Parrot 0.6.2 "Reverse Sublimation" Released | parrotcode.org/ | 20/672 new/open tix
Set by moderator on 21 May 2008.
00:06 teknomunk joined
wknight8111 noted. 00:08
DietCoke tries to figure out where we test the calling conventions 00:16
t/op/calling ?
(why is that under -op-?)
pmichaud probably because it was 'invoke' at one time. 00:19
draft of article about P6object : www.pmichaud.com/sandbox/post3.txt
comments welcome
(but right now I have to go to the grocery store, so bbiah)
00:20 donaldh joined
DietCoke "new class exists in as a normal Parrot class" 00:26
dalek r27757 | Whiteknight++ | trunk: 00:38
: [docs] update/improve chapter 2 of docs/book/.
diff: www.parrotvm.org/svn/parrot/revision?rev=27757
DietCoke I run gdb on parrot. I try to set a breakpoint on Parrot_process_args; it claims to not know anything about it. any clues? 00:41
dalek r27758 | Whiteknight++ | trunk: 00:42
: [docs] kill trailing whitespace in docs/book/ch01_overview.pod
diff: www.parrotvm.org/svn/parrot/revision?rev=27758
00:48 cognominal joined
wknight8111 how many actual registers of each type are there now? Is it still limited to 32 per type? 01:12
I seem to remember that the limit was raised, but I dont remember what it was raised to 01:13
Eevee last I heard it was raised to infinity 01:16
wknight8111 That's what I thought, but I couldn't remember if it was "$P0" registers or actual "P0" 01:17
I take it also that P31 is not "reserved for spilling" then, and that extra registers don't need to be stored on the user stack? 01:22
I mean, I know the user stack has been deleted
dalek r27759 | coke++ | trunk: 01:27
: Add test for RT#39844
: mdiep++ for the original bug report.
diff: www.parrotvm.org/svn/parrot/revision?rev=27759
bacek_ vany: ping 01:31
dalek r27760 | coke++ | trunk: 01:52
: RT #45189
: Remove some deprecated PGE syntax. Modified version of patch originally supplied by chromatic++ with suggestions from pmichaud++.
: prove t\\library\\File_Spec.t , all tests pass.
diff: www.parrotvm.org/svn/parrot/revision?rev=27760
DietCoke digs out a 13 year old o'reilly lex & yacc signed by levine himself to try to figure something out. 01:55
cognominal but we want to kill them :) 02:06
DietCoke kill whom?
cognominal lex and yacc
DietCoke anyone here understand them?
cognominal ask your question, may be I know 02:07
DietCoke I don't know enough to formulate the question. =-)
(looking at rt.perl.org/rt3/Ticket/Display.html?id=46499). I have found "sub_params" in imcc.y and am trying to add in "LINECOMMENT" in the proper location . 02:08
tetragon Bleh, seems like I still need to apply my -arch patch (#52214) for parrot to build
cognominal formulating the question is the pass to the answer
DietCoke tetragon: That patch would be more like to be applied it if was just in the darwin section. 02:09
ls config/init/hints/darwin.pm
tetragon So, how would you suggest having that file only stop a fat build when the -arch flags come from the Perl 5 config 02:10
cognominal looks at imcc.y 02:11
DietCoke it's all run at config time. but if this is a darwin only thing, the place for it is in the darwin hints, I think. 02:12
(at least for where you have $^O eq 'darwin')
DietCoke edges closer on his imcc issue! 02:13
cognominal sub_param should allow a comment. 02:14
DietCoke but it doesn't.
as I said, that's where I am. (ah, I forgot to mention I was in that sub.) yes. 02:15
I have it working for an initial comment, now trying to get it working for an inter-param comment. 02:16
... crap. initial comments always worked. 02:19
cognominal I should be in bed
I don't see how comment can be other than in a line by themselves 02:20
DietCoke it can't.
but .param 's only take up one line.
look at the example in t/compilers/imcc/syn/regressions.t (last tiest)
cognominal I have broken my tree, so I nedd to get at that first :( 02:21
it seems that I got \\h \\v \\H and \\V to compile 02:23
nope :(
dalek r27761 | Whiteknight++ | trunk: 02:37
: [docs] update the first third of /docs/book/ch03_pir_basics. more updates needed.
diff: www.parrotvm.org/svn/parrot/revision?rev=27761
02:45 Eevee joined
pmichaud \\h \\v \\H and \\V supposedly work in PGE. 02:51
do you have a case that shows them not working?
