaak-tutorial: fd7d28d | tcurtis++ | src/Squaak/ (2 files):
EXPR not expression in example compiler.
 23:49
uaak-tutorial: 5225c4e | tcurtis++ | doc/tutorial_episode_3.pod:
Update explicit mention of expression in Ep 3 to EXPR.
uaak-tutorial: edde33a | tcurtis++ | doc/tutorial_episode_4.pod:
s:ai(1) /expression/EXPR/ in Ep 4.
uaak-tutorial: 64a7fff | tcurtis++ | doc/tutorial_episode_5.pod:
expression -> EXPR in Ep. 5.
uaak-tutorial: 2fde2ee | tcurtis++ | doc/tutorial_episode_6.pod:
expression -> EXPR in Ep. 6.
23:56 Psyche^ joined

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!