»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!
Set by sorear on 4 February 2011.
00:00 Trashlord left, abercrombie joined 00:01 rgrau left 00:02 Trashlord joined 00:13 risou is now known as risou_awy 00:21 cexsum joined
benabik std: /<.*>/ 00:32
p6eval std e3c970e: OUTPUT«===SORRY!===␤Unable to parse metachar at /tmp/rRNjsvYMDv line 1:␤------> /⏏<.*>/␤Couldn't find final '>'; gave up at /tmp/rRNjsvYMDv line 1:␤------> /<.⏏*>/␤ expecting any of:␤ assertion␤ dotty method or post…
00:36 cexsum left
soh_cah_toa perl6: say 'foobar' if "foo" ~~ m:P5/[a-z]/ 00:38
p6eval niecza v8-62-g9e4b0e4: OUTPUT«Unhandled exception: Autoloading NYI␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 685 (CORE die @ 2) ␤ at /home/p6eval/niecza/src/STD.pm6 line 4606 (STD Regex.tweak @ 6) ␤ at /home/p6eval/niecza/src/STD.pm6 line 489 (STD C191_ANON @ 2) ␤ at /home/…
..rakudo a55346: OUTPUT«===SORRY!===␤Confused at line 22, near "if \"foo\" ~"␤»
..pugs: OUTPUT«foobar␤»
soh_cah_toa had a feeling :P5 was nyi 00:39
00:42 cexsum joined 00:43 jdhoreOTG is now known as jdhore 01:09 woosley joined 01:40 cexsum left 01:46 whiteknight left 02:31 soh_cah_toa left 02:32 soh_cah_toa joined 02:40 Reaganomicon left 02:52 lichtkind joined
lichtkind nom does not @.categorize 02:52
02:55 mkramer left 02:56 mkramer joined
abercrombie Hi, in P5 we can do radix conversion using functions like oct(). What's the according one in P6? I know adverb :8() works for single target. How can I convert a batch of numbers? 03:06
phenny abercrombie: 28 Aug 19:08Z <moritz> tell abercrombie about irclog.perlgeek.de/perl6/2011-08-28#i_4337460 (calling functions by name)
TimToady well, there's always map 03:08
abercrombie yeah, so there is no conversion function taking an array as argument? 03:09
TimToady not yet
abercrombie Oh, Thanks. 03:10
sorear why should there be?
03:11 [Coke] left
TimToady it might happen by accident :) 03:11
like it did in the other direction
rakudo: say <1 3 5 9>.fmt('%b') 03:12
p6eval rakudo a55346: OUTPUT«1 11 101 1001␤»
03:14 imbafrog joined
abercrombie How to use adverb with map please? 03:16
oh, i see. my bad 03:18
03:18 [Coke] joined
sorear I don't think map takes adverbs 03:19
abercrombie I mean @a.map({:8($_)}) 03:20
sorear ah. 03:24
FWIW, that's not considered an adverb
Adverbs are when a colon-pair is used to modify a call, such as %hash{$key} :exists
abercrombie Isnt :8($var) an adverb? 03:25
03:28 wolfman2000 left
sorear abercrombie: no 03:29
:8($var) is a colonpair
a colonpair only becomes an adverb when it is used in a syntactic infix position
abercrombie Ah, then I just misunderstood the SYN. It calls :8() "adverbial forms" and I treated it as "adverb". Thanks for correcting me. 03:30
sorear hmm, some oddity 03:31
03:35 frhodes joined
abercrombie (TimToady,sorear,moritz)>>++ for helping. Good night #perl6. 03:37
sorear bye
03:38 abercrombie left 03:42 lichtkind left
[Coke] masak: (testing) how do you test your web apps? 03:52
04:06 DarthGandalf left 04:07 HarryS joined 04:13 DarthGandalf joined 04:19 kaare_ joined
dalek ecza: 4cbf060 | sorear++ | lib/ (2 files):
Reimplement &warn using resumable control exceptions; allows for CONTROL blocks to catch warnings, and adds backtrace capability
 04:20
sorear rakudo: ~Any; 04:21
p6eval rakudo a55346: ( no output )
sorear rakudo: Any.Str
p6eval rakudo a55346: ( no output )
sorear rakudo: +Any
p6eval rakudo a55346: OUTPUT«Use of uninitialized value in numeric context in 'Any::Numeric' at line 1498:src/gen/core.pm␤»
04:21 soh_cah_toa left
sorear nom: ~Any 04:22
p6eval nom 184833: OUTPUT«Use of uninitialized value in string context␤»
04:23 frhodes left
sorear evalbot rebuild niecza 04:23
p6eval OK (started asynchronously)
sorear niecza: ~Any 04:25
p6eval niecza v8-63-g4cbf060: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/lib/CORE.setting line 588 (CORE warn @ 2) ␤ at /home/p6eval/niecza/lib/CORE.setting line 21 (CORE Mu.Str @ 12) ␤ at line 0 (ExitRunloop @ 0) ␤ at /tmp/V7CPMCOatu line 0 (MAIN mainline…
04:26 araujo left 04:27 birdwindupbird joined 04:31 jevin left 04:33 jevin joined 05:00 [Coke] left 05:07 jevin left 05:09 jevin joined 05:31 [particle]1 joined 05:34 [particle] left 05:37 odoacre joined 05:39 wamba joined 05:40 [particle]1 is now known as [particle] 05:48 wtw joined 05:53 araujo joined 05:56 HarryS left
dalek ecza: a39a953 | sorear++ | src/ (5 files):
Suppress all undef warnings triggered by the compiler reboot procedure
 05:57
05:58 cognominal left 06:01 HarryS joined 06:02 wamba left 06:04 imbafrog left 06:07 ZaphrodZenovka left, ZaphrodZenovka joined 06:14 koban joined, koban left 06:17 sayu joined 06:21 lue left 06:28 Vlavv joined 06:37 frhodes joined, frhodes left 06:52 mberends left 06:54 baest joined 06:55 mj41 joined 06:58 mberends joined
mberends phenny, tell lichtkind the reason nom does not have categorize() is probably that no tests exist for it at all to remind the developers (I plan to fix that today) 07:06
phenny mberends: I'll pass that on when lichtkind is around.
