Parrot 3.6.0 "Pájaros del Caribe" released | parrot.org | Log: irclog.perlgeek.de/parrot/today | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 23 July 2011.
davidfetter ? 00:02
jay ping bubaflub 00:10
bubaflub pong jay
jay I'm having some odd NCI things, can I show you something? Need a set of eyes...
bubaflub jay: please do 00:11
jay I'll make a gist
gist.github.com/1112869 I have working code for getting the max of a float array, but the obvious changes for int arrays isn't working. I also tried the 'long' versions, and that didn't work, either. 00:13
The changes just seem so trivial, I'm at a loss.
dalek kudo/nom: 34bf1be | jonathan++ | src/ (3 files):
Add a place where we can persist state variables between call frames.
 00:15
kudo/nom: 6850cd0 | jonathan++ | src/pmc/perl6lex (2 files):
Get state variables essentially working; don't handle binding and closure cases correctly yet.
bubaflub jay: i'm not seeing a function named 'gsl_stats_int_max' in docs (www.gnu.org/software/gsl/manual/htm...lues.html) 00:17
jay www.gnu.org/s/gsl/manual/html_node/...stics.html
See the note at top... but I figure it would crash or give an invoke() error if the function weren't in the library. 00:18
00:18 fperrad left
bubaflub jay: hmmm, ok. well, you can check if what it tries to dlfunc is null or not 00:18
jay Not sure what you mean. 00:19
bubaflub jay: sorry for being unclear. the first line loads something into the variable named "nci" 00:20
jay It actually returns an answer but a garbage answer. At first I suspected it was expecting 4-byte ints but getting longs or something, but I think I tried every combination.
bubaflub the second line throws that stuff into the namespace 00:21
jay Right... and yes, gotcha.
bubaflub between those two lines you can check if the variable NCI is null
jay oh, ok.
bubaflub though that might not be the problem; lemme check
jay er: say nci ? what's the trick, my PIR on the spot isn't good enough 00:22
bubaflub you can just say 00:23
unless nci goto no_function
no_function:
die 'oh noes'
or something like that
jay: you've tried it with ipii and ipll and both don't work, right?
jay I'm using the right signatures from the header files... and the ll works with the float version, so I doubt that is it. 00:24
kid51 soh_cah_toa++ hits the IMCC wall!
soh_cah_toa yeah, big time 00:25
jay unless nci goto ok_now
no_function:
die 'oh noes'
ok_now:
kid51 soh_cah_toa: Which branch has your most recent code?
soh_cah_toa kid51: soh-cah-toa/hbdb
dalek kudo/nom: 296d6b8 | jonathan++ | src/Perl6/Metamodel/BOOTSTRAP.pm:
Get the closure semantics for state variables correct (bit of copy-paste here, but it's hot-path-ish...).
jay It is barfing it seems (unless my unless is backwards).
bubaflub: Ok, so it seems that you are right: it isn't getting the dlfunc correctly? 00:27
bubaflub jay: i'm looking at the source
gsl.sourcearchive.com/documentation...ource.html
00:27 kid51 is now known as kid51_at_dinner
bubaflub and it looks like they might be doing something tricky with macros 00:27
though again, i don't think that's the problem
because it works for floats
jay Maybe there is a build flag that needs to be set properly for the non-float versions or something.
bubaflub: nm libgsl.so | grep stats definitely shows that it thinks it is in the .so file, if that helps. 00:29
bubaflub jay: it does. though i'm at a loss of how to debug further - is plobsing around? 00:30
jay Hm. Well, at least I don't think I've missed something obvious. I can live with floats for now, honestly, and come back to this later. I should try a trivial bit of C code working with the library.
bubaflub jay: i wonder if some of the difficulty is the way that these functions are generated 00:31
jay You're looking at the source gsl code now? 00:32
bubaflub yeah
here is minmax.c
gsl.sourcearchive.com/documentation...ource.html
which does some stuff with #define's and some templates
the actual source minmax_source.c
gsl.sourcearchive.com/documentation...ource.html
looks like it uses some templating features
for example 00:33
BASE
FUNCTION (gsl_stats,min) (const BASE data[], const size_t stride,
const size_t n)
{
jay yes, exactly what I'm looking at.
bubaflub and i imagine each time it's included BASE is set to int or float or whatever
i mean, it's a clever way of saving having to write a lot of code 00:34
but i'm not sure how our NCI system handles it
though if it's in the .so, we should be able to load it
jay Ok. Thanks for looking. I'll verify that I can use it with a simple C code example, and if that works I'll come back to it again in Parrot.
