Topic for #parrot is: Parrot 3.3.0 released | parrot.org | Log: irclog.perlgeek.de/parrot/today
Set by moderator on 20 April 2011.
whiteknight Parrot has so much unlearning to do 00:11
packfiles are starting to keep me awake at night
benabik Can't sleep, packfiles will eat me?
whiteknight we're working so hard to get them working correctly, and it's a huge uphill battle
dukeleto ~~ 00:13
00:14 Coke left, Coke joined
whiteknight packfiles are so fundamental. They're used everywhere 00:15
but so many places don't use them the same ways 00:17
I have no idea why Parrot-Instrument isn't working now 00:24
it's saying the code to execute is outside the current bytecode segment 00:25
but I'm calling the same exact functions as every Parrot program calls on startup to run the bytecode, so I know it should work
soh_cah_toa while looking at src/library.c, it looks like runtime/dynext is the only place libs are searched for (after my fix). is that right? 00:26
for libraries, that is. not .include 00:30
just loadlib's
whiteknight there should be some items from config too 00:37
dalek p/ctmo: 0936e24 | jonathan++ | src/pmc/sixmodelobject.pmc:
Turns out the last fix wasn't quite the whole story; this should be closer to it.
 00:39
p/ctmo: 7c92758 | jonathan++ | src/ (2 files):
Install subs in packages through the SC.
p/ctmo: 5a37cdf | jonathan++ | src/stage0/ (7 files):
Update bootstrap.
p/ctmo: 7d0bcd9 | jonathan++ | src/Regex.pir:
Fix up PAST and _dumper imports so they make entries in GLOBALish. Need to do this until they're converted to NQP.
soh_cah_toa config only has perl modules though
whiteknight no, the config PMC 00:40
soh_cah_toa oh
well, i'm not too familiar w/ the config pmc. would it suffice to test that cwd fails and that runtime/parrot/dynext passes? 00:41
00:41 kid51 joined
soh_cah_toa that's why i was asking 00:41
and then do the same for include dir
and runtime/parrot/library 00:42
i was gonna test that .include, load_bytecode, and loadlib all don't search the cwd and that they do search their respect dirs - runtime/parrot/include, runtime/parrot/library, and runtime/parrot/dynext respectively 00:43
01:02 kid51_ joined 01:06 kid51 left
dalek rrot: 9c08d31 | Whiteknight++ | src/ (2 files):
test that the PackFile* is not null inside the place where we actually attempt to use it
 01:13
rrot: d7b3258 | Whiteknight++ | / (3 files):
add in a new packfile api function to get the main Sub from a PBC pmc. Use this from the embedding API to get the main sub
01:14 whiteknight left, Coke left, Coke joined 01:15 woosley joined 01:33 Drossel left 01:34 Kulag joined 01:51 Kulag left 01:52 Kulag joined 01:58 jrtayloriv joined 01:59 kid51_ left 02:09 Drossel joined 02:11 Kulag left 02:17 Drossel left 02:18 ShaneC joined 02:20 atrodo joined 02:22 Kulag joined 02:30 ShaneC left 02:37 plobsing joined 02:48 ShaneC joined 03:03 bubaflub left 03:12 soh_cah_toa left 03:16 plobsing_ joined 03:20 hudnix left, plobsing left
dalek rrot: 5d9ed81 | petdance++ | lib/Parrot/Pmc2c/ (4 files):
fixing up includes
 03:44
dukeleto blog.zpao.com/post/4620873765/about...key-engine 03:48
losing marketshare makes strange bedfellows 03:49
atrodo that is rather odd 03:51
dukeleto atrodo: indeed 03:55
cotto: looks like I am talking at LFNW after all www.linuxfestnorthwest.org/sessions...ration-git
dalek rrot/leto/embed_grant: e8c9f86 | dukeleto++ | t/src/extend_vtable.t:
Improve test for Parrot_PMC_set_pmc_keyed
 04:05
rrot: f73898f | petdance++ | src/platform/ (2 files):
fixing up SHIM interpreters and function annotations
 04:17
05:06 alin joined 05:09 alin left 05:13 theory left 05:27 soh_cah_toa joined 05:29 birdwindupbird joined 05:31 soh_cah_toa left 05:37 utsl left, utsl joined 05:57 utsl left, utsl joined
dalek Heuristic branch merge: pushed 170 commits to parrot/opsc_llvm by bacek 06:45
06:55 rohit_nsit08 joined 06:56 woosley left
rohit_nsit08 hello #parrot 07:11
tadzik hello rohit_nsit08 07:19
rohit_nsit08 tadzik: Hello 07:20
07:27 rohit_nsit08 left 07:28 birdwindupbird left 07:29 rohit_nsit08 joined 08:04 SHODAN joined 08:09 woosley joined, woosley left
rohit_nsit08 less than 11 hours left in gsoc result. 08:18
08:34 rohit_nsit08 left 09:39 ambs joined 09:59 ambs left
dalek rrot/opsc_llvm: 9a03493 | bacek++ | config/gen/makefiles/root.in:
Add CXX compilation rules.