sorear mm, 4 failing test files 07:11
will fix those tomorrow and ship
also need to write an announce
sorear out.
07:12 wamba joined 07:17 mishin joined, drbean left
moritz good morning 07:52
\o/ new HPMoR chapter!
08:02 im2ee joined 08:03 domidumont joined 08:13 daxim joined 08:14 wamba left 08:22 zamolxes left 08:23 zamolxes joined, cexsum joined 08:32 sayu left 08:34 agentzh joined
mls morning perl6! 08:49
08:49 wamba joined
moritz good morning mls 08:52
08:53 SHODAN joined
bbkr_ good morning() 08:55
09:08 cexsum left 09:10 orafu left, orafu joined 09:11 cexsum joined
moritz phenny: tell [Coke] that I've blogged a new grant progress report at perlgeek.de/blog-en/perl-6/grant-re...ors-3.html 09:14
phenny moritz: I'll pass that on when [Coke] is around.
09:15 cexsum left 09:20 frhodes joined
daxim moritz, $blog_entry =~ s/Bactrace/Backtrace/ 09:20
moritz daxim: thanks, fixing... 09:24
fixed.
09:32 woosley left
daxim can I see some examples for error messages in action? 09:32
09:36 Searle3 joined, Searle3 left
moritz daxim: sure, just use nom to hack some Perl 6 scripts 09:37
moritz prepares an example
daxim: nopaste.snit.ch/75672 09:39
I have no idea why it reports something for line 1 09:40
and it doesn't yet know about the mainline 09:41
so not perfect yet
09:47 masak joined
masak [Coke]: "how do you test your webapps?" is almost as general a question as "how do you test your programs?". a webapp could be anything. :) 09:48
but my general principle is always to seek out and isolate, as much as possible, the side-effect-free "business logic". this often turns out to be the reason for building the webapp in the first place, and the thing where tests have the highest ROI. 09:49
I was surprised today to discover that &exit in Perl 5 isn't a listop, but acts more like a prefix op, like &defined. 09:51
perl6: exit 0 || (say "OH HAI")
p6eval pugs: ( no output ) 09:52
..rakudo a55346, niecza v8-64-ga39a953: OUTPUT«OH HAI␤»
masak rakudo++ niecza++
frettled_ masak: Why would exit() operate on lists?
masak no need, of course. 09:54
but I didn't expect it to be a prefix.
just like, say, prefix:<-> 09:55
frettled_ But it isn't really an operator, it's a function, so, uhm? 09:56
(IDGI)
masak frettled_: perl -e 'exit 0 || 7'; echo $? 10:00
frettled_: what's the exit code?
10:00 cognominal joined
moritz frettled_: nearly all built-ins in Perl 5 are operators 10:01
for example open()
masak I think using Perl 6 has given me the expectation that things should be listops most of the time. 10:02
10:05 whiteknight joined
masak whiteknight: greetings. 10:06
whiteknight hello masak
masak whiteknight: how's things going with 6model? 10:08
whiteknight masak: it's going reasonably well. I need to take another look at it today 10:10
masak \o/ 10:13
10:14 uasi joined
mberends what's the p6 equivalent of isupper() ? 10:21
10:23 frhodes left
mberends rakudo: my $a = "A"; say ?($a ~~ /<[A..Z]>/) # ugly, not Unicode friendly and probably inefficient 10:26
p6eval rakudo a55346: OUTPUT«Bool::True␤»
10:26 dukeleto left 10:27 dukeleto joined
snarkyboojum could do $a ~~ $a.uc? 10:28
phenny snarkyboojum: 25 Aug 18:18Z <moritz> tell snarkyboojum about the nom branch in github.com/moritz/Perl6-MIME-Base64 (but it breaks it on master :/ -- I don't think there's a way to fix it for nom without breaking it on rakudo/master)
masak yes, $a ~~ .uc would do it, I think. 10:29
snarkyboojum ah.. ~~ .uc lovely
10:30 im2ee left
mberends snarkyboojum++ 10:30
10:33 im2ee joined 10:39 koban` joined, koban` left
mls argh! the mem leak is gone if I put a "sweep 1" op in the code! 10:39
(or parrot is just so slow afterwards that I don't see it anymore...) 10:41
mberends mls: what would that imply about parrot's GC routines?
