Parrot 2.5.0 Released! | parrot.org | Log: irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | GSOC Students: trac.parrot.org/parrot/wiki/GSoCersStartHere | Priorities: merge gc_massacre, remove deprecated items, close tickets.
Set by moderator on 15 June 2010.
dalek rrot: r47667 | darbelo++ | branches/gsoc_nfg/src/string/encoding/nfg.c:
Commit WIP iterator additions. The first step towards NFG literals has been taken.
 00:10
00:16 kid51 joined
whiteknight ash_: ping 00:22
ash_ hi
whiteknight ash_: where is your gsoc repo, again?
ash_ github.com/ashgti/parrot
purl github.com/ashgti/parrot are in there
ash_ um... purl remember that? /confused 00:23
whiteknight oh okay, the branch. the commit graph there looks like there's been no activity
ash_ gsoc_nci is the branch
whiteknight ok 00:24
you still having that issue from earlier?
ash_ yeah
everything i have tried hasn't made a difference
are you on linux? 00:26
whiteknight yeppers
ash_ i had an issue earlier on linux with it finding the header, they didn't put it where pkg-config said it was, i ended up doing --ccflags='-I/usr/include/i486-linux-gnu/' to get it to work 00:27
so you know
thats the libffi header ffi.h and ffitarget.h
whiteknight src/nci/core_thunks.c:32:12: fatal error: 'core_thunks.str' file not found 00:28
ash_ um, i haven't ever gotten that problem...
it shouldn't be building the thunks anyway
whiteknight that's what I get on a fresh checkout
ash_ one sec let me try 00:29
cotto_work possibly a c2str problem?
whiteknight ash_: core_thunks$(o) is still included as part of a list of targets in the makefile 00:30
ash_: which linux distro are you on?
ash_ hmmm, that doesn't sound right, i changed the config/gen/makefiles/root.in to make it not if you have libffi
ubuntu 10.04
purl ubuntu 10.04 is on slavak's laptop. Need perl 5.12 repo. ;)
ash_ you'll need the libffi headers 00:31
whiteknight ash_: the folder for me was /usr/include/x86_64-linux-gnu. I created some symlinks to get around that 00:32
building fine now
ash_ mine always dies at: ./miniparrot -Iruntime/parrot/include config_lib.pir > runtime/parrot/include/config.fpmc 00:33
mikehh whiteknight: nopaste.snit.ch/21316 00:39
whiteknight mikehh: don't you ever have anything good to say? :) 00:44
mikehh: I'll fix it in a minute, I'm building ash_'s project first 00:47
ash_ did it build whiteknight?
whiteknight ash_: yeah, failed at the miniparrot stage 00:48
ash_ any idea whats up?
whiteknight not yet. Let me fix that pod error mikehh++ reported and I'll dig into it
ash_ kk 00:49
thanks a lot for helping me, i am stumped
whiteknight no problem 00:51
00:53 snarkyboojum joined
dalek rrot: r47668 | whiteknight++ | trunk/docs/book/pir/ch07_objects.pod:
[docs] fix POD error that was breaking t/examples/pod.t. mikehh++ for the find
 00:59
whiteknight ash_: This problem is weird as hell
ash_ lol, and whats even more strange is on occation if i remake after its failed, it gets past that miniparrot point
so i have a feeling its sorta random, but i can't figure out whats wrong 01:00
whiteknight ash_: this is very weird 01:05
ash_ I don't think i made any changes to anything that would be causing this...
i thought
whiteknight ash_: When did this bug first appear? Which commit? 01:06
ash_ 7f2a03aad8 01:07
whiteknight why are you using string_make instead of Parrot_string_new_init? 01:10
ash_ because of the docs? docs.parrot.org/parrot/latest/html/...s.pod.html said "string_make 01:11
A crippled version of a string initializer, now replaced with the full version Parrot_string_new_init."
although thats wrong too, i found out later, it should be Parrot_str_new_init 01:12
but i went back to string_make
since it didn't seem anyone else was using Parrot_str_new_init much directly, everyone seems to be using string_make so i figured the document was outdated? 01:13
whiteknight so before 7f2a03aad8, you didn't have this problem, and starting at that commit you did?
ash_ yeah
whiteknight I suggest you back out that commit (because it was pretty large) and figure out exactly which change is giving you trouble 01:15
the casts are not likely the problem, unless there's some weird pointer arithmetic happening that I'm not aware of 01:16
so back out that change, re-add all the casts and try, redo the makefile changes and try, etc 01:17
I suspect the string_make line, since that's the most substantive change you made
also, strings in Parrot are immutable, so you don't need to mark them constant or treat them constantly 01:20
ash_ alright, i just saw some places where people did use the constant flag 01:21
whiteknight ash_: yeah, it's probably a leftover. Constant flags on strings were made obsolete recently 01:22
ash_ k
01:25 plobsing joined
whiteknight I can't even find that damn string_make line in HEAD. too much stuff has moved 01:26
plobsing ping ash_ 01:27
ash_ hi plobsing
plobsing how goes it? 01:28
ash_ still having problems with miniparrot dying, whiteknight was helping me figure out whats wrong 01:29
plobsing have you tried eliminating all uses of CONST_STRING you introduced? 01:31
ash_ i'll try that real quick 01:33
plobsing also I see you're still trying to give thunks to raw ncis. that will cause hard to track problems (although these symptoms aren't likely caused by that) 01:34
ash_ what do you suggest as an alternative?
plobsing either create a separate class for them (more work), or do what we do now and use a flag (private2) to give them an alternate code path. 01:36
whiteknight I'm heading to bed now. I haven't seen an error like this, so I don't know quite what to make of it. I'll take another look at this bug tomorrow. BAck out those changes like I suggested and see if you can narrow it down at all
ash_ you don't check pbc's into the repository do you? 01:37
plobsing we do check in some for testing purposes, but in general no.
