Parrot 1.9.0 "Blue-fronted Amazon" released! | parrot.org | Priorities: Add deprecations for 2.0; test platforms; check with HLL implementors | Roadmap: icanhaz.com/parrotroadmap | Latest modified TT's: icanhaz.com/parrotbugs
Set by moderator on 5 January 2010.
plobsing rakudo doesn't like my non-standard --prefix 00:00
ash_ oh, i see why mines doing that, i was trying to track down why opengl doesn't work right in os x, so those are my fault, let me revert those changes 00:01
darbelo plobsing: really? I install parrot to ~/parrot/instdir/ and have no issues there.
plobsing I put it in /usr/local/parrot-$branchname 00:02
will try with ~/parrot-$branchname
darbelo plobsing: is that dir on your path or did you use --parrot-config ? 00:03
plobsing added to path. 00:04
as needed
darbelo Then I have no ide why it fails.
plobsing "Can't read '/usr/local/lib/parrot/1.8.0-devel/include/config.fpmc' : No such file or directory" 00:05
ash_ all tests passed successfully, doing make testr 00:10
testr passed too 00:12
do you want the print out from ./parrot examples/benchmarks/freeze.pasm
darbelo sure 00:13
But just knowing that it didn't segfault is cool. 00:14
ash_ i only ran it once, was it randomly segfaulting?
plobsing nope, it was pretty consistent when it didn't like something
darbelo it was pretty consistent on my box.
ash_ gist.github.com/269884 00:15
chromatic alias freezemem='sudo echo 0 > /proc/sys/kernel/randomize_va_space'
usually helps
ash_ there are a few runs on my computer
00:17 nopaste joined
ash_ need anything else on os x? 00:17
Coke gets an OSX box back. 00:18
now with shiny 10.6
(sadly without several years worth of photos. :(
ash_ timemachine?! 00:19
purl i think timemachine is www.apple.com/macosx/features/timemachine.html
ash_ o.0
Coke timemachine didn't exist in 10.4 00:20
ash_ hey, since your on 10.6 now
try running examples/opengl/shapes.pir 00:21
hmm i have to go, but Coke, if your on 10.6 i think you should be experiencing the opengl bug that parrot has, I haven't found the problem completely yet, i'll let you konw if i find anything important 00:24
Whiteknight Are we going to support Ubuntu 8.04? I can champion Ubuntu8.04-x86 if there is value in doing it 00:26
00:28 TonyC joined 00:31 ZeroForce joined 00:50 PacoLinux joined 00:52 nopaste joined 00:53 TonyC joined 00:58 abqar joined 01:03 payload joined
plobsing cardinal passes w/ pmc_freeze_cleanup on linux x86_64 01:10
01:10 ewilhelm joined
ewilhelm anybody know if there's work underway on a perl6 cpan client? 01:11
Coke ewilhelm: japhb is the guy to talk to from the parrot side of that. 01:12
also, see #perl6 on freenode
ewilhelm how about pir and etc packages? 01:13
japhb ewilhelm, so far installers for Perl 6 (proto and Plumage) have been mostly about pulling from VCS and building, or (using Parrot distutils) making RPMs, DEBs, etc. from the above. 01:14
Plumage installs anything that will install on Parrot.
proto is Rakudo-specific.
plobsing pynie passes w/ pmc_freeze_cleanup on linux x86_64 01:18
Coke plobsing: if you're bored, you could check partcl-nqp. =-)
ewilhelm japhb, what does proto do that plumage doesn't?
plobsing Coke: I'll add it to the list 01:19
japhb ewilhelm, a few UI things, some help with creating new projects (Parrot has a different tool for that), ability to download parrot and rakudo source using a Perl 5 script, build them, and bootstrap itself. A few other bits. Plumage will likely learn to do many of those things. 01:20
01:22 nopaste joined
plobsing rakudo passes w/ pmc_freeze_cleanup on linux x86_64 01:36
rakudo-ng passes w/ pmc_freeze_cleanup on linux x86_64 01:43
01:47 nopaste joined, eternaleye joined 01:50 bacek joined
dalek rrot: r43391 | coke++ | branches/one_make/config/auto/cpu.pm:
remove unused import
 01:54
02:07 TonyC joined
dalek rrot: r43392 | coke++ | branches/one_make (7 files):
Don't use := in makefiles, remove := > = translation.
