Parrot 1.8.0 Zygodactyly released | Latest modified TT's: icanhaz.com/parrotbugs | Parrot Languages: icanhaz.com/parrotlang
Set by moderator on 23 November 2009.
00:08 zak_ joined 00:20 eternaleye joined 00:26 patspam1 joined 00:30 theory joined 00:33 KatrinaTheLamia joined 00:35 theory joined 01:33 eternaleye joined 01:40 tetragon joined 01:55 davidfetter joined 02:49 cconstantine joined 03:27 preflex joined 04:28 davidfetter joined 06:28 TiMBuS joined
dalek TT #1333 created by bubaflub++: [PATCH] convert t/compilers/json/from_parrot.t to PIR 06:41
TT #1334 created by bubaflub++: [PATCH] convert t/compilers/pge/02-match.t to PIR 06:58
07:02 theory joined
dalek TT #1335 created by bubaflub++: [PATCH] convert t/compilers/pge/04-compile.t to PIR 07:08
07:13 uniejo joined 07:26 szabgab joined 07:52 theory joined 07:58 bogen joined 08:21 iblechbot joined
dukeleto 'ello 08:34
08:35 eternaleye joined 08:48 nopaste joined 08:52 slavorg joined 08:56 elmex joined 08:57 JimmyZ joined 09:07 uniejo joined
Tene hi leto 09:14
09:23 nopaste joined
JimmyZ hello 09:36
09:50 cognominal joined 09:51 payload joined 10:05 theory joined 10:32 payload joined 10:50 zak_ joined 11:12 bacek joined
bacek o hai 11:12
11:14 elmex joined 11:41 cotto joined
mikehh All tests PASS (pre/post-config, smoke (#30196), fulltest) at r42806 - Ubuntu 9.10 amd64 (g++ with --optimize) 12:15
12:23 bacek joined
Infinoid good morning #parrot 12:27
Coke: I will look at it. 12:30
Unfortunately, dalek has a pitifully small log rotation, so I need to restart it to see what it thinks is happening with partcl-nqp 12:31
12:32 dalek joined
Infinoid hmm. "partcl-nqplog github ATOM parser autoloaded." 12:33
Coke: according to dalek, everything is fine and dandy and it should emit commit messages for you 12:35
12:40 ruoso joined 12:45 bacek joined 12:49 lucian joined
dalek rrot: r42807 | mikehh++ | trunk/t/op/time.t:
re-instate copyright, ID, doc header and coda
 12:55
13:02 payload joined 13:26 bluescreen joined, plobsing joined 13:35 iblechbot joined 13:46 nopaste joined
kthakore plobsing: I see you are from toronto 13:48
dalek rrot: r42808 | mikehh++ | trunk/t/library/uuid.t:
convert test to pir - based on patch by bubaflub++ (TT #1311)
 13:50
13:58 payload joined 14:05 riffraff joined 14:37 cconstantine joined
dalek rrot: r42809 | mikehh++ | trunk/t/library/protoobject.t:
convert test to pir - based on patch by bubaflub++ (TT #1312)
 14:46
14:53 JimmyZ joined 14:54 patspam joined 14:57 fperrad joined 15:03 Psyche^ joined 15:04 fperrad_ joined 15:10 masak joined
NotFound Can some nice gut make me a review of this post? notfound.posterous.com/self-hosted-...hy-and-how 15:22
moritz NotFound: nice post 15:26
NotFound Thanks 15:27
moritz NotFound: there are more reasons for bootstrapping a compiler, though
NotFound: for example testing - doing two bootstrap compilations, and then binary compare the results
(unless there are IDs with timestamps in it)
and it forces you not to regress on important features 15:28
NotFound moritz: I just try to explain it to people not very versed in compilers, so I avoided providing too much information.
15:29 cconstantine joined
moritz ok 15:30
NotFound: then maybe the "all respected languages are self-hosting" argument is the right for you
C, Pascal, C++, various Lisp dialects, Eiffel etc. 15:31
15:31 cconstantine joined
NotFound Yes, but I'm sure not all people agree with such strong assertion. 15:31
moritz heh :-)
s/all/most/
anyway, it's a good post, I liked it 15:32
NotFound Larry Wall, for example X-)
moritz I'm sure we'll see a self-hosting Perl 6 compiler in a few years or decades :-) 15:33
NotFound Hope so.
dalek nxed: r222 | julian.notfound++ | trunk/examples/parser.winxed:
implement using statements in example parser, making it able to pass some tests
 15:38
15:45 payload joined
kthakore moritz: hi 15:53
moritz kthakore: hola 15:54
kthakore moritz: is there a way to vizulize the NFA of a PGE grammar file
moritz: a la Antlrworks?
