»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend! | Rakudo Star Released!
Set by diakopter on 6 September 2010.
00:03 Chillance joined
dalek ecza/master: e4c7a4c | sorear++ | / (7 files):
Turn GetMO() into an instance field

This avoids one virtual call for most REPR calls.		 00:07
ecza/master: e151b74 | sorear++ | src/ (2 files):
Refactor metamodel cross-references a bit
ecza/master: 8dcdd25 | sorear++ | / (4 files):
Run strings and syscalls into portability layer
00:19 justatheory left 01:00 meppl joined 01:23 agentzh joined 01:25 masonkramer left, masonkramer joined 01:30 hercynium joined 01:32 pythonian4000 is now known as pythonian4000afk 01:36 clkao_ left 01:42 justatheory joined 01:43 jferrero left 01:47 molaf__ joined 01:51 molaf_ left 01:52 PerlJam left 01:53 PerlJam joined 02:12 Schwern joined 03:01 Schwern left 03:09 am0c^ joined 03:18 agentzh left 03:19 Italian_Plumber joined 03:37 meppl left
diakopter quietness 03:41
03:50 meppl joined
sorear I am pondering to separate the front/middle and back ends of niecza into separate processes and have them communicate using text files 03:53
rather than passing an AST around
04:00 buubot left, buubot joined 04:12 smartcat99s joined 04:13 hercynium left 04:14 Italian_Plumber left 04:20 smartcat99s left 04:22 obra joined 04:25 meppl left 04:28 am0c^ left 04:43 pythonian4000afk is now known as pythonian4000 04:53 [Sec] left 05:01 envi joined 05:09 risou joined 05:13 [Sec] joined 05:16 Trashlord joined 05:17 Chillance left 05:19 clkao joined 05:24 plobsing_ left 05:31 _kaare joined 05:34 risou_ joined 05:35 risou left 05:55 justatheory left 05:59 satyavvd joined
dalek ecza/master: 06cf3ef | sorear++ | / (4 files):
Add \$foo and \|$foo parameters
 06:07
ecza/master: c2f02c8 | sorear++ | / (5 files):
Porting redirector for clr:Frame and clr:Cursor
ecza/master: 12112fb | sorear++ | / (3 files):
Remove last CLR dependencies from setting
06:25 molaf__ left 06:41 satyavvd_ joined 06:44 satyavvd left, satyavvd_ is now known as satyavvd 06:49 Trashlord left 06:51 _kaare left 06:58 kjeldahl joined 07:26 kjeldahl left, Trashlord joined 07:35 bluescreen left 07:36 risou_ left, bluescreen joined 08:07 mavrc joined 08:14 am0c^ joined 08:34 Mowah joined 08:35 am0c^ left 08:54 M_o_C joined
dalek ecza/master: dc6fe7d | sorear++ | lib/SAFE.setting:
Use \$var through setting
 09:01
ecza/master: 29aab1d | sorear++ | / (6 files):
And some more yaks
moritz_ rakudo: say " one two three four " 09:04
p6eval rakudo 668719: OUTPUT« one two three four ␤»
moritz_ rakudo: say " one two three four ".words.reverse.join(' ')
p6eval rakudo 668719: OUTPUT«four three two one␤»
09:05 zulon joined
sorear out 09:07
09:15 risou joined 09:16 masak joined
masak lol hai 09:16
moritz_ it's masak. LOL! 09:17
masak just gave my premier talk
moritz_ did the audience like it?
and did you like it?
masak I distinctly got that feeling, yes.
yes, I am very happy.
I promised to put up the slides software and the slides. here they are: github.com/masak/osdc-fr-2010-hardc...-reporting
Tene: ^ 09:18
dukeleto: ^^
BooK in particular was impressed and asked questions about the software afterwards. 09:19
moritz_
.oO( why does that system(qw<rm -rf>, $gs_dir); make me slightly uneasy? :-)
 09:20
masak moritz_: $gs_dir is effectively a constant in the program. 09:21
moritz_ I know :-)
masak moritz_: I'm no Perl 5 expert; maybe there are better ways to do the same without 'rm -rf'
moritz_ masak: there's use File::Path qw(remove_tree);
colomon masak++
masak moritz_: I'll consider that.
moritz_++ 09:22
moritz_ installs the prereqs for building the slides...