 02:10
02:17 nopaste joined
plobsing partcl-nqp fails t/cmd_lindex.t 11-12 w/ pmc_freeze_cleanup on linux x86_64 02:17
partcl-nqp passes w/ (pmc_freeze_cleanup merged with latest trunk) on linux x86_64 02:18
cotto_work something added to trunk since the last branch point? 02:19
plobsing that's my guess
pmc_freeze_cleanup branched at r43030 02:20
I tried testing partcl-nqp against r43029 but it didn't like that 02:21
(too old)
Coke msg chromatic - are you the guy for ext/Parrot-Embed ? 02:22
purl Message for chromatic stored.
02:33 TonyC joined 02:37 ZeroForce joined 02:49 cognominal joined 02:54 kid51 joined 03:01 patspam joined 03:14 bacek joined
dalek rrot: r43393 | jkeenan++ | trunk/config/auto/cpu.pm:
Remove unneeded import per suggestion from Coke++.
 03:15
03:15 Austin joined 03:23 Austin_Hastings joined 03:28 jhelwig joined
plobsing anyone have windows for testing pmc_freeze_cleanup? 03:41
04:19 pdcawley_ joined 04:31 ZeroForce joined 04:41 Essobi joined 04:49 jhelwig joined 04:52 jhelwig joined 05:26 bacek joined 06:31 dduncan left 06:36 JimmyZ joined
JimmyZ msg plobsing Windows XP `make smoke` passed with strawberry perl for testing pmc_freeze_cleanup. 06:39
purl Message for plobsing stored.
06:46 Essobi joined 07:06 uniejo joined 07:13 bacek joined 07:16 Essobi joined 07:31 brrant joined 07:52 chromatic joined 08:00 iblechbot joined
chromatic Coke, yes. 08:13
08:18 payload joined 08:25 Essobi joined 08:31 payload joined 08:48 fperrad joined 09:42 plobsing joined 09:43 bacek joined 10:01 ruoso joined 10:15 payload joined 10:16 cognominal joined 10:46 riffraff joined 11:04 cognominal joined 11:12 bacek joined 11:32 bluescreen joined, payload joined 11:48 cotto joined 12:00 iblechbot joined 12:17 ruoso joined 12:18 bluescreen joined 12:21 bluescreen joined 12:22 mj41_ joined, estrabd joined
Coke hey, is ext/ considered experimental? 12:25
12:29 Hunger joined 12:35 barney joined 12:53 mj41_ joined
Util Coke: nqp-rx is in ext/, and ISTR that it is not experimental; it is the next-gen|successor|not-ready-for-primetime-replacement of nqp, but being develeoped in Git (*ext*ternal to our SVN) and synced. Looking also at Parrot::Embed and SQLite, I would think that ext/ just means "external", but if so, I would expect OpenGL to live there too. 13:00
Scanning the source for /experimental/i, especially DEPRECATED.pod, does not show much overlap between experimental status|intent and ext/ residence.
IOW, "I don't know, but no one else answered, so here is my brain dump" :)
13:07 whiteknight joined 13:09 gaz joined 13:18 lucian joined
Coke Util: not-ready-for-primetime-replacement === "experimental replacement" 13:35
but, point taken that the answer is, in general, no. 13:39
14:03 riffraff joined 14:04 plobsing joined
Coke autocompleteme.com/2010/01/06/and-i...els-great/ which I would normally not post here, but this particular auto-complete might be amusing to parrot hackers. 14:25
szbalint it refers to the comment made by Stephen Fry :) 14:27
www.youtube.com/watch?v=9T1vfsHYiKY 14:28
14:37 mj41 joined
Util ++Coke, ++szbalint. Funniest thing I have seen all year. 14:41
14:46 payload joined 15:01 renormalist joined
renormalist Hi! Someone is asking me whether there are *scientific* (academic style) papers about parrot. Are there? 15:03
I remember that in the beginning when it was about register vs. stack based that academic papers were consumed, but not sure whether some were written. 15:06
15:07 bubaflub joined 15:18 theory joined 15:29 ash_ joined 15:40 payload joined 15:47 Psyche^ joined 15:51 darbelo joined
japhb Coke, Util: nqp-rx is quite definitely "ready for prime time" -- it's in ext/, because as Util surmised, its official VCS repo is *ext*ernal to Parrot's repo. OpenGL is not external yet (it will be eventually), because the Parrot NCI system is not yet capable of building thunks on the fly, so OpenGL has to be part of the config process in order to make a list of needed thunks, which are then built as part of the build process. I can't wa 16:01
it until this goes away, but I'm entirely blocked on NCI improvements.