bubaflub jay: i know dukeleto has some Perl5 code that uses GSL - perhaps he'll have some pointers 00:36
jay ok... bubaflub++ for the old college try... ! 00:39
bubaflub heh. sorry i couldn't be of more help - i had trouble getting GSL on Mac OS X previously
jay Ah. Baby steps. float is good enough for now. 00:40
whiteknight ...I found a bug in my :tag implementation already
ah, nevermind. bug was between couch and keyboard 00:44
soh_cah_toa whiteknight: ew, new house has bugs in it? yuck 00:46
:) 00:47
00:49 preflex joined
whiteknight actually...there is a bug still 00:50
and it's a doozie
00:53 preflex_ left
whiteknight imcc is changing the interp->current_pf and not resetting it 00:54
jay ping bubaflub 00:58
bubaflub pong jay
soh_cah_toa shudders every time he sees that word
bubaflub jay: i'll be around for about another hour
jay My unless was wrong. gist.github.com/1112918 So both say True, so both functions are dlfunc-ed it appears. So it isn't that.
Anyway, I'm back to where I started, I'll leave it for now and just test it from C code later. Just wanted to let you know. 00:59
bubaflub jay: ok. well, that's at least one thing we know that hasn't failed.
jay Yup. Good to have that sanity check, though. 01:00
dalek rrot/whiteknight/imcc_tag: e828461 | Whiteknight++ | / (3 files):
IMCC changes interp->current_pf for compilation but did not reset it. Fix that. Add some tests for new :tag syntax to t/pmc/packfileview.t
rrot/whiteknight/imcc_tag: 5e3743f | Whiteknight++ | / (7 files):
Merge branch 'whiteknight/imcc_tag' of github.com:parrot/parrot into whiteknight/imcc_tag
soh_cah_toa whiteknight: what's the imcc_tag branch for?
whiteknight soh_cah_toa: adds a new :tag syntax, which is an intended replacement for flags like :load and :init 01:01
soh_cah_toa oh ok
whiteknight instead of having a small number of predefined flags, like :load and :init, we can associate any arbitrary string with any pmc, and look them up by group 01:02
soh_cah_toa interesting
Felipe hi guys 01:05
jay Bubaflub: it appears that there are multiple libgsl.so files on my system:
soh_cah_toa Felipe: hey
jay libgsl.so (libc6,x86-64) => /usr/local/lib/libgsl.so
libgsl.so (libc6) => /usr/lib32/libgsl.so
If it was finding the 32-bit version, I imagine this might be the issue, no?
bubaflub jay: no bueno. that may be causing some troubles.
jay Ok, well, that is the best guess we have at this point I think. 01:06
cotto_work tt #2165 is even more fun if you omit the print statement 01:12
it fails with a 0-byte alloc 01:13
jay bubaflub: I gave the explicit path to the 64-bit version in the loadlib, and got the same failures. Good try though.
bubaflub jay: hmmm. ok. could be a mismatch between how parrot was configured and how that library was compiled... 01:14
dalek rrot/whiteknight/imcc_tag: 0fab325 | Whiteknight++ | src/pmc/imccompiler.pmc:
some much-needed cleanups to imccompiler, to make sure we enter and exit compilations in a standard way and reset values we change
 01:25
rrot: 2a0b46c | soh_cah_toa++ | t/library/osutils.t:
Made the code in t/library/osutils.t a little more readable and the test descriptions more consistent.
 01:26
rrot: 0434949 | soh_cah_toa++ | t/library/ (2 files):
Added test for cp() to t/library/osutils.t. Also created a 'testlib' directory so that mkpath() and others don't operate on the root directory.
whiteknight msg plobsing I added some tests for :tag to t/pmc/packfileview.t. I think I'll have a few more to add, but this is basic functionality that I want to preserve as a baseline. 01:28
aloha OK. I'll deliver the message.
01:31 kid51_at_dinner left 01:33 GodFather left
dalek rrot/whiteknight/imcc_tag: 6952d3f | Whiteknight++ | t/pmc/packfileview.t:
Add test to prove the IMCC bug I just fixed is actually fixed and stays that way
 01:38
01:39 whiteknight left 01:40 contingencyplan left 01:44 kid51 joined
dalek rrot/soh-cah-toa/hbdb: 8ef69b3 | jkeenan++ | config/gen/makefiles/root.in:
Add dependencies to Makefile so that checkdepends.t passes.
 02:02
soh_cah_toa kid51++ didn't realize that was failing 02:04
kid51 'make test' is your friend :-)
A cage-cleaner's got to do what a cage-cleaner's got to do. 02:05
soh_cah_toa :)
dalek R: 958a1bd | (Jay Emerson)++ | src/nqr (4 files):
Expanded c() for strings; added min and max for float arrays
 02:07
02:15 bubaflub left
dalek rrot: 7b9b14c | soh_cah_toa++ | t/library/osutils.t:
Added tests for install() and unlink() to t/library/osutils.t.