09:22 jhuni joined
masak oh! non-Mac users will want to change the Inkscape path at the end. 09:22
masak adds that to the README
moritz_ yes, seen that
moritz_ got bit by that shebang line again 09:23
my default perl is in ~/perl5/bin/ somewhere, my /usr/bin/perl didn't find those modules... 09:24
masak would adding 'env' help?
moritz_ yes
masak does that
moritz_ I must say that I'm increasingly impressed by the p5 toolchain
09:24 zulon left
moritz_ I've installed a new Debian testing here, and used perlbrew to build a local perl 09:24
09:25 tadzik joined
moritz_ and from there on 'cpan -i App::cpanminus', and then cpanm $lots_of_modules 09:25
and while the first perl compilation failed for reasons unknown to me, it remained the only failure
so far all those modules with hundreds of dependencies installed just fine 09:26
(for example Dist::Zilla has a huge depenency tree)
tadzik hi
moritz_ o/
masak tadzik: \o 09:27
there; pushed the changes to the README and the script.
I'll go offline now and concentrate on the slides of my second talk -- they're not done yet. :/
o/
09:27 masak left
moritz_ have fun! 09:27
09:33 am0c^ joined 09:47 MayDaniel joined 09:59 Axius joined 10:01 am0c^ left
moritz_ phenny: ask masak if it's normal that slide generation takes a Really Long Time[tm]? 10:02
phenny moritz_: I'll pass that on when masak is around.
10:05 am0c^ joined 10:06 masak joined
masak moritz_: on my laptop it takes about 3 minutes. 10:07
phenny masak: 10:02Z <moritz_> ask masak if it's normal that slide generation takes a Really Long Time[tm]?
10:07 masak left
moritz_ hm, seems that inkscape has a unreasonably high startup cost here 10:09
masak++ # nice slides 10:18
tadzik slides? 10:19
moritz_ though a bit of the text is truncated here
tadzik: backlog :-)
tadzik how long?
:)
moritz_ today :-)
it's not so long
tadzik I see
I can has pdf? 10:20
moritz_ phenny: tell masak about pdftk slides*.pdf cat out all-slides.pdf
phenny moritz_: I'll pass that on when masak is around.
moritz_ tadzik: I'm still building a PDF, inkscape is sooo slow here that the generation takes ages
now at 98 of 111 10:21
rakudo: my Int @a; @a[0] = 3
p6eval rakudo 668719: OUTPUT«maximum recursion depth exceeded␤ in 'at_pos' at line 1␤ in 'at_pos' at line 5:CORE.setting␤ in 'Any::postcircumfix:<[ ]>' at line 1697:CORE.setting␤ in 'at_pos' at line 5:CORE.setting␤ in 'at_pos' at line 5:CORE.setting␤ in 'Any::postcircumfix:<[ ]>' at line
..1697:CORE.settin…
moritz_ rakudo: my Int @a; @a[0] = 'f'
p6eval rakudo 668719: OUTPUT«maximum recursion depth exceeded␤ in 'at_pos' at line 1␤ in 'at_pos' at line 5:CORE.setting␤ in 'Any::postcircumfix:<[ ]>' at line 1697:CORE.setting␤ in 'at_pos' at line 5:CORE.setting␤ in 'at_pos' at line 5:CORE.setting␤ in 'Any::postcircumfix:<[ ]>' at line
..1697:CORE.settin…
10:25 satyavvd left 10:27 am0c^ left 10:28 Holy_Cow left
dalek kudo: 5bdbd4e | moritz++ | t/spectest.data:
don't run S09-typed-arrays/arrays.t

  mikehh++ reports that it sometimes takes ages to run,
and typed arrays are so broken at the moment that it's not really worth running the tests.		 10:30
10:31 amkrankruleuen joined 10:34 risou_ joined 10:35 risou left
moritz_ tadzik: moritz.faui2k3.org/tmp/masak-slides.pdf is the generated PDF... the layout is a bit off in places, I hope masak eventually uploads the generated PDF 10:35
tadzik moritz_++ 10:36
10:47 Axius left 10:50 tadzik left 10:52 szabgab left 10:53 szabgab joined 10:54 Axius joined 10:56 timbunce joined 11:02 Axius left 11:03 jedai_ joined 11:04 Axius joined 11:06 jedai left 11:11 masak joined
masak SVG is platform-independent and nice in many ways. shame about the fonts, though. 11:12
phenny masak: 10:20Z <moritz_> tell masak about pdftk slides*.pdf cat out all-slides.pdf
masak I've merged them (using another tool). now attempting to upload 17M file to a branch on github.