they would break far too frequently 01:38
ash_ k, just checking, removing the CONST_STRING might of worked... (feels dumb)
plobsing why do you ask about checking in PBCs? 01:40
ash_ i just did a make and it got past the miniparrot but was complaining about pbc versions PackFile_unpack: This Parrot cannot read bytecode files with version 6.15. so i deleted the offending pbc's 01:41
just wanted to make sure they were generated
plobsing probably some PBCs don't have their dependencies quite right and aren't being remade
bacek_at_work ash_, make realclean or reconfig usually helps
ash_ i'll do that real fast 01:42
make realclean, then config and make 01:43
nope, after a make clean it still fails 01:45
i am going to go with what whiteknight suggested and just backtrack to a point were it didn't fail and then work fowards 01:46
plobsing good idea
ash_ anyone know why pdd28 (strings) suggests string_make is "A crippled version of a string initializer, now replaced with the full version Parrot_string_new_init" ? should i not use string_make? 01:49
plobsing for one, it doesn't accept an encoding argument 01:51
ash_ which one? 01:52
plobsing string_make
ash_ i thought it was string_make(interp, "CString", /* strlen */, NULL | "ascii" | "unicode" ... etc., flags); 01:54
plobsing that's charset. Parrot strings have both charset and encoding (don't ask me why). 01:55
ash_ got ya 01:56
plobsing Also, string_make throws an exception when the charset is not found. Parrot_str_new_init, OTOH, accepts CHARSET and ENCODING parameters directly, meaning you can fetch them yourself and deal with errors in possibly a more appropriate fashion.
ash_ well, i am going to try to get this working, if i get it done and make some tests for it i'll try to switch the llvm_detect branch detection over to llvm-config and start work there 02:01
plobsing once you declare it "done", I'll try and get it into an svn branch for wider testing and eventual merging 02:03
ash_ I know parts of it work, I manually forcing a signature in parts of it and it seemed to do alright 02:06
plobsing manually forcing a signature? where and how? 02:10
02:13 snarkyboojum joined
ash_ well, for a while i manually crafted the signature vJP and forced that 02:18
and the raw_nci stuff worked
(thats kinda why i think i have things close to done)
in the invoke of the NCI pmc i just built that specific signature and always used it
kinda to prove I had things right, then i switched to forcing the signature to "vJP" and having my signature parser do the work to make sure that was right too 02:19
plobsing sounds close then. 02:20
02:21 TiMBuS joined
ash_ i am going to add a switch, if the private2 flag is set i am going to use the old nci stuff, so it will only be there for the raw_nci stuff 02:23
Coke tries to come with a decent json representation for these html pages. 02:24
ash_ now it doesn't die when it initializes
like it was
dalek parrot: 8267b71 | dukeleto++ | plparrot.c:
Refactor and remove dead code from plparrot.c
 02:39
ash_ plobsing: i got it working again, or part of it, its back to the PGE error 02:44
i know why miniparrot died all the time now
i changed the makefile to not build one of the thunk files because i didn't think it was needed, putting it back fixed the miniparrot error 02:45
03:13 snarkyboojum joined 03:32 janus joined 03:48 hercynium joined 04:07 snarkyboojum joined
dukeleto How do I turn a Rakudo data structure into a Parrot data structure ? 04:10
dukeleto has some Perl 6 code running on PL/Parrot, and is now trying to marshall data
sorear We haven't quite figured out how cross-HLL data sharing works, yet 04:11
dukeleto actually, if I could, for instance, turn a Rakudo integer into a C integer, that would help 04:12
so Perl 6 objects have a .perl method, maybe Rakudo objects should have a .parrot method or something?
plobsing do Rakudo objects not respond to get_integer?
dukeleto plobsing: i will try that. Currently my code does a Parrot_PMC_isa on the datastructure, and none of the Rakudo datastructures return true for Integer, String or Float 04:14
sorear isa is harmful; use does instead 04:16
dukeleto sorear: touche
plobsing $isa does Harm; use does; 04:17
dukeleto sorear: Parrot_PMC_does seems to break my code which works with Parrot_PMC_isa.
dukeleto goes and reads some source
such as, what is the difference between Parrot_PMC_does and Parrot_PMC_does_pmc ? 04:18
ah, I see, _pmc takes a PMC instead of a String
plobsing: get_integer seems to just be returning 0 on a Rakudo integer 04:19
plobsing well 0 is an integer. :P 04:21
dukeleto plobsing: yes, one of my favorites. But it would be nice if 42.get_integer == 42 instead of 0 :) 04:22
so it seems that _get_intval was removed, but I don't see it in DEPRECATED.pod . Am I missing something?
plobsing _get_intval sounds like one of the duplicated vtable-calling methods formerly in src/extend.c. 04:25
I removed them in favour of using the equivalent functions in src/extend_vtable
TT #1587 04:27
04:28 mikehh joined
dukeleto plobsing: that sounds good, I was wondering if that should be mentioned in DEPRECATED.pod ? 04:31
plobsing it was. in 2.3. 04:33
04:39 LoganLK joined
dukeleto plobsing: i forgot that stuff gets deleted from DEPRECATED.pod once it is gone 04:41
Parrot_PMC_get_class on a Rakudo object returns the empty string => :( 04:42
dalek rrot: r47669 | dukeleto++ | trunk/docs (2 files):
[docs] Update docs referring to get_intval which is now get_integer
 04:49
04:50 abqar joined
dukeleto abqar: welcome 04:51
sorear Why don't we have preproc/configure magic to support __attribute((deprecated)) or however you spell that? 05:15
plobsing because you hadn't volunteered yet ;-) 05:16
sorear I'm immune to volunteering under the CLA policy. 05:17
Coke dukeleto: I think you want $I0 = $some_Rakudo_Thing 05:20
if in c, INTVAL foo = VTABLE_get_integer(INTERP, RakudoThingee); (whetever that expands to for embedders)
dukeleto Coke: that is Parrot_PMC_get_integer in C, and that doesn't seem to work 05:22
Coke: where would a test for that even go?