 02:20
R: ace65bc | (Jay Emerson)++ | src/nqr (3 files):
Implemented var and sd
 02:37
02:42 rdesfo joined
jay My first Segmentation fault! 02:46
dalek R: 1bef9ae | (Jay Emerson)++ | src/nqr (2 files):
Added covariance and correlation
 02:48
02:55 particle1 joined, rdesfo left 02:58 particle left
Coke (partcl-old is using some pasm. FYI) 02:59
dalek R: 2d7a9e4 | (Jay Emerson)++ | src/nqr (2 files):
exp() and log()
 03:02
kudo/nom: cd07404 | Coke++ | t/spectest.data:
track spectest file err statuses
 03:03
rrot: 702c3ee | jkeenan++ | t/library/osutils.t:
Merge branch 'master' of git@github.com:parrot/parrot
 03:09
rrot: 35fae10 | jkeenan++ | config/gen/makefiles/ (2 files):
[cage] hard-tabs on Makefile turnover lines.
03:12 kid51 left
jay NQP question if anyone is up: if $a is a ResizableFloatArray, does $b = $a actually make a copy or would it get the reference only? If a reference only, is there a built-in "deep copy" or do I have to write the loop myself? 03:19
soh_cah_toa jay: nqp only has binding, no assignment 03:20
sorear jay: that ambiguity is the exact reason why NQP forbids "="
soh_cah_toa so it would be $b :=$a
and yes, it would get a reference
sorear := is explicitly a pointer copy operation
jay Right. Typo. So I need to write the loop in other words.
Not hard, obviously, but I wanted to ask in case there was a speed-optimized deep copy waiting to be used. 03:22
sorear I beleive that is spelled pir::asign 03:23
assing
assign
I can spell
jay sorear++ thanks I'll check that out. 03:24
soh_cah_toa what's a deep copy? never heard of that before
sorear assign is probably only one level deep
jay Maybe we made it up, but we use the term for actually making duplicate copies of data. 03:25
sorear soh_cah_toa: deep_copy($x) { $x is aggregate ?? map(&deep_copy, $x) !! $x }
soh_cah_toa ah, ok cool
sorear the "deep copy" in Parrot is pir::clone 03:28
but I doubt it works
jay lol
sorear it's based on freeze/thaw, which has been in a sorry way for years
dalek R: 0140076 | (Jay Emerson)++ | src/nqr (2 files):
sort() done (need decreasing option, though)
 03:30
soh_cah_toa sorear: is there a p6 equivalent for p5's caller()?
sorear soh_cah_toa: p6 has a &caller function, it doesn't work quite the same but it provides the same functionality 03:35
soh_cah_toa ok 03:36
sorear give me a usage to translate
soh_cah_toa oh, i was just trying to figure what would be involved in porting Carp to p6 and was looking at `my $something = defined &{"CORE::GLOBAL::caller"} ? &{"CORE::GLOBAL::caller"}() : caller()` for a minute 03:37
sorear I don't think porting Carp to p6 makes much sense 03:38
the major functions of Carp are built-in in p6
soh_cah_toa oh ok, didn't realize that
i find that a lot w/ the "essential" perl 5 modules: p6 does it by default 03:40
which is obviously a good thing
dalek R: 65318ac | (Jay Emerson)++ | src/nqr (2 files):
Attempted order(); have integer passing problem with GSL
 03:41
sorear soh_cah_toa: that's sort of the point 03:42
soh_cah_toa: a lot of perl 5 core modules are stuff that should be (or in some cases, technically is) builtin, but can't be exposed by default for compatibility reasons 03:43
cotto ~~
sorear soh_cah_toa: perl 6 was invented as a clean break from the past
soh_cah_toa right 03:44
03:56 soh_cah_toa left 03:57 soh_cah_toa joined
dukeleto ~~ 04:02
cotto hio dukeleto 04:06
dukeleto cotto: howdy
04:18 woosley joined
dafrito I seem to have found a bug in parrot-lua. It generates different PIR output for identical lines of code (with the second generated PIR causing an exception): pastebin.com/c585b6xJ 04:20