moritz_ branch? they have a download section 11:13
masak ooh
masak aborts
and here it is: github.com/masak/osdc-fr-2010-hardc.../downloads 11:17
moritz++ 11:18
11:21 pythonian4000 is now known as pythonian4000afk
moritz_ masak: what's your second talk about? 11:24
masak act.osdc.fr/osdc2010fr/talk/3051
rakudo: role Bottle[::T] { method Str { "a bottle of {T}" } }; class Wine { ... }; say Bottle[Wine].new 11:28
p6eval rakudo 5bdbd4: OUTPUT«a bottle of Failure()␤»
masak oh la la
masak submits rakudobug
oh, I guess it's the dots. 11:29
rakudo: role Bottle[::T] { method Str { "a bottle of {T.perl}" } }; class Wine {}; say Bottle[Wine].new
p6eval rakudo 5bdbd4: OUTPUT«a bottle of Wine␤»
11:29 hudnix joined 11:30 nsh left, rjbs left, rjbs joined 11:31 nsh joined
masak would doing 'class Wine { ... }' without actually creating the class for realz later be considered a compile-time erreur? 11:32
moritz_ erreur? are you Frenchly assimiliated? :-)
moritz_ not sure 11:33
11:34 am0c^ joined
masak moritz_: pas du tout! 11:34
rakudo: class Car { has @.wheels }; say Car.^attributes 11:37
p6eval rakudo 5bdbd4: OUTPUT«@!wheels␤»
masak rakudo: class Car { has @.wheels }; say Car.^attributes.perl
p6eval rakudo 5bdbd4: OUTPUT«Method 'attributes' not found for invocant of class ''␤ in 'Mu::attribs' at line 1216:CORE.setting␤ in 'Mu::perl' at line 1225:CORE.setting␤ in main program body at line 1␤»
masak o.O
masak submits rakudobug
x3nU how can i convert str to int? Only operator i can find is '+' but it converts to Num 11:38
moritz_ x3nU: .Int
x3nU thanks
11:39 Axius left 11:40 timbunce left 11:46 timbunce joined
masak x3nU: in the long run, even prefix:<+> will convert to Int if the number converted is integrale. 11:47
a whole number, I mean.
zut, all this French subitement!
11:50 fda314925 joined
masak time to focus on slides again. 11:51
11:51 masak left 11:53 tadzik joined 11:58 dju joined 12:02 meraxes left, dju left, satyavvd joined, dju joined 12:09 am0c^ left 12:14 satyavvd left 12:21 M_o_C left, Schwern joined, MayDaniel left 12:29 zulon joined 12:31 fuad joined 12:33 fuad left 12:34 ruoso joined 12:35 nymacro joined 12:36 risou joined 12:37 risou_ left 12:39 Holy_Cow joined 12:44 zulon left 12:47 Axius joined 12:50 rhr left 12:51 am0c^ joined, am0c^ left 12:56 maddingue_ joined
maddingue_ masak speaking about Perl 6 at OSDC.fr 2010 (Paris), live at cite-sciences.ubicast.eu/lives/thd-live/ 12:57
12:57 zulon joined
moritz_ notices a typo in the slides 13:00
=+
instead of +=
13:02 Schwern left
tadzik masak speak louder! :) 13:08
13:08 awwaiid joined 13:09 maddingue_ left
moritz_ too bad the sound doesn't work here at all :( 13:09
13:10 maddingue_ joined
maddingue_ I forgot to say that jnthn will make two talks after Carl (same url) 13:11
moritz_ is there a raw video stream URL somewhere? 13:12
13:13 risou left, risou_ joined 13:14 icwiener joined
x3nU damn, writing something that uses sockets in perl6 seems to be impossible, there's no examples nowhere ;/ 13:15
tadzik LWP::Simple, HTTP::Server::Simple, tests... 13:16
maddingue_ moritz_: no. but everything is recorded. we will upload video on the tubes at some point
moritz_ maddingue_: great, please share the URL when you do 13:17
13:17 cottoo left 13:19 Patterner left
maddingue_ moritz_: sure :) 13:21
13:25 Schwern joined, masonkramer left, masonkramer joined 13:30 Psyche^ joined, Psyche^ is now known as Patterner 13:36 risou joined, risou_ left 13:45 MayDaniel joined 13:47 Schwern left 13:56 Axius left 14:06 ruoso left, timbunce left 14:19 justatheory joined
x3nU i have written simle script which reads everything from socket but threre's problem 14:21
when connection closes
gist.github.com/78b2bfa363efd05606a8
Method 'bytes' not found for invocant of class '' in 'recv' at line 5597:CORE.setting in 'receiveLine' at line 30:januszpawlacz.pl6 in main program body at line 38:januszpawlacz.pl6
14:21 hudnix left 14:22 ruoso joined 14:24 isBEKaml joined
isBEKaml hi, folks! 14:25
tadzik hi isBEKaml!