Coke: actually, it should probably go in t/02-embed in the Rakudo repo 05:23
05:27 mj41_ joined 05:30 kjeldahl_ joined 06:14 dngor joined 06:18 ttbot joined, LoganLK joined, hercynium joined, janus joined, plobsing joined, tcurtis joined, wizard joined, theory joined, Patterner joined, PacoLinux joined, integral joined, ash_ joined, jan joined, preflex joined, tewk joined, pjcj joined, simcop2387 joined, Hunger joined, Util joined, darbelo joined, NotFound joined, ruoso joined, gaz joined, elmex joined, slavorg joined, Ryan52 joined, workbench joined, Khisanth joined, TonyC joined, japhb joined, bacek joined, zibri joined, s1n joined, spinclad joined, szbalint joined, moritz joined, knewt joined, ascent joined, sri joined, TimToady joined, jnthn joined, sjn joined
pmichaud prepare for commit flood 06:18
moritz tcurtis: pong 06:19
cotto and suddenly 06:20
dalek kudo: ddb39d0 | (Solomon Foster)++ | t/spectest.data:
Regress range-iterator.t, because it conforms to the now-obsolete former
 06:21
kudo: 6b2af44 | moritz++ | src/core/MAIN.pm:
fix all but one test in main-usage.t
kudo: 1cce3a3 | pmichaud++ | src/builtins/Junction.pir:
Adjust Junction.new so that it accepts slurpy arguments, stores $!eigenstates as
kudo: 5e523b1 | pmichaud++ | (2 files):
Merge branch 'list' of github.com:rakudo/rakudo into list
kudo: b6cc36c | pmichaud++ | src/builtins/Parcel.pir:
Make list assignment more robust -- assigning to lists containing hashes now
kudo: d982db1 | pmichaud++ | src/ (5 files):
Refactor postcircumfix:<{ }> in terms of .at_key and to have proper
kudo: 7e58a35 | pmichaud++ | src/ (6 files):
Refactor Positional to follow the model developed for Associative.
kudo: 8f09a6e | pmichaud++ | src/core/Positional.pm:
Fix typo in Positional slices.
kudo: f015a83 | pmichaud++ | src/builtins/Parcel.pir:
Assignment to Whatever acts like assigning to a scalar.
kudo: 465ef30 | pmichaud++ | src/core/Array.pm:
Typo in Array at_pos method.
kudo: 48e2e8d | pmichaud++ | src/builtins/Junction.pir:
Update eigenstates handling to use Seq for value semantics.
kudo: 4cb5c39 | pmichaud++ | src/core/Match.pm:
Finally, Match does Positional.
kudo: 62f2219 | pmichaud++ | src/ (5 files):
Eliminate obsolete FETCH methods.
kudo: 41191e9 | pmichaud++ | src/builtins/Proxy.pir:
Switch Proxy to create Perl6Scalar instead of ObjectRef.
kudo: ffd450c | pmichaud++ | src/ (5 files):
Significant refactor of Parcel, !STORE, assignment, and the 'scalar' flag.

and lays a good foundation for an even better solution (resolving our hash and array vivification issues) that I'll do shortly after the branch merge and June release.
kudo: e4e9822 | pmichaud++ | src/glue/regex.pir:
Fix handling of regex assertions (wasn't using a flat list).
kudo: 70d91d4 | pmichaud++ | src/glue/dispatch.pir:
Clean up dispatch list handling (&flat instead of &list).
kudo: 5832221 | pmichaud++ | (53 files):
Merge branch 'list'
kudo: e3eb608 | pmichaud++ | src/core/List.pm:
Restore List.rotate method.
tcurtis moritz: Is "my &foldable := sub ($opName) {...}" valid in Perl 6? NQP doesn't appear to support using &foldable to refer to "sub foldable ($opName) {...}". 06:23
moritz tcurtis: yes, it is 06:25
tcurtis moritz: I'm about to update my blog post and the associated example code repository with that, then. 06:28
moritz tcurtis++ 06:29
tcurtis I also fixed the lack of escaping of the angle brackets in the "--target=constantfold" output in the post. 06:32
dalek rrot: r47670 | plobsing++ | branches/dynop_mapping/compilers/imcc (2 files):
handle const pointer symregs slightly more appropriately
 06:44
kudo: cca4a84 | pmichaud++ | src/Perl6/Compiler.pir:
Restore backtrace handling.
kudo: a5f844b | moritz++ | t/spectest.data:
enable quite a few passing test files (some of them were regressions of the list
 07:09
07:14 fperrad joined 07:21 fperrad_ joined 07:53 jan joined 08:04 Coke joined
dalek kudo: 012ea2b | snarkyboojum++ | src/core/Any-list.pm:
Get rotate working for Parcel again
 08:18
08:37 JimmyZ joined 09:04 tcurtis joined
dalek kudo: d1a7442 | moritz++ | src/core/CallFrame.pm:
fix callframe().line and .file by creating a Perl 6 hash from the annotations
 09:22
kudo: cb0e95e | moritz++ | src/core/CallFrame.pm:
simplify CallFrame.pm, and make a parameter optional
bacek aloha, humans 09:28
mikehh bacek: anything happening in gc_massacre recently? 09:58
bacek mikehh, not really. It needs some deep tuning...
