www.parrot.org/ | Parrot 1.4.0 "Mundo Cani" Released!
Set by moderator on 21 July 2009.
00:11 theory joined
Whiteknight urg, I'm still getting one lousy test failure that I'm having trouble tracking down 00:28
chromatic If it's a lousy test, delete it. 00:29
Whiteknight it's one of the PGE tests 00:30
I've already deleted a few lousy ones
t/compilers/pge/perl6regex/01-regex.t 00:31
and it doesn't help that the test is so convoluted that I can't hardly debug it 00:32
00:34 payload joined 00:43 kid51 joined 01:20 hercynium joined 01:34 mikehh joined
mikehh the docs at www.parrot.org are still at 1.3.0 - they should be updated to 1.4.0 or whatever 01:36
01:41 dukeleto joined
dalek cnum-dynpmcs: r120 | darbelo++ | trunk/src/pmc/decnumcontext.pmc:
[DecNum] Initialize the context in the correct order. We were overwriting
 01:46
02:15 wayland76 joined 02:22 TiMBuS joined
dalek rdinal: f988f48 | treed++ | (3 files):
Remove the CardinalAny perlism.
 02:34
02:47 davidfetter joined, janus joined
dalek rdinal: 77d6f8e | treed++ | t/string/ (2 files):
Mark some tests as skip/todo due to issue #22.
 02:57
mikehh All tests PASS (pre/post config, smolder, nqp_test, fulltest) at r40259 - Ubuntu 9.04 amd64 03:07
pmichaud_ The 'ins2' branch in the rakudo repository now builds a working Rakudo from Parrot 1.4.0 on my machine. Any reports from other platforms would be greatly appreciated. 03:10
(there's also a 'make install' target, if people would like to test that) 03:11
cotto from only an installed Parrot? 03:14
pmichaud_ yes. But if you use --gen-parrot then it downloads and builds an installed parrot for you. 03:19
(i.e., the prevous instructions for building Rakudo still work... they just work differently :-)
03:20 donaldh joined
pmichaud_ afk, dinner 03:23
mikehh rakudo (dd5767c) builds on parrot r40259 - make test PASS/make spectest (up to r27720) PASS - Ubuntu 9.04 amd64 03:29
using perl Configure.pl --parrot-config=../parrot/parrot_config 03:31
I build and test parrot separately then cd ../rakudo, make realclean, git pull, perl Configure.pl as above, make, make test, make spectest 03:34
03:39 Zak joined 04:09 jdv joined
dalek rdinal: 0b92deb | treed++ | (19 files):
Add code to Test.rb to permit todo/skip to note the issue number that the problem is associated with. Also update the tests to take advantage of it.
 04:21
04:25 wayland76 joined 04:36 eternaleye joined 04:37 theory joined 05:23 bacek joined
dalek rdinal: 4ef71dd | treed++ | t/splat.t:
One test slipped through the last commit.
 05:46
cotto chromatic, ping 05:57
dalek rdinal: fd55266 | treed++ | Rakefile:
Add support for pluralization in the test suite messages.
 06:03
rdinal: be5f3c4 | treed++ | Rakefile:
Make the test harness watch for "See issue" messages and report how many tests failures are related to each issue. Also add a rule to clean? that will deny if there are any failures without associated issues.
06:09 patspam joined
dalek ose: r77 | Austin++ | trunk/ (12 files):
Updated parser to deal with parrot 1.4 changes
 06:40
ose: r78 | Austin++ | trunk/src/parser/ (7 files):
Updated parser to deal with parrot 1.4 changes
ose: r79 | Austin++ | trunk/library/pct/P (2 files):
Added implementation files for 0.2 milestone
rrot: r40260 | allison++ | trunk/tools/dev/pbc_to_exe.pir:
[cage] Provide description for pbc_to_exe (fix to get it to generate a
06:43 asciiville joined
asciiville test 06:44
cotto I'm thinking about moving to asciiville. Life's simpler there.
asciiville :) 06:45
dalek ose: r80 | Austin++ | trunk/src/parser/Namespace.c=:
Added Namespace class
ose: r81 | Austin++ | trunk/src/parser (4 files):
ncf
ose: r82 | Austin++ | tags/0.1.1:
Tagging progress before significant top-level grammar refactoring.