isBEKaml hello, tadzik! how's things? :)
tadzik isBEKaml: crap, the short story 14:26
the long story is: my phone got lost on a concert :(
isBEKaml oh, boy! that sucks.
tadzik like hell. At least the concert was awesome. But still, it sucks
I found a part of it, so it's probably torn into hundreds of pieces by hundreds of metalheads' boots 14:27
isBEKaml I thought you just lost it - hoping you might get it back. but this, I don't know what to say. :( 14:28
14:28 cotto joined
tadzik well, maybe someone found this biggest part (the rest of what I have), but there was not even PIN, so anyone would have alredy turned it off and see the message "looking for my phone, reward" 14:29
14:29 justatheory left
tadzik x3nU: which one is line 38? 14:30
isBEKaml: funny, in a queue to a dressing room, I met a guy who claimed finding a phone without that my part and sold it for a beer to sb else :/
probably was pulling my leg, but still :/ 14:31
14:31 amkrankruleuen left
isBEKaml folks sell mobiles for a glass of beer? :/ 14:31
tadzik hard to believe, hence I didn't bother with the guy
even if he was speaking truth, it'd still be impossible to find the other guy 14:32
isBEKaml If I were you, I won't think too much of getting it back. You never know, mobiles are the easiest to change hands - but hard to believe they sell these all for beer. 14:33
14:33 Axius joined
tadzik so well, I found some ol' Nokia in my home, will buy some cheap starter 14:33
isBEKaml: especially where beer is for 6 PLN (like 1.5 euro)
isBEKaml tadzik: yeah, that's why I find it hard to believe. Anyway, let's hope everything ends well. (optimistically, you get your mobile back in working condition) 14:35
tadzik duh, I very doubt it. But a simcard would be nice
my subtenant is moving in on monday, and the only contact she has is that phone number 14:36
hopefully she reads her emails
isBEKaml duh. no phone but email. I very much doubt it. 14:37
In the worst case, you got to be at home all day until she turns up. 14:38
tadzik she knows that I have it, so hopefully she will check it out, once she will see that phoning me is impossible
x3nU tadzik: 7 14:39
14:39 Schwern joined, satyavvd joined
tadzik no idea 14:39
14:40 Axius left
isBEKaml rakudo: my $str="a b c d \n"; $str.bytes.say; 14:41
p6eval rakudo 5bdbd4: OUTPUT«9␤»
isBEKaml is chop in Cool-str OS agnostic? 14:42
I kind of vaguely remember jnthn++ adding windows specific code in there when it failed.
it says this: self.substr(0,-1) 14:43
moritz_ isBEKaml: iirc the thing that converts bytes to Str should unify all newlines to \n
chop is always one character off; it's chomp that removes newlines
isBEKaml moritz_: ah, right. confused {me} :| 14:44
moritz_: I was also surprised to find chomp implemented in PIR. Can't it be done like: $str.flip.substr(1).flip; ? 14:45
14:45 zulon left
isBEKaml assuming \r\n would count as 1 byte on Windows? 14:45
rakudo: my $str="a b c d \n"; $str.chomp.say; $str.flip.substr(1).flip.say; 14:46
p6eval rakudo 5bdbd4: OUTPUT«a b c d ␤a b c d ␤»
14:46 meppl joined
isBEKaml rakudo: my $str="a b c d \n"; $str.say; $str.chomp.say; $str.flip.substr(1).flip.say; 14:46
p6eval rakudo 5bdbd4: OUTPUT«a b c d ␤␤a b c d ␤a b c d ␤»
isBEKaml moritz_: ? 14:47
is that a slow implementation?