mikehh bacek: 'k, anything I can do to help? 10:00
bacek wanna tune GC? :) 10:02
mikehh :-} 10:03
wizard hi, is somewhere document with references to all predefined classes in parrot like ResizablePMCArray, FizedPMCArray etc ? 10:09
JimmyZ wizard: docs.parrot.org/parrot/latest/html/pmc.html 10:12
wizard tnx 10:13
I see parrot has class Timer but I see no classes like Date or DateTime so has parrot something to operate with dates? 10:22
moritz wizard: don't think so. dukeleto++ has been planning some work on them, iirc 10:24
wizard: I'm sure your contribution would be welcome
github.com/rakudo/rakudo/blob/maste...re/Date.pm might serve as an inspiration 10:25
github.com/rakudo/rakudo/blob/maste...emporal.pm
10:26 khairul joined
wizard hmm. I'm not sure my knowledge of parrot allow me to mess up with it's source 10:31
10:31 lucian joined
wizard but I can try :) 10:31
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#34450), fulltest) at r47670 - Ubuntu 10.04 amd64 (g++) 10:37
rakudo (cb0e95e) builds on parrot r47670 - make test PASS, spectest_smolder -> #34452 (pugs r31326) FAIL - Ubuntu 10.04 amd64 (g++ with --optimize) 11:14
11:39 ruoso joined 11:47 JimmyZ joined
dalek r: 266a544 | bacek++ | (6 files):
Create "mini-framework" for testing POST output.
 11:51
bacek msg cotto If you are bored and want to work on PIRATE check t/post/sub.txt from github.com/bacek/pir/commit/266a544...0b263e0182 :) 11:52
purl Message for cotto stored.
12:02 whiteknight joined
Coke dukeleto: when you say Parrot_PMC, do you mean Parrot_<NAME OF PMC>_ ? 12:16
dalek kudo: 19e7264 | moritz++ | src/core/CallFrame.pm:
fix typo in CallFrame.pm, moritz--
 12:32
kudo: f17fe96 | moritz++ | docs/ChangeLog:
extend ChangeLog a bit
kudo: 17e43e8 | moritz++ | t/spectest.data:
turn on the series tests; those that fail are regressions from the "list" merge,
12:38 JimmyZ joined 13:12 ambs joined
whiteknight good mornin, #parrot 13:15
13:22 gbacon joined 13:28 atrodo joined 13:35 bluescreen joined 13:56 plobsing joined
bacek aloha, whiteknight 13:57
13:57 bubaflub joined
whiteknight good morning bacek 13:58
mikehh howdy whiteknight
whiteknight hello mikehh
mikehh whiteknight: no complaints for you at the moment :-} 13:59
whiteknight mikehh: I don't mind if you do. Somebody has to keep an eye on the crap code I write :) 14:02
dalek r: 3c2550f | bacek++ | src/PIR/Actions.pm:
Remove debug output.
r: 6f53e70 | bacek++ | (4 files):
Factor out common POST::Value
mikehh whiteknight: hey, it's much better than most, no complaints on that score
r: 7f320bd | bacek++ | t/post/subs.txt:
More tests.
purl i think more tests is trivial.
r: 35e344b | bacek++ | (3 files):
Implement POST::Sub.symbol similar to PAST::Sub.symbol
r: f9aa791 | bacek++ | (3 files):
Store .param in symtable
r: 0da913b | bacek++ | (4 files):
Implement proper handling of undeclared named registers
r: 9bf18ef | bacek++ | src/PIR/Actions.pm:
Remove debug, fix typo.
r: 318b524 | bacek++ | t/test_post.pir:
Handle fail_like adverb in test_post for validating error messages.
r: 2e19911 | bacek++ | t/ (2 files):
Add test for not predeclared registers.
mikehh whiteknight: and i really enjoy your blog _1 there or more 14:05
that should hve been +1 - whiteknight++
whiteknight thanks! I've been lazy with writing new posts lately
14:05 patspam joined
bacek ok, on this positive note I'm departing to bed 14:06
mikehh I don't spend much time on television these days - fully sympathise on that score
whiteknight good night bacek 14:07
mikehh see you later
nopaste "bacek" at 192.168.1.3 pasted "PIRATE rocks!" (7 lines) at nopaste.snit.ch/21327
bacek Good night, humans 14:08
dalek rrot: r47671 | khairul++ | branches/gsoc_instrument/src/dynpmc/instrument.pmc:
Cleanup hook list on destroy.
r: 65d4cd7 | bacek++ | t/post/subs.txt:
Add test for declared params usage.
whiteknight bacek++ # PIRATE is getting quite impressive! 14:23
atrodo So now that there's Pirate, which project is going to be Ninja? 14:24
bubaflub ninja should be the name of the next nqp 14:26
whiteknight nqp isn't sneaky enough 14:27
moritz perl 6, the language of ninjas and pirates alike!
even chuck norris uses it
atrodo moritz++ 14:32
whiteknight Code that Chuck Norris writes is valid input to all compilers, and always solves the halting problem 14:34
particle what is this pirate you speak of? 14:51
because it's not the pirate of old... python on parrot 14:52
Coke bacek stole the name.
particle for what?
purl for fun.
Coke pirate?
purl pirate is, like, a career that is washed up these days...and raping and pillaging just doesn't have a good reputation these days.... or www.eff.org/IP/DRM/piratead/CEA_ad.png or www.wired.com/politics/security/mag...li_pirates or PIR parser (and compiler) implemented in NQP github.com/bacek/pir
Coke the last one.
particle ah
is that on the road to lorito? 14:53
Coke I imagine it could be.
if iI knew where lorito was, I could google map it.