mls dunno. must investigate further... 10:42
10:43 im2ee left 10:50 satyavvd joined 10:53 satyavvd left, satyavvd joined 10:56 pernatiy left, satyavvd left 10:57 satyavvd joined
frettled_ masak: Well, according to perldoc -f exit, exit expects an expression as its argument. As with other listing EXPR as its argument, I expect exit to follow expression precedence rules. So while print() accepts LIST as its argument, exit accepts EXPR. (There may be precedence goblins that I don't se here, though.) 11:03
moritz: What you mean is that exit(), open() etc. are implemented as operators, although they are not "specified" as such in the Perl documentation? 11:06
Perl 5, that is
masak: Regarding your example, though, I'd consider it bad style, and a high risk of gremlins, since 0 || b can give you 0, "true" or b, depending on context. :) Is it more consistent with Perl 6? 11:08
11:08 dakkar joined
mberends because exit is flow control I would want it to behave like return, only stronger 11:09
11:10 cosimo joined 11:13 MayDaniel joined 11:18 Reaganomicon joined 11:20 dakkar left 11:24 Tedd1 joined
masak frettled_: you didn't answer my question and I don't understand your objection. 11:25
frettled_ masak: which objection?
masak that it's bad style. 11:26
and might give different results in different contexts.
frettled_ masak: I'd expect gremlins when you use a notation like that, and I'd probably avoid typing it out like that in the first place.
masak I see that it can, I just think the context is clear here.
but that's not at all pertinent to what I wanted to show with my example.
frettled_ Okay 11:32
masak my point is that it parses as (exit 0) || 7, not exit(0 || 7) 11:33
frettled masak: to answer your question with another question: how do you expect: open(my $F, '|filename') || 7; to work in Perl 5? 11:34
With or without parentheses :)
Synopsis: open FILEHANDLE,EXPR
masak I generally expect things to be listops. 11:36
frettled Surprise: it behaves exactly as exit().
And AFAICR, it has to do with the EXPR thing
masak the only exception I can think of is that I liked that defined is actually an operator.
frettled So yes, Perl 5 holds some surprises, because different builtins behave differently, and you can't program «intuitively». 11:37
mberends so I write this first test for categorize(), and only then discover that it's not even implemented! D'oh! 11:51
masak no, it's fairly newly spec'd. 11:52
11:52 JimmyZ_ joined
mberends phenny, tell lichtkind I'll get you for this! (categorize) 11:52
phenny mberends: I'll pass that on when lichtkind is around.
11:53 satyavvd left
JimmyZ_ is there a destructor method in perl 6? 11:53
11:53 wamba left
masak JimmyZ_: a DESTROY submethod, yes. 11:54
12:00 JimmyZ_ left 12:03 renormalist|work left 12:04 icwiener joined 12:17 [Coke] joined 12:19 uasi left
djanatyn *yawn* 12:20
I woke up early today, specifically to have some time for playing with perl6. :)
12:22 cexsum joined
masak yay 12:23
12:25 JimmyZ_ joined 12:26 JimmyZ_ is now known as JimmyZ
djanatyn ...hmm. Think I just overwrote my .emacs file. 12:27
12:38 lateau_ joined
[Coke] I am reminded that I need to put all that into a git repo. 12:39
phenny [Coke]: 09:14Z <moritz> tell [Coke] that I've blogged a new grant progress report at perlgeek.de/blog-en/perl-6/grant-re...ors-3.html
12:40 im2ee joined
[Coke] what, am I everybody's grant cop? 12:46
that reminds me.
phenny, ask pmichaud if that grant thing is still ongoing and if I should poke him about it.
phenny [Coke]: I'll pass that on when pmichaud is around.
12:50 [Coke] left 12:52 [Coke] joined 12:55 bluescreen10 joined 13:11 lateau_ left 13:19 drbean joined 13:24 hatseflats joined
mls hah! found the while(1) {} mem leak 13:28
masak \o/ mls++
mls I don't think it's responsible for the slowness, though 13:29
moritz mls: 1.Bool leaks
mls: since the patch that added :D to the invocant 13:30
mls all multis leak
13:30 lateau_ joined
moritz extra points for a patch :-) 13:30
mls since jnthn changed the way protos get called
13:31 wamba joined
mls patch: gist.github.com/1178393 13:31
BUT: parrot currently doesn't export Parrot_pcc_free_registers 13:32
the invoke() call clobbers the current registers, so they need to be freed first
that was really hard to track down... 13:33
moritz I can imagine 13:35
moritz tests 13:36
iirc some windows compilers or linkers are stricter about non-exported symbols
so that might get us into trouble
[Coke] I can try a build on windows if you need. 13:37
mls you need to add an "EXPORT" to parrot
moritz already fails on linux
mls yes.
[Coke] is happy moritz has commit bits there. ;)
mls actually rakudo is doing some evil parrot hacks in the perl6_enter_multi_dispatch_from_onlystar_block op, maybe we should ask the parrot guys to add an "official" way for us 13:40
we basically want to do a tailcall without changing the already set arguments 13:41
13:41 whiteknight left
mls i.e. we want to reuse the callcontext 13:41
13:44 JimmyZ left 13:47 SHODAN left 13:49 mtk joined, JimmyZ joined 13:53 drbean left 13:59 PacoLinux joined 14:07 woosley joined, wtw left 14:10 M_o_C joined 14:13 uasi joined 14:21 uasi_ joined 14:24 uasi left 14:28 woosley left 14:30 woosley joined, thou joined 14:35 M_o_C left 14:38 uasi_ left 14:40 jevin left 14:41 jevin joined 14:47 Holy_Cow joined, sjn left 14:49 tokuhirom joined 14:51 jevin left 14:54 jaldhar left 14:56 jevin joined 14:58 mtk left 14:59 mtk joined 15:01 mtk left 15:03 mtk joined 15:04 birdwindupbird left 15:05 im2ee left 15:06 risou_awy is now known as risou, aloha left 15:11 mkramer left, aloha joined 15:14 pernatiy joined 15:16 im2ee joined 15:19 agentzh left, mtk left 15:33 wamba left, mkramer joined 15:34 mkramer left 15:35 mtk joined 15:36 im2ee left 15:38 masak left 15:39 Holy_Cow left 15:45 wamba joined
TimToady today seems very...mondayish... 15:46
[Coke] TimToady: funny thing about that.
