Parrot 2.11.0 Released | parrot.org | Log: irclog.perlgeek.de/parrot/today | Keep up with GCI
Set by moderator on 5 January 2011.
bacek briefly here 00:04
cotto_work, basically we can treat most of PIRATE as part of PCT. What we need is just small standalone "PIRATE compiler" built on top of it. 00:05
00:08 gbacon left
cotto_work ok 00:09
dalek website: jkeenan++ | On Roadmaps and Teams, Part Three 00:31
website: www.parrot.org/content/roadmaps-and...part-three
00:51 dmalcolm left
cotto_work kid51, good definition of "introduced". 00:58
also, very good idea about having a requisite bus number >1 for roadmap items. 00:59
arhipov.blogspot.com/2011/01/java-b...ntals.html 01:05
kid51 cotto_work: Thanks. 01:07
In a way, this is like a situation in parliamentary procedure.
In legislatures, anyone can introduce a motion. 01:08
01:08 davidfetter joined
kid51 But a motion requires a second. Motions without seconds fail automatically. 01:08
This, of course, is different.
We need at least two people to make a public commitment to a goal before we can consider placing that goal on our selective list of roadmap goals. 01:09
Roadmap goals should also meet the criterion: "On or after their target dates, the Project Architect, the Product Manager, the board members, etc., should be prepared to publicly advocate for Parrot's use on the basis of the achievement of the roadmap goals." 01:10
That way, people like you and me have some skin in the game. 01:11
Comments welcome on the website. 01:12
01:12 kid51 is now known as kid51_at_dinner 01:18 particle left 01:49 davidfetter left
dalek r: 683dccc | bacek++ | t/post/sub-modifiers.txt:
Fix test to use binary encoding
 01:52
r: d6e98de | bacek++ | src/ (2 files):
Add 'defined' pir sub
r: a3c36dd | bacek++ | src/ (2 files):
Add and use 'new' sub for aesthetic reasons
 01:57
Heuristic branch merge: pushed 144 commits to parrot/html_cleanup by mikehh 01:58
bacek seen pmichaud 01:59
aloha pmichaud was last seen in #perl6 17 mins 14 seconds ago saying "ah yes, you have.".
bacek aloha, good girl 02:00
02:01 cotto joined, kid51_at_dinner is now known as kid51
chromatic Has anything happened on the "please mail us full diffs" GCI task? 02:01
bacek chromatic, do we have such task? 02:02
cotto, we have to have way to emit PIR from new-POST for seamless transition. 02:08
cotto chromatic, which tasks are those? 02:12
bacek, how so?
bacek cotto, for transition period until PIRATE will be sufficient to handle all code.
In the nutshell: 02:13
1. Migrate POST -> newPOST.
cotto 5 minutes, tops
bacek 2. Migrate PCT::Compiler to change last step from "pir" to "pbc"
cotto, little bit more than 5 minutes :) 02:14
newPOST is implemented in nqp
chromatic cotto, dukeleto had a task to fixup the Git reporting mailer 02:15
bacek So, we have to sort out .pir for new nodes. Bootstrapping. Etc.
cotto chromatic, ah. that got completed. We just need someone to set it up. 02:16
it's in the repo now 02:17
tools/dev/github_post_receive.pl
bacek, is there a way to do the POST->newPOST transition without a sudden abrupt change? 02:21
cotto suspects not
also, can you add stuff to PirateTODO?
bacek cotto, nope.
(For transition) 02:22
cotto no way around it though 02:23
bacek cotto, nope. But it's not a huge task. 02:25
cotto, trac.parrot.org/parrot/wiki/PirateT...;version=6
afk # making some lunch 02:26
cotto thanks
happy caturday
dalek tracwiki: v5 | bacek++ | PirateTodo 02:36
tracwiki: Wiki-syntax fix
tracwiki: trac.parrot.org/parrot/wiki/PirateT...ction=diff
tracwiki: v6 | bacek++ | PirateTodo
tracwiki: Add POST transition steps.
tracwiki: trac.parrot.org/parrot/wiki/PirateT...ction=diff
02:40 particle joined
dalek rrot/spaces_in_paths: 61350bf | jkeenan++ | / (3 files):
Remove two tests from t/tools/install/.
 02:51
r: 4fd1888 | bacek++ | src/POST/Compiler.pm:
Remove useless multi declarators.