16:08 cognominal joined 16:17 mikehh joined
Coke I think I will leave ext/nqp-rx as a combined makefile (since we are shipping it by default), but may leave everything else in ext alone. 16:31
(and remove the ability to invoke them from the top level make.) 16:32
msg chromatic - it looks like we can avoid generating P::E's .pl file at config time if we just use Parrot::Config. that seem reasonable to you?
purl Message for chromatic stored.
Coke looks like the only usage of dynpmc.pl in the repo now is ncigen 16:36
16:48 dalek joined
dalek rrot: r43394 | coke++ | branches/one_make/tools/build (2 files):
Add deprecation note (from TT #338)
 16:48
rrot: r43395 | coke++ | branches/one_make (4 files):
Remove dynoplibs.pl (TT #338)
16:54 plobsing joined
dukeleto 'ello 17:36
cotto_work hi
Coke go to google. type in 'what is' . the first entry is "what is nqp?" 17:37
dukeleto: hi
purl hola, Coke.
Coke purl, difa 17:38
purl Coke: excuse me?
Coke purl, diaf
purl it has been said that diaf is Die in a Fire
cotto_work Coke, that must be customized. I don't get that even with "what is nq". 17:39
darbelo same here. 17:40
Coke checks. 17:41
whoops, you're absolutely right. 17:42
ash_ Coke: are you logged into a gmail account?
it tailors results based off your search history
Coke yup, and obviously marks them as such. was just having fun playing with autocomplete.
ash_ searching parrot for me doesn't give me a real parrot until like halfway down the first page of search results 17:43
Coke 9and missed it.)
cotto_work There's much fun to be had.
Tene Coke: you've seen the "WTF CENSORSHIP" about google autocomplete, yes?
It doesn't offer results for "Islam is"
Coke nope, missed that. 17:44
dukeleto Tene: interesting
17:47 theory joined
bubaflub dukeleto, Coke: i.imgur.com/BL3Lo.gif is what Tene was referring to 18:01
Tene www.blaghag.com/2009/12/update-goog...rship.html and the earlier post it links to were my source. 18:02
dukeleto bubaflub: thanks. 18:03
bubaflub: howdy!
bubaflub dukeleto: howdy. long time, no chat. how'd the holidays treat ya? 18:04
dukeleto bubaflub: good. just got back to town 18:06
bubaflub dukeleto: nice. how's pdx? we're suppose to get the worst blizzard in 5 years here in IL tonight. 18:07
18:19 payload joined
cotto_work clock? 18:22
purl cotto_work: LAX: Wed 10:22am PST / CHI: Wed 12:22pm CST / NYC: Wed 1:22pm EST / LON: Wed 6:22pm GMT / BER: Wed 7:22pm CET / IND: Wed 11:52pm IST / TOK: Thu 3:22am JST / SYD: Thu 5:22am EST /
dukeleto bubaflub: pdx is dry and in the 40's today 18:26
bubaflub dukeleto: that sounds nice. it was about -2 when i woke up today. 18:27
dukeleto bubaflub: whoa!
bubaflub dukeleto: yeah, i'm feeling it since a few days ago i was in San Diego and Mexico for a bit. 18:28
dukeleto bubaflub: are you going to get snowed in?
bubaflub dukeleto: that's what "they" say 18:29
i heard 7 - 10 inches of snow tonight.
cotto_work good time to build up your snowball arsenal 18:32
bad time for skinny dipping
18:33 chromatic joined
dukeleto bubaflub: make sure your local git mirror is up to date. you might lose internet, but you can still hack on stuff! 18:34
bubaflub haha. if only my iphone tethered then i would have a little more robust internet.
didn't they promise tethering by end of year 2009?
oh well.
dukeleto bubaflub: i had iphone tethering for a year 18:42
bubaflub jailbreak?
dukeleto bubaflub: then my carrier profile expired. got to fix that.
bubaflub: nope, no need to jailbreak
bubaflub: you just download a new carrier profile
bubaflub oh, that's cool. 18:43
does it work with an unlimited data plan?
dukeleto bubaflub: yes
bubaflub: but there are limitations. it only works on older iphone OS's 18:44
bubaflub oh. i think i have the latest
dukeleto bubaflub: you can do it with the latest, but that requires jailbraking 18:51
bubaflub ugh. can't AT&T just come through on its promises?
</whining> 18:59
19:08 jan joined
chromatic Coke, I like the idea. 19:13
dukeleto so what needs doing before 2.0? 19:17
chromatic Fix bugs! 19:19
List deprecations!
Write documentation!
