Parrot 1.7.0 "African Grey" is out! | Testing hackathon November 14 and 15 -- trac.parrot.org/parrot/wiki/Opcode...gHackathon | Latest modified TT's: icanhaz.com/parrotbugs
Set by moderator on 14 November 2009.
calculus cconstantine: newer patch based off of svn r42521, pastebin.com/m1a40cfbc 00:02
cconstantine 'patch -p0 -i patch-file' doesn't do it 00:03
I'm sure I'm doing it wrong
calculus try -p1 instead of -p0 00:04
cconstantine that did it
now what am I testing?
snprintf from above?
the test gave the same output as kid51, but it built 00:08
Whiteknight my console window died and I can't create a new one 00:09
"There was an error creating the child process for this terminal" 00:10
cconstantine Whiteknight: that's one hell of a parrot bug 00:11
Whiteknight not parrot 00:12
brb 00:15
cconstantine dukeleto: are there docs for _dumper?
00:22 mikehh joined
nopaste "mikehh" at 81.149.189.7 pasted "build failure with g++" (78 lines) at nopaste.snit.ch/18705 00:28
dalek rrot: r42522 | mikehh++ | trunk/ext/nqp-rx/src/stage0 (2 files):
fix codetest failure - temp fix
 01:16
mikehh messages 01:22
cotto You have no new messages waiting. 01:25
diakopter @messages 01:29
>messages
!messages 01:30
messages
hm.
cotto try /msging purl 01:47
cconstantine how would I go about explicitly converting some AST into executable code in an action? 01:48
cotto You don't have to. Just put it into the PAST tree and the next stage of the compiler will take care of it. 01:52
cconstantine right, I'm working on a lisp, and for macros I need to be able to take part of the AST, run it, and insert the result back into the AST
01:55 Whiteknight joined
cotto lisps are fun like that 01:55
Whiteknight I knew a kid with a lisp once, he didn't think so highly of it
cconstantine hehe 01:56
PerlJam I think you can do PAST::Op.new( :inline($pir_code), :pasttype('inline') ) if that helps
Whiteknight matrixy is starting to really be wanting a Complex matrix type from parrot-linear-algebra 02:03
so that may be a project to work on soon
02:03 Patterner joined
cconstantine PerlJam: I don't think that's quite it. 'inline' puts code into the ast, I want to convert an instance of a AST node into code I can execute. 02:04
02:09 Zak joined 02:11 Whiteknight joined
Whiteknight ubuntu-- 02:12
I thought the "exit" opcode was supposed to throw an uncatchable exception 02:15
because I have an exit call that keeps getting handled and my program doesn't exit
so how do I exit a program in a guaranteed, unhandleable away? 02:16
cotto Whiteknight, segfault? 02:18
(though that can also be handled) 02:19
Whiteknight cotto: right idea, just more graceful
cotto You're leaving. Who cares if you leave a mess.
;)
Whiteknight well, if it's part of a pipe we would want it to exit gracefully
cotto Whiteknight, I think you want die(.EXCEPT_DOOMED, n) with n being your exit code 02:20
There's a special case to keep anything from handling that. 02:21
Whiteknight ok 02:23
where is EXCEPT_DOOMED defined?
cotto except_severity.pasm 02:26
Is there a nice way to suppress all pir-level stdout/stderr output? 02:29
especially stdout 02:30
ah. setstdout looks like my friend 02:31
02:39 kid51 joined 02:45 TiMBuS joined
diakopter hi, does anyone know how to get the memory address (or some other unique identifier) of an object in NQP? (or pir; I can add an inline pir sub) 03:20
03:24 wknight8111 joined
wknight8111 ubuntu-- 03:24
cotto diakopter, you can use get_pointer in pir 03:32
diakopter cotto: cool, thanks 03:35
wknight8111 INCOMING 03:38
purl duck!
cotto quack. 03:40
cotto sees nothing incoming 03:42
kid51 I wonder why dalek is not color-coded as a bot 03:44
cotto That'd be a nice feature. 03:45
diakopter there you go 03:46
03:48 janus joined
Coke ah, I was looking for that earlier. 03:57
also buildbot.
guessing ilbot2.
ttbot
and possibly particle. I was never sure about that guy.
wknight8111 ...at least I thought there was going to be some incoming messages...