I'm working on debugging it, but I'm sort-of stumped, was hoping for some advice/ideas 04:21
soh_cah_toa dafrito: lua is fperrad's territory so he'd be the first person i'd go to 04:23
04:23 daniel-s joined
dafrito soh_cah_toa, thanks, I guess I'll write em an email (or should I post it on the mailing list?) 04:25
soh_cah_toa dafrito: parrot-dev is always good 04:26
that way the other brainiacs who aren't here right now can take a look as well 04:29
dafrito good call, will do :) 04:30
04:43 ascent_ left, ascent joined 06:07 contingencyplan joined, GeJ left, GeJ joined, woosley left, particle joined, Hunger- left, eternaleye_ joined, woosley joined, Kulag left, benabik_ joined, simcop2387 left, Kulag joined, tempire_ joined, tempire left, simcop2387 joined, silug left, jjore left, slavorgn left, Hunger joined, particle1 left 06:08 eternaleye left, jjore joined, benabik left, benabik_ is now known as benabik 06:09 mj41 joined 06:14 slavorgn joined 06:32 ppant joined 06:34 soh_cah_toa left 06:50 fperrad joined 07:03 particle left, particle joined 07:17 cosimo left 07:38 fperrad left 07:44 fperrad joined 08:00 silug joined 08:01 particle1 joined 08:04 particle left 08:24 AzureSto_ joined 08:26 AzureStone left
dalek kudo/nom: 1f2a242 | moritz++ | src/core/ (2 files):
start to expand on atan2 implementation
 08:37
kudo/nom: 926a0ba | moritz++ | src/core/Mu.pm:
Mu.so
kudo/nom: cbdfa61 | moritz++ | README:
[README] add pointer to status page
08:46 cosimo joined 08:53 plobsing left 08:58 treed joined 08:59 treed left
dalek kudo/nom: 2176102 | moritz++ | / (2 files):
fix Hash.pairs etc. for empty hashes; run 4 more test files
 09:00
09:07 particle joined 09:11 particle1 left 09:22 ppant left
dalek kudo/nom: c57bfad | jonathan++ | src/Perl6/Actions.pm:
A couple of state variable fixes.
 09:27
09:32 woosley left
dalek kudo/nom: 9ec3312 | jonathan++ | t/spectest.data:
Run S12-class/inheritance.t.
 09:45
kudo/nom: bb18757 | jonathan++ | src/Perl6/Actions.pm:
Undo something I didn't really want to commit.
 09:55
10:27 ambs joined
dalek kudo/nom: 2416f7a | moritz++ | src/core/ (2 files):
copy Str.subst over from master
 10:35
kudo/nom: 84ca50d | moritz++ | / (2 files):
Merge branch 'nom' of github.com:rakudo/rakudo into nom
kudo/nom: a5aaddb | moritz++ | docs/compiler_overview.pod:
start to tell fewer outdated things in docs/compiler_overview.pod
kudo/nom: 98198e8 | moritz++ | src/Perl6/ (2 files):
try to get s/// working again.

Currently dies with Method 'clone' not found for invocant of class 'Sub'
11:34 contingencyplan left
allison turns out Storable isn't the best format for persistent data storage 11:47
having a bit of trouble in the Perl 5.10->5.12 transition for Parrot PMC build tools in Debian packaging 11:48
12:03 daniel-s left, kurahaupo joined 12:06 JimmyZ joined 12:10 whiteknight joined
whiteknight good morning, #parrot 12:11
jay ~ whiteknight
whiteknight good morning jay 12:13
interesting warning about the new java release: www.lucidimagination.com/blog/2011/...-anything/ 12:28
12:31 woosley joined 13:02 bubaflub joined
bubaflub ~ 13:04
13:08 ambs left 13:13 Coke left 13:15 bubaflub left
whiteknight good morning, bubaflub 13:15
whiteknight is smoldering a win7_x64 parrot with --optimize now 13:16
13:16 bubaflub joined
bubaflub morning whiteknight 13:19
13:31 bubaflub left 13:32 bubaflub joined
bubaflub whiteknight: you mentioned getting 6model as a roadmap goal 13:33
whiteknight I just sent out an email to that effect 13:34
but yes, I think it's an important goal
bubaflub whiteknight: you have my axe
whiteknight awesome!