19:26 joeri joined 19:35 plobsing joined
ash_ i have my iphone hacked for tethering, it wasn't that hard 19:37
19:44 bubaflub_ joined 19:47 bacek joined
TimToady phone in 1 minute? 19:58
chromatic yes 19:59
Coke chromatic: eek! 20:06
dalek TT #1403 created by bacek++: [DEPRECATED] OrderedHash PMC 20:13
rrot: r43396 | bacek++ | trunk/DEPRECATED.pod:
Add OrderedHash PMC into DEPRECATED.pod
 20:19
PerlJam bacek: when is this supposed reimplementation taking place? 20:22
20:23 ash_ joined, Hunger joined 20:24 jsut joined 20:35 ash_ left 20:37 bluescreen joined 21:00 bacek joined
bacek o hai 21:00
PerlJam, it almost done in orderedhash_revamp branch. I just forgot to add deprecation note in previous deprecation cycle. 21:01
PerlJam bacek: oh. cool. 21:02
dukeleto bacek: o hai 21:18
bacek dukeleto, aloha 21:24
21:25 ash_ joined
ash_ can parrot be built with a cross-compiler? 21:25
darbelo not easily. 21:26
The only cross-compiling case I kinow of is the RTEMS port and that pretty much sidesteps most of parrot's own build infrastructure. 21:27
ash_ as in yes possible but you have to re-write the make files or ... ?
hmm 21:28
dukeleto ash_: what do you want to do?
ash_ i have a programming class coming up this spring where we have to make an andriod app and i was thinking of trying to build parrot for it and make an app to teach rakudo
i know thats probably out of my league but i was just wondering if its even in the realm of possible 21:29
dukeleto ash_: it is possible, but you would have to hand-write some Makefiles
ash_ 2 issues i see is when you run the simulator its actually building your code for x86, when you build it for the actual phone it targets arm
dukeleto ash_: you can never trust a simulator 21:30
ash_ yup, i know, but that does raise issues, since you'd only really need to target arm for the live app and not for local simulations 21:31
dukeleto ash_: yes, that is an interesting situation
darbelo Hm. Have we ever built a parrot for arm before?
ash_ yeah, i saw a report that someone has at one point 21:32
cotto_work There seems to have been a successful ARM build on NetBSD
dukeleto cotto_work: cool 21:34
ash_ maybe i'll try the iphone first, its also an arm processor, but it uses the gcc out of the box, technically andriod stuff is written in java but there are ways around that to write C if you want, plus i have an iphone so i can actually test it on the hardware
dukeleto ash_: go for it! we will help you as much as we can 21:35
21:36 mikehh joined
dalek rrot: r43397 | chromatic++ | branches/tt389_fix:
Don't store :method subs in namespace
 21:41
21:44 theory joined, bacek joined
chromatic PGE doesn't build, but I think that's a simple fix. 21:44
All of the test failures are the same thing: no more get_global on :methods.
bacek chromatic, trac.parrot.org/parrot/ticket/1393#comment:18 21:48
I would like to kill RetContinuation. It's way too dangerous.
21:52 Austin_Hastings joined
bacek $dayjob time 21:52
c u
dalek rrot: r43398 | chromatic++ | branches/tt389_fix/src/pmc/namespace.pmc:
[PMC] Made NameSpace PMC skip adding subs marked with :method to the namespace

   * where there's an existing class into which the methods go, or
   * where the NameSpace's name does not match a built-in PMC type
This resolves TT #389. Some tests fail. They need fixing; they're wrong.		 21:57
rrot: r43399 | chromatic++ | branches/tt389_fix/t/oo/mro-c3.t:
[t] Fixed test not to grab methods from the namespace directly, as they aren't
 21:58
rrot: r43400 | mikehh++ | trunk/lib/Parrot/Harness/Smoke.pm:
apply patch from Michael Peters to fix smolder uploads response from server - works for me
rrot: r43401 | mikehh++ | trunk/DEPRECATED.pod:
fix codetest failure - pod syntax (=back seems to need a blank line before it)
 22:14
22:18 bacek joined
mikehh mpeters++ for patch applied in r43400 22:20
22:27 pjcj joined
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#31506), fulltest) at r43401 - Ubuntu 9.10 i386 (g++ with --optimize) 22:54
22:55 mikehh joined 22:56 japhb joined 23:04 cognominal joined 23:05 patspam joined 23:18 bacek joined
mikehh pmc_freeze_cleanup branch - All tests PASS (pre/post-config, make corevm/make coretest, test, fulltest) at r43401 - Ubuntu 9.10 i386 (g++ with --optimize) 23:29
however - t/pmc/eval.t - TODO passed: 12, in coretest, test and all cores in fulltest 23:32
23:57 eternaleye 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!