03:57 TiMBuS joined
cotto you can always diy 03:58
wknight8111 matrixy is passing most of it's test suite again. That makes me happy 04:02
all the rest of the failing tests rely on Complex number handling, which we don't support yet 04:03
so, doing well
kid51 wknight8111: Are you (or is anyone) still working on garbage collection? 04:10
wknight8111 kid51: not actively working on it right now, no
been reading a lot of literature though, and putting together some rough ideas mentally
kid51 Does that depend on something else getting done first?
cotto: what command do you issue to voice the bots? 04:12
wknight8111 kid51: no, no dependencies
cotto kid51, I used xchat's gui
wknight8111 it's just daunting and I'm trying to scratch a few other itches before I jump into it
diakopter /voice dalek 04:13
kid51: ^^
kid51 diakopter: Thanks, that's what I was looking for
kid51 must sleep 04:18
purl $kid51->sleep(8 * 3600);
cotto Hey. If pprof2cg were written in nqp or pir, I could reuse that code for testing. 04:20
cotto is tempted
pmichaud is tempted 04:21
any chance it'd be faster?
cotto on idea
*no
cotto goes afk
feel free to experiment though. 04:22
wknight8111 leaves too. goodnight
dalek p-rx: ad79935 | pmichaud++ | src/PAST/Compiler-Regex.pir:
Adjust PAST::Regex literal matching to handle runtime literals, PAST::Val constants.
 04:43
p-rx: 26ad71a | pmichaud++ | src/HLL/Actions.pm:
Eliminate isaPAST in favor of Protoobject.ACCEPTS.
04:46 elmex joined
Coke pmichaud: hio 04:46
pmichaud Coke: wb 04:56
05:09 Essobi joined 05:10 TiMBuS joined
Coke zzzzzzzzzz 05:20
japhb msg mikehh It sounds like g++ is more picky about headers than gcc is. I've been including parrot/parrot.h in glut_callbacks.c, but perhaps parrot/extend.h or parrot/embed.h works better. Can you try swapping the include on line 1023 of config/gen/opengl.pm (which generates glut_callbacks.c during Configure), and see which one gets a clean g++ compile for you? 05:38
purl Message for mikehh stored. 05:39
dalek p-rx: b99043c | pmichaud++ | src/Regex/P6Regex/Actions.pm:
Only auto-concatenate PAST::Regex literal nodes with non-PAST children.
 06:04
p-rx: 9c899a5 | pmichaud++ | src/Regex/P6Regex/ (2 files):
Switch P6Regex quoting to use HLL quoting.
p-rx: 45b57ff | pmichaud++ | src/Regex/P6Regex/ (2 files):
Allow some double-quoted arguments in subrules.
p-rx: ba0fcdc | pmichaud++ | src/stage0/ (4 files):
Update bootstrap.
rrot: r42523 | pmichaud++ | failed to fetch changeset:
Add better quote handling to nqp regexes.
 06:10
06:27 mokurai joined 06:43 mikehh joined
cotto pmichaud, does nqp-rx support marking a sub with :main? 06:45
or is there some other way to pass in arguments from the cli? 06:47
bah. I can't use Q:PIR to declare a stub main either. 06:54
07:02 xenoterracide joined
diakopter hi, how can I test for Undef, in nqp (or pir) 07:13
07:21 davidfetter joined
japhb cotto, see the onload sub in Plumage's src/lib/Glue.pir -- $PROGRAM_NAME and @ARGS are set up in there, pulling from the interpreter's IGLOBALS_ARGV_LIST 08:05
And for NQP, I just create a sub MAIN, but call it from the mainline code manually.
cotto thanks 08:09
now it's time for sleep
night
08:13 fperrad joined
mikehh t/pmc/filehandle.t - Failed test: 18 - smoke (#29919) (also #29914) - fails fulltest in all cores - Ubuntu 9.10 amd64 08:17
dalek rrot: r42524 | mikehh++ | trunk/config/gen/opengl.pm:
fix g++ build (and warnings in gcc build)
 08:53
mikehh japhb: swapping bdidn't work - adding #include parrot/extend did 08:55
s/bdidn't/didn't/
08:58 iblechbot joined 09:02 davidfetter joined 09:13 desertm4x joined 09:18 mikehh joined 09:28 chromatic joined 09:32 desertm4x_ joined
dukeleto spread much parrot lore and interest amongst opensqlcamp conference attendees 10:03
dukeleto gave a lightning talk about pl/parrot that was well-received 10:04
i met a guy that is trying to write pl/haskell and we have both given each other some good pointers. it is kind of hilarious.
i also talked to the drizzle people and they are keen on embedding parrot as well 10:07
mikehh t/pmc/filehandle.t - Failed test: 18 - smoke (#29920) - fails fulltest in all cores - Ubuntu 9.10 amd64 g++ 10:09
all other tests PASS (pre/post-config, smoke, fulltest) - in testr t/pmc/eval.t - TODO passed: 12 - see TT #1142 10:11
dukeleto mikehh: interesing, it passes on darwin/x86+gcc 10:18
mikehh: do you --optimize ?