cotto ncf? 06:46
no cookies found? 06:47
sounds rough
chromatic, reping 06:48
asciiville License question. I wrote a C wrapper library to allow the SQLite3.pir to talk to the sqlite3.dll/libsqlite3.so. Does that go under a Perl artistic license or Parrot artistic license? 06:49
cotto Why would Perl's Artistic License come into play (although I don't know if it'd even matter). 06:54
s/\\./?/
asciiville Want to put it up on Github. My PM group is going to do a Rakudo project and I got self-tasked with making the SQLite xface work. 06:55
I have a workaround that allows the DBDI classes to work under the /ext/SQLite subdir. 06:56
cotto PM? 06:57
purl rumour has it PM is usually better than google
chromatic pong
asciiville Perl Mongers
purl Perl Mongers is at www.pm.org/ or "longer sperm" or "Peg, or Mr. Lens"
cotto chromatic, I'm unclear about how the profiling runcore will know how long a PIR opcode takes. Since we're not running in a emulator like with Callgrind, do we have a way to get instruction counts, or are we stuck with simple timing information?
chromatic Each PIR opcode counts as one instruction; we'll treat them that way. 06:58
cotto right. I mean for relative costs.
chromatic Best we can do I fear is microsecond timing. 06:59
I don't know if we can stuff that in the callgrind format somehow.
cotto asciiville, you might ping allison about that, but I think the only difference between the licenses is the name of the entity at the top.
(even then, it looks like LICENSE says "Perl Foundation" 07:00
)
chromatic Because the Perl Foundation wrote AL 2.0.
cotto chromatic, ok. that'll be way better than nothing, even if it's not ideal.
asciiville ah. much obliged. I wasn't really sure since it took some PIR changes as well as Rakudo changes to make it work. 07:01
chromatic Right. At some point people have to benchmark their code for iterations per second.
cotto also, the callgrind format looks non-iterative. We might want to spit out data in a format that can be written in realtime then postprocess it into something callgrind-compatible. 07:03
chromatic I have tried and failed to understand the format three times.
cotto it'll take me another pass before I feel good about it, but I think I have the basic idea. 07:04
I'm glad it's documented, but it's not the most readable spec. 07:05
I think part of getting it solid will be redocumenting it on our wiki. 07:06
chromatic At least the parts of it we need to produce.
cotto Yeah.
chromatic Part of my trouble was that so much of it seems optional.
dalek rrot: r40261 | allison++ | trunk/ports/debian (3 files):
[debian] Update for the 1.4 release.
 07:07
cotto It's very fuzzy in that way. Nailing it down to the necessary subset will help.
Who feeds docs.parrot.org? 07:09
It's still on 1.3.0
heh. I imagine that we could set up a cron job to update it after the monthly release. 07:10
dalek rdinal: db841ed | treed++ | Rakefile:
Move the issue stats report to a separate target: test:stats, which depends on test:all. All other stats still come with test:all.
 07:16
rrot: r40262 | allison++ | trunk/ports/debian/changelog:
[debian] Simplified description of change.
 07:17
07:20 donaldh joined 07:24 iblechbot joined 07:27 wayland76 joined 07:34 mokurai left 07:51 flh joined
nopaste "tene" at 24.10.252.130 pasted "Curses unicode FAIL on parrot" (13 lines) at nopaste.snit.ch/17385 08:06
dalek rdinal: 741da70 | (Joeri Samson)++ | (3 files):
Implement Array.reject and .reject!; tests included.
rrot: r40263 | fperrad++ | trunk/t/tools/pgegrep.t:
[pgegrep] fix test on Windows.