cognominal see Expr.pir::1042 they are missing 02:55
pmichaud those aren't implemented in Exp.pir
they're in Perl6Regex.pir
cognominal arf 02:56
so my rules are wrong
tetragon Bleh, I have a build getting past the first gcc invocation now with a darwin hints flag munger, but ld is failing (for what would most likely be a reason unrelated to the munger) 02:57
pmichaud see also the tests for \\h \\v \\H \\v in t/compilers/pge/perl6regex/rx_metachars 02:58
tetragon ld: duplicate symbol _Parrot_exec_rel_addr in src/jit.o and src/exec.o
cognominal indeed, really go to bed now 03:04
dalek r27762 | coke++ | trunk: 03:05
: RT#46499
: Update syntax tests for comments and whitespace before and in the middle of
: .param lists
diff: www.parrotvm.org/svn/parrot/revision?rev=27762
DietCoke pmichaud: do you know lex/yacc? 03:08
purl lex/yacc is my friend when parsing xml and sgml
wknight8111 that reminds me, I need to install lex/yacc on my new system 03:09
PerlJam purl: perl6 will be your friend :)
purl PerlJam: sorry...
DietCoke RT#45499 looks like it should be trivial to implement if one knows yacc, so I figured I had a shot at getting it to work. =-)
anyone else lying about, feel free to have at it. 03:10
wknight8111 let me look at it
what's the number on it? 45499 seems to be pointing to dead space 03:12
03:12 Zaba joined
DietCoke rt.perl.org/rt3/Ticket/Display.html?id=46499 03:12
my bad.
wknight8111 thanks
dalek r27763 | coke++ | trunk: 03:13
: ... Make sure the whitespace we're supposed to be testing is present.
diff: www.parrotvm.org/svn/parrot/revision?rev=27763
wknight8111 I assume we're patching IMCC and not PIRC? 03:14
well, i guess there is nothing stopping me from doing both except a few tuits 03:15
tetragon Hrm... exec.c lost the extern on "char **Parrot_exec_rel_addr;" (line 64 of exec.c) since my last successful build (I think it was r27690) 03:19
03:27 starc joined
wknight8111 I could tackle #48549 too, now that i have lex/yacc installed! 03:28
cotto_home DietCoke, I'll be glad to do a release if I can ever get clearance from work. 03:44
sorry I can't be more immediately helpful 03:45
03:50 mdiep_ joined, mdiep_ left
tetragon So, r27736 broke building on my box 03:52
04:11 tetragon joined
tetragon notices that #54602 was updated as she was relocating her system 04:13
04:39 Coke joined
moderator Parrot 0.6.2 "Reverse Sublimation" Released | parrotcode.org/ | 22/666 new/open tix 04:56
Coke pmichaud: can you look at rt.perl.org/rt3/Ticket/Display.html?id=53956 ? I'm not sure what cognomial is asking for, but it looks rakudo related. 04:59
pmichaud I think that ticket stems from a misunderstanding of what copy is supposed to do and how types will be implemented in Rakudo. So I'm not even sure how to respond. 05:00
but I'll come up with something. 05:01
done. 05:04
05:05 gmansi joined
Coke -> z 05:06
cotto_home tutorial 05:16
purl tutorial is doing exactly what the helper does and exactly what I'm telling you to do
cotto_home pct tutorial?
purl hmmm... pct tutorial is done! See parrotblog.org.
cotto_home nopaste, pct tutorial is <reply>www.parrotblog.org/2008/03/targetin...ot-vm.html 05:17
purl i already had it that way, cotto_home.
cotto_home pct tutorial is <reply>www.parrotblog.org/2008/03/targetin...ot-vm.html
no, pct tutorial is <reply>www.parrotblog.org/2008/03/targetin...ot-vm.html 05:18
purl okay, cotto_home.
cotto_home take that, autocomplete
pct tutorial?