I can't distract you from GSOC though, not till the project ends
bubaflub whiteknight: yeah, that's a few weeks into August
whiteknight how is your project going, by the way? You on schedule or close enough? 13:35
bubaflub whiteknight: i'm still a bit ahead of schedule
whiteknight: it's completely useable as it is right now - just need to finish out the docs and the tutorials and examples 13:36
(which is my goal for this week)
13:37 bluescreen joined
whiteknight bubaflub: awesome 13:38
those are the really important parts. The quality of the docs, examples, and tests will determine whether anything uses your software in the future, and if they use it, whether they can rely on it
bubaflub yup. i finally got the keys to our new place last night; we'll be moving in today and the weekend, and get internet on monday
whiteknight lucky bastard
bubaflub whiteknight: i agree. i'd like some NQP examples and more indepth examples. also a tutorial on *how* to do what i did with a library 13:39
jay < will be reading it >
dalek R: e1e90d1 | (Jay Emerson)++ | src/nqr (3 files):
Cleaning out junk; did is.array() properly
 13:41
jnthn__ needs to find tuits to fill out more of the 6model docs he started 13:43
whiteknight I need to re-read the most recent versions of them
jnthn__ Need to at least do a working example of class-based and prototype-based OO. 13:44
jay Looking like a language? --> 13:46
> set.seed(1,2)
> a <- rexp(30, 1.0)
> print(paste("ok", which.max(sort(a))))
ok 29
bubaflub jay: very nice. 13:47
dalek R: 2245594 | (Jay Emerson)++ | t/00-sanity.t:
Added test of some new functions
13:52 plobsing joined
dalek R: 3ec3307 | (Jay Emerson)++ | VERSION:
Updated minor version number
 14:03
R: de51f2b | (Jay Emerson)++ | / (2 files):
Fixed is.array() return bug
R: d9194fd | (Jay Emerson)++ | / (2 files):
Added warning messages on some string operations
R: 2b30bc2 | (Jay Emerson)++ | / (2 files):
More tests added
 14:21
14:28 mj41 left 14:37 bubaflub left, bubaflub joined 14:52 kurahaupo left, bubaflub left 14:53 bubaflub joined 15:15 jay left 15:24 jay joined 15:26 dod left
dalek R: 1172038 | (Jay Emerson)++ | src/nqr/ (2 files):
Cleaning house
 15:31
tadzik oh, NQP lives in a house :) 15:33
JimmyZ it's not NQP, it's NQR 15:35
tadzik ah, a typo on my side :) 15:37
15:41 woosley left 15:42 bubaflub left, bubaflub joined
dalek Heuristic branch merge: pushed 138 commits to rakudo/podparser by tadzik 16:00
kudo/nom: df778cc | jonathan++ | src/Perl6/Metamodel/TypePretence.pm:
Add a type_check method to TypePretence; ParametricRoleHOW wants to play smarter, but it'll suffice for normal pretenders.
 16:19
kudo/nom: 9f3d032 | jonathan++ | / (4 files):
PackageHOW and ModuleHOW do TypePretence, so now they act as if they're subtypes of Any (and thus Mu) for purposes of type checking.
kudo/nom: 08945fc | jonathan++ | / (5 files):
Make module and package type objects have the same methods as Mu and Any, so you can do 'say SomePackage' or pass package type objects around; sould also make SomePackage.WHY work for tadzik++ who reported the issue.
R: 3971ef1 | (Jay Emerson)++ | / (4 files):
May have fixed up ResizableStringArray usage
 16:21
R: 0d7034b | (Jay Emerson)++ | / (2 files):
Fixed up string testing in a few cases
R: 21ee2d1 | (Jay Emerson)++ | STATUS:
Updated status to reflect string fixing and testing
16:31 jay left
dalek kudo/podparser: ab7dc6b | tadzik++ | / (43 files):
Merge branch 'nom' into podparser
 16:35
kudo/podparser: df778cc | jonathan++ | src/Perl6/Metamodel/TypePretence.pm:
Add a type_check method to TypePretence; ParametricRoleHOW wants to play smarter, but it'll suffice for normal pretenders.
kudo/podparser: 9f3d032 | jonathan++ | / (4 files):
PackageHOW and ModuleHOW do TypePretence, so now they act as if they're subtypes of Any (and thus Mu) for purposes of type checking.
kudo/podparser: 08945fc | jonathan++ | / (5 files):
Make module and package type objects have the same methods as Mu and Any, so you can do 'say SomePackage' or pass package type objects around; sould also make SomePackage.WHY work for tadzik++ who reported the issue.
kudo/podparser: 3a7c5f6 | tadzik++ | / (6 files):
Merge branch 'nom' into podparser
kudo/podparser: 6947907 | tadzik++ | t/pod/why.t:
Use .WHY on packages and modules properly, jnthn++
16:41 JimmyZ left
dalek rrot/whiteknight/load_bytecode: 2c4c7db | Whiteknight++ | src/packfile/api.c:
Add the LOADED_PBCS cache to the load_bytecode_s op. Don't use PackFile_append_pmc, it's not needed here.