 10:06
rrot/opsc_llvm: 2973176 | bacek++ | / (2 files):
Add llvm_extra.cpp with supplimentary LLVM functions.
rrot/opsc_llvm: 7ad3085 | bacek++ | runtime/parrot/library/LLVM (4 files):
Bind and use new dumping functions
rrot/opsc_llvm: 90b57bd | bacek++ | t/library/llvm/05-module.t:
Add usage/testing of new LLVM::Module.dump.
rrot/opsc_llvm: 48694c4 | bacek++ | t/library/llvm/99-example.t:
Add more LLVM::Module.dump tests. Because we can!
10:16 lucian joined
dalek rrot/opsc_llvm: 67c4b8b | bacek++ | src/llvm_extra.cpp:
Use template specialization for dumping modules. Also rename 'do_print' to 'dump_to_string'
 10:20
bacek msg cotto Can you please check code.google.com/p/llvm-py/source/br.../extra.cpp and github.com/parrot/parrot/blob/opsc..._extra.cpp and add proper copyright notice? 10:23
aloha OK. I'll deliver the message.
dalek p/ctmo: ef63ac2 | jonathan++ | src/NQP/Actions.pm:
Correct and clean up global lookup logic.
 10:26
p/ctmo: 6f58a6d | jonathan++ | src/HLL/SerializationContextBuilder.pm:
Make serialization Event class lexical inside of the serialization context builder. Partly because we can, also because it avoids a NYI in the new package handling.
p/ctmo: 3a9acf3 | jonathan++ | src/NQP/Actions.pm:
Don't hardcode KnowHOW in the actions, but grab it from the current %*HOW.
rrot/opsc_llvm: 1fa34b4 | bacek++ | src/llvm_extra.cpp:
Specialize only related part of dumping of LLVM::Stuff.
 10:43
11:02 lucian left
dalek rrot/jit_prototype: c65ee16 | bacek++ | runtime/parrot/library/LLVM/Value.pm:
Add metadata accessors
 11:04
11:05 lucian joined
dalek p/ctmo: c67bd2d | jonathan++ | / (7 files):
The compiler so far has been using the HOWs that it is implemented with rather than the ones defined by the meta-objects library the program we're compiling uses, which was obviously wrong. This fixes it, by introducing a mechanism for exporting stuff in a way that will see it end up in %*HOW. Note, not official API for this. And yes, this means you can now customize what class means in NQP by writing a meta-object in a module and stuffing it into
 11:16
11:22 estrabd left, estrabd joined
dalek p/ctmo: 26f67b1 | jonathan++ | src/HLL/SerializationContextBuilder.pm:
A little chop with the circularity saw. :-)
 11:52
p/ctmo: 0c29c72 | jonathan++ | src/NQP/Grammar.pm:
We never actually install KnowHOW in the package (yet), so just obtain it directly from the 6model core for now.
p/ctmo: 9e9d521 | jonathan++ | src/Regex/P6Regex/Actions.pm:
Add a missing 'our' that we somehow got away with before.
11:57 Patterner left 12:04 Psyche^ joined, Psyche^ is now known as Patterner
dalek Heuristic branch merge: pushed 53 commits to parrot/jit_prototype by bacek 12:15
12:20 woosley joined 12:22 mtk joined
dalek p/ctmo: 5613e50 | jonathan++ | src/NQP/Actions.pm:
Switch over the majority of lookups to the new packages model. A bunch of new regressions to hunt down, but most things work.