purl www.parrotblog.org/2008/03/targetin...ot-vm.html
05:23 Andy joined
pmichaud purl, forget nopaste, pct tutorial 05:34
purl pmichaud: I forgot nopaste, pct tutorial
cotto_home thanks 05:36
can someone please add documentation for commandline_banner and commandline_prompt to compilers/pct/src/PCT/HLLCompiler.pir? 05:37
06:00 donaldh joined
spinclad post3.txt++ pmichaud++ 06:08
06:21 uniejo joined 06:22 Zaba joined 06:51 bsb joined 06:52 AndyA joined 07:01 Zaba_ joined 07:04 Ademan joined 07:46 IllvilJa joined 07:47 Zaba joined 08:02 iblechbot joined 08:56 cognominal joined 09:38 ank joined 09:39 wknight8111 joined 09:41 particle joined
dalek r27764 | fperrad++ | trunk: 09:54
: [Lua]
: - bytecode translation : fix segfault
diff: www.parrotvm.org/svn/parrot/revision?rev=27764
10:12 ruoso joined
bacek evening 10:15
cognominal is there a official logo for rakudo? 10:16
10:29 Ademan joined
cognominal 惩ć‚Æ惉 10:49
donaldh :)
cognominal I play with pixelmator to get a camel with rakudo in romaji and tatakana 10:53
Zaba katakana? 10:59
purl rumour has it katakana is for non-japanese words, Hiragana is for native ones
11:00 kj joined
donaldh or is that Kanji ? 11:04
Ah, perhaps both. en.wikipedia.org/wiki/Japanese_writing_system 11:07
cognominal katakana for rakudo
kanas are graphically simpler than kanjis. And hiraganas are rounder than the katakanas. 11:08
so you can guess what you get without knowing anything. 11:09
donaldh :D
cognominal ę„½åŗ¦
11:09 masak joined
donaldh Been there. Designing onscreen displays for televisions using 'guessed' katakana. 11:10
cognominal apparently my system can translate some words in kana to kanji. 11:11
He did it from his own volition :)
I have yet to know every feature of Mac OS X
ć‚‰ćć© 11:12
this is rakudo in hiragana
moritz it's nice, but it doesn't make a good logo 11:14
cognominal katakana is used for the translitteration of foreign words. So writing it in katakana conveys the idea that the design has been made by strangers (I knew that word). 11:16
I am learning pixelmator to make it.
seems like inkscape but using Apple core image 11:17
Juerd It would be funnier to transliterate it to yet another script.
Korean? :)
masak Tibetan
11:18 askie joined
cognominal don't make it harder on me 11:18
gaijin, that's the japanese word for strangers 11:19
masak maybe Mongolian too...
cognominal when PGE will support Unicode, you will be able to use whatever alphabet for you identifiers and subclass the Perl6 grammar to do the same for keywords. 11:20
It is probably the way for Perl 6 to be taught in schools of non english speaking countries. 11:22
not that would be a good thing.
I meant countries with a non roman alphabet. 11:23
masak that might impede code reuse somewhat :/ 11:29
moritz masak: I found that even comments in a foreign language kills code reuse 11:30
masak I have indications of that too
s/have/have seen/ 11:31
cognominal when I save in jpeg, I loose the font layers. But I am close to it. This could before a graphist comes. 11:34
tetragon looks at pixelmator 11:37
I do all of my Core Image stuff in Quartz Composer and Python
cognominal I got it in a macheist bundle but never used it before 11:38
psd? 11:41
purl well, psd is PhotoShop Document
donaldh mmm, it looks like pixelmator gets the usual mac eyecandy award. 11:43
tetragon Anyone know if there is a good reason for there to be a couple of "#if PARROT_EXEC_OS_AIX" in ppc/jit_emit.h instead of using ifdef in those cases? (lines 23 and 683) 11:55
They're triggering compiler warnings on my OS X box. 11:56
cognominal objectif-brousse.org/rakudo.jpg 11:59
8666 bytes. It is low quality but ultralight 12:00
that will do for the slides I prepare for fpw2008 12:01
jonathan? 12:02
purl rumour has it jonathan is mailto:jnthn@jnthn.net
tetragon Hrm... Subtest 8 of t/op/debuginfo.t passes if I pass parrot any option that specifies a runcore (other than --slow-core), but fails if I don't (for the set of all values that I have tested so far) 12:15
pmichaud (when PGE supports Unicode) PGE already supports Unicode. 12:18
you probably mean "When Parrot supports Unicode at a reasonable speed... " 12:19
cognominal pmichaud: you can tell rakudo to build with unicode? 12:27
btw for \\h and the other escapes, I think I was wrong.
the problem is with utf-8 , we could use the 32 bit encodin whatever his name. 12:28
pmichaud 32-bit doesn't work on systems w/o icu 12:29
(actually, ucs2 would be more likely -- 16 bit. still doesn't work w/o icu.)
cognominal I am surprised. I thought one needed icu for utf-8. I learn a lot here. 12:30
pmichaud Parrot has some minimal support for utf-8 operations. Unfortunately, utf8 strings are slow largely because of the variable-length encoding 12:31
about the only thing we can't reliably do on utf-8 at the moment w/o icu is downcase, upcase, and character properties for codepoints > 255 12:32
cognominal I should read the book written in French with a name whichs greek. Book I got en english cause it is cheaper.
moritz how does perl5 handle that?
cognominal Anyway, O'Reilly France is dead. 12:33
pmichaud I think perl5 just deals with variable-length encoded utf8 strings and takes the speed hit.