 16:58
dukeleto does anybody know *exactly* how the parrot binary detects the difference between PASM and PIR ? Is it file extension? Or "deep content inspection" ?
whiteknight file extension
dukeleto whiteknight: indeed. Renaming a .pasm to .pir makes it bork 16:59
whiteknight: is there an automated way to go from PASM -> PIR ? Or is it only a one-way operation to go from PIR->PASM ? 17:01
whiteknight there's no way to go from either to the other
no way that works
there was a way to output badly-broken PASM-like code from PIR. I don't know if we still have it 17:02
17:02 theory joined
atrodo whiteknight> pbc_disassemble 17:02
dukeleto whiteknight: oh yeah, i was thinking of PIR->PBC.
whiteknight I may have ripped that out as part of the IMCC refactors
dukeleto atrodo: so PIR->PBC->PASM is possible?
the more I learn about PASM, the worse of an idea it seems
atrodo dukeleto> I'm not saying that. I'm saying pbc_disassemble exists and outputs PASM like code 17:03
s/like/looking/
I've never tried to feed it back into parrot
whiteknight pbc_disassemble does not output working code
it outputs a code-like substance
17:04 soh_cah_toa joined
soh_cah_toa ~~ 17:05
dukeleto that is horrible
whiteknight: do we have a good document that describes all the differences between pasm and pir ?
whiteknight i would be very surprised 17:06
cotto_work ~~
atrodo While I'm here, maybe someone can save me time. What kind of PMC constants can be in a PBC? 17:09
whiteknight any kind, if it implements freeze/thaw
17:11 soh_cah_toa left
dukeleto karma pasm 17:12
aloha pasm has karma of 0.
dukeleto pasm--
tadzik karma imcc 17:13
aloha imcc has karma of -25.
17:15 jay joined
dalek kudo/nom: 66b3d21 | jonathan++ | src/Perl6/Actions.pm:
Should be able to mention self in attr init.
 17:17
kudo/nom: 0455eb5 | jonathan++ | src/ (2 files):
Implement can. Unfortunately, ?() turns out to be True...if it's meant to be, will have to do this differently somehow.
whiteknight s/pbc_disassemble with no arguments is not useful/pbc_disassemble is not useful/ # Fixed that for you
dalek R: 4be85ca | (Jay Emerson)++ | / (3 files):
Updated notes
 17:21
17:21 Drossel joined, Kulag left, soh_cah_toa joined
soh_cah_toa msg fperrad thanks for "distutils-izing" the plumage manpage i made. it's a whole lot cleaner that way. what exactly is the `inst_data` key for anyway? how does it know the manpage goes in /usr/share/man? the distutils perldoc is pretty poor 17:22
aloha OK. I'll deliver the message.
whiteknight soh_cah_toa: inst_data is used to install data files to a shared directory 17:23
Rosella is using it now to install template files
soh_cah_toa whiteknight: what defines a "data file" and what do you mean "shared directory"?
whiteknight a data file is a thing that isn't an executable file 17:25
and a shared directory, on linux, would be something like /usr/local/share/....
soh_cah_toa ah ok
dalek TT #2166 created by dukeleto++: pbc_disassemble with no arguments is not useful 17:26
TT #2166: trac.parrot.org/parrot/ticket/2166
rrot/whiteknight/load_bytecode: 4545413 | Whiteknight++ | src/packfile/api.c:
Remove commented out code. We don't need it
17:36 theory left
dalek R: ad936ba | (Jay Emerson)++ | src/nqr/Grammar.pm:
Grammar notes
 17:37
17:38 jay left
atrodo cotto_work> Has any work happened on Mole? Or is it just the one gist? 17:41
cotto_work atrodo: it's still just the gist. I'm still finding enough holes in M0's spec and implementation that I don't want to start actually hacking out a compiler quite yet. 17:49
atrodo cotto_work> okay, thanks 17:50
cotto_work atrodo: why do you ask? 17:51
atrodo cotto_work> curiosity more than anything. I'm trying to work through pbc->m0 still and I thought of it 17:52
cotto_work atrodo: I'm glad you're thinking about that. It'll become very important. 17:54
atrodo: if you see anything that M0 needs to do, please let me know.
atrodo cotto_work> I feel like I'm getting closer, but there's so much magic bottled up in pbc
cotto_work actually, reviewing PIR to make sure I can hand-wavingly convince myself that M0 can do what's needed would be a good item for the M0 todo list 17:55
stupid list keeps getting longer
atrodo cotto_work> My current theory is having the m0 system, with m0 pmc's, by itself then have a parrot compatibility pmc 17:57
17:58 dmalcolm joined
cotto_work atrodo: using PMCs as the compatibility layer to start using M0 in Parrot seems like it'd be a pretty thick layer. 18:01
I was thinking something more lower-level, more like a normal shared library. 18:02
atrodo cotto_work> I'm interested in this idea 18:03
*to hear more about this idea
18:09 tempire_ left
dalek kudo/nom: 95168b5 | pmichaud++ | src/core/control.pm:
Fix multi-argument control functions (return, take, etc.).