14:48 awwaiid left 14:50 hercynium joined 14:52 awwaiid joined 14:53 fglock joined 14:54 satyavvd left 14:56 _twitch joined
isBEKaml moritz_: I take that back. that's slow. :( 14:57
x3nU like everyting else written in perl 6 :D
isBEKaml rakudo: my $str="a b c d e \n\n\n\n"; $str.chomp.say; ## is this right? 15:00
p6eval rakudo 5bdbd4: OUTPUT«a b c d e ␤␤␤␤»
isBEKaml it just removed the last new line, not all of it. IIRC, chomp removes _trailing_ new lines. 15:01
15:01 zulon joined 15:02 dju left, dju joined 15:03 dju left, nymacro left, dju joined, maddingue_ left 15:04 dju left, dual left 15:05 dju joined
moritz_ chomp removes at most one trailing newline 15:09
15:09 Italian_Plumber joined
moritz_ buubot: eval: $_ = "ab\n\n\n"; chomp; $_ 15:09
buubot moritz_: ab
isBEKaml wrong again. :( S32's Str.pod says "Returns string with one newline removed from the end."
moritz_ buubot: eval: "ab\n\n"
buubot moritz_: ab
moritz_ buubot: eval: $_ = "ab\n\n\n"; chomp; length
buubot moritz_: 4
15:10 ggoebel joined
isBEKaml moritz_: about that chomp in p6 implementation. Can we have something like that? speed up some parts so we can have it in p6? 15:10
moritz_: if we are going by flip.substr.flip - flip needs to be sped up. this method is really slow, though. 15:11
moritz_ why not use plain substr to start with? 15:12
sorear good * #perl6
moritz_ good * sorear 15:13
isBEKaml hi sorear
moritz_: not sure, maybe I was put off by the comment in there, perhaps? "in PIR for speed" So thought you guys must have already thought of this. 15:14
moritz_ isBEKaml: there's an easy way - write an alternative version, and benchmark it 15:15
isBEKaml moritz_: I'm not even sure. :) 15:20
moritz_: if we are just using substr, why not chop? bombs on windows?
colomon isBEKaml: wasn't that long ago chomp was written in pure perl 6. there was a huge performance improvement by switching to the PIR. 15:30
I don't think it was done using substr, however, so it's probably worth trying to time your version. 15:31
isBEKaml colomon: thanks. will try to rustle up something on my (somewhat ancient) machine. :)
sorear moritz_: looking at your last blog post, (\w+)\w?$0 seems... wrong 15:34
rakudo: say "abcdabc" ~~ / (\w+) \w? $0 /
p6eval rakudo 5bdbd4: OUTPUT«abcdabc␤»
sorear rakudo: say "abcdcba" ~~ / (\w+) \w? $0 /
p6eval rakudo 5bdbd4: OUTPUT«cdc␤»
15:35 plobsing joined 15:37 risou left, risou joined
moritz_ sorear: should be (\w) \w? $0 15:39
sorear ...Oh. 15:40
moritz_ and that's what it is :-)
sorear hmm, there's a STD bug which causes role Foo[$x] { ... }; Foo[2] to be misparsed 15:44
15:45 wamba joined
sorear ::Foo[2] is handled as a syntactic special case but Foo[2] winds up an ordinary postcircumfix call 15:45
15:46 flatwhatson_ left 15:47 Axius joined 15:48 flatwhatson_ joined
isBEKaml rakudo: my $str = "a b c d \n "; $str.chomp.say; $str.bytes.say; 15:51
p6eval rakudo 5bdbd4: OUTPUT«a b c d ␤ ␤10␤»
15:53 obra left
colomon rakudo: my $str = "a b c d \n"; $str.chomp.say; $str.bytes.say; 15:56
p6eval rakudo 5bdbd4: OUTPUT«a b c d ␤9␤»
moritz_ chomp is not in-place
isBEKaml colomon: that's alright. notice the trailing space after \n 15:57
15:58 awwaiid left 15:59 Axius left
isBEKaml rakudo: my $str = "a b c d \n "; $nstr = $str.chomp; say $nstr.bytes ~ " -- " ~ $nstr; 15:59
p6eval rakudo 5bdbd4: OUTPUT«===SORRY!===␤too few positional arguments: 2 passed, 3 (or more) expected␤»
diakopter phenny: tell TimToady sorear> hmm, there's a STD bug which causes role Foo[$x] { ... }; Foo[2] to be misparsed
phenny diakopter: I'll pass that on when TimToady is around.