Andy Morning all 15:05
purl morning, Andy
Andy I haven't done anyting in Parrot for weeks. It's very sad.
15:11 ash_ joined
whiteknight good morning Andy 15:28
Andy: don't feel bad. I've been slacking too
dalek rrot: r47672 | NotFound++ | trunk/t/pmc/stringiterator.t:
fix StringIterator out of bounds tests that were passing for wrong reasons
 15:30
Coke Andy: if you want some LHF, you could prove tools/*/checkdepend.pl and fix a makedepend or two. 15:31
particle s/prove/run/ ? 15:32
Andy Oh, didn't mean that I had time for any F, LH or not. 15:33
Coke particle: , no prove. 15:34
(it's a test.)
Andy added to my task list 15:35
FWTW
Coke Andy: ah well. going to yapc:na 2010?
Andy nope
Coke but I so miss you yelling at me!
mikehh rakudo (17e43e8) builds on parrot r47670 - make test PASS, spectest_smolder -> #34457 (pugs r31332) FAIL - Ubuntu 10.04 amd64 (g++ with --optimize) 15:36
rakudo: t/spec/S32-io/IO-Socket-INET.rakudo reports a failure, but when i run it separately (4 times) it passes 15:37
also t/spec/S09-typed-arrays/arrays.rakudo fails
bunch of TODO passes 15:38
particle a test that lives in tools/ and ends in .pl? 15:41
Andy When have I yelled at you, Coke? 15:43
COME ON, WHEN?
I'd be tempted to go just for a good hackathon.
Did I really yell at you at some YAPC past? I apologize. 15:44
Coke Andy: the first Chicago, I think. 15:46
I think you were just goofing around and speaking in ALL CAPS, though.
you are hard to read, both online and in person. 15:47
particle: yes. it's not intended to be run as part of the test suite. it's more of a guide.
and will eventually be part of a "make depend" step.
but since it was test like, I made it act like a test. 15:48
15:49 ambs joined 15:51 bubaflub joined
dalek rrot: r47673 | NotFound++ | trunk/src/pmc/string.pmc:
drop pointless check for PMC_data
 16:02
Coke Andy: no worries.
purl Ha!
Coke no worries, mate?
Andy Back 16:14
Coke: Yeah, no worries. I'm mostly goofing at any given point in time.
Is there indeed a post- or pre-YAPC hackathon? 16:15
Coke we have a room for perl6 & parrot the run of the con. nothing schedule, just as you drop in. 16:19
dalek rrot: r47674 | NotFound++ | trunk/t/pmc/arrayiterator.t:
test for ArrayIterator wrong direction
Andy oh crud, it's next week. I can't go. 16:20
Coke ugh. why did java go with single dash for long options? no wonder I can't ever find the version of the damn jvm. 16:21
would anyone care if docs.parrot.org/parrot/latest/html/pmc.html did not distinguish beteen abstract and concrete? 16:25
eh. I'll leave it for now. easier to verify I haven't borked anything. 16:28
16:35 ambs joined
dukeleto Coke: no, Parrot_PMC_* is the prefix for embedd API functions that deal with PMC's 16:40
cotto_work happy Thursday 16:43
Coke dukeleto: ah, ok. don't know what hte wrapper in that func is doing, butit is calling the right vtable. 16:45
dukeleto cotto_work: happy $day 16:48
16:48 Essobi joined
dukeleto Coke: yeah, i think i am dealing with a Rakudo issue involving return values 16:48
ash_ dukeleto: did you get your issue figured out? 16:49
16:50 ruoso joined
dukeleto ash_: not really. I can't seem to get a Rakudo return value from PIR 16:52
jnthn dukeleto: I think Rakudo may always return a Parcel. 16:53
If which case you can look inside it. 16:54
NotFound Can someone tell me what's the difference between src/nci/extra_thunks.c and src/extra_nci_thunks.c and what is used for whatever?
Coke I think that was... plobsing? 16:55
ISTR one is for core, and one is for embedding.
Coke finds code for generating HTML pages that is never run.
dukeleto jnthn: when I use dumper.pbc to dump the return value I get, it is: "VAR1" => undef with-properties: Hash { "rw" => PMC 'Bool' { ... } } 16:56
NotFound I think both are for annoying dvelopers.
ash_ I don't have a src/extra_nci_thunks.c ... is that a left oer?
over*
dukeleto jnthn: is that a Parcel in disguise? Or an actual undef ?
Coke it's generated.
jnthn dukeleto: That looks like a real undef. 16:57
dukeleto: Which is a tad odd.
What's your Perl 6 code look like?
Coke dumper can't dump a Bool?
oh,that's a p6 bool. nevermind.
dukeleto jnthn: github.com/leto/plparrot/commit/409...d539feb0cf 16:58
jnthn: you can see in the diff what I wrap code in. I have just been trying to get a function body with a simple integer to work.
jnthn: I know the code is running, because if I put a "say" in the function body, it shows up in my logs
NotFound The question really is: how should be the src/nci/extra_thunks.c regenerated? 16:59
dukeleto jnthn: try { my $r = eval ' 42 '; return $r };
Coke ... holy crap is make html broken. 17:00
jnthn dukeleto: havign just tried in evalbot...erm...something looks broken. :-S
dukeleto jnthn: yeah, returning values from Rakudo to PIR has some wonkiness 17:01
ash_ will return do something useful there?
doesn't return throw a control exception?
dukeleto ash_: i have tried with just a plain $r as well
ash_ (and won't try catch that?)