 18:10
umage: e15a437 | soh_cah_toa++ | / (3 files):
Changed 'info' command to 'metadata' and now 'info' prints out a nice summary similar to `yum info`. Added new tests for this and changed manpage to reflect new changes.
 18:17
umage: d5a5428 | soh_cah_toa++ | src/plumage.nqp:
Renamed print_metadata() to print_project_summary() as the previous name was a bit misleading.
 18:21
soh_cah_toa if anyone else has plumage installed, does the 'status' command throw a 'too many named arguments: 1 passed, 0 used' exception for you too? 18:27
w/ or w/o the optional 'project' argument. it doesn't matter 18:29
dalek umage: 2b7c49d | soh_cah_toa++ | src/plumage.nqp:
Added :$command argument to command_status() so that it doesn't throw a 'too many named arguments: 1 passed, 0 used' exception.
 18:40
18:46 ash_ joined 18:49 Coke joined 19:04 Eclesia joined
Eclesia hello 19:04
soh_cah_toa Eclesia: howdy 19:05
19:05 Coke left 19:06 Coke joined
Eclesia I'm still searching to compile the opengl module of parrot. if someone has the list of requiered libs or do know what is missing that would help 19:08
soh_cah_toa Eclesia: sorry, don't know much about opengl. if you do `git blame runtime/parrot/library/OpenGL.pir`, you can see who has edited the file in the past. i'd asked them first since they'd obviously know at least a little bit about that library 19:17
Eclesia Peter Lobsinger seems to be the last one to work on it 19:24
soh_cah_toa yeah 19:25
Eclesia ping plobsing, sorry to disturb you, are you there ? 19:26
19:32 contingencyplan joined
bubaflub whiteknight: would i put my generated NCI thunk shared library nonsense in inst_data? 19:35
whiteknight I don't think so. That would only be for things like data files and docs and stuff 19:36
bubaflub nuts.
plobsing Eclesia: pong 19:39
Eclesia plobsing: hi
plobsing: I would like to know what is missing on my system to compile the opengl module. is there some compile options to show that ?
plobsing so you're trying to do something with opengl? what problem are you running into?
ah 19:40
Eclesia plobsing: so far, the compilation is always skipped :/
plobsing that would likely mean Configure.pl doesn't think you have opengl.
if you are comfortable with Perl, the code to detect that is in config/auto/opengl.pm 19:41
Eclesia plobsing: that's what I'm suspecting, but I installed several mesa/gl/glu packages and still have the module skipped
plobsing what OS are you on? 19:42
Eclesia plobsing: sorry but I'm not at all a friend of perl . I come from the java world ^^
ash_ it tries to compile github.com/parrot/parrot/blob/mast...pengl_c.in and link against "-lglut -lGLU -lGL"
Eclesia plobsing: Lubuntu 11.04 x64
plobsing ok, 'cp config/auto/opengl/opengl_c.in test.c', then 'cc -lglut -lGLU -lGL test.c' 19:45
finally, './a.out'
this is what the configure process does, so doing it manually should show where it fails
Eclesia ./a.out 19:46
4 freeglut
plobsing ok, so you have the headers installed. 19:47
do you have libpcre available? shouldn't have to be libpcre-dev.
Eclesia checking 2sec 19:48
ash_ make reconfig ? (forces it to re-run the configure scripts)
Eclesia plobsing: I have libpcre3 installed but not libpcre3-dev
plobsing what about './parrot_config HAS_PCRE'? 19:49
Eclesia 0 19:50
plobsing so configure isn't picking up your pcre install (which is currently a dependancy for our opengl bindings) 19:51
so a similar drill as before - manually running the configure detection steps to see where it fails 19:52
Eclesia plobsing: success :) I added the libpcre3-dev and It didn't skipped the opengl module 19:53
plobsing first, 'cp config/auto/pcre/pcre_c.in test.c', then 'cc -lpcre test.c', finally './a.out'
oh it worked? sweet~!