TimToady: how's things? Looking forward to the Hobbit movie? 15:47
TimToady it's not really on my radar yet--still recovering from my spring and summer
JimmyZ will be tuesday after 13 mins
TimToady JimmyZ: I thought about you specially when I said it :) 15:48
[Coke] I have no idea when it's even coming out, but hopefully I will have finished reading the book to my youngest before then. ;)
JimmyZ yeah, I'm from the near futher
TimToady though I guess it's already tomorrow in .jp 15:49
15:51 baest left
PerlJam [Coke]: I hope Peter Jackson gets the Hobbit righter than he got the ending of The Return of the King 15:51
TimToady "a novel is a prose narrative of some length that has something wrong with it." --Randall Jarrell 15:55
15:56 jedai joined 15:58 whiteknight joined 16:01 im2ee joined 16:07 woosley left
TimToady btw, niecza++ runs rosettacode.org/wiki/Resistor_mesh 96 times faster than rakudo; couldn't get it to work under nom (but didn't try very hard) 16:13
C, of course, runs it much faster than niecza :) 16:15
JimmyZ parrot--
16:18 mj41 left 16:23 whiteknight left 16:24 donri joined 16:29 whiteknight joined 16:37 JimmyZ left
flussence I'm about ¼ finished with these graphs now - github.com/flussence/specgraphs/ra...-tests.png 16:38
(would've been slightly further along, but some tests were getting stuck in infinite loops so I had to rig something up to babysit the process) 16:39
daxim svg plox
flussence k 16:40
plobsing oooh shiny
16:40 sjn joined
flussence here you go, I don't think github serves the correct mimetype though... github.com/flussence/specgraphs/bl...-tests.svg 16:41
16:47 thou left
mls vi 16:56
(oops, wrong window)
sjohnson vi == good
16:59 explorer__ left
mls about the setting compilation being slow: it seems to use 20% of the time for find_method 17:00
plobsing vii == better ;)
mls (i.e. the method_cache is often not set)
17:01 avi_ joined
mls jnthn, where art thou? ;) 17:01
tadzik oh, the backlogging... 17:02
phenny tadzik: 26 Aug 14:01Z <moritz> tell tadzik that github.com/perl6/modules.perl6.org...ed348b4f9b makes IO::Select show up on modules.perl6.org again.
tadzik hello #perl6
flussence yay it's tadzik o/
tadzik moritz: good catch, moritz++ 17:03
[Coke] grumbles about grumbling about parrot's speed. 17:04
moritz tadzik: btw I think your last Pod patch broke a test 17:05
tadzik oh?
moritz S02-whitespace_and_comments/begin_end_pod.t iirc 17:06
no wait, S02-whitespace_and_comments/end-pod.t
tadzik what did I introduce last time, hmm 17:07
specialcasing of =config I believe
I'll look into that, let me just sort out dozen of emails and dozens of RSSes ;)
moritz don't hurry :-) 17:08
hope you had a nice weekend
17:08 masak joined, ZaphrodZenovka left
masak g'devening. 17:08
tadzik oh, masak-san!
masak tadzik! \o/ 17:09
I've *missed* you! how long've you been away? three days.
tadzik ;)
don't worry, I'm back for some timez
masak feels like four days, at least.
tadzik my physics exam is starting to look angry at me from the calenda
r
masak oh, you studious you. 17:10
mls afk...
17:10 avi_ left 17:12 sftp left
tadzik oh, new hpotamor 17:12
masak I got halfway through it on the bus home. 17:13
17:13 sftp joined
tadzik I just read the previous one on the metro back home 17:14
Please contact your mentor to discuss the results of 17:20
your evaluation and to plan your goals and development plan for the rest of
the program
intredasting
masak hi :) 17:21
tadzik (: 17:22
17:23 sftp left 17:24 sftp joined
dalek nda: 88c01da | duff++ | bin/panda:
Add ability to search module names/decriptions using regex
 17:28
nda: c57f79b | duff++ | bin/panda:
Add simple word-wrap on searched descriptions
nda: f0748e6 | tadzik++ | bin/panda:
Merge pull request #10 from perlpilot/search

Search
17:30 sftp_ joined 17:31 sftp left
moritz tadzik: I feel contacted. :-) 17:34
the result is awesome, but not perfect
tadzik still swims through the high seas of emails
masak .oO( LTP )
tadzik okay, I'm all ears then
moritz somebody (flussence?) noted that formatting codes don't nest correctly 17:35
sorry, I'm being called to dinner, bbl 17:36
tadzik oh, look, chromatic is bashing Perl 6 again
masak url?
tadzik www.modernperlbooks.com/mt/2011/08/...erl-6.html 17:37
masak actually, that looks like reasoned consideration, not bashing. 17:38
tadzik maybe I detect too much offence in that
flussence: pingy
masak I disagree that Rakudo Star isn't/wasn't what it was supposed to be. 17:39
but beyond that, it was a good read.