dukeleto jnthn: if you can point me in the direction of a test, I can write some 17:02
jnthn: i am guessing t/02-embed can use a few more tests :)
Coke ash_: try doesn't eat CONTROL exceptions, I don't think. 17:03
dalek kudo: 96af91d | pmichaud++ | src/Perl6/Actions.pm:
Remove fossil: 'for' should use &flat on its source, not &eager.
ash_ Coke: rakudo: sub foo { try { return 13213 }; return 1 }; say foo; 17:04
[12:03pm] <p6eval>
rakudo 17e43e: OUTPUTĀ«1ā¤Ā»
maybe i am misunderstanding return there, but try seems to have caught it
Coke ash_: certainly does, ja. 17:05
ash_ i wouldn't expect try to catch CONTROL exceptions, but i think it might be
17:07 tcurtis joined
ambs Coke: next week, don't forget schwern. 17:08
17:09 clinton joined
Coke ambs: I will endeavor to remember. 17:09
ambs Coke: if I find you, I will remember :) 17:10
thank you
dalek kudo: 7df145b | jimmy++ | src/core/metaops.pm:
removed unused code
 17:15
kudo: 1cfec1c | moritz++ | src/Perl6/Actions.pm:
Merge branch 'master' of github.com:rakudo/rakudo
Coke seen jhorwitz? 17:40
purl jhorwitz was last seen on #parrot 282 days, 1 hours, 4 minutes and 39 seconds ago, saying: Tene: not at all. i should have the tuits this week to work on your compile issue [Sep 8 16:35:38 2009]
Tene Coke: I find he's often responsive on aim
dalek rrot: r47675 | NotFound++ | trunk (3 files):
use STRING_IS_NULL instead of NULLness checks in STRING * to char * NCI thunks
 17:41
Coke he would have seen the email I just got. no worries.
Tene 'k
dalek nxed: r504 | julian.notfound++ | trunk/examples/Mysql.winxed:
stop using deprecated 'v' arguments NCI signature in example Mysql
 17:42
NotFound At least!
17:47 patspam1 joined
dalek rrot: r47676 | dukeleto++ | trunk/t/op/sprintf2.t:
[docs][TT# 1667] Get rid of wrong documentation
 17:57
cotto_work We need to start GPG signing our releases. Maybe I'll take that on while I wait for OSUOSL to get the trac/git test site working.
I can't complain about the cost, but their speed leaves me unimpressed. 18:01
dukeleto cotto_work: i would like to see md5/sha1s of Parrot releases as well 18:02
cotto_work Happily we've trained our release managers to follow the documentation, so a patch to that should be sufficient. 18:04
dalek TT #1667 closed by dukeleto++: t/op/sprintf2.t: Misleading documentation 18:06
TT #1667: trac.parrot.org/parrot/ticket/1667
kudo: 7a230d9 | pmichaud++ | docs/spectest-progress.csv:
spectest-progress update through 06-14; still waiting on later dates to be
 18:07
Coke I recall needing the sums for making the macport release. 18:32
cotto_work ooc, has anyone tried cutting a release on windows? 18:33
18:52 contingencyplan joined 18:58 lucian_ joined
dukeleto Coke: yes, many packaging systems require a md5 or sha1. We should provide those. 18:58
dalek kudo: fdb5ca4 | (Solomon Foster)++ | src/core/operators.pm:
Unify the infix:<...>($lhs, $rhs) case with the main infix:<...> code. Add
 19:06
whiteknight it bugs the hell out of me when a big fancy IDE doesn't have some of the basic capabilities built-in that my shitty notepad-replacement software does 19:50
Tene What is it this time, and which fancy IDE are you using? 19:51
One of my co-workers has started renaming programs in our subversion repo to add .pl and .sh suffixes, because he says Eclipse can't figure out the syntax hilighting without it. 19:52
whiteknight I'm using VisualStudio, and it doesn't have some things like guide lines, decent syntax highlighting for some languages, etc 19:54
yeah, being able to autodetect those kinds of things would be nice, but I've never seen an editor do that properly 19:57
even just detecting the shebang line would be nice, though incomplete
dalek kudo: 2be1d04 | (Solomon Foster)++ | src/core/operators.pm:
Handle the infix:<@lhs, @rhs> case we forgot about last time.
20:03 davidfetter joined 20:04 Psyche^ joined 20:06 silug joined
davidfetter oh hai 20:10
20:15 Chandon joined
whiteknight hello davidfetter 20:18
...and goodbye davidfetter
davidfetter i hope it wasn't something i said
darbelo I doubt he finds 'oh hai' offensive. 20:19
cotto_work I dunno. He's killed people for less. 20:20
Though I might be thinking of someone else.
Could someone going to YAPC::NA make sure that good notes get taken at the Lorito meeting? 20:27
atrodo When is the Lorito meeting? 20:32
kthakore atrodo: now!
atrodo: you are missing it! OMG RUN! 20:33
atrodo: no time for that youtube video! RUN! RUN!
atrodo ! I forgot to drive to Columbus this morning!
kthakore OH NO!
atrodo Srsly, can everyone wait two hours while i get there?
kthakore I will.
atrodo: I will wait in canada 20:34
20:35 silug joined
dalek kudo: af53808 | pmichaud++ | docs/spectest-progress.csv:
spectest-progress.csv update: 489 files, 33280 (83.4% of 39887) pass, 37 fail
 20:38
davidfetter tries to figure out how quickly he could get to columbus
kthakore davidfetter: 20 hr by count 20:39
21:01 eternaleye_ joined
kthakore wow ... 21:16
updating parrot repo after a couple months
cotto_work yeah. We've got some changes. 21:17
kthakore cotto_work: I saw that ... I wonder if parrotSDL broke
cotto_work: isn't NCI scheduled for workage next? 21:18
darbelo Probably.
kthakore yay!