mikehh dukeleto: I have run the tests with g++ and gcc with --optimize and without on gcc (smoke #29920, #29914 and #29919) 10:32
dukeleto: all on Ubuntui 9.10 amd64 - I am going to reboot to i386 laster to test t/op/comp.t on i386 g++ --optimize where it failed before (with .pasm - the test is now .pir) 10:35
later
got to take my son to the station - he's going down to Edinburgh, I'll bbl 10:38
10:46 barney joined 12:08 hachi_ joined 12:38 bacek joined 12:57 joeri joined 12:59 Whiteknight joined
Whiteknight perlide.org/poll200911/ 13:03
Everybody should vote for PIR :)
moritz already did 13:04
Whiteknight w00t 13:05
13:09 mikehh joined, kid51 joined, payload joined 13:17 kj joined
kj it seems that parrot.org is down 13:17
kid51 Confirmed 13:25
kj: They just got it back up. 13:33
kj kid51: thanks 13:34
for the update :-)
13:40 mikehh joined
mikehh ah ha - just managed an svn update :-} 13:41
szabgab Whiteknight, pingo 13:59
Whiteknight szabgab: pong
szabgab oh I just saw you previous comment regarding voting for PIR :-) 14:00
but I was ringing you for another reasons
I see you are running github.com/Whiteknight/Matrixy ?
Whiteknight yessir
szabgab that's great, are you also on the pdl mailing list ? 14:01
Whiteknight no, not on the PDL list. probably should be eventually
szabgab especially as I was just about to post a link to Matrixy
Whiteknight oh awesome!
it's pretty messy right now, I'm in the middle of a huge refactor 14:02
szabgab in the meantime you might want to update the link from www.parrot.org/languages to point to the github thing ?
so here is the thing
mikehh t/pmc/filehandle.t - Failed test: 18 - smoke (#29926) at r42524 - fails fulltest in all cores - Ubuntu 9.10 amd64 g++ (no optimize)
all other tests PASS (pre/post-config, smoke, fulltest) - in testr t/pmc/eval.t - TODO passed: 12 - see TT #1142
szabgab I am trying to turn PDL into the next best thing after frozen water
Whiteknight ah yes, I forgot about that page. I'll update it soon
PDL is a pretty great product 14:03
szabgab which would be an easy task khm, :-) if I was not told that noone will move away from Matlab due to the tons of legacy code
so I asked if anyone new of a Matlab parser
and here I am talking to the author now 14:04
Whiteknight it's a very incomplete parser, but I see where you are headed
szabgab right into the wall :-)
Whiteknight so you'd be interested in an M->PDL translator?
szabgab I think so
Whiteknight ok. At the moment the parser is tied in pretty closely to the runtime but it shouldn't be too too hard to tease them apart 14:05
mikehh going to reboot to i386 to do some tests there - bbiab 14:06
14:06 mikehh joined
szabgab it might also be grate thing to use Matrixy for my world domination needs 14:07
it just needs a good IDE ;-)
if you are planning to register to the pdl list then I wait a few minutes with my reply so you will have a chance to reply with some details about Matrixy 14:09
Whiteknight where do I register? link?
szabgab pdl.perl.org
mailman.jach.hawaii.edu/mailman/listinfo/perldl 14:10
dalek rrot: r42525 | barney++ | trunk/NEWS:
Added recent news.