benabik Looks more like "I'd like to use Perl 6 but can't because ___", which is useful as it provides input as to what modules R* should add.
masak getting to Perl 5's level of module goodness takes time, that's all.
tadzik maybe my angriness meter is playing silly after a few days of not-caring
masak maybe. 17:42
with me, it's the other way around. if I'm here too much, I start caring too much.
no, we need more users like chromatic, who say "I'd use Rakudo, but it currently lacks these things". 17:43
tadzik agreed 17:44
sorear good * #perl6 17:45
flussence I hit a brick wall while working on Pod::To::HTML... but I was on my netbook at the time so it was more of a slight bump :) 17:46
colomon \o, colomon
sorear benabik: it's pretty clear to me that async IO and full-featured networking are the #1 enabling factors we're missing. I'll attack that next month.
colomon sorear++
benabik sorear++, indeed
sorear btw, undef warnings in niecza have backtraces now. 17:47
oh, the reason niecza no longer passes p5chop and p5chomp... the test files are gone 17:48
flussence 3 things I've come across: the nested stuff is misparsed as moritz mentioned for me, various things ignore config and either parse too much or too little (C<...> mainly), and it'd make my life much saner if the Pod parser parsed L<foo|bar> and D<foo|bar> label/thing bits instead of having to do it myself 17:49
17:49 cognominal_ joined
dalek ast: 8f9a98d | (Martin Berends)++ | S32-list/categorize.t:
[S32-list/categorize.t] four tests for categorize (added to spec 2011-01-14)
 17:49
tadzik flussence: I'm quite sure everything ignores =config, if that's what you mean :) 17:52
17:52 cognominal left
tadzik +1 on L<foo|bar> and friends 17:53
could you provide a testcase for failnesting?
flussence C<C<...>> misparses, I think
tadzik let me see that 17:54
flussence checks local copy of s26.html...
tadzik flussence: well, C<> doesn't treat it's content as verbatim, that's correct 17:55
but is that what you mean?
the insides look fine to me
I tried C<C<foo>>, and $=POD[0].content[0].content[1].content[1].content[0] indeed gives you "foo" :P 17:56
dalek kudo/nom: ebd4957 | (Martin Berends)++ | / (2 files):
[src/core/List.pm] categorize() method and sub, add test to spectest.data
flussence hm, I might be looking at the wrong thing
oh, found one 17:57
"<p>The <code>U&lt;</code>&gt; formatting code specifies[...]"
tadzik hmm
flussence from "The C<U<>> formatting code" 17:58
tadzik I assume that's "The C<U<>>" formatting code specifies"?
yeah, I can understand the misparsing issue
flussence oh, a few lines down there's "C< <em>...<em/> >" which gets pretty messed up in output 17:59
sorear rakudo: 1/0; say "alive"
p6eval rakudo a55346: OUTPUT«alive␤»
sorear perl6: 1/0; say "alive"
p6eval rakudo a55346, niecza v8-64-ga39a953: OUTPUT«alive␤»
..pugs: OUTPUT«*** Illegal division by zero␤ at /tmp/U_5PcLLx4G line 1, column 1-4␤»
flussence (if they're not meant to be verbatim, is that a bug in S26.pod?)
tadzik hmm 18:00
sorear I'd like a show of hands for "pugs is right"
benabik rakudo: say 1/0
p6eval rakudo a55346: OUTPUT«Inf␤»
tadzik verbatimness is less of an issue here, it's more like the document should be probably using «»
so it can nest >
otherwise, it's quite logical that a > ends what < has started 18:01
benabik … Isn't 1/0 NaN, not Inf?
nom: say 1/0
p6eval nom 184833: OUTPUT«Inf␤»
flussence but S26 says I I<can> use balanced delimiters :)
tadzik hmm 18:02
18:02 sftp_ left
tadzik I'll look into that in a second 18:02
18:02 sftp_ joined
flussence sorear: I agree with pugs. I don't think we should be emulating IEEE854 float semantics with arbitrary-precision numeric objects.... 18:02
(maybe with native types though) 18:03
tadzik fix memory leaks, patch by mls++ 18:06
woo!
you mean it doesn't need 1.5 GB anymore? :P
colomon oooo, is it verified the patch works?
flussence oh, I'm compiling right now actually... 18:07
oh wait, this is one of those old revisions, never mind :)
mberends memory size is down some, but still high, almost 800MB on 32 bit, more on 64
tadzik nihil novi it seems
18:08 jevin left 18:11 jevin joined
sorear apparently, I accidentally made lives_ok/dies_ok treat warnings as "die" 18:26
18:26 bbkr joined 18:27 icwiener left
sorear wonders if we want routines for testing warning-nature 18:29
flussence I want one of those!
sorear ? 18:30
flussence (there's one test in Text-Tabs-Wrap that checks whether a thing that's supposed to warn actually does. It doesn't work properly)
sorear flussence: I'd like your input on a sensible API 18:31
moritz sorear: maybe lives_ok/dies_ok need to become throws_ok/lives_ok with an optional :warn flag 18:32
flussence that sounds a bit more sensible than what I had in mind
moritz sorry, throws-ok/lives-ok
tadzik :) 18:33
sorear no, it's lives_ok
18:33 REPLeffect left
moritz doesn't care, as long as we are strangely consistent in some way 18:34
masak \o/
moritz (straight forward consistent would be OK too) 18:35
moritz -> afk agian 18:36
sorear the user should design this
masak I'm busy 18:38
:P 18:39
dalek ast: 27b15f5 | sorear++ | S04-declarations/my.t:
[S04-declarations/my.t] Niecza unfudging, warning suppression, do not rely on 1/0 throwing, SETTING::<> has two :s
 18:41
sorear masak: the user is flussence.