\\o/
|o/
darbelo It's one of our gsoc projects.
kthakore \\o?
cotto_work no idea
kthakore darbelo: cool
darbelo: where is the guys blog so I can cyber stalk him
cotto_work It is if it becomes a pain point or if someone feels like working on it. 21:19
21:19 whiteknight joined
kthakore cotto_work: darbelo just said it was a gsoc 21:19
cotto_work That's true.
kthakore whiteknight: hola! just finished my exam finally! so I am tracking the NCI gsoc guy
darbelo I think ash_ is doing that one.
whiteknight kthakore: nice 21:20
kthakore whiteknight: and updating git-svn repo of parrot
darbelo And plobsing is mentoring him.
kthakore oh cool
whiteknight ash_ is doing it, yes.
kthakore I wonder is parrotSDL will be broke again
but then I didn't even release it yet
or get much done
cotto_work only one way to find out
kthakore cotto_work: yeah I am exploring today 21:21
s/win 2 21:23
wow 21:33
still going
did you guys rewrite all of it?
whiteknight: so ... where is ash_'s blog on NCI ? 21:38
whiteknight: also where can I learn how NCI works and junk and sutff
darbelo We have junk and suff all over the place. Pick a direction and start digging >( 21:39
;) 21:40
darbelo curses at his misconfigured keyboard.
kthakore darbelo: ok but where is this guys blog! ash_'s NCI gsoc stuff
cotto_work ash's blog? 21:41
kthakore yeah
cotto_work I was seeing if purl knew. 21:42
kthakore oh ok
FAIL! bit.ly/b4B2GP 21:43
how are people supposed to know what is being done?
man that site sucks
google should know better
is he submitting directly to trunk? 21:45
or a branch or something
GeJ Bonjour everyone.
kthakore hi GeJ 21:46
nci?
purl nci is Native Call Interface or the National Cancer Institute, a part of the NIH or national cancer institute
kthakore wowo ..
cotto_work ash's blog is www.parrot.org/ash
kthakore am I stupid today or waht
that should have been ovious
oh he is using libffi 21:47
cool
now where is da code!
dalek kudo: 736bef8 | masak++ | docs/ChangeLog:
[docs/ChangeLog] some further updates
 21:48
kthakore h mmm ...
darbelo I think his code is on github.
kthakore darbelo: any search terms to narrow it down? 21:49
darbelo His github id escapes me now.
cotto_work eriku iirc
darbelo Let me check in the irclogs.
kthakore wow this should really be on his blog or a wiki
or something
purl something is, like, really wrong out there :)
darbelo github.com/ashgti/parrot 21:50
purl github.com/ashgti/parrot are in there
kthakore darbelo: wow thanks
darbelo: lot harder then it should be to find this info
cotto_work ash's code?
purl ash's code is a circle
cotto_work no, ash's code is github.com/ashgti/parrot 21:51
purl okay, cotto_work.
darbelo The gsoc_nci branch is the one you want.
I think.
kthakore cotto_work: heh. well I was refering to a page
darbelo: how does one do multiple parrot installs 21:53
multiple parrots?
maybe a parrot brew is in order
darbelo Never tried it, but as long as you install them to different dirs, you should be fine. 21:54
kthakore darbelo: he is commiting to master
darbelo: I think I can follow most of what was done
jeez you guys have a like a million tests 21:55
darbelo: cotto_work thanks
cotto_work dukeleto, ping 21:58
21:58 eternaleye joined
cotto_work kthakore, np 21:59
dukeleto cotto_work: pong 22:00
kthakore hola dukeleto
cotto_work dukeleto, could you write a version of docs/project/branching_guide.pod for git?
dukeleto cotto_work: yes 22:01
kthakore yay! tests pass
cotto_work thanks
kthakore cotto_work: almost to finding out if parrotSDL is broken
dukeleto cotto_work: can you make a TT and assign it to me?
cotto_work not only can I, but I will
dukeleto kthakore: hello
purl what's up, dukeleto.
dukeleto kthakore: managing multiple parrots would be useful for me as well 22:02
kthakore oh yay! ... SUPER SUPER SUPER FAIL :( 22:03
dukeleto: yeah parrot-brew would be sooo nice 22:04
cotto_work trac.parrot.org/parrot/ticket/1681
kthakore printerr changed? 22:05
printerr "Hint: create a link from libSDL-1.2..." 22:06
we need a ( now?
cotto_work It's a dynop now. 22:07
dukeleto kthakore: printerr is not loaded by default now 22:08
cotto_work .loadlib 'io_ops' #take one of these and call me in the morning
dukeleto kthakore: you can use getstderr to get the proper filehandle without loading dynops, if you want
kthakore ???
wow I am so lost now
dukeleto kthakore: .loadlib 'io_ops' is probably what you want 22:09
kthakore: but in the case where you don't want to use dynops, there is a core op to get the stderr file handle
kthakore great! all cmod are no dynops too ... 22:10
wow
dukeleto kthakore: .loadlib 'bit_ops'
cotto_work do you use those?
kthakore it is in StopWatch.pir
dukeleto: I have a feeling I might have to start parrotSDL from scratch now 22:11
dukeleto kthakore: why so?
kthakore dukeleto: is this stuff likely to change again?
dukeleto: stopwatch.pir is pointless for a core SDL library
dukeleto: frankly the SDL pirs in parrot core are a joke 22:12
dukeleto: they expose nothing of SDL for a parrot end user
button.pir .... um no
dukeleto: and there is no configure to find SDL libraries
you can't even do export SDL_LIBS_HERE_MON 22:13
dukeleto: so that the NCI can find them
*sigh* ...