 03:00
r: 2774fc5 | bacek++ | src/POST/Compiler.pm:
Remove explicit usage of OPLIB. It's all encapsulated inside BytecodeSegment now.
r: 9da2eb6 | bacek++ | src/POST/Compiler.pm:
Move REGALLOC inside %context
r: 460b682 | bacek++ | src/POST/Compiler.pm:
Pass %adverbs to create_context for future encapsulating $DEBUG into it.
r: c7d7b54 | bacek++ | src/POST/Compiler.pm:
Move DEBUG inside %context
03:20 Matt221 left 03:44 kid51 left
kapace_ cotto: ping 03:47
04:11 Matt221 joined, prabu joined
prabu hi do anyone know how to write pge regex to parse single quote string 04:12
bacek proto token quote { <...> } 04:17
token quote:sym<apos> { <?[\\']> <quote_EXPR: ':q'> }
prabu, this is from nqp-rx
prabu cool, i cant find the docs for explaining this .. where i can find? 04:18
bacek, ? 04:19
bacek prabu, hmm... Interesting question. Just ask here :)
basic syntax described in Perl6 synopsis. 04:20
prabu i can't understand what is <quote_EXPR ...?
04:20 Matt221 left 04:21 Matt221 joined
prabu what is the difference between <?[\\']> and <[\\']>? 04:21
04:22 fbrito left
prabu bacek, this is confusing <quote_EXPR: ':q'> 04:22
bacek prabu, this came from nqp-rx.
Basically it's invocation of method "quote_EXPR" in Grammar, passing :q as argument 04:23
04:25 theory left
prabu bacek, why we dont go for this : token string_constant { <?[\\"]> <![\\"]>* } 04:29
{ <?[\\"]> <![\\"]>* <[\\"]> }
bacek prabu, double-quoted strings interpolates content 04:30
and they are parsed into little bit more complex structure.
and in Perl world "quoting" is broad term.
that's main reason to have "quote_EXPR" in nqp 04:31
OTOH, if your grammar is much simpler you can go with your regex.
prabu bacek, thanks 04:35
bacek prabu, you welcome 04:43
dalek r: 3882c26 | bacek++ | src/POST/Compiler.pm:
Switch fixup labels to use hashes as todo.
 04:53
r: c628a66 | bacek++ | src/POST/Compiler.pm:
Remove now unused method.
r: 502cf1f | bacek++ | src/POST/Compiler.pm:
Add compiler to context
r: 6c81d5d | bacek++ | t/ (21 files):
Start adding tests for IMCC to preserve compatibility.
r: 2b62b87 | bacek++ | t/pbc/imcc/syn/clash.t:
Convert more tests
r: 7d25510 | bacek++ | t/test_pbc.pir:
Redirect stderr into stdout in testing pbc.
r: 2bbf4b6 | bacek++ | t/test_pbc.pir:
Handle "likes" in "test_post"
r: 9fdd207 | bacek++ | t/pbc/imcc/syn/clash.t:
Update test
 04:54
r: 67d1858 | bacek++ | src/PIR/Actions.pm:
Handle unknown ops in Actions.
tracwiki: v7 | bacek++ | PirateTodo 05:05
tracwiki: trac.parrot.org/parrot/wiki/PirateT...ction=diff
05:06 nwellnhof_ joined 05:09 Matt221 left 05:11 nwellnhof left, nwellnhof_ is now known as nwellnhof 05:12 prabu left
dalek rrot/nqp_pct: 97f43f7 | bacek++ | compilers/pct/src/POST/ (11 files):
Initial drop of PIRATE's POST modules
 05:18
rrot/nqp_pct: f0303c4 | bacek++ | compilers/pct/Rules.mak:
Prototype for bootstrapping PCT.
rrot/nqp_pct: c5927bb | bacek++ | compilers/pct/Rules.mak:
More bootstrapping of PCT. POST::Compiler left for now.
rrot/nqp_pct: 18d71ff | bacek++ | compilers/pct/src/POST/ (10 files):
Add bootstrapped files
rrot/nqp_pct: 27fe160 | bacek++ | compilers/pct/Rules.mak:
Add more files to Node.pbc
rrot/nqp_pct: d86af94 | bacek++ | compilers/pct/Rules.mak:
Add VanillaAllocator to POST::Compiler.pbc
rrot/nqp_pct: a3f6d58 | bacek++ | compilers/pct/Rules.mak:
Add NQP_RX into bootstrap-pct dependencies.