 12:29
p/ctmo: 628e050 | jonathan++ | src/NQP/ (2 files):
Load the Regex library by default; NQP users expect Regex::Match and so forth to be to hand by default. Recovers some of the broken tests.
p/ctmo: f4a923d | jonathan++ | src/NQP/Actions.pm:
Fix GLOBAL::foo and contextual fallback. Fixes regressions in package var test script, apart from one that is dubious.
p/ctmo: c17da73 | jonathan++ | build/Makefile.in:
Need to actually respect declaration order of things now. This fixes the regressed roles tests - it's just a reordering in the meta-objects file.
12:33 benabik left
dalek p/ctmo: 12f14b1 | jonathan++ | t/hll/05-commandline.t:
We don't import HLL automatically any more, so explicitly do so in the 05-commandline.t tests. They all pass.
 12:36
12:39 BloodyCoin joined 12:40 whiteknight joined
whiteknight good morning, #parrot 12:41
redline.st/
12:41 hudnix joined
whiteknight eventually, we should revive smalltalk on parrot 12:43
12:43 S_Arms left 12:44 mtk left 12:49 mtk joined
dalek TT #2102 created by coke++: test failure in latest macport. 12:54
TT #2102: trac.parrot.org/parrot/ticket/2102
12:57 ambs joined 13:02 mtk left 13:05 bluescreen joined 13:09 mtk joined 13:10 alin joined
dalek p/ctmo: fb745e0 | jonathan++ | t/hll/01-language.t:
Another missing use HLL.
 13:16
p/ctmo: dcc0d96 | jonathan++ | src/Regex/Method.pir:
Partial fix for broken regex interpolation tests.
13:19 bubaflub joined 13:21 lucian_ joined 13:24 lucian__ joined 13:26 darbelo joined, lucian left 13:27 lucian_ left 13:28 lucian joined 13:32 lucian__ left
dalek rrot: 4856c36 | Whiteknight++ | / (4 files):
break out the bytecode-related routines from the embedding API into their own file, bytecode.c. Bytecode is a very important part of the embedding API, and it's worthwhile to have all the routines for it in a single place. After this update, you are probably going to need to reconfig to get the makefile changes
 13:52
p/ctmo: 3fdc2de | jonathan++ | build/Makefile.in:
Keep CLEANUPS up to date, so make clean actually cleans.
 14:01
p/ctmo: eec5353 | jonathan++ | src/ModuleLoader.pm:
D'oh. It's all well and good every module getting a clean slate (aka separate compilation), but it really shouldn't enforce its limited world view on its loader. Particularly when it's loader is the compiler, which would really rather its own symbols don't go missing. Fixes the broken bootstrap.
p/ctmo: e9a452a | jonathan++ | src/stage0/ (7 files):
Update the bootstrap.
 14:02
14:05 Andy_ joined 14:08 particle joined 14:11 particle1 left 14:15 JimmyZ joined 14:17 mj41 joined 14:23 ambs left 14:27 alin_ joined 14:30 alin left, alin_ is now known as alin, lucian_ joined 14:35 lucian left
whiteknight grr... these packfile problems in parrot-instrument are really starting to get on my nerves 14:42
14:45 JimmyZ left
whiteknight I know I'm getting close 14:48
14:54 theory joined 15:02 GodFather joined
whiteknight okay, calling Parrot_pf_execute_bytecode_program somehow modifies the current bytecode 15:11
and I feel like that's wrong
dalek rrot: f5401ed | Whiteknight++ | src/packfile/api.c:
to execute bytecode, we MUST have a valid packfile and a valid packfile->cur_cs.
 15:20
rrot: 793d0b7 | Whiteknight++ | src/embed/bytecode.c:
save and restore the current executing bytecode when we execute a packfile, just in case we are doing a nested call
rrot: d52cc32 | Whiteknight++ | src/ (2 files):
save and restore the previous bytecode in the correct place
rrot: 3934d20 | Whiteknight++ | / (3 files):
cleanup Parrot_pf_set_current_packfile, preserve old pf over call to IMCCompiler.compile_file
15:27 sigue left 15:32 bubaflub left 15:34 sigue joined 15:35 sigue left, sigue joined 15:40 sigue left, sigue joined 15:43 dmalcolm joined
whiteknight blah, i give up for now 15:44
Andy_ my src/embed/bytecode.o is blowin' up 15:45
Which I'm assuming is yours.
from that batch of upushes
whiteknight blowing up? how?