moritz kthakore: I don't think so... but it's really pmichaud you should ask
kthakore ok
pmichaud: hi
16:46 iblechbot joined
kthakore moritz: pretty quite today US thanksgiving eh 16:55
dalek nxed: r223 | julian.notfound++ | trunk/examples/parser.winxed:
prefix ++ operator in example parser
17:40 bogen left 17:50 jan joined 17:51 fperrad joined
dalek nxed: r224 | julian.notfound++ | trunk/examples/parser.winxed:
example parser: <, >, <= and >= operators, and refactor of == and !=
 17:58
18:07 ruoso joined 18:08 payload joined
dalek rrot: r42810 | mikehh++ | trunk/t/compilers/json/from_parrot.t:
convert test to pir - based on patch by bubaflub++ (TT #1333)
 18:09
18:14 cognominal joined 18:20 darbelo joined
mikehh messages 18:22
darbelo You have no messages waiting. 18:23
;)
mikehh darbelo: :-} 18:24
18:48 eternaleye joined
dalek rrot: r42811 | mikehh++ | trunk/t/compilers/pge/02-match.t:
convert test to pir - from patch by bubaflub++ (TT #1334)
 18:55
nxed: r225 | julian.notfound++ | trunk/examples/parser.winxed:
parser example: get labels from outer blocks, on line arrays
 18:56
dukeleto 'ello
dalek nxed: r226 | julian.notfound++ | trunk/examples/parser.winxed:
parser example: fix string add
 19:01
darbelo \\o
NotFound o/
Looks like today all people is busy killing turkeys. 19:02
ETOOMUCHUSA 19:03
darbelo I'm just getting back to speed after the last round of exams. 19:04
I have so much time now I slept like 8 full hours last night.
NotFound Good 19:05
darbelo I'll probably break something big in the next 24-48 hours.
;) 19:06
NotFound If you have plenty of time, please take a look at my last Winxed post: notfound.posterous.com/self-hosted-...hy-and-how 19:07
darbelo (bootstrapping)++ 19:08
Wait, popularity of Pascal? I didn't know it had that ;) 19:09
NotFound I intentionally don't put time references ;) 19:10
But I can quote '80 magazines if you want ;)
I just try to show that world don't started with the JVM 19:11
PacoLinux curiosly : James Gosling cites UCSD Pascal as a key influence (along with the Smalltalk virtual machine) on the design of the Java virtual machine. 19:17
from : en.wikipedia.org/wiki/UCSD_p-System
NotFound But surely many young programmers didn't even know who is Gosling ;) 19:20
mikehh I did a lot of work with UCSD p-syatem in the 1980's and early '90s
dalek rrot: r42812 | mikehh++ | trunk/t/compilers/pge/04-compile.t:
convert test to pir - from patch by bubaflub++ (TT #1335)
 19:28
cotto seen chromatic 19:32
purl chromatic was last seen on #parrot 5 days, 2 hours, 20 minutes and 43 seconds ago, saying: msg kid51 RT #56110 can stay closed. The final message in RT #60000 can become a new TT. [Nov 21 17:10:18 2009]
mikehh just curious - has purl stopped updating karma? 19:35
darbelo karma mikehh
purl mikehh has karma of 295
darbelo mikehh++
karma mikehh
purl mikehh has karma of 297
darbelo Seems to work.
japhb NotFound, just started reading your self-host post. Just as a side note, it is unusual to refer to a computer program with a gendered pronoun in English ... "it" and "itself" are more common. 19:36
mikehh don't know - I had a bunch of commits and it usually updates karma which is why I had bubaflub++ in the comment and it didn't update 19:37
or is purl ignoring dalek? 19:38
darbelo Hmm. purl can only handle one "++" per line.
karma nobody
purl nobody has karma of -1
darbelo mikehh++ nobody++
karma nobody
purl nobody has neutral karma
NotFound japhb: my english is not good enough to locate the sentences with that error.
japhb NotFound, I'd say search for 'him', 'his', and 'himself', and replace with 'it', 'its', and 'itself' respectively. 19:39
mikehh comments from dalek are not on the same line anyway
NotFound japhb: nice, thanks.
japhb NotFound, np. :-) 19:40
darbelo japhb: "take part in his development"?
NotFound BTW my english has improved a lot if that are the only things that called your attention :)
japhb darbelo, that one depends on whether 'his' refers to the author or the project. 19:41
I read it as the latter.