05:18 nwellnhof left 05:27 Yuki`N left 05:44 rurban_ joined 05:47 rurban left, rurban_ is now known as rurban
dalek rrot/nqp_pct: 36f69e4 | bacek++ | compilers/pct/ (2 files):
Include new POST::* files in POST/Node.pir
 06:10
rrot/nqp_pct: d431d88 | bacek++ | compilers/pct/src/POST/ (3 files):
Migrate POST::Label to nqp
rrot/nqp_pct: e8f21d9 | bacek++ | compilers/pct/src/POST/Label.pm:
Remove misleading comment
 06:12
bacek msg pmichaud Can you check parrot/nqp_pct branch when you'll have time? This is WIP for migrating PCT POST to NQP (and PIRATE's version of POST) 06:13
aloha OK. I'll deliver the message.
dalek rrot/nqp_pct: eee3846 | bacek++ | compilers/pct/src/POST/ (3 files):
Move POST::Sub accessors to Sub.pm
 06:20
rrot/nqp_pct: 9fdc791 | bacek++ | compilers/pct/src/POST/ (3 files):
Migrate Sub.subid into Sub.pm
 06:36
rrot/nqp_pct: 18a500f | bacek++ | compilers/pct/src/POST/ (3 files):
Migrate Sub.add_directive to Sub.pm
07:28 chromatic left
dalek r: dbe9d3b | bacek++ | / (4 files):
Move validation of call variable into POST::Compiler
 07:28
08:27 bacek left 08:37 fperrad joined 09:06 bacek joined 09:36 TonyC left 09:55 rfw left
dalek r: ba986d4 | bacek++ | src/POST/Compiler.pm:
Allocate constants for Subs upfront. It will be useful for handling
 10:13
r: 65286ff | bacek++ | src/POST/VanillaAllocator.pm:
Skip constants in VanillaAllocator
r: 65d27c8 | bacek++ | src/PIR/Actions.pm:
Create POST::Constant instead of String for foo() calls
r: 88bbfb3 | bacek++ | t/pbc/call.txt:
Add tests for .const Sub handling.
r: 75c0232 | bacek++ | src/POST/Compiler.pm:
Initial implementation of .const Sub handling.
r: b19f5dc | bacek++ | / (3 files):
Store only actual value of sub pragmas.
r: 03c409c | bacek++ | / (2 files):
Fix PASTing :nsentry and :vtable
 10:42
r: 932e899 | bacek++ | / (2 files):
s/src/gen/ for generated files
 10:43
jnthn bacek: ping 11:02
bacek jnthn, pong
jnthn bacek: Working on POST nodes => NQP?
bacek jnthn, yes. 11:03
jnthn bacek: OK, nice. :)
bacek jnthn, :)
Just one of the tasks for nqp->past->post->pbc flow
tadzik I'm more excited about 'after 3.0' than 3.0 itself
jnthn bacek: Planning to do PAST nodes to NQP too?
bacek jnthn, not just yet.
but in future - yes, of course :) 11:04
jnthn Doing everything in the parrot/pir repo for now? 11:05
bacek jnthn, parrot/pir + branch in parrot 11:06
jnthn OK
bacek parrot/pir is for pir compiler
bacek crossed fingers waiting for pirate to compile it self 11:07
This bird isn't particularly fast... 11:14
11:17 JimmyZ joined
moritz maybe two month ago there was a patch that added ipv6 support... has it ever been applied? 11:19
bacek moritz, no idea. 11:21
dalek TT #1938 closed by moritz++: rakudo on parrot RELEASE_2_11_0-777-gef82f5c segfaults on role related ... 11:22
TT #1938: trac.parrot.org/parrot/ticket/1938
11:33 fbrito joined
JimmyZ moritz: it's applied, IIRC 11:34
moritz cool 11:35
then I should apply the rakudo patch too
bacek moritz, nope. trac.parrot.org/parrot/ticket/1798 11:36
12:00 JimmyZ left 12:07 whiteknight joined 12:51 gbacon joined 13:30 kid51 joined 13:39 whiteknight left 13:45 rurban_ joined 13:46 TonyC joined 13:47 rurban left, rurban_ is now known as rurban
dalek rrot: 61350bf | jkeenan++ | / (3 files):
Remove two tests from t/tools/install/.