Andy_ src/embed/bytecode.c: In function ‘Parrot_Int Parrot_api_load_bytecode_bytes(Parrot_PMC, const unsigned char*, Parrot_Int, PMC**)’:
src/embed/bytecode.c:90:52: error: cannot convert ‘PMC**’ to ‘PMC*’ for argument ‘4’ to ‘void do_sub_pragmas(Parrot_Interp, PMC*, pbc_action_enum_t, PMC*)’
15:45 sigue left
whiteknight hmm...weird. I didn't see that on my machine 15:46
i'll fix it in a bit
dalek rrot: c13d4ef | Whiteknight++ | src/embed/bytecode.c:
fix typo, should be PMC* not PMC**
 15:48
whiteknight Andy_: try that, please
plobsing_ Does anyone know how OrderedHash is supposed to work? 15:50
or for that matter, why we still have it in core?
Andy_ ok, rebuidling 15:51
15:51 bubaflub joined
plobsing_ nopaste.snit.ch/41768 shows a program I would expect to print 'baz', but it prints '1' 15:51
dalek p/ctmo: cd12868 | jonathan++ | src/ (2 files):
Refactor to ensure proper handling of --module-path.
 15:55
p/ctmo: 04274af | jonathan++ | build/Makefile.in:
Use --module-path in stage2 building. This should mean that we're achieving real separate compilation.
p/ctmo: 6080095 | jonathan++ | src/stage0/ (6 files):
Update bootstrap with latest fixes.
p/ctmo: cbaf9e0 | jonathan++ | build/Makefile.in:
Use --module-path in stage 1 build, so it's also doing separate compilation.
Andy_ whiteknight: now it's not linking. 15:56
wait, that might be me 15:57
whiteknight it is linking fine here 16:00
moritz plobsing_: I guess it returns an RPA with one element
Andy_ whiteknight: All good here in linking. Now running tests. 16:01
whiteknight plobsing_: no idea. I don't know how OrderedHash works, or why it's in core
plobsing_: if I had my druthers, it would be gone
Andy_ Mmmm, code deletion.
whiteknight Andy_: awesome. Thanks for the report!
16:01 sigue joined
Andy_ I'm running splint now. 16:01
i'll see if anything new pops up.
whiteknight Andy_++
plobsing_ turns out the test that was testing for that was added by a GCI student just to up our coverage. I feel no guilt skipping it now. 16:04
whiteknight As best I remember it, OrderedHash was manually walking the linked list of hash entries
although maybe I'm confusing that with UnManagedStruct
All the FAIL parts of parrot combine together in my head 16:05
plobsing_ moritz: nah. it has some odd, tightly-coupled interaction with Key that always returns '1', no matter what the subsequent key entries are
Andy_ whiteknight: Only line number changes.
whiteknight Andy_: okay, that's good.
16:05 sigue left 16:07 sigue joined 16:14 GodFather left 16:15 sigue left, sigue joined
dalek rrot/tt1931-nci-parameters-deprecation: df298c1 | plobsing++ | src/nci/libffi.c:
implement generic call-by-reference

Any argument type can be called by reference.
Call-by-ref arguments are returned as multi-returns, after the native return value if not void, in the order they occur in the signature.
Also, clean up libffi dynamic thunk generator. Should have less allocator-churn.		 16:15
rrot/tt1931-nci-parameters-deprecation: c8b0124 | plobsing++ | t/pmc/nci.t:
update test from UnManagedStruct API to Ptr/StructView API
rrot/tt1931-nci-parameters-deprecation: 73ef4d0 | plobsing++ | src/pmc/nci.pmc:
add interface to allow creation of NCI objects from Ptr objects from HLL
rrot/tt1931-nci-parameters-deprecation: 35f5d9c | plobsing++ | src/pmc/key.pmc:
add aggregate interface to Key

Allows keyed (by int only to avoid recursion problems) access on Key objects. This in turn allows code expecting Keys to be more general and accept other objects and more general, aggregate-expecting code to operate on keys correctly.		 16:16
16:16 darbelo left
rrot/tt1931-nci-parameters-deprecation: 0ac9739 | plobsing++ | t/pmc/orderedhash.t:
skip silly defacto test
16:20 sigue left 16:21 sigue joined
dalek p/ctmo: a60d76e | jonathan++ | t/nqp/43-package-var.t:
Toss a bogus test; TimToady++ for confirming bogusness.