 08:15
08:27 bacek joined 08:32 masak joined 08:51 TonyC joined 08:58 nopaste joined 09:53 jimmy joined 10:04 Whiteknight joined
mikehh All tests PASS (pre/post config, smolder, nqp_test, fulltest) at r40263 - Ubuntu 9.04 amd64 10:15
rakudo (dd5767c) builds on parrot r40263 - make test PASS/make spectest (up to r27722) PASS - Ubuntu 9.04 amd64 10:16
dalek rrot: r40264 | whiteknight++ | branches/bsr_jsr_ret/runtime/parrot/library/PGE/Dumper.pir:
[bsr_jsr_ret] fix a stupid mistake on my part in PGE/Dumper.pir, which fixes the last failing test. However, this test now hangs for some reason I think, so that needs diagnosing
 10:39
mikehh looking at smolder - a lot of packfile tests seem to be failing on i386 10:48
11:13 wayland76 joined 11:19 payload joined 11:20 donaldh joined 11:46 kid51 joined 11:49 payload1 joined 11:50 bacek joined
bacek o hai 11:56
12:43 HG` joined
kid51 Middle of the summer. Everything very quiet here. 12:45
Other channels silent as well. 12:46
bacek kid51: erm... It was winter this morning.
kid51 well, middle of summer in northern hemisphere
bacek :) 12:47
kid51 bacek: Does winter in Australia ever imply .... snow?
bacek hmm... You have to search it. Really hard. 12:48
Today was about +24. 12:49
www.bom.gov.au
Sydney\t17°\tPatchy rain developing.
(And I lied... +24 was on Thursday) 12:50
kid51 24 C is winter temperature that's warmer than current summer temperature in NYC
bacek kid51: it's very humid here.
So, something like +10 is worse than -10 in Moscow. 12:51
kid51 Where in Oz are you?
(Current NYC temp is 21 C/69 F)
bacek Sydney 12:52
Currently ~12C. 12:53
jonathan read that as -12C at first!
bacek At -12C good half of Sydney will freeze to death 12:54
13:02 iblechbot joined
wayland76 You can get snow in Tasmania, or in the mountains 13:02
We have a mountain somewhere
And lots of hills that we call mountains because we don't have any *real* mountains
Night all
afk &
13:46 fdorothy joined 14:07 tetragon joined
pmichaud_ good morning, #parrot 14:20
kid51 pmichaud_ good morning. Where are you located terestrially today? 14:31
terrestrially 14:32
pmichaud_ At the moment I'm at the San Jose airport 14:33
San Jose seems to have a thing for very long, straight hallways
I thought the hallways at the convention center were long. They're short compared to the gates at the airport :-)
Anyway, 7 hours from now I should be back in Texas 14:34
kid51 They built it out of lemon groves, where all the trees were in straight rows. Consistency in design.
7 hours California to Texas???
pmichaud_ My flight doesn't leave for another hour (got here early -- no point in hanging around the hotel)
and I'm flying Southwest, which means it's actually SJC <-> SAN <-> DAL 14:35
and there's undoubtedly a stop for SAN <-> DAL, probably in PHX, ELP, or ABQ
(actually, not PHX)
kid51 pmichaud_ So, I guess you don't sleep much. For most of us, that's what hotels are for ;-)
pmichaud_ I don't sleep much. At least, not for long stretches at a time. 14:36
by 5:45am I was already awake
kid51 Are there no direct flights SJC - DFW ?
pmichaud_ oh, there are, but they would've cost me more
this one was a frequent flyer ticket that if left unused would've expired in 10 days
kid51 How did talk/conference go? (I saw your lightning talk slides.) 14:37
pmichaud_ the conference was outstanding
very good talks all around, lots of interest in Rakudo and Parrot
lots of good discussions in hallway track and other places
kid51 Was there any time to resolve Parrot issues with allison et al.? 14:38
pmichaud_ No, both allison and I were pretty busy. But I didn't really have any outstanding issues to discuss with allison 14:40
I've already filed the tickets for the things that are high-priority at the moment, so they're just waiting for her to have some time to work on them :-)
Gloria (Wall) and I were chating with allison immediately after the final keynote session, and Allison said "after today I have nothing to do for the next 3 months". To which I responded "Oh, really?!?" :-) :-) 14:41
(she of course meant she had nothing "non-Parrot" to do for the next 3 months :-)
kid51 okay. I know that both wayland76 and I would like some guidance on install-related issues -- preferably in the form of well-defined tasks that we could grind out
moritz that's great news for us, I think
pmichaud_ it's very good news here. And Allison is expected to be at YAPC::EU, so we'll have more time to meet then (when she's not distracted by running a conference) 14:43
(install issues): I put my updates for Rakudo install stuff into its 'ins2' branch
The branch appears to be working for (1) building Rakudo from an installed Parrot and (2) installing Rakudo itself 14:44
moritz but not with the parrot build dir?
pmichaud_ So, what Rakudo really wants there is more testing on more platforms.