 14:02
rrot: b614102 | jkeenan++ | / (7 files):
Merge branch 'spaces_in_paths'
moritz needs somebody with MacOS fu for rt.perl.org/rt3/Ticket/Display.html?id=76808 14:17
kid51 So I take it that was never solved for Rakudo, either 14:19
dalek TT #1939 created by jkeenan++: darwin: installed parrot linking to build directory 14:22
TT #1939: trac.parrot.org/parrot/ticket/1939
kid51 A dylib *is* being installed but parrot is looking for it in the wrong place. 14:26
ll lib/ 14:27
total 54760
-rwxr-xr-x 1 jimk jimk 13263108 Jan 8 08:50 libparrot.2.11.0.dylib
-rw-r--r-- 1 jimk jimk 14764440 Jan 8 08:50 libparrot.a
lrwxr-xr-x 1 jimk jimk 22 Jan 8 08:50 libparrot.dylib -> libparrot.2.11.0.dylib
dalek p-rx/nom: f613fcf | bacek++ | src/setting/IO.pm:
Improve c<slurp> to be more consistent with Perl6 and work in utf8/binary mode
 14:48
p-rx/nom: 0fc4e19 | bacek++ | src/stage0/nqp-setting.nqp:
Reboostrap setting only.
14:48 Matt221 joined 14:52 contingencyplan left 15:07 Patterner left, Psyche^ joined, Psyche^ is now known as Patterner 15:24 wesjdj joined, wesjdj left 15:26 wesjdj joined, wesjdj left 15:27 whiteknight joined
whiteknight good morning, #parrot 15:28
dalek rrot/gci_fill_params_reduce2: 0bb0c0e | (Matt Rajca)++ | src/call/args.c:
Refactored parts of fill_params
 15:31
15:33 wesjdj joined 15:34 wesjdj left, wesjdj joined, Matt221 left, Matt221 joined
dalek rrot/gci_embed_api_test: f30aa2f | kapace++ | / (2 files):
Added some tests for embed api, get/set compiler and commented set_std_handles
 15:35
rrot: 958e2a3 | Whiteknight++ | / (2 files):
Merge remote branch 'origin/gci_embed_api_test'
 15:49
rrot: 696a71d | Whiteknight++ | t/src/embed.t:
t/src/embed.t is for the old API. t/src/embed/api.t is going to be tests for functions in src/embed/api.c
15:58 aantn joined
dalek TT #1939 closed by coke++: darwin: installed parrot linking to build directory 16:16
TT #1939: trac.parrot.org/parrot/ticket/1939
kapace_ morning whiteknight 16:29
whiteknight good morning, kapace 16:34
kid51 Coke, moritz: please see my patch in trac.parrot.org/parrot/ticket/1890
kapace_ whiteknight, iirc, share_ro needs to die in a fire, correct?
kthakore hello 16:35
whiteknight kapace_: if you're feeling nice
kapace_: if not, you can do something much more evil
kapace_ whiteknight, I'm feeling generous: github.com/kapace/parrot/tree/kapa...n_share_ro 16:37
for object.pmc, that increases coverage by 10%
whiteknight nice. Is that a GCI task?
kapace_ yes
16:37 fbrito left
dalek rrot/gci_kill_share_ro: b6ce88c | kapace++ | src/pmc/object.pmc:
Burn share_ro!
 16:38
whiteknight awesome pawsome
kapace_ cool beans :) 16:39
whiteknight link to GCI task?
kapace_ well I still have the other task open
www.google-melange.com/gci/task/sho...9336697097
whiteknight I just closed your other task 16:40
I think
kapace_ ah ok
then heres the object pmc one www.google-melange.com/gci/task/sho...9360665850 16:41
whiteknight accepted. Move it to NeedsReview 16:42
kapace_ ok
whiteknight this is an easy one to review
...and closed 16:43
thanks
kapace_ thanks again whiteknight :)
now lets see if I can track down that pesky sha bug 16:45
Matt221 whiteknight: can you review: github.com/parrot/parrot/pull/104#...ent-662718 16:49
whiteknight Matt221: approved. Thanks 16:55
Matt221: we might not merge that branch. There's some argument about what we want to do with that fill_params function
but the work you did looks good
Matt221 whiteknight: thanks, yeah, cotto pointed out there might be some performance issues since the code is called often
whiteknight not just performance issues, we may want to change that entire algorithm in the near future 16:56
dalek rrot: bb10b36 | plobsing++ | config/gen/platform (3 files):
remove dead memexec code (JIT fossil)
 17:14
17:23 theory joined
nopaste "kapace" at 192.168.1.3 pasted "bnot broke" (7 lines) at nopaste.snit.ch/27658 17:28
Kapace i think bnot is broken ^
nvm, just forgot the bitops 17:29
17:45 aantn left 17:54 Kristaba joined
Kapace ... "# If the byte oder (byte sex)" ... 17:54
I didn't know bytes had odour nor sex
Kristaba cotto: Sorry for the delay, I was at the hospital from yesterday so I haven't the time to work on parrot :/ 17:58
cotto: But in all case I still don't fix the PCT issue, it's really out of my skill (or I miss something stupid when I try to understand it) 18:01
18:10 TiMBuS left 18:11 TiMBuS joined, fbrito joined 18:28 aantn joined 18:29 aws joined 18:30 aws left 18:32 whiteknight left 18:43 aws joined 18:53 aantn left
dukeleto ~~ 18:54
18:56 aws left, davidfetter joined
tadzik o/ 19:09
Matt221 dukeleto: Can you accept this: www.google-melange.com/gci/task/sho...9899#c3001 19:13
dukeleto looks 19:16
Matt221: accepted, good luck! 19:17
Matt221 dukeleto: Submitting pull request: github.com/parrot/parrot/pull/105
dukeleto cotto: did you want me to approve another config.log task? I see one that is unapproved. 19:18
Kapace hello dukeleto 19:19
dukeleto Kapace: howdy
Matt221: testing your code now
how are the GCI students doing?