 16:24
p/ctmo: 319fc03 | jonathan++ | src/NQP/Actions.pm:
Make typename rule lookup symbols properly, not using legacy packages.
p/ctmo: ace800f | jonathan++ | src/NQP/Actions.pm:
Add parent class using compile time meta-object. Also give decent compile time error indicating missing parent classes.
p/ctmo: ec8a1e6 | jonathan++ | src/HLL/SerializationContextBuilder.pm:
Locate meta-object via the SC. This way, we'll be able to have lexically imported/exported meta-objects, etc...plus it's a faster lookup.
p/ctmo: aab21a5 | jonathan++ | src/stage0/ (6 files):
Update bootstrap.
16:27 woosley left 16:28 sigue left 16:33 sigue joined, sigue left 16:35 sigue joined 16:38 Andy_ left 16:39 sigue left 16:40 jevin left, jevin joined 16:45 sigue joined 16:49 dodathome joined 16:50 sigue left
dukeleto ~~ 16:51
tadzik o/ 16:52
dukeleto tadzik: how goes it?
tadzik dukeleto: oh, I'm killing time waiting for the GSoC results :) 16:54
bubaflub tadzik: same here...
cotto_work ~~
dukeleto cotto_work: mornin' 16:55
tadzik and in Parrotland, submitted a fix for the ipv6 detection I mentioned the Select PMC to... cotto, hello :)
dukeleto tadzik: ooh, nice!
cotto_work dukeleto: I saw that you'll be speaking at lfnw too
dukeleto wishes there was an escape hatch for email
16:55 jrtayloriv left
dukeleto cotto_work: yep! Should be fn. 16:55
16:55 sigue joined
dukeleto fun, even. 16:55
16:56 sigue left 16:58 sigue joined
dalek p/ctmo: 5375140 | jonathan++ | src/Regex/Cursor.pir:
Make sure Regex::Cursor ends up in an SC.
 16:59
p/ctmo: e42c76e | jonathan++ | src/HLL/SerializationContextBuilder.pm:
Improved error.
p/ctmo: ea71689 | jonathan++ | src/NQP/Actions.pm:
Set default parent via the SC.
p/ctmo: 9f893e0 | jonathan++ | src/NQP/Actions.pm:
Add roles via compile time meta-object; good error if role not found.
17:01 Andy_ joined 17:02 sigue left, sigue joined 17:09 benabik joined 17:11 sigue left 17:12 sigue joined
dukeleto cotto_work: have you thought about lodging for LFNW ? 17:21
17:22 rohit_nsit08 joined
cotto_work dukeleto: nope. I haven't had much time to think this weekend. 17:23
17:24 ShaneC left
rohit_nsit08 hello #parrot 17:24
whiteknight hello rohit_nsit08 17:25
tadzik hello rohit_nsit08
rohit_nsit08 whiteknight: hello, waiting for the results 17:26
whiteknight cotto_work: not to clog your thinker any more, but I am making progress on Parrot-Instrument. We're running into some weird packfile-related bugs that I think are inside Parrot
bubaflub rohit_nsit08: same here
rohit_nsit08 goo.gl/JPGx3
whiteknight rohit_nsit08: yes. That's soon. I'm in the middle of drafting emails right now
exciting times!
rohit_nsit08 the countdown
cotto_work whiteknight: no worries. The weekend was crazy, but it's over now. 17:27
tadzik this counter is weird. One hour _and_ 90 minutes?
oh, I'm stupid 17:28
benabik Same confusion here. But each line is separate...
rohit_nsit08 tadzik: they are not related
whiteknight dukeleto: I'm only going to enter a short message for the acceptance emails. I'm drafting a larger email with all the necessary information and send that out separately
rohit_nsit08 minutes onlyl
whiteknight dukeleto: I'll send you a copy first for proofing
17:28 ambs joined
dukeleto whiteknight: where is this acceptance email stuff? 17:28
whiteknight: i don't see it
benabik tadzik: The "Alternative version" has the more normal version of "1 hour, 31 minutes"
whiteknight dukeleto: it's under the foundation, if you edit it
tadzik benabik: yeah, I noticed
whiteknight there are two text boxes at the bottom.