 14:11
14:12 barney joined 14:15 cconstantine joined 14:17 masak joined
japhb mikehh, thank you for fixing config/gen/opengl.pm for me. :-) 14:18
mikehh japhb: np - I needed it too :-} 14:22
I need to run some tests on i386 - was going to switch now - but I just remembered I have a LUG meeting in about half an hour- will work on it when I get back 14:25
14:27 davidfetter joined 14:29 integral joined 14:33 payload joined
mikehh szabgab: just installed PDL - haven't used it for a couple of years now - will be trying it out again :-} 14:49
szabgab I just stared to play with a few weeks ago and I never used any such tools
mikehh It was very useful for some work I was involved in some years ago - haven't used it recently - don't know why - it would have been very useful for a project I did last year 14:52
szabgab I would be very interested hearing about project where PDL can be useful 14:54
besides plain engineering stuff that I don't do 14:55
mikehh I'll see if I can locate some notes on what I did - got to go to the LUG meeting now - bbl
cconstantine I'm trying to use _dumper from docs.parrot.org/parrot/latest/html/...od.html... and I keep getting a 'Could not find non-existent sub _dumper'. Could someone help? 14:57
plobsing looks like it comes from here: ./runtime/parrot/library/dumper.pir 15:02
but all the examples in ./t/library/dumper.t do load_bytecode "dumper.pbc" 15:04
cconstantine hmm, loading directly from runtime/parrot/library appears to work 15:05
15:05 allison joined
Whiteknight szabgab: subscribed to pdl-l now 15:12
(sorry about the delay)
szabgab oh, I sent the mail a few minutes ago
never mind you can see it in the archive and respond
dalek rrot-linear-algebra: 3af9c98 | Whiteknight++ | src/pmc/ (4 files):
major cleanup. the x and y monikers were confusing. Rename them throughout to row and col for more clarity. Cleanup some macros, add some comments, and add some stub macros for dealing with complex matrices (to come)
 15:13
15:15 Psyche^ joined 15:19 integral joined 16:03 theory joined 16:09 kthakore joined
dalek rrot: r42526 | pmichaud++ | failed to fetch changeset:
[nqp]: Some prefix, postfix, and nullterm fixes.
 16:33
16:33 desertm4x joined 16:44 payload joined 17:03 tetragon joined 17:15 payload joined 17:29 mikehh joined 17:32 desertm4x_ joined 17:33 iblechbot joined
Coke is filehandle.t failing for anyone else? 17:41
(I don't see a trac)
dalek rrot: r42527 | coke++ | trunk/t/compilers/pge/pge_examples.t:
untodo passing test;
TT #1274 created by coke++: failure in t/pmc/filehandle.t 17:46
Coke msg cotto TT#1274 might be you. 17:49
purl Message for cotto stored.
17:49 allison joined 17:58 barney joined
dukeleto 'ello 18:01
18:05 chromatic joined 18:09 nbrown joined 18:10 mikehh joined
dalek lscript: 12330b2 | fperrad++ | t/helpers.pir:
helpers for test in PIR
 18:10
18:12 jan joined 18:15 fperrad joined 18:29 payload joined
dalek p-rx: aa07d85 | pmichaud++ | src/cheats/hll-grammar.pir:
In HLL::Grammar.EXPR, make a better distinction between <termish> and <term>.
 18:52
p-rx: 9b01eaa | pmichaud++ | src/cheats/hll-grammar.pir:
Convert termstack and opstack from lexicals to parameters for reusability.
 18:53
p-rx: a984e53 | pmichaud++ | src/ (3 files):
Eliminate nullterms in source code.
p-rx: 49e36a8 | pmichaud++ | src/ (5 files):
Update <EXPR> to properly handle prefix/postfix and terms.
p-rx: 9ded5d3 | pmichaud++ | src/stage0/ (3 files):
Update boostrap with fixed EXPR and nullterm handling.
plobsing hi #parrot! 19:14
19:16 desertm4x_ joined 19:18 nbrown joined
dukeleto plobsing: hello 19:31
19:31 nbrown_ joined
plobsing dukeleto: what else has to be done to get libjit_framebuilder merged? 19:32
it passes on linux x86/x64
I assume I need someone to test it on mac + windows
anything else?
purl hmmm... anything else is going to be even worse.
19:33 cconstantine joined
plobsing also, who would be good to poke about testing on those platforms? 19:33
19:38 Andy_ joined 19:48 particle joined, bacek joined
dukeleto plobsing: i can test on darwin/x86 19:48
plobsing: what am i testing?
19:48 Whiteknight joined
plobsing dukeleto: svn co svn.parrot.org/parrot/branches/lib...amebuilder 19:49
make sure it works without libjit installed. If you could install libjit and retest, that would be super awesome too.
dukeleto plobsing: checking it out now 19:56
plobsing: compiles fine without --optimize on darwin/x86, testing now 20:12
plobsing: that is without libjit 20:14
20:19 cconstantine joined
dukeleto plobsing: 'make test' passes on darwin/x86, no libjit 20:20
plobsing dukeleto: great! 20:24
dukeleto: can I convince you to install libjit and retest?