Eclesia it also added the libpcrecpp0 package. don't know which one was in fault
plobsing: If I may suggest : I think the list of requiered packages should be added somewhere in the wiki/doc 19:54
plobsing patches welcome ;) 19:56
ash_ wouldn't one of the port files list the dependencies? 19:58
under /ports/debian for instance
Eclesia hmm ... I know javadoc structure and mercurial ... but perl and git . I will take me a few hours ^^ 19:59
19:59 writtencode1 left
Eclesia it* 19:59
plobsing Eclesia: we tend to keep that kind of doc away from our code. I'm not sure where it exactly belongs, but likely nowhere near any Perl code. 20:01
ash_: dependancies get a little fuzzy for infrequently used optional extensions
ash_ maybe under PREREQUISITES there should be a list of optional dependencies 20:06
such as PCRE and OpenGL
Eclesia plobsing: does that mean the opengl module was some kind of experiment, not intended but used widely ? 20:09
20:09 kurahaupo joined
whiteknight opengl extensions haven't been maintained for a while 20:13
Patterner I can't use PCRE at all, because /usr/lib64/libpcre.so is a linker script that points to /lib64/libpcre.so.0 and parrot cannot handle that (the reason is an old gcc bug...) 20:16
NotFound Uh? Opengl works fine for me.
nopaste "Eclesia" at 192.168.1.3 pasted "opengl bug" (7 lines) at nopaste.snit.ch/65817
Eclesia not yet working for me ... 20:17
20:17 mj41 joined, ambs joined
Eclesia plobsing: you have an idea on what's causing this error ? 20:19
20:21 bluescreen left 20:24 whiteknight left, perlite_ joined
NotFound Eclesia: What says glxinfo in your display? 20:25
nopaste "Eclesia" at 192.168.1.3 pasted "glxinfo" (404 lines) at nopaste.snit.ch/65818 20:26
Eclesia NotFound: here it is 20:27
NotFound Looks fine. 20:28
20:28 perlite left, perlite_ is now known as perlite
NotFound Eclesia: Have you done a realclean after solving the pcre issue? Maybe some missing dependency is causing a faulty build. 20:31
Eclesia NotFound: I did 20:32
haha, good news for parrot : www.lucidimagination.com/blog/2011/...-anything/ 20:51
plobsing Eclesia: that looks like an xwindows error. I'm not sure what parrot can do about that. 21:01
you could try bisecting the code with debug prints, or running with --trace to determine where to problem is (although I strongly suspect it is in glutCreateWindow). 21:03
Eclesia :( 21:05
21:08 jay joined
dalek umage/gh-pages: 384621a | soh_cah_toa++ | index.html:
Added self to 'Authors' section and changed reference to `usage` command to `help`.
 21:13
umage: 75ec150 | soh_cah_toa++ | / (4 files):
Added a new 'help' command with both one and two argument forms. Removed it from TODO list and updated manpage. Should be used in favor of 'usage'.
umage: 6c6daad | soh_cah_toa++ | CREDITS:
Alphabetized and added self to CREDITS. Also added a description of each field.
umage: 11ee1d5 | soh_cah_toa++ | / (2 files):
Added the optional [COMMAND] argument for `help` to manpage and output message.
21:14 bubaflub left 21:16 ambs left 21:25 fperrad left
dalek tracwiki: v5 | soh_cah_toa++ | GettingStartedWithPlumage 21:25
tracwiki: Changed reference to usage</tt> command to <tt>help</tt>.
tracwiki: trac.parrot.org/parrot/wiki/Getting...ction=diff 21:26
soh_cah_toa agh, stupid wiki formatting
21:35 Eclesia left 21:44 Psyche^ joined 21:49 Patterner left, Psyche^ is now known as Patterner 21:54 mj41 left 22:06 soh_cah_toa left 22:47 zostay_ left 23:04 dmalcolm left 23:09 Khisanth left 23:12 GodFather joined 23:13 Khisanth joined
dukeleto anybody ever messed with pygments before? bitbucket.org/birkenfeld/pygments-main 23:26
i am toying with the idea of adding syntax highlighting for PIR and possibly Winxed, which would mean we would get syntax highlighting on Github, since that is what they use
tadzik they don't use something Ruby? 23:27
23:28 whiteknight joined
dukeleto tadzik: nope, they use pygments 23:28
tadzik: github.com/github/linguist/pull/58 23:29
23:29 ash_ left 23:30 ash_ joined 23:34 ash_ left
whiteknight good evening, #parrot 23:34
dalek rrot: 8d5cbd7 | petdance++ | src/gc/fixed_allocator.c:
consting an art in pool_is_owned(), and removed an unused var
 23:37
rrot: f726b3c | Whiteknight++ | src/packfile/api.c:
Add the LOADED_PBCS cache to the load_bytecode_s op. Don't use PackFile_append_pmc, it's not needed here.
 23:42
rrot: 2e5eb2a | Whiteknight++ | src/packfile/api.c:
Remove commented out code. We don't need it

Powered by Cro and the Raku® Programming Language.

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

Thank you!