tadzik just the first thing I saw were the biggest letters (who would've guessed?), and that made me wonder 17:29
whiteknight Damnit Jim, I'm a coder not a proof-reader
benabik whiteknight: Just write the acceptance letters in PIR. ;-) 17:30
dukeleto whiteknight: i see it now
rohit_nsit08 benabik++
dukeleto whiteknight: i am working on the acceptance/denied emails. 17:31
whiteknight: you were being way to nice ;)
whiteknight dukeleto: okay, awesome. Like I said I have a large information email that I'm drafting. I'll send you a copy of that soon 17:32
dukeleto too nice, even.
whiteknight: coolio
17:32 sigue left
dukeleto whiteknight: please include this link google-opensource.blogspot.com/2011...-code.html 17:32
whiteknight: which is for students
whiteknight: google-opensource.blogspot.com/2011...de_21.html is for mentors
17:38 ShaneC joined 17:39 sigue_ joined 17:40 benabik left 17:41 sigue joined, sigue left 17:44 sigue_ left
dukeleto whiteknight: ok, I beefed up our acceptance email. Sadly, working on the rejection one now. 17:46
whiteknight plobsing: ping 17:50
plobsing_ whiteknight: pong
whiteknight plobsing_: do you have time today or soon to take a look at Parrot-Instrument?
17:50 sigue_ joined
whiteknight I'm seeing some packfile-related problems there, and I suspect an eye more educated on the topic than my own could help solve it 17:50
17:50 sigue joined, sigue left
plobsing_ whiteknight: I'm finishing up the tt1931 branch. I intend to have a look at Parrot-Instrument while I wait on the pre-merge testing for that branch. 17:51
whiteknight plobsing_: okay, awesome. No huge rush on it. I've fixed a few bugs since the last time I messaged you, so I can fill you in on the details of the failures I am seeing 17:52
plobsing_ I just need to fix up the few core NCI libs that use the now-removed functioinality
whiteknight okay, you need any help with that?
dukeleto whiteknight: ok. acceptance+rejection stuff is done
whiteknight dukeleto++
dukeleto whiteknight: please take a look and see if you agree with it
whiteknight I like it, yes 17:53
dukeleto: I just emailed you that draft I've been working on. Take a leisurely glance at it when you have time 17:54
plobsing_ whiteknight: I should be good. PCRE has gone mostly smoothly. Plus, without those examples, it might be a little hard to understand how things are supposed to work now.
whiteknight it's a bigg'un
17:55 sigue_ left, sigue joined 17:57 mj41 left 18:00 sigue left, benabik joined, sigue joined 18:11 whiteknight left, whiteknight joined 18:17 sigue left, sigue joined 18:20 lucian_ left 18:22 sigue left 18:23 sigue joined 18:26 lucian joined 18:28 sigue left, sigue joined 18:33 sigue left 18:37 sigue joined
dukeleto runs some meatspace errands 18:39
benabik 1000 seconds until GSoC announcement 18:43
whiteknight THE SUSPENSE!
I suspect everybody who is going to be accepted already knows whether their proposal was good enough or not
18:43 sigue left
whiteknight I don't think there are going to be any surprises 18:44
benabik There's always that little nagging doubt.
whiteknight of course!
benabik has butterflies in his stomach... That or the leftovers he had for lunch were older than he remembered. 18:48
18:50 sigue_ joined 18:53 lucian_ joined
whiteknight zomg zomg 4 minutes left! 18:55
actually, where are results being posted?
atrodo I'm excited and I'm not a mentor or student 18:56
18:57 lucian left
benabik whiteknight: That's an awesome question. 18:57
19:00 soh_cah_toa joined, sigue_ left
rohit_nsit08 got the acceptance mail :-) 19:00
soh_cah_toa yayy!!!!!!!!!!!!!!!!!!!!!!
dukeleto allison: ping
rohit_nsit08++
soh_cah_toa++
rohit_nsit08 dukeleto: thanks 19:01
soh_cah_toa rohit_nsit08: good for you man. i'm so excited!