dukeleto dyld: lazy symbol binding failed: Symbol not found: _Parrot_pcc_trace_flags_test Referenced from: /Users/leto/svn/parrot-libjit_framebuilder/blib/lib/libparrot.1.7.0.dylib Expected in: dynamic lookup 20:25
dyld: Symbol not found: _Parrot_pcc_trace_flags_test Referenced from: /Users/leto/svn/parrot-libjit_framebuilder/blib/lib/libparrot.1.7.0.dylib Expected in: dynamic lookup
plobsing: that is a build fail with --optimize, still no libjit
plobsing thats weird. I didn't touch the tracing code 20:29
20:29 mokurai joined
dukeleto plobsing: have you tried it with --optimize? 20:30
plobsing yes, I've tried it on x86_64 optimize. 20:31
have there been any fixes recently in trunk for that kind of thing? recently ~= 1 week 20:33
dukeleto i don't know 20:35
20:42 bacek joined
plobsing dukeleto: the problem seems to stem from the failure to include the macros introduced in r41987 20:51
Whiteknight dukeleto: you're going to love this next parrot-linear-algebra commit 20:55
bacek good morning
Whiteknight hello bacek
bacek alloha Whiteknight 20:56
plobsing, (undefined symbol) I fixed it in trunk few days ago
plobsing oh good
dalek rrot-linear-algebra: eb2c012 | Whiteknight++ | src/pmc/pla_matrix_types.h:
fix all the complex macros to be copies of the new real macros. Other misc cleanups
 21:00
rrot-linear-algebra: 8a8d868 | Whiteknight++ | (4 files):
Add a new ComplexMatrix2D PMC prototype. A copy+paste from NumMatrix2D with some changes and omissions. Missing some features for now
plobsing how do I get those changes merged into my branch? 21:01
dukeleto plobsing: 'make test' passes with libjit, no --optimize on darwin/x86
bacek plobsing, ask Whiteknight, he can do it. 21:02
plobsing dukeleto: did it identify that libjit was installed?
Whiteknight: can you get the fixes bacek made on trunk into the libjit_framebuilder branch? 21:03
afk for a bit 21:05
21:14 cconstantine joined 21:19 theory_ joined 21:40 cconstantine joined 21:44 payload joined
plobsing hi again 21:49
purl oh, you're back!
dalek parrot: 1fb44db | dukeleto++ | Configure.nqp:
make Configure.nqp play nice with the new nqp
 21:58
parrot: e6a1220 | dukeleto++ | t/test.sql:
example test script that creates the plparrot language and creates a void function
cotto Coke, ping 22:02
Is anyone one who's seeing the t/pmc/filehandle.t test failure? 22:03
Coke cotto;pong.
nopaste "cotto" at 74.61.2.46 pasted "possible fix for filehandle.t failure" (41 lines) at nopaste.snit.ch/18708 22:04
cotto That just patches the .t, so no need to rebuild.
Can you test that?
Coke? 22:06
purl Coke is Will Coleda <mailto:will@coleda.com> or perpetually annoyed. or magical ticket robot or (if Diet), something that turns into formaldehyde in my blood.
22:14 payload joined, mikehh joined
Coke testing. 22:16
purl i guess testing is the best thing ever ZOMG
Coke cotto;was the bug that it assumed . in PATH?
or rather, that a parrot was in your path?
cotto it did make that assumption 22:17
Coke cotto: works now.
cotto Awesome. I'll commit the fix and close the ticket. Thanks for reporting that. istr getting another OSX report but I guess that OSX wasn't wacky enough. 22:18
done and done 22:20
dalek rrot: r42528 | cotto++ | trunk/t/pmc/filehandle.t:
[t] fix a FileHandle test on systems where . isn't in the PATH
TT #1274 closed by cotto++: failure in t/pmc/filehandle.t 22:23
22:31 mikehh joined
dukeleto is getting plparrot to do stuff 22:41
22:42 Whiteknight joined
diakopter is failing at getting nqp to do stuff 22:42
dukeleto diakopter: what is up? 22:46
purl A direction away from the center of gravity of a celestial object. or the y-axis, unless you're using a strange coordinate system.
diakopter oh, just the first-few-hours-using-a-language[-implementation]-uncertainty-whether-bugs-are-in-my-code-or-language[-implementation] 22:47
dukeleto always uses a toroidal coordinate system
diakopter: here is your gold watch 22:48
diakopter sweet, thanks. how does it work 22:49
dukeleto diakopter: you put it on your wrist and it does nothing
dalek parrot: 9785ba3 | dukeleto++ | HOWTO:
Add a HOWTO
 23:02
parrot: 561630c | dukeleto++ | t/test.sql:
Basic sanity test for plparrot
23:30 mikehh joined 23:39 payload1 joined 23:41 cconstantine joined 23:49 preflex joined
dalek rrot: r42529 | mikehh++ | trunk/t/op/comp.t:
fix test so failure logic is correct
 23:58

Powered by Cro and the Raku® Programming Language.

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

Thank you!