dukeleto: I need to learn PIR from scratch the new one again
dukeleto kthakore: hard to say. More things will become dynops, I highly doubt that any dynops will ever become core ops, but it could happen
kthakore right that stuff is fine
dukeleto: but the SDL/*.pir being a joke are unrelated 22:14
dukeleto: I mean you can't do anything with them right now ...
dukeleto: I might use Rect.pir and so on but the other stuff Button.pir and Stopwatch.pir are going 22:15
dukeleto: I would suggest you guys get rid of them in the trunk too
dalek TT #1681 created by cotto++: git version of branching guide 22:16
TT #1681: trac.parrot.org/parrot/ticket/1681
dukeleto kthakore: if you create a TT with an explanation, we can do that stuff
kthakore dukeleto: better yet I update parrotSDL new API shortly in github.com/kthakore/parrotSDL and bug you guys then 22:17
dukeleto: meanwhile I should take a break after a 5 hr exam and 6 hrs of work ....
dukeleto: peace out 22:18
purl Yeah homey.. I'm outtie
kthakore cotto_work: cya thanks
whiteknight: bye
ps is there a new guide for PIR and NCI for the new parrot some place? 22:19
bit.ly/9LIBzS seems 15 months oldie 22:21
cotto_work: ?? doc.parrot.org is at 2.4.0 ?
no 2.5.0 doc yet? 22:22
dukeleto kthakore: bug Coke about that
yeah, that PIR guide is pretty hoary
kthakore *sigh* ... perl6 SDL just got longer ... 22:23
cotto_work I'm capable of fixing it now. I'll try to remember to do it when I get home tonight.
kthakore cotto_work: oh thx!
is there a section of NCI somewhere?
cotto_work docs/pdds/draft/pdd16_native_call.pod may be useful 22:24
kthakore ... ok
wow ... that was short 22:25
cotto_work the key word being "may"
kthakore adds getting some real docs done for NCI to the list 22:26
cotto_work: I will try to doc my misadventures in perldoc format
cotto_work: do people do Doc work for parrot or am I just in a sparse land of parrot? 22:27
dalek kudo: 22b71e0 | masak++ | docs/announce/2010.06:
[docs/announce/2010.06] added
 22:35
22:42 cotto_work joined
cotto_work kthakore: people write docs when they write docs. More is better. 22:43
kthakore cotto_work: actually more of outdated docs is never better IMO
cotto_work: more of actually updated docs for each release makes sense
cotto_work sure, but hopefully docs aren't out of date while they're being written
kthakore cotto_work: which is the point many perl5 guys make about parrot and perl6 22:44
cotto_work: you guys seem to change the syntax so often that when are we going to be sure an app written now won't work in 4 months
cotto_work: I made parrotSDL work 3 months ago .. and now it is all broken in so many ways 22:45
cotto_work That's a fair criticism.
kthakore when do you guys just say ok this is the syntax for a long term release
cotto_work: are you guys even close to that?
dalek kudo: 8a1a1a5 | masak++ | docs/ (2 files):
[ChangeLog, announce] improvements
 22:46
kthakore oh well it is as it is 22:48
may in another 10 years it would have stablized
cotto_work: night for now
22:50 bubaflub joined
cotto_work kthakore: night 22:50
dalek kudo: 6b12f89 | masak++ | docs/announce/2010.06:
[announce] more fixes
 22:52
22:55 eternaleye joined
dalek kudo: 5eb3875 | masak++ | docs/announce/2010.06:
[announce] more recent passing-tests figure
 22:58
cotto_work kthakore, I'd encourage you to be noisy (especially on parrot-dev) when you get bitten by backwards-incompatibility. The better we know how our changes cause pain for users, the better we can be at avoiding that pain. 23:00
23:03 davidfetter joined, mikehh joined
dukeleto cotto_work: maybe it would be nice if we could provide a document with each release geared towards helping people update their code after deprecations 23:04
cotto_work: for instance, DEPRECATED.pod only lists *future* deprecations, and such, it totally useless after the fact
cotto_work I was thinking something similar. Drupal did something like that that made migrating modules between versions minimally painful. 23:05
drupal.org/update/modules 23:10
23:11 eternaleye_ joined
cotto_work drupal.org/node/244569 is a decent example 23:11
dukeleto yeah, Parrot really needs "how to upgrage" documents. That is a big sore point with HLL devs 23:14
23:20 d4l3k_ joined
kthakore cotto_work: will do. I hope I can be noisy without being considered rude 23:20
cotto_work We'll understand if you're annoyed. 23:21
dukeleto kthakore: you are not alone, it is just that nobody has been squeaky yet 23:22
kthakore: i will back you up on parrot-dev, no worries :)
kthakore ok 23:23
23:23 snarkyboojum joined
kthakore feels like a goat lead to the slaughter house 23:23
cotto_work Don't. You're developing on top of Parrot. We want you to be happy. 23:24
kthakore then wear are my hookers?
where* 23:25
cotto_work They must have the wrong address. 23:27
This could get awkward in a hurry.
kthakore heh 23:28
kthakore waits patiently at his mailbox
I will make an email tonight. 23:29
darbelo I-ll start thinking up excuses. 23:32
dalek kudo: cbff7d0 | masak++ | docs/release_guide.pod:
[docs/release_guide.pod] updates and improvements
 23:51
whiteknight has to merge his branch soon, now that rakudo is out 23:56
dalek kudo: 30db218 | masak++ | build/Makefile.in:
[Makefile.in] added missing '.' in error message
 23:57

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!