and now I have to leave.
flussence hm, I was originally thinking of a warns_ok function, but having an exception type mask on lives_ok/etc. sounds better for everyone 18:43
(except implementors :) 18:45
tadzik oh, I thought you were volunteering? 18:46
:P
18:47 REPLeffect joined
flussence I'm keeping myself busy already, trying to turn this Pod tree-walking/html-emitting stuff into two slightly more sane modules... 18:48
18:48 birdwindupbird joined
tadzik ooh, ++ 18:49
we really need something to detect Lists properly 18:50
for now, we can detect elements
how much that sucks for HTML or LaTeX, you know already :) 18:51
masak huh? it sucks for HTML or LaTeX? 18:52
flussence masak: =item doesn't have a grouping thing, so the best we've got at the moment is wrapping each list item in a <ul>
masak oh, that little problem. :) 18:54
tadzik yeah, to say it gently 18:55
masak I believe it has a nice solution, and that you'll find it. 18:56
18:58 wamba left 19:07 dorlamm joined, daxim left
flussence maybe I could borrow from the html5 parsing algorithm, they'll have something for loose <li> tags... 19:12
19:17 [particle]1 joined, dukeleto left 19:20 [particle] left 19:21 [particle]1 is now known as [particle] 19:32 dorlamm left 19:34 TiMBuS joined, lichtkind joined
lichtkind masak ping 19:35
phenny lichtkind: 07:06Z <mberends> tell lichtkind the reason nom does not have categorize() is probably that no tests exist for it at all to remind the developers (I plan to fix that today)
lichtkind: 11:52Z <mberends> tell lichtkind I'll get you for this! (categorize)
19:35 dorlamm joined
lichtkind mberends: you want tests? 19:35
lichtkind jsut finished his slides, last one is a big thank you to jnthn++ and pmichaud++ 19:36
mberends lichtkind: I wrote 4 tests today, but more might be useful if they cover different code 19:37
lichtkind mberends: link pls
masak lichtkind: you pang? 19:38
lichtkind hahah
mberends lichtkind: github.com/perl6/roast/blob/master...tegorize.t 19:39
lichtkind masak: i just wanted show you the slides
masak \o/
lichtkind masak: kepher.kilo028.server4you.de/nightl...ctures.pdf 19:43
its up in 30 sec or so
mberends: thanks 19:44
masak looks 19:45
lichtkind: those are excellent slides :) 19:47
19:48 thou joined
masak *@( ... ) ? really? 19:48
lichtkind masak: im happy you like it
19:48 dukeleto joined
lichtkind but still tweaking it a bit 19:48
masak isn't flattening just prefix:<|>?
lichtkind masak: *@ doesnt work in nom 19:49
but thats what i got from the syn
| is something different
thats just positional context for parcels
thats slightly different to *@ 19:50
or it was changed, iwasnt around for some month
masak lichtkind: s/How many elements has this list\?/How many elements does this list have?/ 19:51
lichtkind: I thought prefix:<*> for flattening arrays was killed looooong ago. 19:52
tadzik where did you get Lego Larry?
lichtkind tadzik: from the interwebs, i stole it from one of my older slides 19:54
tadzik tee hee
masak lichtkind: @digits = 0..9 without the parentheses works in Perl 5 too.
the common term for a "linear skipping" sequence is "arithmetic". you already have "geometric" in there. 19:55
lichtkind masak yes 19:56
i corrected the other too
+otehr changes
its 10% funnier in 2 minutes
masak lichtkind: "Works Ob Seq() Too"... "Ob"? 19:57
whoa, those pictures of jnthn and pmichaud both appear to be a few years old. 19:58
19:59 whiteknight left
masak anyway, lichtkind++ 19:59
lichtkind masak 20:00
you have better
that of jnthn i would change 20:01
but not that one of pm because there hes looks more fresh :)
masak which slides is ob?
20:02 whiteknight joined, birdwindupbird left 20:03 mj41 joined
lichtkind masak: vound and corrected to : Works On Parcel() Too 20:03
masak goodie
mberends lichtkind: categorize() implemented today in nom 20:05
lichtkind good so i remove the npot implemented yet warning too 20:06
20:06 soh_cah_toa joined
masak whoa, mberends++! 20:07
20:07 envi left
masak lichtkind: I still think you're working on very old information with prefix:<*> and prefix:<**>, and that you should check that with someone. 20:07
lichtkind yes i planed to do that 20:08
masak .oO( spec changes are not to be taken lichtly ) 20:09
mberends :) 20:11
lichtkind if i can find it in syn i change it 20:12
20:12 bluescreen10 left 20:13 kaare_ left
lichtkind allright everything corrected unless the **, reload the pdf and spot the 2 new jokes 20:14
its up in 20 sec 20:15
looks like lol ist now written LoL 20:18
but *@ is still used in syn3 ? 20:21
ist just slurpy or fossils?
moritz *@ in a signature is slurpy, and not a fossile 20:22
lichtkind mberends++ for implementing categorize
moritz but that's not prefix:<*>
lichtkind moritz: just wanted to clear that up :) 20:24
moritz (signatures are parsed sepcially, not with the operator precdence parser
*specially
)
masak they're like... a DSL... :)
20:26 jferrero joined 20:28 molaf_ left
lichtkind | seems to not be implemented in nom+ 20:28
masak nom: my @a = 1, 2, 3; sub foo($x, $y, $z) { say "$x $y $z" }; foo(|@a) 20:29
p6eval nom ebd495: OUTPUT«1 2 3␤»
masak seems it is.