But there the speed hit is much less because we're typically matching a single pattern, and can keep track of the position (both logical and absolute) without having to re-calculate it on every operation. 12:34
12:42 wknight8111 joined 12:59 Andy joined
donaldh Is the empty 'stack opcodes' page intentional? www.parrotcode.org/docs/ops/stack.html 13:13
cognominal these opcodes have been removed so this page should be removed to. 13:23
kj cognominal, donaldh: I'll file a ticket 13:38
donaldh thanks 13:39
kj thanks for reporting
donaldh Is there a way to specify a return parameter in an NCI signature? 13:42
s/return/out/ 13:43
e.g. int sqlite3_open(const char *filename, sqlite3 **ppDb);
13:49 gryphon joined 14:05 rdice joined
Coke donaldh: web page fixed. 14:06
donaldh I really should help with the rt backlog
DietCoke we'd be most appreciative. =-) 14:07
donaldh setting up an account now ... 14:13
DietCoke there are some guidelines for going through the tickets. moment. 14:17
rt hints?
rt hints is 14:18
no, rt hints is www.parrotblog.org/2008/05/700-tick...lenge.html 14:19
there you go.
thanks.
14:19 Andy joined
DietCoke ANDY! ZOMG BBQ ACK! 14:20
Andy WHAT? WHAT WHAT WHAT?
DietCoke hey. 14:21
wknight8111 Failed test 'Pod syntax correct'
got: '/languages/perl6/src/pmc/mutable.pmc.template' 14:22
expected: ''
DietCoke Wie viel fur den Andy im Fenster?
er..
what is mutable.pmc.template ?
wknight8111 never heard of it
i assume it's some temporary for the mutable pmc that pmichaud and jonathan have been working on yesterday 14:23
DietCoke what does svn st tell you about it?
donaldh DietCoke: thanks.
wknight8111 haven't even checked, the test suite is still running
DietCoke I suspect it's just you. 14:24
oh wait!
there it is in svn up. wtf.
oh, it's a -perl6- PMC. =-)
wknight8111 Last changed author: jonathan
DietCoke I am blind.
wknight8111 Last changed rev: 27755 14:25
DietCoke lowers his screen resolution to compensate.
ah, that's better.
wknight8111 for some reason it's being treated as a POD file, and it's failing the POD tests 14:26
Failed test 'Copyright statement exists' 14:27
No copyright statement found in 1 files: languages/perl6/build/get_mutable_pmc.pl
Failed test 'No multi-dot filenames' 14:28
Found in 1 files: 'languages/perl6/src/pmc/mutable.pmc.template'
A few other failures with these two files as well, mostly codingstd
actually, all codingstd 14:29
DietCoke I'd open a ticket and let jonathan deal with it when he returns.
(if it was just one thing, I'd fix it.)
wknight8111 yeah, i'm working on it right now
14:42 cjfields joined
dalek r27765 | Whiteknight++ | gsoc_pdd09: 14:47
: Creating branch for GSoC PDD09 garbage collector work
diff: www.parrotvm.org/svn/parrot/revision?rev=27765
particle wknight8111: re 48549 (.namespace)... 14:51
don't include a patch to imcparser.c, it's a generated file 14:52
apply the patch to imcc.y and run perl Configure.pl --maintainer
that will regenerate imcparser.*
wknight8111 okay, well my previous patch did exactly that. I'll test it with the --maintainer flag 14:57
do you think kjs would mind if i steal the ticket? 15:00
kj kjs doesn't mind :-)
wknight8111 okay, that makes it easy 15:01
the word "steal" always seems so negative
when I commit the patch though (assuming I commit it before somebody else does), I should commit imcparser.c too, right? 15:03
because most developers won't have lex/yacc
moritz if it's under version control you should submit it 15:04
wknight8111 ok, that's what I thought
moritz (at least that's what my common sense tells me ;) 15:05
wknight8111 how long do I give something like this? A day? 15:06
particle moritz++ # common sense
kj I always committed both the .y file and the regenerated .c/.h files, but i'm on windows (using cygwin's yacc) 15:07
particle wknight8111: start with three days
wknight8111 okay, good guideline. that will give me plenty of time to write tests
particle tests++
DietCoke a day is fine. if you get a +1 or two in the meantime from other committers, go for it. 15:12
and we reserve the right to yell at you for screwing things up.
particle hey, let him write his tests! :P
DietCoke oh!
wknight8111 it's a good right to reserve!
DietCoke plugs his laptop into his new lcd tv. that's a hell of a second monitor. 15:13
15:14 tewk joined, jhorwitz joined 15:26 Theory joined
particle coke: ping 15:29
er, DietCoke
DietCoke yes? 15:38
particle re gsoc and #parrotsketch 15:39
should auzon participate in #parrotsketch, do you think?
pmichaud auzon is primarily working on rakudo test suite, yes?
particle spec test suite
pmichaud sorry, Perl 6 test suite
particle yes
pmichaud I'd say it's up to auzon. 15:40
particle or perl 6 design meeting?