diakopter phenny: tell TimToady sorear> ::Foo[2] is handled as a syntactic special case but Foo[2] winds up an ordinary postcircumfix call 16:00
phenny diakopter: I'll pass that on when TimToady is around.
isBEKaml rakudo: my $str = "a b c d \n "; $nstr = $str.chomp; $nstr.say;
p6eval rakudo 5bdbd4: OUTPUT«===SORRY!===␤too few positional arguments: 2 passed, 3 (or more) expected␤»
isBEKaml colomon, moritz_ The initial benchmark results just came out. The difference between chomp and substr is huge! 16:06
colomon in favor of chomp? 16:07
isBEKaml colomon: yes.
chomp 1286725626 1286725630 4 0.004
substr 1286726304 1286726348 44 0.044
chop was close.
chop 1286725630 1286725676 46 0.046
and flip was downright horrible. 16:08
flip 1286725676 1286726304 628 0.628
moritz_ ugh. 16:10
isBEKaml colomon: should I post my benchmark code somewhere? (I know it's really horrible, maybe you guys can point out where things can be sped up) 16:11
colomon sure, how about gist.github.com ?
isBEKaml colomon: sure.
16:11 dual joined
colomon though for what it's worth, your results correspond almost exactly with what I would have guessed. :) 16:12
moritz_ parrot strings have a .reverse method
colomon but it's always very good to verify those instincts!
16:12 Italian_Plumber left
isBEKaml gist.github.com/619356 16:13
16:15 tadzik left
moritz_ rakudo: say pir::reverse__SS('abc') 16:15
p6eval rakudo 5bdbd4: OUTPUT«error:imcc:syntax error, unexpected SREG, expecting '(' ('$S126')␤ in file 'EVAL_10' line 89␤===SORRY!===␤syntax error ... somewhere␤»
16:18 mavrc left
isBEKaml colomon: find anything off in the benchmark code? 16:25
colomon isBEKaml: haven't had a chance to look yet, I'm cooking sunday dinner. :) 16:26
isBEKaml colomon: I've used \r\n in there, not sure how that would behave on windows.
colomon: Enjoy! no problem. take your time. :)
16:27 fridim joined, risou left, risou joined 16:30 ggoebel left
dukeleto sorear: re: your comment about niecza processes communicating via text files: C'Dent stores AST's in YAML, you might want to look at that : github.com/ingydotnet/cdent 16:35
16:37 mavrc joined
isBEKaml moritz_: it seems reverse is no longer available. parrot/examples/shootout/revcomp.pir:77 16:39
moritz_: I might be looking at an older file though. 16:40
16:45 hercynium left
moritz_ isBEKaml: right; I've opened a parrot bug report 16:46
isBEKaml moritz_++ 16:47
dukeleto moritz_: i will ask parrot-dev what is up with String.reverse
16:50 [particle] left 16:51 ggoebel joined 16:53 _twitch left
sorear phenny: ask jnthn Where in S14 does it talk about body run time of parametric roles? 16:57
phenny sorear: I'll pass that on when jnthn is around.