NotFound, there were a couple minor things (mostly bits that were implied rather than stated, which I often find in non-native English writing), but the pronouns were the only thing that rose above the level of an "accent" 19:42
19:42 chromatic joined
darbelo Spanish has no 'it', it takes actual effort to not use gendered pronouns. 19:44
I actually read that and saw nothing wrong with it until you mentioned that. 19:45
japhb darbelo, I'm not surprised. I have the opposite problem using romance languages. I have to remember French nouns with the articles attached, or I'd never remember which gender to use. 19:46
NotFound That's the problem, I don't fix that error because I don't see it X-)
chromatic eso 19:49
cotto hio chromatic 19:52
chromatic That profile is hard to read; there are a lot of cycles that confuse Callgrind. 19:55
kthakore leto: are you available for a design exaplaing on the blitzkost signature in p5invocation? 20:06
dalek TT #1268 closed by mikehh++: [PATCH] convert t/op/time.t to PIR 20:10
TT #1311 closed by mikehh++: [PATCH] convert t/library/uuid.t to PIR
mikehh purl is ignoring dalek 20:16
I know she was making some rude comments, but that is where karma is picked up 20:18
cotto chromatic, I just re-ran nqp-rx against its own Actions.pm and I'm still not seeing anything. There are about 16 billion instructions, but looking at the callgrind output it looks like only about 2.6 million instructions are spent in Parrot_Capture_mark. 20:21
(That's looking directly at the file, not via kcachegrind.) 20:22
dukeleto kthakore: what do you mean? 20:27
chromatic Do you have cycle detection on or off? 20:28
cotto off
chromatic Let me run it again then.
cotto (though the % time spent in Parrot_gc_mark_and_sweep looks the same either way. 20:29
)
kthakore dukeleto: um I have been looking at the page Allision gave me for accessing signatures 20:33
dukeleto: since the old way C structs are depreceated
dukeleto: but I am trying to understand as an overview what is going on there 20:34
dukeleto: let me get the exact code
dukeleto: src/pmc/p5invocation.pmc :45 20:35
dukeleto: we have PMC* arg and PMC* sig 20:36
dukeleto: from what I get here s(parrot) perldoc src/pmc/callsignature.pmc 20:37
chromatic If I disable cycle detection, Parrot_gc_mark_PMC_alive_fun() costs 632.48%, inclusive. 20:38
kthakore dukeleto: both args and sig are accessible from get_pmc()
chromatic Parrot_Capture_mark() costs 95.01%, inclusive.
kthakore dukeleto: but I have no clue what we are trying to do there ... 20:39
cotto chromatic, something's odd. 20:40
Can you upload your callgrind.out file somewhere? 20:41
chromatic wgz.org/chromatic/tmp/callgrind.out.1156.bz2 20:44
20:45 iblechbot joined
dukeleto hola, backlogging 20:47
kthakore dukeleto: ?? 20:49
cotto chromatic, forbidden
dukeleto kthakore: what are you exactly asking me?
kthakore dukeleto: um ... what p5invocation is supposed to do
chromatic cotto, try again please.
kthakore dukeleto: I need to rewrite getargs() so wanted to know logic there 20:50
dukeleto kthakore: it turns perl 5 calling conventions into parrot calling conventions
cotto mine at mksig.org/parrot/nqp_actions_build....nd.out.bz2
dukeleto kthakore: actually, get_args seems to convert parrot calling conventions into perl5 cc 20:51
kthakore dukeleto: so like p5method(args) to parrot(args)
??
dukeleto: that is what I was thinking
ok so I am not insane
cotto chromatic, on your file I also see 21.17% of the time spent in Parrot_gc_mark_and_sweep.
dukeleto kthakore: yes, you are correct 20:52
kthakore: it converts the different ways that arguments can be passed to functions in perl 5/parrot
kthakore dukeleto: ah ok
so it takes a signature (just learned this today) p5methd(@ARGS) to 20:53
parrot(@ARGS) ?
but how do parrot signatures look like?
dukeleto kthakore: they look like that :) 20:56
kthakore reallly?
dukeleto kthakore: you can go look at the CallSignature PMC docs
kthakore: they changed a lot recently from some branch merges, so blizkost has not yet been updated 20:57
kthakore dukeleto: from here (parrot) perldoc src/pmc/callsignature.pmc?
dukeleto: ya blitzkost is going to take some effort
dukeleto: but they are no longer goign to change PCC right? 20:58
cotto and now kcachegrind is freezing
kthakore chromatic: right?
1.8 - 2.0 won't destroy PCC anymore?
dukeleto kthakore: this is what things used to look like: docs/pmc/subs.pod
kthakore: that is all wrong now
kthakore: or at least quite different in some ways 20:59
kthakore dukeleto: ok well I will keep trying ... I have to head into a meeting now
dukeleto 'ack -ai CallSignature docs/' returns nothing. that is scary.
kthakore so ttyl
dukeleto: indeed
dukeleto kthakore: learn about CallSignatures!