PerlJam particle: I'd say definitely. Heck, they even let me participate on occasion :-)
pmichaud well, parrotsketch is primarily parrot-focused.
the Perl 6 test suite isn't really parrot.
DietCoke I would say the perl6 te... right.
and the design meeting may be too high level. 15:41
certainly welcome to audit the PS meetings, though.
PerlJam point him at #perl6 and TimToady and friends.
DietCoke (as is everyone else)
moritz PerlJam: too late for that, he already hangs out there ;)
15:42 sjansen joined
pmichaud as far as design meetings go -- I'd say "yes" if there are significant design questions arising from the test suite 15:42
15:42 particl1 joined
particl1 don't know which is better, though. so i'd like your thoughts 15:42
pmichaud but in some sense it might make more sense for those to go to p6l or p6c
moritz from the logs I don't think he'll profit from the design minutes
particl1 stupid freakin comcast
pmichaud I'd prefer to see test suite questions on either p6l or p6c or irc
particl1 questions will be in irc or mailing list 15:43
i want a weekly report
15:43 cjfields_ joined
pmichaud have him write use.perl journal posts 15:43
as a weekly report
particl1 last years, we've had SoCers report in #parrotsketch
i plan to continue that for this year's crowd, but auzon is not a parrot project per se 15:44
pmichaud well, that's why I said it's really up to auzon. Personally I'd prefer to see the weekly reports on use.perl, p6l, or p6c.
larger audience. 15:45
purl larger audience is just a smaller audience with more faces to idly study for emotional reaction, Alias.
particle there will be official weekly reports, via a webapp or blog
due saturday night, every week
pmichaud I think that's good enough, personally. I'd like to see those added to the planet perl six aggregator.
DietCoke I might have privs to that.
(I do to parrot's, anyway) 15:46
pmichaud are you auzon's mentor for gsoc? 15:47
particle moritz is auzon's mentor. i'm backup
pmichaud I'd say it's up to you and moritz then. :-)
particle moritz: ping 15:48
moritz auzon's blog is auzon.blogspot.com/, don't know if he's set up a gsoc channel
particle: pong
particle moritz: do you attend #parrotsketch regularly?
moritz particle: no, I forget it quite often ;)
particle ok
moritz particle: blog is fine by me
pmichaud if it's in an rss feed somewhere I'll certainly see it and keep up with it. 15:49
I finally discovered Google Reader and it has solved my aggregating issues :-)
particle moritz: do you want to have a formal interactive meeting every week, or just meet informally?
(google reader)++
i've been using it for years
pmichaud particle/moritz: if the two of you wanted to set a regular meeting time for test discussions with auzon, I'll commit to attending as well. 15:50
it could probably be done on #perl6
15:50 Auzon joined
particle hi there Auzon 15:50
pmichaud I propose we take this to #perl6. :-)
moritz I just took Auzon here ;) 15:51
pmichaud I saw that. :-)
Auzon Hi all.
moritz ok, over to #perl6 again ;)
15:51 Auzon left, Auzon joined
DietCoke (aggregating issues) newsgator is also nice, and has a decent mobile option. 15:52
(and is now free.)
pmichaud well, I have so many Google tools already enabled, I figure one more doesn't hurt :-) 15:53
DietCoke (and has an awesome mac client.)
true. I've considered switching over. Kills me because I originally paid for NG. =-)
particle wonders if there's a phone app for google reader 15:55
pmichaud particle: yes. Used it yesterday in fact while the kids were swimming at our neighborhood pool :-)
DietCoke I wish google apps would create iphone icons. would make it that much slicker, for very little effort. 15:56
16:13 wknight8111 joined 16:23 wknight8111 joined 16:25 wknight8111 joined 16:26 Zaba_ joined
pmichaud afk, lunch. 16:37
dalek r27766 | pmichaud++ | trunk:
: [nqp]:
: * Add ternary ?? !! operator to NQP.
diff: www.parrotvm.org/svn/parrot/revision?rev=27766
17:07 cotto-work joined
mj41 seen jonathan 17:09
purl jonathan was last seen on #parrot 18 hours ago, saying: so, night from me
particle long night! 17:10
17:29 ejs joined 17:33 slightlyoff joined, slightlyoff left 17:46 kj joined 17:55 ejs joined
dalek r27767 | allison++ | pdd25cx: 18:17
: [pdd25cx] Deleting stray curly bracket from merge.