bubaflub booyah!
benabik Yes, apparently google sends e-mails.
bubaflub cheers!
benabik does a victory dance, as the e-mail suggests
rohit_nsit08 soh_cah_toa: thanks , and congratz
whiteknight heh. google-melange.com is down. 500 error
dukeleto lulz. of course. 19:02
SHODAN www.google-melange.com/gsoc/project...e/gsoc2011
benabik More like "too many students hitting reload" error.
whiteknight yeah, all those students need to calm down and let me look at it first!
bubaflub keep hitting F5 till it shows up
soh_cah_toa homepage works fine but not the above link 19:03
whiteknight awesome
atrodo hits f5 a few more times 19:04
19:04 sigue joined
tadzik mdk.per.ly/2011/04/25/gsoc-the-tpf-...-students/ 19:05
not from Parrotland, but I can't help the excitement, sorry :')
benabik tadzik: Looks like most of that is P5?
atrodo So PaFo got 7? 19:06
19:06 sigue left
tadzik benabik: yeah, I'm all alone in the sixland 19:06
benabik tadzik: You can rest happy in the knowledge that your version of Perl is one higher than any other GSoC student. :-D 19:07
whiteknight atrodo: we ended up with 7, yes
dalek rrot/tt1931-nci-parameters-deprecation: 1c71ff5 | plobsing++ | runtime/parrot/library/ (2 files):
update PCRE bindings to new NCI API
rrot/tt1931-nci-parameters-deprecation: 75e8f19 | plobsing++ | src/nci/ (2 files):
fix call-by-ref problems exposed by PCRE bindings
whiteknight originally, we received 9
rohit_nsit08 whiteknight: btw who is my mentor? 19:08
atrodo whiteknight> thanks, just making sure melange was telling me the right thing
whiteknight rohit_nsit08: we're sending out an email with all the details later. Your mentor is Coke
soh_cah_toa oh yeah, i don't even know who mine is
rohit_nsit08 whiteknight: thanks, going for the party :-) 19:09
whiteknight awesome 19:10
soh_cah_toa: I think cotto is your mentor
jnthn__ Yay, the PAST => NQP migration is in the approved \\o/ 19:11
soh_cah_toa alright, kickass
benabik jnthn__: Yeah, I get to deal with sorting out PAST::Compiler.
jnthn__ benabik: Oh, it's you :) 19:12
whiteknight benabik: your mentor is bacek
19:12 sigue joined
jnthn__ benabik: I somehow hadn't put the two together before. :) 19:12
Awesome :)
benabik jnthn__: That's why I pestered you to figure out what your patch was doing. :-D
whiteknight: Woo!
jnthn__ benabik: Aha! 19:13
benabik: I...er...wrote quite a few NQP patches in the last few days :)
benabik jnthn__: I've been following along since before GSoC, although I'm probably just going to be using nqp-rx (aka parrot-nqp). 19:14
jnthn__ Oh :/
Though, it'll likely almost "just work" on new nqp anyway :)
whiteknight we have to use what we have
if we can get parrot ported over to 6model, we can start using new NQP in core Parrot 19:15
benabik I don't want to try to force parrot to switch over, although I may experiment with ctmo branch on new PAST and vice-versa.
whiteknight ...as if we need any more motivatins
jnthn__ benabik: It should be easy to actually have it work on both.
Or relatively easy, anyway. 19:16
At the language level, not a crazy amount has changed between them.
benabik jnthn__: Since PAST really doesn't care about the object model, that's my hope. The trickier part will be getting the PAST->POST->bytecode chain solid enough to get NQP to compile on it.
jnthn__ Yes.