kthakore dukeleto: I am use perldoc src/pmc/callsignature.pmc for now 21:00
japhb (homemade chili and cornbread)++
kthakore dukeleto: send me links if you have some
thanks
mikehh All tests PASS (pre/post-config, smoke (#30206), fulltest) at r42812 - Ubuntu 9.10 amd64 (gcc with --optimize) 21:07
21:10 cconstantine joined 21:19 bluescreen joined 21:20 eternaleye joined
dukeleto mikehh++ 21:27
dalek nxed: r227 | julian.notfound++ | trunk/examples/parser.winxed:
parser example: fix an error message that doesn't reported position
 21:31
dukeleto mikehh: please close RT tickets! 21:38
darbelo Weren't all of those closed already? 21:39
dukeleto mikehh: i just spent time applying bubaflubs patch and just realized you already did that
TT #1312 was left open
which is was the first one i went for :) 21:40
dalek TT #1312 closed by dukeleto++: [PATCH] convert t/library/protoobject.t to PIR
darbelo s/RT/Trac/ ?
dukeleto ug
yes,
mikehh: please close Trac tickets!
dukeleto deletes RT from memory banks
darbelo gets unconfused. 21:41
mikehh dukeleto: oh dear - I think I have applied the patches bubaflub++ posted
dukeleto it looks like TT 1333,1334,1335 are also left open
but applied
mikehh: can you close all the ones that you have applied?
dalek nxed: r228 | julian.notfound++ | trunk/examples/parser.winxed:
parser example: typed function parameters
mikehh sure - I was doing that when something else came up 21:42
dukeleto mikehh: no worries, i just started going through the patch queue and getting lots of conflicts :)
dukeleto hates conflicts
chromatic cotto, toggle Skip Cycle Detection. 21:43
21:44 joeri joined
mikehh I mentioned before I think, but I commented the commits to give bubaflub++ the karma, but whoever assigns karma (i think purl) is ignoring dalek 21:45
dalek nxed: r229 | julian.notfound++ | trunk/ (2 files):
rename and move example parser to winxedst1
 21:51
TT #1333 closed by mikehh++: [PATCH] convert t/compilers/json/from_parrot.t to PIR 21:57
TT #1334 closed by mikehh++: [PATCH] convert t/compilers/pge/02-match.t to PIR 22:00
TT #1335 closed by mikehh++: [PATCH] convert t/compilers/pge/04-compile.t to PIR
rrot-plumage: 77f5ac2 | japhb++ | :
[LIB,plumage] try() -> try {} or CATCH {}, Tene++ for NQP-rx features
 22:03
rrot-plumage: 0729bc0 | japhb++ | :
[CORE] Glue.pir: Remove old try() sub
rrot-plumage: a1dc6f8 | japhb++ | :
[LIB,plumage] die() -> pir::die(), and get rid of Glue.pir implementa...
rrot-plumage: aee1a1f | japhb++ | :
[LIB,plumage] does() -> pir::does__IPs(), and get rid of Glue.pir imp...
rrot-plumage: e4bed10 | japhb++ | :
[LIB,plumage] split() -> pir::split(); get rid of Glue.pir implementa...
rrot-plumage: 4ea086f | japhb++ | :
[LIB,plumage] join() -> pir::join(); get rid of Glue.pir implementati...
22:07 theory joined
dukeleto japhb: i see some yak shaving! 22:17
mikehh: we will have to see about the karma. we need to just get bubaflub++ a commit bit :) 22:18
dukeleto vouches for bubaflub++ having commit access
mikehh dukeleto: his patches look good - we need to bring it up at the next #parrotsketch - has he done a CLA 22:19
japhb dukeleto, Yaks? What yaks? All I see is a bunch of hairy four legged beasts with horns .... 22:20
mikehh japhb: GNU's perhaps 22:22
japhb It would be gnus to me. 22:23
mikehh naw - we need to parrot it about 22:24
dukeleto stands back in awe 22:26
mikehh: if he has done the CLA, then don't we just need to bug chromatic to turn his bit?
mikehh: i think he hasn't done the CLA yet 22:27
22:31 mikehh joined
mikehh my internet connection is having quite a few problems at the moment 22:31
chromatic The most recent CLA I've seen is from Gerd. 22:40
dalek nxed: r230 | julian.notfound++ | trunk/winxed (2 files):
put stage 1 options on line with driver expectations and add --stage option to
22:42 cconstantine joined
dalek TT #1336 created by kurahaupo++: Convert t/pmc/bigint.t to PIR 23:00
mikehh dukeleto: you want to do that or shall I? 23:03
GeJ Good morning everyone 23:10
dalek rrot: r42813 | mikehh++ | trunk/t/pmc/bigint.t:
convert to PIR - based on patch from kurahaupo++ (TT #1336)
 23:46
23:46 tetragon joined
dalek TT #1336 closed by mikehh++: Convert t/pmc/bigint.t to PIR 23:46

Powered by Cro and the Raku® Programming Language.

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

Thank you!