20:29 bluescreen10 joined
moritz nom: sub f(*@a) { say @a.perl }; f 1, 2, 3 20:29
p6eval nom ebd495: OUTPUT«Array.new(1, 2, 3)␤»
moritz that's slurpy
masak++ showed argument interpolation
masak I never *really* understood why slurpy could be * but argument interpolation couldn't. 20:30
moritz std: |(my @foo)
p6eval std e3c970e: OUTPUT«ok 00:01 120m␤»
moritz perhaps we didn't want a prefix:<*> in the normal grammar? 20:31
masak yes, but why?
moritz oh, because * is a term already? 20:32
having both a prefix and a term of the same name is bound to cause trouble, no?
masak this was before term:<*> existed.
maybe just maybe we had @a[*-1], but we had no conception that it was as unsugared as it is now. 20:33
20:33 sili is now known as railbait_heavy
lichtkind maybe your new is fresher 20:33
i just compile again
but here | doesnt work
masak but no, even @a[*-1] came after argument interpolation switched to being |
lichtkind: well, it works here :) 20:34
20:34 railbait_heavy is now known as sili
lichtkind ist there a operator for LoL ? 20:35
moritz masak: ah, but we do have (|$x) in signatures
djanatyn Do I need to put a semicolon on the ending curly brace of an if statement in perl 6?
moritz djanatyn: only if there's no newline after it 20:36
djanatyn Or do I only have to do that to make cperl-mode happy?
Ah. Is it considered good practice, then?
moritz only if there's no newline after it :-)
tadzik our best practices are built-in in the spec :) 20:37
masak djanatyn: I put in the semicolon after the closing brace sometimes even when it's not needed. but that's not very often at all.
djanatyn: if the statement is an assignment, I might, for example. otherwise it looks so weird.
but it's mostly a nostalgic reaction.
moritz masak: to your p5 trained eye :-)
djanatyn Hmm. Maybe cperl-mode is just freaking out because I'm writing Perl 6.
masak or some foolish consistency.
djanatyn It's also doing *weird* things to my indentation.
tadzik sleeps 20:38
moritz masak: what if you pass a block as argument?
lichtkind tadzik: good night
masak moritz: I'm not *that* well-versed in Perl 5, you know :P
moritz: yes, probably then too. I dunno.
moritz do-something { \n code here \n } \n
semicolon before the final \n ?
masak oh! it depends.
is do-something a quasi-keyword? then no. 20:39
is it just a "normal sub" that happens to take a code block as its argument? then yes.
for me, it's more about usage than about extreme consistency.
20:39 dorlamm left
masak essentially I want the code to lie consistently :P 20:39
lichtkind fixed **@ in tablets too 20:40
20:41 mj41 left
masak yay 20:41
djanatyn does anyone here write perl 6 in emacs?
masak lichtkind: it's infix:<||> now
djanatyn: when I do, I use cperl too.
but mostly I use vim with stubbornly crappy Perl 5 highlighting :( 20:42
djanatyn Hmm.
moritz uses vim + perl6.vim
unless it hits a pathologically slow case (very seldom), then I switch off syntax hilighting entirely
djanatyn Hmm. 20:43
I don't even care about incorrect syntax highlighting very much - cperl-mode is doing some very, very weird things with indetation.
Also, I apologize for saying "Hmm." twice. ;)
I'm in four different IRC channels, sometimes I get confused as to what I've already said. 20:44
moritz no problem
djanatyn (four channels simultaneously, I mean)
moritz 's channel count as at 14, though not all of them are active simultaneously
djanatyn I've heard lots of good things about perl6.vim.
djanatyn checks his channel chat
moritz it is maintained! 20:45
djanatyn 9 right now in ERC.
Still, I really don't want to switch from emacs to vim just to write perl 6. I'm really comfortable in emacs.
I saw an elisp file specifically for perl 6 highlighting and indentation on the emacs wiki. 20:46
It was hosted on the pug site's svn, and I wasn't able to access it. :(
masak ooh, a fellow ERC user! finally! \o/
djanatyn: I use Emacs for prose and vim for programming. YMMV.
djanatyn Yeah, emacs has some really good features, even besides writing code. 20:47
masak yeah.
I don't even look at editors without macro capabilities anymore. 20:48
djanatyn I personally have an emacs session running in tmux on a remote host that I ssh to - it's constantly running.
masak and regex-based commands feel like a requirement, too.
20:48 tokuhirom left
lichtkind i think i delete that flatening and non flatening section completely 20:48
have way too much slides anyway
djanatyn I write my code locally, and use org-mode pretty frequently too.
moritz djanatyn: fwiw the files from the pugs svn repo are now in the perl6/mu repo on github 20:49
djanatyn Hmm, maybe I can find it there.
moritz file util/cperl-mode.el
20:52 wamba joined
djanatyn Awesome! :D 20:52
woo, it's working :D 20:55
masak woo!