diff: www.parrotvm.org/svn/parrot/revision?rev=27767
r27768 | particle++ | trunk: 18:19
: [rakudo] add .perl method to List
diff: www.parrotvm.org/svn/parrot/revision?rev=27768
18:19 ambs joined
dalek r27769 | particle++ | trunk: 18:29
: [rakudo] classes must respect their own apis. converted all C<elements self> to C<self.'elements'()> in List
diff: www.parrotvm.org/svn/parrot/revision?rev=27769
pmichaud particle: .perl on List should use square brackets, not parens 18:33
particle oh? that's not what pugs does. 18:34
parens for list, square for array
pmichaud From S02: # To get a Perlish representation of any object, use the .perl method. Like the Data::Dumper module in Perl 5, the .perl method will put quotes around strings, square brackets around list values, curlies around hash values, constructors around objects, etc., so that Perl can evaluate the result back to the same object.
moritz particle: pugs' .perl isn't very reliable
particle ok, easy fix
purl easy fix is to add [ '-' <ident> ]? , which should suffice for now
pmichaud purl, forget easy fix
purl pmichaud: I forgot easy fix
dalek r27770 | particle++ | trunk: 18:37
: [rakudo] List.perl should use square brackets. pmichaud++
diff: www.parrotvm.org/svn/parrot/revision?rev=27770
pmichaud particle++ ## class method cleanups 18:38
particle yeah, i suspect we need many more cleanups
pmichaud we do.
I've seen lots of issues that have cropped up in the codebase since I was last working on it heavily (early january) 18:39
so I'm doing cleanups now also
but I'm very happy to have others do it :-)
Tene Hm. I need to get the laptop out and git-svn rebase before I leave for the airplane. 18:41
pmichaud I'm doing the same thing before I depart on my trip. :-) 18:44
ambs purl? 18:45
purl ambs?
pmichaud purl?
purl pmichaud?
pmichaud pmichaud?
purl you are, like, www.pmichaud.com/ or "Patrick R. Michaud" <mailto:pmichaud@pobox.com>
Tene tene?
purl you are probably Stephen Weeks
Tene Probably.
purl Really? Probably? Are you Certain it's not certain? Are you sure it's unsure? I think you need to look harder.
ambs purl, shut up!
purl ambs: i'm not following you...
moderator Parrot 0.6.2 "Reverse Sublimation" Released | parrotcode.org/ | 18/672 new/open tix 18:47
18:48 Patterner joined
pmichaud gone for a while (memorial day travel) 18:50
particle enjoy!
dalek r27771 | Whiteknight++ | gsoc_pdd09: 18:52
: [gsoc_pdd09] updating branch to trunk r27766
diff: www.parrotvm.org/svn/parrot/revision?rev=27771
wknight8111 yay! I'm learnding! 18:55
moritz wknight8111++ 18:56
I've seen your first psoc-report, sounds good 18:57
wknight8111: if you have any linux (or tools) related issues feel free to ask
Tene Nice! I think I just recruited someone else to help me with Cardinal. 18:58
Except he can't really start for another two weeks.
:(
kj Tene: how's the cardinal work coming along?
wknight8111 thanks for the offer, moritz! I'm sure I'll have questions eventually
kj is it faster yet? :-)
Tene kj: yes. 18:59
kj cool. Tene++
Tene I got a 50% speedup by replacing many rules with tokens
dalek r27772 | jkeenan++ | trunk:
: Set SVN properties; tag. Add copyright statement.
diff: www.parrotvm.org/svn/parrot/revision?rev=27772
kj gee, couldn't have guessed that's would that a big effect
Tene kj: the big problem right now is tweaking the parser. I think I'll have to figure out how to get help from backtracking.
kj actually, I thought most of the grammar was there already.. wasn't it? 19:00
if not all..
Tene No, a lot of stuff fails to parse or parses incorrectly.
kj ooh ok
Tene I've fixed up most of it since your first commit, but I still have a few bits left over.
Right now I'm trying to get 'var[0]' to parse properly. 19:01
kj I think there's some ruby grammar spec that's incorrect somewhere. I used the yacc grammar input file when I worked on cardinal
19:01 cizra joined
Tene I can get either a function call with '[0]' as an argument, or a parse failure. 19:02
ambs # Failed test 'No multi-dot filenames'
# at t/codingstd/filenames.t line 87.
# Multi-dot filename found in 1 files:
# languages/perl6/src/pmc/mutable.pmc.template
pmichaud, this is with you? :)
Tene I got a lot of other stuff working, though. Check the svn log.
kj ok. I'm sure you know it, but I'll say it anyway: the order of the alternatives in a rule matters
Tene Yes, that's how I can switch between thsoe two options. 19:03
:)
kj and what about a ? clause?