Kapace I'm thinking of working on this ticket trac.parrot.org/parrot/ticket/1682, it has gci keywork but no task 19:20
dalek rrot: b8d20cc | (Matt Rajca)++ | t/pmc/nci.t:
Increased test coverage for NCI PMC
 19:21
dukeleto Matt221: task completed and merged 19:22
Matt221 dukeleto: Thanks for the quick review
dukeleto Matt221: no worries
Kapace: that could be a decent ticket. 19:23
Kapace: task, i mean.
Kapace: i added a useful comment to that ticket 19:27
Kapace, Matt221 : do you have any tasks to work on?
Matt221 dukeleto: nope - was just looking 19:29
dukeleto Matt221: parrot has 9 unclaimed tasks right now
19:30 cognominal left
fbrito I think this video/screencast task is underrated :P 19:32
19:33 cognominal joined
dukeleto fbrito: you mean the difficulty level? 19:33
fbrito Yes
dukeleto fbrito: i am going to change it. You are right. 19:34
19:34 contingencyplan joined
fbrito It involves research and video editing skills 19:34
dukeleto fbrito: i bumped the time to 3 days and made it "difficult" 19:35
19:36 masak joined
masak Configure.pl can't find config/gen/platform/generic/memexec.c and fails during the MANIFEST check. 19:36
hi, by the way :)
Matt221 masak: got the same error after plobsing 's latest commit 19:37
masak good, then I'll back up a bit and try again.
plobsing: might want to see if you have unpushed files there.
Matt221 yup reverting fixed the issue
fbrito dukeleto: nice :D. Unfortunately my video editing skills are really bad and English is not my first language
dukeleto fbrito: i understand. But there are still many coding tasks left
fbrito: and if you have an idea for a task, let me know 19:38
fbrito fbrito: ok, thank you for your attention :). I was going to pick the sha256 bug task, but I got frustrated after looking the source file and having no clue where to start. However, if no one fix it until 10th, it will be a pleasure to try again, without the GCI time pressure that I have now :x 19:41
19:42 kid51 left
fbrito g/fbrito/dukeleto/ :P 19:43
19:43 plobsing left
Kapace fbrito: yeah, same here 19:44
did you see those comments?
fbrito which comments? 19:45
Kapace it kinda sounds like it was half finished, and its really hard to find out what was left
in sha256.pir
dukeleto you guys can feel free to implement some other crypto/hash algorithm in PIR 19:51
fbrito: yes, i hope you and the rest of the GCI students stick around after GCI and help us hack on really cool stuff :) 19:52
Kapace dukeleto: rot13? :P
dukeleto Kapace: something slightly more complicated :) 19:53
fbrito Kapace: we should develop our own crypto algorithm 19:57
it would be called rot26. 19:58
"Hello World" ---> rot26 ---> "Hello World"
Kapace haha 19:59
masak I have a really nice implementation for that encryption algorithm. 20:00
tadzik there is a Parrot opcode for that
fbrito really? :) 20:03
tadzik sure
something starting with 'n'
three letters, iirc 20:04
Kapace nop?
fbrito nnn?
tadzik Kapace: bingo!
Kapace nom?
masak tadzik: though we'd need something more like an identity function here...