16:58 cognominal joined, Schwern left 17:02 ggoebel left 17:06 fglock left 17:07 Axius joined, zulon left 17:11 Axius left 17:13 Axius joined
sorear is now of the impression that "parametric role" is a misnomer 17:16
because EVERY role has at least one type parameter - ::?CLASS 17:17
17:21 kjeldahl joined 17:23 awwaiid joined 17:27 Axius left 17:29 Axius joined, wamba left 17:32 Axius left, mavrc left 17:34 Axius joined 17:37 mavrc joined 17:40 vortex_ joined 17:42 wamba joined 17:47 flatwhatson left
dalek ecza/master: c69ae8b | sorear++ | / (4 files):
Instantiation of parameterized roles
 17:56
ecza/master: 87fd722 | sorear++ | / (6 files):
Implement runtime mixins
17:57 tadzik joined 17:59 flatwhatson joined 18:01 envi left 18:07 Chillance joined
colomon isBEKaml: I believe your substr test isn't actually legal p6. At least, I've heard TimToady rant at length about getting rid of the "-1 means one from the end" trick. 18:09
actually, I guess they're all illegal for that reason. hmmm.
(Rakudo implements it despite TimToady's objections...) 18:10
tadzik it's overfeatured!
isBEKaml all of them illegal? :) 18:11
colomon: so how do we go about substr without reversing it? 18:13
colomon: taking length on long strings is ineffficient, i guess.
18:14 kjeldahl left
moritz_ *-1 18:15
colomon though *-1 isn't implemented yet 18:16
:\
isBEKaml rakudo: my $str="abcdefgh"; $str.substr(*-1).say;
p6eval rakudo 5bdbd4: OUTPUT«maximum recursion depth exceeded␤ in 'Cool::Num' at line 1801:CORE.setting␤ in 'Cool::Num' at line 1801:CORE.setting␤ in 'Cool::Numeric' at line 1788:CORE.setting␤ in 'Cool::Numeric' at line 1788:CORE.setting␤ in 'Cool::Num' at line 7168:CORE.setting␤ in 'Cool::Numeric'
..at li…
colomon (or spec'd, as far as I can see.)
18:17 Axius left
isBEKaml colomon: all I wanted to do was check the last character to see if there's a newline(wasn't quite sure about \r\n - added that to junction) and run the actual code if that passes. 18:17
18:17 [particle] joined
colomon isBEKaml: I understand that. It's just tricky. 18:17
the spec doesn't mention *-1, but I know that's what TimToady wants. 18:20
18:20 flatwhatson_ left
colomon and it does mention StrPos and StrLen helper types, neither of which is actually defined anywhere so far as I know. 18:20
18:20 leprevost left 18:21 Axius joined
sorear StrPos and StrLen are synonyms for Int in any implementation with only one supported unicode level 18:22
moritz_ thinks it's much saner to support exactly one Unicode level per type 18:24
having a separate type for Buf is a big win in terms of clarity
having Str for graphemes and maybe some third type for codepoints would be my preference 18:25
isBEKaml moritz_: when you say Unicode level, are you talking about planes?
moritz_ isBEKaml: no
I'm talking about the bytes/codepoints/graphemes distinction 18:26
isBEKaml oh, that. got ya.
btw, git segfaults when I try to clone the specs repo.
tadzik oh funny 18:27
isBEKaml tadzik: you got that too?
tadzik isBEKaml: nah
18:27 Gothmog_ left
tadzik but it's funny anyway :) 18:27
isBEKaml zsh: segmentation fault git clone github.com/perl6/specs.git
tadzik which git?
isBEKaml no more than that.
v1.6.4 18:28
moritz_ isBEKaml: try the git:// URL
tadzik 1.7.3.1 hier
moritz_ or the git@... ssh URL
tadzik oh, I lied
18:29 vortex_ left
isBEKaml git clone git://git@github.com:perl6/specs.git 18:29
fatal: Unable to look up [email@hidden.address] (port perl6) (Servname not supported for ai_socktype)
tadzik "port perl6" sounds wrong
isBEKaml this is getting weird
moritz_ isBEKaml: wront URL
git://github.com/perl6/specs.git
18:30 Gothmog_ joined
isBEKaml moritz_: isn't perl6 the organisation that can be given in there? 18:30
:perl6/specs.git
tadzik :perl6 is the port no
moritz_ isBEKaml: I don#t understand 18:31
tadzik I think anything after : in an URI is a port number
isBEKaml github.com/masak/yapsi.git
like above.