Correct, it doesn't work with the parrot build dir. That's the bigger component of "install-related issues" that needs to be addressed. 14:45
Ideally we want Parrot's "make" target to create an install image within the build directory
and then "make test" should test that install image
and "make install" should copy that image to the install location 14:46
Allison and I are definitely in agreement on that point. 14:48
when that takes place, it will then be easier for HLLs to be able to run from either an installed parrot or from the build-copy of Parrot (because the build-copy will have the same structure as an installed parrot) 14:50
okay, time to board 14:56
15:20 donaldh joined 15:21 Whiteknight joined 15:26 MoC joined 15:28 MoC joined 15:41 payload joined, mokurai joined
dalek rrot: r40265 | allison++ | trunk/ports/ubuntu (3 files):
[ubuntu] Preparing for 1.4 packages.
 15:44
15:45 cognominal joined
dalek rrot: r40266 | allison++ | trunk/ports/ubuntu/control.in:
[ubuntu] No longer need dependencie on quilt, since not applying
 16:26
purl hmmm... patches is trac.parrot.org/parrot/newticket
16:48 Psyche^ joined 16:51 Whiteknight joined
dalek kudo: 240b984 | (Kyle Hasselbacher)++ | Test.pm:
report expected and actual results after "is" test failure
 17:03
17:18 mikehh_ joined 17:25 kj joined 17:31 kid51 joined 17:35 chromatic joined
dalek rrot: r40267 | whiteknight++ | branches/bsr_jsr_ret (90 files):
[bsr_jsr_ret] updating to trunk r40266. All tests are passing
 17:48
Whiteknight hello #parrot 17:51
dalek l: 03a4f6d | fperrad++ | config/makefiles/root.in:
update config variable names
 17:57
l: c547886 | fperrad++ | src/builtins/say.pir:
use the opcode 'iter' instead of a instantiation of 'Iterator'
Whiteknight holy shit, the book is out! 18:05
www.onyxneon.com
jonathan Cool retro cover! 18:08
cotto Nice.
treed Looks like a college textbook. 18:09
chromatic That was the inspiration. 18:10
"Design Patterns Explained" 18:11
treed How long before it's out of date? :-P 18:14
Whiteknight is so happy he's about to plotz
treed: it probably already is :)
treed Haha.
And thus the problem with print in Computer Science.
cotto How much of each purchase does PaFo get? 18:15
Whiteknight I would assume 100% of profits 18:17
well, after my exorbitant royalties are taken out :)
cotto Heh. It must be pretty cool to see your name on the cover. 18:18
even if the link for your name is 404
Whiteknight yeah, it's cool to the max 18:19
I'm going to have to buy like a dozen copies now just because family members are going to want copies
chromatic If you send me a bio, I'll update your page on the site. 18:21
cotto Whiteknight, send one in lolspeak. 18:22
Whiteknight "Andrew is freaking awesome. One time he killed a bear with a fork and a rolled-up newspaper. Oh, and he programs like a wizard too"
:)
chromatic I can't find my suspenders of disbelief!
Wait, wrong window....
Whiteknight posting things to the correct window is so burgeousie
cotto It works in this window too. 18:24
Whiteknight As they say on Wikipedia: [Citation Needed] 18:25
chromatic Funny that!
Whiteknight amazon is already selling the book for 4$ less 18:26
18:27 mikehh__ joined
jonathan notes that next to "get copy of Parrot book" in his todo list 18:28
;-) 18:29
treed Tempting.
purl i guess Tempting. is there wifi there?
jonathan wonders if allison will bring any to YAPC::EU
kid51 But where's the "Buy it now" button? 18:30
treed Mine's right up top. 18:31
And on the right. 18:32
purl i think on the right is steph, lindsay's friend
treed There's a marketplace selling selling it for 13 something.
kid51 bookpool.com's web page has expired: www.bookpool.com/ 18:33
Are they out of business?