Kapace lol
masak rakudo: sub id($x) { $x }; say id("Hello World")
p6eval rakudo e7e9d5: OUTPUTĀ«Hello Worldā¤Ā» 20:05
masak I'm sure that can be written more succinctly. :)
davidfetter rakudo say "hello" to my little friend! 20:08
hrm
rakudo: say "hello" to my little friend!
p6eval rakudo e7e9d5: OUTPUTĀ«===SORRY!===ā¤Confused at line 22, near "say \\"hello"ā¤Ā» 20:09
dukeleto begs github for stuff : support.github.com/discussions/acco...-dashboard
fbrito perl6: exec('ls') 20:13
p6eval rakudo e7e9d5: OUTPUTĀ«Could not find sub &execā¤ in main program body at line 22:/tmp/LijAvySgWUā¤Ā»
..pugs: OUTPUTĀ«*** Unsafe function 'Pugs::Internals::exec' called under safe modeā¤ at Prelude.pm line 37, column 9-59ā¤Ā»
dukeleto rakudo: say ~([+] (1 .. 10) 20:15
p6eval rakudo e7e9d5: OUTPUTĀ«===SORRY!===ā¤Confused at line 22, near "say ~([+] "ā¤Ā»
dukeleto rakudo: say ~([+] (1 .. 10))
p6eval rakudo e7e9d5: OUTPUTĀ«55ā¤Ā»
fbrito :o
dukeleto rakudo: say [+] (1 .. 10) 20:16
p6eval rakudo e7e9d5: OUTPUTĀ«55ā¤Ā»
dukeleto fbrito: ~ is the stringification operator in perl 6
masak: you have any ideas for GCI tasks? It is over on Monday.
masak: perl 6 GCI tasks, that is.
20:19 rfw joined
masak dukeleto: I have ideas, but I fear I might not have the tuits to put any of them down in a useful form. :/ 20:19
fbrito I guess I will take the sha256 task. Will find a implementation of it that actually works, put "prints" everywhere and see what is happening 20:29
20:30 plobsing joined
Kapace fbrito++ good luck! 20:31
fbrito I hope I have enough time to work on it :). Tomorrow and the day after tomorrow I will write a 5-hours test to apply to a university
plobsing ~~
dukeleto masak: what is the simplest of the ideas that you have ? 20:43
masak: explain to me and maybe i can make the task
masak dukeleto: we have a number of LHF tickets in RT. some of those could probably be written up as GCI tasks.
dukeleto fbrito: i can extend the task, but GCI ends on monday ;) 20:44
masak: LHF ?
tadzik "Find 20 open tickets on RT which bugs are alredy fixed"
masak dukeleto: the LHF tickts are marked as "LHF" in RT.
tadzik should be hell easy
masak dukeleto: "Low-Hanging Fruit"
dukeleto masak: ah
masak: can you provide a link to one that is very easy ? Let's start with just one GCI task
dukeleto is adding a few tasks right now 20:45
fbrito dukeleto: if I take the german translation task, do you think there is someone available to review it? 20:53
dukeleto fbrito: which task? i think it already taken :) 20:54
fbrito: i am about to create a bunch more code coverage tasks 20:55
fbrito: those are more fun :)
Kapace yay!
dukeleto fbrito: good luck with your tests
fbrito wow, that was a quick claim request :D
21:00 Andy left 21:06 contingencyplan left
dukeleto just added a bunch of parrot tasks 21:07
21:08 contingencyplan joined
Kapace wasn't there already a few bit test coverage tasks already? 21:11
guess not enough :P 21:12
fbrito: are you going to claim it? 21:18
fbrito Oh, forgot to mark the action 21:19
kapace would you like to try it?
Kapace yes
fbrito aren't there 2 of them? 21:20
dukeleto yep
Kapace one is already claimed :|
fbrito ah, the other was claimed
Kapace: you can have it 21:21
Kapace thank fbrito :)
fbrito I can also try the pbc_dump task. I have never wrote tests for neither of them (only for PMCs) so its not a big deal to me
written*
Matt221 yup I claimed the first one. Will be done soon so work doesn't get duplicated 21:22
Kapace Matt221: we're probably going to have to coordinate which function who is going to cover?
cotto ~~ 21:23
Kapace hey cotto
21:23 fperrad left
fbrito ok, I have to go now. Good luck guys :D 21:23
Kapace dukeleto: theres also those footer functions, which makes it really hard to get to 100%
(hash_str, store_op, get_op etc) 21:24
bye fbrito
dukeleto Kapace: which task are you talking about?