moritz_ just try the same URL with s/http/git/
isBEKaml [email@hidden.address] 18:32
oh, I gave the wrong url - that's what I use when pushing commits.
tadzik ah, right, I failed again :)
isBEKaml anyway, all's well now. :) 18:33
18:33 flatwhatson_ joined
isBEKaml speaking of which, I should be thinking of doing something in there. :) (been idle/away for too long) 18:34
colomon: S06 shows substr(-1) usage. Why'd TimToady put it there if he didn't want it to be used? 18:37
18:39 risou left
isBEKaml rakudo: my $string="abcdef"; substr($string, -1, Inf).say; 18:40
p6eval rakudo 5bdbd4: OUTPUT«␤»
isBEKaml uhh
weird, that's what S06 shows. 18:41
specifically, S06:2873
pugs: my $string="abcdef"; substr($string, -1, Inf).say; 18:42
p6eval pugs: OUTPUT«␤»
18:44 PerlPilot joined 18:45 PerlJam left 18:49 risou joined 18:55 mavrc left 18:57 Axius left 18:58 Axius joined 19:00 risou left
colomon isBEKaml: S06 is talking about something completely different, I wouldn't trust what it says about substr 19:03
19:07 Axius left
isBEKaml colomon: I don't understand. The only difference I see is, doing $var.substr would involve a 2 parameter signature while substr would involve a 3 paramter signature. Is the latter a direct call into pir? 19:07
s/pir/parrot/
19:08 icwiener left
isBEKaml colomon: If I'm misled in believing they are the same, please point that out. 19:11
colomon isBEKaml: the problem is that you're looking at something which has chosen substr as an arbitrary example, not something which defines substr. 19:13
isBEKaml colomon: well, in that case, I don't see any definition anywhere except perhaps S32/Str.pod 19:15
colomon right, S32/Str.pod is where it is defined, and the definition is imcomplete 19:18
or at least, not properly tracking TimToady++'s mind.
19:23 c9s_ joined, c9s_ left 19:30 mavrc joined
isBEKaml poke TimToady, then? :) 19:34
19:40 xabbu42 joined 19:47 awwaiid left 19:49 Holy_Cow left 20:12 MayDaniel left 20:17 Mowah left 20:30 eternaleye left
isBEKaml rakudo: my $str="abcdefgh abcdefghijkl\n\n"; my @ww = $str.words; @ww.join(" ").say; 20:34
p6eval rakudo 5bdbd4: OUTPUT«abcdefgh abcdefghijkl␤»
20:34 kavakava joined
isBEKaml rakudo: my $str="abcdefghijkl\n\n"; my @ww = $str.words; @ww.join(" ").say; 20:34
p6eval rakudo 5bdbd4: OUTPUT«abcdefghijkl␤»
isBEKaml colomon: ^^ no substr!
colomon: but then, it does break the rule that only one trailing newline char should be deleted. 20:39
rakudo: my $str="abcdefgh,abcdefghijkl\n\n"; my @ww = $str.words; @ww.join(" ").say; 20:45
p6eval rakudo 5bdbd4: OUTPUT«abcdefgh,abcdefghijkl␤»
21:00 wamba left, isBEKaml left 21:11 ggoebel joined 21:25 fridim left 21:30 pythonian4000afk is now known as pythonian4000 21:33 xabbu42 left 21:40 pothos joined 21:47 nymacro joined 22:03 pothos left, pothos joined
dalek ecza/master: 2aa5397 | sorear++ | / (2 files):
Implement :dba
 22:04
sorear needs a way to implement nextsame; that isn't stupidly inefficient for calls that DON'T use it 22:40
dalek ecza/master: 3170c97 | sorear++ | / (7 files):
Implement :lang
 22:41
diakopter as long as nextsame isn't allowed in a stringeval... 22:55
22:56 mfollett joined, mfollett left, mfollett joined 23:12 rhr joined 23:17 mfollett left 23:21 mavrc left
cotto Is t/spec/S05-metasyntax/unknown.rakudo expected to be slow? I've been running Rakudo's spectest under the profiling runcore, and that test has been going for most of a day. 23:30
diakopter coughs 23:35
cotto need a cough drop? 23:39
23:40 Eevee left 23:44 mfollett joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!