apparently so 18:36
18:39 iblechbot joined
kid51 Yuk. Wikipedia deleted my entry on 'phalanx' as a verb, claiming it was unsourced and not a frequently used term. 18:40
Whiteknight I'm at a Wikipedia conference right now 18:44
literally, I could raise the issue with Jimmy Wales himself if you wanted :)
cotto Is it me or did it just get *more* nerdy in here. 18:45
Whiteknight yes, +5 more nerdy, to be exact
chromatic +10 creepy 18:46
Whiteknight yay! enough points to become a paladin
kid51 Whiteknight: where exactly is that conference? 18:58
Whiteknight NYU, the vanderbuilt building 19:00
kid51 Jeez, we could have arranged an emergency ny.pm social meeting for you! 19:01
Whiteknight haha, yeah. I wish I could stay up here for longer, but it's only a day trip 19:02
I'm not nearly as big in perl as you are either
kid51 I once said to David Adler, "David, my ambition is to become so well noted in the Perl world that London.pm schedules an emergency social meeting when I come to town." 19:03
dha replied: "Jim, you're setting the bar very low, aren't you?"
Whiteknight some people will take any excuse to party 19:04
19:12 masak joined
kid51 Whiteknight: re bsr_jsr_ret branch: Is there any reason to expect test failures to show up in 'make fulltest' than would not show up with 'make test'? 19:13
i.e., with other cores?
Whiteknight i don't think so, no. I only removed those three opcodes
although I haven't run a fulltest, so maybe I should to make sure something crazy like JIT wasn't relying on it 19:14
kid51 I'll give it a fulltest on Linux, but only regular make test on Darwin 19:15
Whiteknight i'm on linux and running a fulltest now, so you don't need to worry with it 19:18
19:19 mberends joined 19:20 donaldh joined 19:25 mberends joined
kid51 bsr_jsr_ret passes 'make test' on Darwin/PPC: smolder.plusthree.com/app/public_pr...ails/25539 19:38
Whiteknight yay! that makes me happy 19:50
dalek rrot: r40268 | whiteknight++ | trunk (102 files):
[bsr_jsr_ret] merging this branch into trunk. bsr, jsr, and ret opcodes are out now, any code in external repos that were using them need to update. Probably will require a make realclean/Configure to update everything.
 19:54
rdinal: 63aa1d8 | (Danius Michaelides)++ | (3 files):
Implement Array.index/rindex, with tests.
 19:57
chromatic No such file: t/native_pbc/integer_1.pbc
dalek rrot: r40269 | whiteknight++ | trunk/DEPRECATED.pod:
[bsr_jsr_ret] remove note from DEPRECATED
chromatic Oh, that was my mistake.
Whiteknight next step is to rip out stacks.c 19:58
and that's going to be awesome
dalek rdinal: cc4f210 | (Danius Michaelides)++ | (3 files):
Implementation of Array.values_at, with tests.
 20:04
cotto Whiteknight, I think you forgot the PBC_COMPAT bump 20:06
kid51 Whiteknight: I got two FAILs in branch on fulltest 20:10
nopaste "kid51" at 71.247.43.247 pasted "make fulltest failures in bsr_jsr_ret branch" (46 lines) at nopaste.snit.ch/17388 20:14
cotto Too late. It's been merged. All further problems will be ignored. 20:20
kid51 Same failures in trunk: t/examples/catchall.t t/examples/subs.t t/benchmark/benchmark.t 20:24
mikehh benchmark_tests, examples_tests FAIL, All others PASS (pre/post config, smolder, nqp_test, fulltest) at r40269 - Ubuntu 9.04 amd64 20:25
failures the same as kid51 in nopaste.snit.ch/17388 20:26
chromatic Hm, config failed to find my bison, which is why I couldn't rebuild IMCC's parser. 20:31
20:38 fperrad joined
dalek rdinal: 8f8f5d6 | tene++ | src/ (2 files):
Disallow use of bare blocks
 20:39
mikehh rakudo (240b984) builds on parrot r40269 - make test PASS, 1 FAILure in make spectest (up to 27729) - t/spec/S03-operators/range.rakudo passes tests then exits with Segentation fault 20:44
fperrad Whiteknight, have you seen my recent patch in TT #113 ? 20:46
dalek rdinal: e6bea57 | tene++ | (4 files):
Add a lambda function that just returns its arguments. Clean up several test failures
 20:50
chromatic hates IMCC 20:57
20:57 fdorothy joined
Tene pirc? 20:58
purl it has been said that pirc is kjs's
chromatic If I thought it would work, yes.