Tene That's the other option I was going to look into this weekend.
19:03 mire joined
kj maybe it helps ifyou insert some assertions 19:04
Tene welcome, cizra
cizra hullo
kj actually, maybe that's nonsense what I just said :-P
19:04 Zaba joined
dalek r27773 | jkeenan++ | trunk: 19:07
: Change filename to conform to 'no multiple-dot filenames' coding standard.
diff: www.parrotvm.org/svn/parrot/revision?rev=27773
19:07 mire joined
dalek r27774 | jkeenan++ | trunk: 19:07
: Fix POD formatting error: '=over' without corresponding '=back'.
diff: www.parrotvm.org/svn/parrot/revision?rev=27774
PerlJam kj: re alternation order matters-- once the longest token matcher gets integrated that may only be true of || (as it should be) So, people who are writing grammars should probably keep that in mind and use | or || appropriately. 19:08
kj PerlJam: yes you're right
I keep being confused in what I should use: single or double
hard to remember
PerlJam if you want order to matter, you use ||
if you want the longest match to win, you use | 19:09
well, if you want the longest match to win *first*, you use |
(backtracking still applies)
kj I thought rules didn't backtrack...
Tene I think that pmichaud said that LTM will make the things I'm having trouble with work better.
kj: but regexes do. 19:11
PerlJam kj: okay, back tracking *may* apply :)
You can turn it on/off at will withing a rule/regex/token
s/ing/in/
kj ah ok
PerlJam but, yes you're right. rules don't backtrack by default. 19:12
DietCoke -> install memory 19:14
ambs question: is it relevant to add a configure step to detect operating system and operting system version? 19:15
particle ambs: why do you need that info? and isn't it there already in init/defaults or somesuch? 19:16
if parrot currently pulls that info blindly from perl's config, then we should have a step that does it ourselves
ambs particle, ATM I need it do decide wether to use dlopen or the NSfoobar API to open shared libraries under MacOS 19:17
but let me look into init/defaults
particle likely you want to modify init::hints &| init::defaults
ambs probably 19:19
purl Really? Probably? Are you Certain it's not certain? Are you sure it's unsure? I think you need to look harder.
moritz purl: forget probably
purl moritz: I forgot probably
ambs moritz, thanks
Auzon ls 19:20
ambs & # dinner
Auzon, no such file or directory 19:21
Auzon now notices this is not a terminal
ambs++
19:25 Coke joined
Coke quadruples the RAM on his desktop. 19:26
confound did you paint it mauve? 19:27
moritz RAM came and diet went?
19:28 DietCoke joined
confound CokeZero 19:28
DietCoke there we go. I never told my irc client about my new nick. CokeZero can be when I'm done. =-)
confound DietCoke: you don't have to /quit to change your nick
DietCoke confound: no, but I do have to quit to test to make sure that my default nick is setup properly for startup. kthnxbye. 19:29
confound wtf? I was just trying to be helpful
DietCoke I have a hard time of thinking of a way of saying "thanks, but I knew that." without sounding like an ass. sorry. 19:30
confound oh. it was the 'kthnxbye' that did it :)
DietCoke sadly, the memory win is only a triple, not a quadruple. (stupid imac) 19:31
(can't address the last gig.) ah well, still much better.
DietCoke tries a parrot build.
confound really? bleah
DietCoke older models had an issue; you can put in 4gb, but only use 3.
Tene exam over. go-go-gadget airport! 19:35
cognominal my macintel thanked me for upgrading to 2Go. He keeps always 500Ko free to be ready to swap in something 19:36
The difference with the 512Ko config was amazing.
oops you speak bytes, non octets
20:08 ruoso joined
dalek allison@perl.org | A foundation for Parrot: 20:10
link: www.perlfoundation.org/parrot/index...for_parrot
shorten dalek's url is at xrl.us/bkxq5
20:26 Psyche^ joined 20:30 davidfetter joined 20:55 donaldh joined 21:04 peepsalot joined
donaldh DietCoke: re rt.perl.org/rt3/Ticket/Modify.html?id=39827 is_abs_path is implemented in src/library.c 21:16
21:18 Zaba_ joined 21:19 lichtkind joined
lichtkind congrats for 0.6.2 21:19
ist has different bugs than 0.6.1 :)
21:20 autarch joined, autarch left
lichtkind where to file bugreport? 21:20
Infinoid purl, parrotbug?