Kapace: dynlexpad doesn't have those functions
21:25 fbrito left
Kapace coverage of bit ops by 20% t/dynoplibs/bit.t 21:25
dukeleto cotto: prepare for a spike in test coverage this weekend :)
cotto dukeleto, I think the other config log task was just there as a template. 21:26
dukeleto, I'm ready.
all pending claims accepted
well, both, but "all" sounds better 21:27
dukeleto cotto: i added about 10 new tasks today 21:31
Kapace dukeleto: bit_ops had already like 40ish coverage
dukeleto cotto: i am thinking there will be a last minute GCI surge, if I know anything about student procrastination
Kapace: yeah, the % listed in the task is wrong
Kapace: type
Kapace: but the goal of raising 20% is still fine :) 21:32
Matt221 dukeleto: its 57% here
already got it up to 60... :)
dukeleto Matt221: that is what i thought. That is why I created 2 tasks for 20% each
Kapace but I doubt we'll be able to get it to 100 (57+20+20) with the footer functions 21:33
maybe just cover as much as the bit ops as possible?
Matt221 yeah
those static functions will be hard to cover
Kapace Matt221: which ones are you going to do? So we don't write the same tests? 21:35
dukeleto Matt221: don't worry about the static functions 21:36
Matt221 just did the rest of the bor_* ones
dukeleto Matt221: if you each do 15%, i that will be fine.
cotto dukeleto, you're probably right about the surge. 21:37
Matt221 Kapace: Ill also do the shr_* and shl_* ones
thats should be enough
Kapace ok 21:39
dukeleto cotto: i added 10 tasks today. 21:40
cotto dukeleto++ 21:41
Kapace 34 hours left, let's get to them! 21:42
rfw oh, are there more parrot tasks? 21:43
\\o/
dukeleto rfw: yep 21:45
21:45 rurban_ joined
Kapace hey rfw, how's it rolling? 21:46
rfw Kapace: not bad
dukeleto: www.google-melange.com/gci/task/sho...9452207376 :)
21:47 rurban left, rurban_ is now known as rurban
cotto accepted 21:47
rfw thanks
Kapace is lsr_p_p_p supposed to do nothing with the first preg? 21:57
21:58 avms joined
sorear it should use it as an output 21:58
Kapace file:///home/david/Data/GCI/kapace-parrot/cover_db/src-dynoplibs-bit_ops-c.html line 1690 only touches PREG(2) and PREG(3)
cotto Kapace, that's no a helpful link. ;] 22:01
aloha, coverage?
aloha cotto: coverage is cv.perl6.cz or tapir2.ro.vutbr.cz/cover/cover-results/
rfw jm i'm getting "Multiple Dispatch: No suitable candidate found for 'i_add', with signature 'PP'" instead of "BigInt: no multiple dispatch variant 'i_add' for %Ss", 22:02
that's odd
Kapace oops lol
tapir2.ro.vutbr.cz/cover/latest-c_c...ops-c.html 1690 22:03
cotto that does look broken
that's bizarre 22:04
take a look at all the lsr ops in the bit.ops file
see if one sticks out
Kapace, feel free to fix that in your pull request. 22:11
Kapace I don't see how "$1 = (INTVAL)((UINTVAL)$2 >> $3);" ignores the first arg.. 22:12
how is bits.ops coverted to C?
oh, ops2c heh
Matt221 can someone review: github.com/parrot/parrot/pull/106
bacek good morning, humans
dalek r: 5f8dedf | bacek++ | src/POST/Compiler.pm:
Reduce syntax noice - use postfix for.
r: e4b6eff | bacek++ | src/POST/Compiler.pm:
Remove useless pirop signatures. PAST::Compiler knows them anyway.
Kapace morning bacek 22:13
bacek hi Kapace
tadzik morning bacek, The Fast Commiter
rfw cotto: set_pmc in bigint segfaults, lol
bacek tadzik, :)
Kapace cotto: the problem is in ops2c right? It doesn't sound like bit.ops is wrong.. 22:14
nopaste "rfw" at 192.168.1.3 pasted "BigInt set_pmc backtrace" (13 lines) at nopaste.snit.ch/27659
cotto Kapace, look at lsr(invar PMC, invar PMC, invar PMC) 22:15
in bit.ops
rfw cotto: ^ backtrace
cotto note that it doesn't ever refer to $1
Kapace cotto: I must be looking at the wrong bit.ops
nopaste "cotto" at 192.168.1.3 pasted "broken lsr" (13 lines) at nopaste.snit.ch/27661 22:16
Kapace cotto: which file is that? 22:18
like path?