It's still incomplete, last I know.
moritz let me guess... right now it has a bus number of 0? 21:01
chromatic Yep.
Tene I mostly understand it. 21:04
I wasn't able to successfully fix a bug with it, though.
So maybe bus number of 0.8
cotto seen kj? 21:10
purl kj was last seen on #parrot 44 days, 3 hours, 4 minutes and 8 seconds ago, saying: (forgiving, or maybe just keeping silent) [Jun 11 18:00:25 2009]
cotto seen kjs?
purl kjs was last seen on #perl 1 years, 32 days, 5 hours, 31 minutes and 5 seconds ago, saying: yo [Jun 23 15:33:30 2008]
chromatic Ah, maybe that did it. 21:17
I still hate IMCC. 21:19
Fiddly, fiddly.
cotto The callgrind output for a simple 3-function program has a ton of noise. 21:21
21:28 bacek joined
cotto Mmmm. 17M of debugging symbols. 21:30
dalek rrot: r40270 | chromatic++ | trunk (9 files):
[IMCC] Changed sub lookup by lexical to use find_sub_not_null. See TT #536,
 21:32
chromatic There, there's a Rakudo blocker fixed. Now I can run off again. 21:33
dalek TT #536 closed by chromatic++: imcc optimization interferes with lexical subs
jonathan chromatic++ # thank you thank you! :-) 21:34
chromatic How many tests does that bump Rakudo?
jonathan chromatic: At the moment, probably only a couple.
chromatic: However, it was a blocker for a refactor we need to do.
chromatic: A lot of stuff we are looking up via packages, we should be looking up via the lex pad. 21:35
And this woulda royally got in the way.
21:35 payload joined
jonathan Since we need to do that refactor soon, this is well timed. Thanks again. :-) 21:35
dalek rrot: r40271 | chromatic++ | trunk/t/op/lexicals.t:
[t] Added tests for TT #536, looking up lexical subs.
 21:36
chromatic That's why I'm here. 21:37
There may be a bug in nested lookups. I don't think it's the case, but it may be there. Let me know if so.
jonathan Will keep that in mind, thanks. 21:40
22:07 skv joined
bacek good morning 22:29
purl And good moroning to you, bacek.
22:30 asciiville joined 22:37 rg1 joined
dalek rtcl: r544 | coke++ | wiki/SpecTestStatus.wiki:
Update after recent unfudge run.

several regressions remain.		 22:47
rtcl: r545 | coke++ | trunk/tools/tcl_test.pl:
Remove unhelpful CHLD handler.
rtcl: r546 | coke++ | trunk/config/makefiles/root.in:
fix "checkout from tcl cvs repo" target.
22:49 fperrad left
dalek TT #875 created by aardvark++: [PATCH] Eliminate duplicate "2009" in PIR book documentation 22:51
rtcl: r547 | coke++ | trunk/tools/tcl_test.pl:
fix test harness for normal usage. (without passing in --skip)
 22:52
23:10 theory joined 23:20 donaldh joined
dalek TT #875 closed by bacek++: [PATCH] Eliminate duplicate "2009" in PIR book documentation 23:30
rrot: r40272 | bacek++ | trunk/docs/book/pir/ch01_introduction.pod:
[docs] Remove redundant "2009" in pirbook. Spotted by aardvark++
 23:32
23:33 Maddingue joined 23:43 tetragon joined
Coke docs.parrot.org has been updated to 1.4.0 23:50
msg bacek I'd put out a call to whoever is generating those smolders and ask for more intel.
purl Message for bacek stored.
bacek Coke: I'm almost ready to close #823 as duplicate of #357 23:52
Coke msg chromatic link to b&n from onyxneon is a 404.
purl Message for chromatic stored.
Coke msg chromatic Also, www.amazon.com/gp/help/customer/dis...d=14061791 may or may not be nifty. 23:56
purl Message for chromatic stored.
Coke (*@#&$@#(*$&@# 23:57
it is exceedingly hard to get a "clean" spectest run when things are segfaulting one way and not another.
dalek rtcl: r548 | coke++ | trunk/docs/spectest- (2 files):
log another run of spectests.

run standalone.		 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!