19:17 whiteknight left
jnthn__ I wrote a little langauge compiler recently and ran it on both nqp-rx and nqp and I think it Just Worked. :) So if you stick to normal NQP syntax, it should all work out pretty easily. :) 19:18
benabik \\o/
19:19 whiteknight joined
whiteknight I'm starting to get very excited about GSoC 19:20
I can't wait to see some of the rest of the students
19:21 lucian_ left
moderator Parrot 3.3.0 released | parrot.org | Log: irclog.perlgeek.de/parrot/today” | Accepted GSoC Students announced! | GSoC student information emails coming out soon
19:22 sigue left 19:23 sigue joined 19:31 bubaflub left
benabik Looking forward to working with you all! But for right now, I have to head to class. 19:33
19:34 sigue left, sigue joined 19:35 benabik left 19:41 sigue left, sigue joined 19:46 sigue left, sigue joined
allison dukeleto: pong 19:47
19:48 rohit_nsit08 left 19:57 SHODAN left 20:00 bubaflub joined
dalek p/ctmo: fb7d90b | jonathan++ | src/NQP/Actions.pm:
Toss unneeded old package installation code; remove a XXX comment that's been long dealt with.
 20:05
p/ctmo: 8f36dee | jonathan++ | src/ModuleLoader.pm:
Explicitly stash the module loader object away somewhere we'll be able to find it even after old package installation code is dropped.
p/ctmo: 25b6538 | jonathan++ | src/ (2 files):
Get the regex interpolation tests passing again, after I busted 'em earlier.
20:12 sigue left 20:13 sigue joined
dalek rrot/tt1931-nci-parameters-deprecation: b362d1a | plobsing++ | runtime/parrot/library/libpcre.pir:
minor fixups to pcre nci interface
 20:15
20:15 perlite left
20:17 perlite joined 20:18 sigue left, sigue joined 20:23 sigue left, sigue joined, theory left 20:26 theory joined 20:28 bluescreen left 20:30 dodathome left 20:35 sigue left, whiteknight left 20:38 bluescreen joined 20:56 Coke left 20:57 Coke joined
tcurtis is very excited about GSoC. 20:58
21:00 sigue joined 21:01 davidfetter joined 21:07 Coke left, Coke joined 21:15 Coke left, Coke joined
dalek p/ctmo: 4579af8 | jonathan++ | t/hll/0 (3 files):
Skip a few tests for now that need significant revision due to design/spec changes.
 21:16
Heuristic branch merge: pushed 134 commits to nqp by jnthn 21:19
21:23 benabik joined 21:24 ambs left, coke_ joined
coke_ rohit? 21:25
aloha, rohit?
aloha coke_: Sorry, I don't know.
21:25 jrtayloriv joined
coke_ needs to get his act together, since he ended up a frontline mentor this time. :) 21:27
dalek p/ctmo: 6e8751b | jonathan++ | src/ops/nqp.ops:
Remove a now-unrequired workaround.
cotto_work coke_: he's the one working on a js compiler 21:31
21:32 sigue left, sigue joined
coke_ cotto_work: yes, he's my studen.t 21:32
was trying to find him on irc. 21:33
benabik coke_: I think his nick is rohit_nsit08 21:34
21:47 hercynium joined 22:02 bubaflub left 22:07 jrt4 joined, TiMBuS left
cotto_work blog.mozilla.com/dmandelin/2011/04/...ript-2011/ - note the second feature 22:08
22:10 jrtayloriv left 22:15 jrt4__ joined 22:16 Andy_ left 22:19 jrt4 left 22:21 sigue left, sigue joined 22:31 hercynium left 22:34 kid51 joined 22:35 whiteknight joined
bacek ~~ 22:36
whiteknight hello bacek 22:37
bacek aloha, whiteknight
whiteknight dukeleto: ping 22:41
22:46 sigue left, sigue joined 22:59 sigue left 23:01 PacoLinux left 23:05 sigue joined
whiteknight doesn't see GSoC students running around in a panic yet 23:10
23:14 bubaflub joined
davidfetter well, they did sign up for GSoC, so they must be immune to panic 23:14
...or don't know what they're in for ;)
benabik is too busy panicking about school to really get into panicking about GSoC yet. 23:16
cotto_work your day will come
whiteknight yeah, school sucks 23:17
they should stop school so students can get on with important stuff: GSoC
23:27 soh_cah_toa left 23:30 soh_cah_toa joined 23:32 pranq joined 23:36 ybit joined
ybit <3 this project 23:36
whiteknight hello ybit 23:40
ybit hi whiteknight 23:41
23:50 benabik left 23:54 benabik joined
davidfetter whiteknight, that's an awesome idea! .oO(GSoC Internship) 23:58
whiteknight yeah, sure. I'll take credit for that idea 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!