purl i guess parrotbug is mailto:parrotbug@parrotcode.org or svn.perl.org/parrot/trunk/docs/submissions.pod or see also "rakudobug"
lichtkind purl, rakudobug? 21:21
purl i guess rakudobug is mailto:rakudobug@perl.org
lichtkind thanks In
thanks Infinoid 21:22
Infinoid lichtkind++ # finding and reporting bugs
lichtkind Infinoid: it wasnt that hard when simple say fails :)
Infinoid well, that's more testing than I've been able to do for the last week or so... :) 21:23
dalek allison@perl.org | Bylaws:
link: www.perlfoundation.org/parrot/index.cgi?bylaws
allison@perl.org | Bylaws: 21:24
link: www.perlfoundation.org/parrot/index.cgi?bylaws
allison@perl.org | Bylaws: 21:26
link: www.perlfoundation.org/parrot/index.cgi?bylaws
lichtkind k send
good night
purl sleep fast, sleep tight, don't let the bed bugs bite
lichtkind haha
21:27 IllvilJa joined
bacek morning everyone 21:36
spinclad afternoon, bacek. 21:39
dalek allison@perl.org | Bylaws: 21:57
link: www.perlfoundation.org/parrot/index.cgi?bylaws
21:58 Ivatar joined 22:02 Zaba joined
dalek allison@perl.org | Bylaws: 22:02
link: www.perlfoundation.org/parrot/index.cgi?bylaws
allison@perl.org | Bylaws: 22:06
link: www.perlfoundation.org/parrot/index.cgi?bylaws
allison@perl.org | Bylaws: 22:08
link: www.perlfoundation.org/parrot/index.cgi?bylaws
22:12 teknomunk joined
bacek rakudobugs? 22:22
purl, rakudobugs? 22:23
purl bacek: i don't know
moritz without the 's'
bacek purl, rakudobug?
purl well, rakudobug is mailto:rakudobug@perl.org
moritz rakudo has only one bug at a time!
bacek moritz, :) 22:25
Eevee I suppose "not done yet" could be construed as a single bug
bacek fixed implementation of List.map during morning coffee 22:28
Can anyone review #54740 and #54742 please 22:29
22:29 mire joined
moritz just sent patch to #54734 22:30
DietCoke donaldh++
bacek thinks that other List's methods should use Iterator instead of indexed access aswell
DietCoke bacek: are they parrot or perl6?
Infinoid moritz: I had just made an identical change :) 22:31
dalek allison@perl.org | Bylaws: 22:32
link: www.perlfoundation.org/parrot/index.cgi?bylaws
DietCoke dalek: you should wait a minute before reporting on those updates. :|
dalek allison@perl.org | Bylaws:
link: www.perlfoundation.org/parrot/index.cgi?bylaws
DietCoke Seriously.
purl Seriously. are you some kind of python programmer?
Infinoid *ouch*
dalek r27775 | infinoid++ | trunk:
: [rakudo] Fix compilation (RT #54734), moritz++.
diff: www.parrotvm.org/svn/parrot/revision?rev=27775
allison@perl.org | Bylaws: 22:33
link: www.perlfoundation.org/parrot/index.cgi?bylaws
22:33 paco joined
bacek DietCoke, rakudo 22:35
purl, python rulez? 22:42
purl bacek: wish i knew
bacek purl, python?
purl i guess python is THE SCOURGE OF MAN
Eevee I wouldn't go THAT far 22:44
purl, php?
purl it has been said that php is really badly designed or swahili for "child molestor" or or Philippine pesos or the best or PHP: Hardly Programming or brought to you by the teletubbies or proof there is no god or xrl.us/t6wx or more modular than perl or the little transgender brother of perl or a WOMBAT or tnx.nl/php or easy or or was better when it was a templating system or blamed on Rasmus Lerdorf
Eevee purl, perl?
purl i heard perl was or a rather nice compromise between lisp, c and my native spoken language or a 4 letter word
Auzon purl, purl? 22:45
purl i am going alone. or almost an anagram of Donaudampfschiffahrtskapitaensmuetzenkordel or a perv or an auto-triage bot or a she or so stupid or a smartass or the sixth beatle. or www.infobot.org or dumb or a butt sniffer
Eevee purl, purl is also me 22:46
purl okay, Eevee.
Eevee purl, purl?
purl i am going alone. or almost an anagram of Donaudampfschiffahrtskapitaensmuetzenkordel or a perv or an auto-triage bot or a she or so stupid or a smartass or the sixth beatle. or www.infobot.org or dumb or a butt sniffer or me
Tene I updated cardinal to track the rakoo object model changes 23:00
23:00 mire_ joined 23:03 tetragon joined 23:11 mire__ joined 23:21 mire_ joined 23:25 confound joined 23:27 mire__ joined 23:30 confound_ joined, confound joined 23:34 mire_ joined 23:39 mire__ joined 23:53 mire__ joined

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!