cotto src/dynoplibs/bit.ops 22:19
cotto forgot that we had two of those 22:20
Kapace ah, just realized that now, looking at the wrong bit.ops
alright, should be easy enough from there, thanks cotto
Matt221 cotto: can you review: github.com/parrot/parrot/pull/106
cotto Matt221, I need to run some errands. I'll do that when I get back. 22:21
afk &
rfw who the hell wrote these horrible bigint tests 22:22
Kapace Matt221: can I have your coverage report for bitops? 22:31
rfw During configuration the following steps failed: 22:37
01: init::manifest
uh oh
Kapace :O
rfw: git clean and reset? 22:38
rfw init::manifest - Check MANIFEST...No such file: config/gen/platform/generic/memexec.c 22:39
No such file: config/gen/platform/openbsd/memexec.c
22:39 kid51 joined
Kapace rfw: nuke your computer and start over? 22:39
rfw Kapace: can you check master
Kapace sure..
rfw: same error here 22:40
rfw :O
someone broke master
i regenerated a manifest 22:41
:/
Matt221: you ripped out the exact same parts of fill_params as i di :| 22:46
except i didn't double pointer named_used_list, lol
Matt221 rfw: lol. did you get a head start on that?
rfw Matt221: yeah i finished before you started 22:47
github.com/parrot/parrot/commit/18...0640270929
Matt221 lol nearly identical :P 22:48
rfw yep :P
dalek rrot: f85159f | jkeenan++ | MANIFEST:
MANIFEST should have been updated after removal of memexec files.
rrot: f7fed3d | nwellnhof++ | t/library/test_more.t:
[t] Fix t/library/test_more.t without BigNum support
 22:49
23:00 masak left
rfw seen fbrito 23:04
aloha fbrito was last seen in #parrot 1 hours 40 mins ago saying "ok, I have to go now. Good luck guys :D".
rfw :(
does anyone have any idea how to invoke bigint_mod_bigint_in 23:11
23:11 wesjdj left
rfw +t 23:11
sorear $P1 %= $I1 ? 23:12
rfw Multiple Dispatch: No suitable candidate found for 'i_modulus_int', with signature 'PI'
23:15 mtk joined
plobsing $P1 %= $P2 ? 23:16
rfw plobsing: nope :/
sorear $P1 = $P2 % $I3 ? 23:23
rfw Multiple Dispatch: No suitable candidate found for 'modulus_int', with signature 'PIP->P' 23:24
23:32 mtk left
cotto ~~ 23:33
Matt221 cotto: can you review :D = www.google-melange.com/gci/task/sho...0224#c3002 23:36
cotto Matt221, deal 23:38
rfw cotto: also www.google-melange.com/gci/task/sho...9452207376 :D 23:40
23:41 aantn joined 23:42 wagle left 23:43 contingencyplan left
dalek rrot: b1459cf | cotto++ | t/dynoplibs/bit.t:
Merge branch 'master' of github.com/mattrajca/parrot into mattrajca-bit-coverage
 23:44
cotto rfw, have you filed ticket on set_pmc segfaulting? 23:45
cotto wonders if there's some super-sneaky way to get around a segfault in PIR. 23:46
of course if there is, it's probably a Bad Idea 23:47
rfw cotto: not yet
cotto you should do that 23:48
rfw on it right now
cotto rfw++
plobsing cotto: use mmappedbytearray to map the null page. no more segfaults. 23:49
or far fewer at least
23:49 wagle joined
rfw done 23:49
kid51 is puzzled as to why no GCIers have claimed www.google-melange.com/gci/task/sho...9393919283
cotto which task is that? 23:50
rfw cotto: could you review
kid51 trac.parrot.org/parrot/ticket/1925#comment:2
Evaluate risk of removing a config step, auto::jit 23:51
cotto rfw, you should use diag() instead of say in tests.
kid51 I'm very much tempted to knock off that TT tonight
rfw cotto: oh, i was fixing some tests that i think fbrito wrote 23:52
i should probably change his says to diags
mikehh this html_cleanup is driving me crazy, keep findin something else that needs fixin'
almost there with the index page generation, just a couple of things to do 23:53
taken me a lot longer thab I thought it wouls :-{
than 23:54
and would
bah still have to get used to this new keyboard
rfw okay cotto, converting say to diag 23:55
cotto great
rfw cotto: done
mikehh think another couple of hours should sort it out, then I will commit and you all can look at it 23:56
cotto mikehh, I'm very happy to hear that. Thanks for your continued work on that branch.
It'll be really nice not to have a nasty mess responsible for generating our html docs.
23:57 nwellnhof joined
dalek TT #1940 created by rfw++: set_pmc segfaults in BigInt 23:58
TT #1940: trac.parrot.org/parrot/ticket/1940
nwellnhof rfw: can you add the code that causes the segfault to the ticket? 23:59
mikehh it should be all in .json files now, much easier to edit and change, though it takes descriptions/titles from the .pod itself at the moment
rfw nwellnhof: hold on

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!