20:56 bbkr left 20:57 frhodes joined, frhodes left
djanatyn So, I was bored, and read an artice on coding horror that said 199/200 people that interview for coding jobs couldn't even write FizzBuzz. 21:01
So I said, "Huh. That's crazy."
"I wonder if I could do that in perl 6..."
masak so you implemented it.
ooh, let's spend the rest of the evening writing FizzBuzz in different ways! 21:02
plobsing is there a p6 fizzbuzz implementation on rosettacode yet?
masak I'd be surprised if not.
I mean, if there's an article on RC about FizzBuzz, and I'd be surprised if there wasn't :)
djanatyn: shall I post a solution, or shall I wait for you to do it? ;) 21:03
djanatyn gist.github.com/1179403 <-- is it possible to get rid of those "$_"?
masak djanatyn: well, you can do .say
in the last case.
djanatyn ".say"?
what is that? 21:04
plobsing shorthand for $_.say
flussence nom: 'a'.say
p6eval nom ebd495: OUTPUT«a␤»
djanatyn it gives a binary subroutine/function-thing the $_ as an argument?
...oooohhh
flussence nom: 'a'.print
djanatyn I see
p6eval nom ebd495: OUTPUT«Method 'print' not found for invocant of class 'Str'␤ in <anon> at /tmp/oQSH8nIGMa:1␤ in <anon> at /tmp/oQSH8nIGMa:1␤␤»
flussence eh
masak hah!
flussence nom: 'a'.^methods(:local)
p6eval nom ebd495: ( no output )
flussence nom: 'a'.^methods(:local)».name.say
p6eval nom ebd495: OUTPUT«Int Num chomp chop substr pred succ ll-match ords lines samecase samespace trim-leading trim-trailing trim words capitalize Bool Str ACCEPTS Numeric gist perl comb match subst split␤»
masak and you people say nom is ready for prime time! it doesn't even have .print! :P 21:05
plobsing wonders how many spec failures nom could fix by implementing .say
flussence zero!
plobsing really?
it is untested?
flussence I think you meant .print :)
masak :P
plobsing yes. that would be it
masak flussence: wise-ass. I like you :) 21:06
flussence huh, it seems there really aren't any Str.prints in roast 21:08
there's filehandle.print but no strings AFAICT
plobsing isn't that a feature thats used fairly frequently? why is it untested? 21:09
masak nom: sub fizzbuzz { return "FizzBuzz!" if $^n %% 15; return "Fizz!" if $n %% 3; return "Buzz!" if $n %% 5; return $n }; say .&fizzbuzz for 1..100 21:10
p6eval nom ebd495: OUTPUT«1␤2␤Fizz!␤4␤Buzz!␤Fizz!␤7␤8␤Fizz!␤Buzz!␤11␤Fizz!␤13␤14␤FizzBuzz!␤16␤17␤Fizz!␤19␤Buzz!␤Fizz!␤22␤23␤Fizz!␤Buzz!␤26␤Fizz!␤28␤29␤FizzBuzz!␤31␤32␤Fizz!␤34␤Buzz!␤Fizz!␤37␤38␤Fizz!␤Buzz!␤41␤Fizz!␤43␤44␤FizzBuzz!␤46␤47␤Fizz!␤49␤Buzz!␤Fizz!␤52␤53␤Fizz!␤Buzz!␤56␤Fizz!␤58␤59␤Fiz…
masak \o/
djanatyn: TIMTOWTDI.
djanatyn runs away to #python
flussence plobsing: not that frequent when there's a shorter function that DWIM more often...
djanatyn what's that "$^n"?
masak djanatyn: oh, it just says "this is a parameter". 21:11
flussence $^n says to the compiler "I'm too lazy to put a signature on this sub, go figure it out yourself"
plobsing djanatyn: it is like $a and $b in perl5's sort
flussence and they work in alphabetical order
masak well, Unicode order :)
djanatyn so, there's also a $^a?
plobsing djanatyn: if you type it, it is there 21:12
flussence std: $^Æ
p6eval std e3c970e: OUTPUT«===SORRY!===␤Placeholder variable $^Æ may not be used outside of a block at /tmp/xljdL1Dui5 line 1:␤------> $^Æ⏏<EOL>␤Check failed␤FAILED 00:01 118m␤»
flussence std: {$^Æ}
p6eval std e3c970e: OUTPUT«ok 00:01 118m␤»
plobsing masak: is that locale-specific?
masak plobsing: no.
plobsing: the code points are ordered.
PerlJam djanatyn: variables like $^a automatically create a signature for the sub with those vars in unicode-order as parameters
plobsing aw. that was an opening for so much evil... 21:13
masak plobsing: it's not evil enough!?
flussence I think (or at least hope) the Unicode people have all that sorting stuff defined already...
plobsing masak: it is never enough
flussence (unicode's evil enough already :)
masak nom: say { $^n %% 15 ?? "FizzBuzz!" !! $n %% 3 ?? "Fizz!" !! $n %% 5 ?? "Buzz!" !! $n }($_) for 1..100 21:15
p6eval nom ebd495: OUTPUT«1␤2␤Fizz!␤4␤Buzz!␤Fizz!␤7␤8␤Fizz!␤Buzz!␤11␤Fizz!␤13␤14␤FizzBuzz!␤16␤17␤Fizz!␤19␤Buzz!␤Fizz!␤22␤23␤Fizz!␤Buzz!␤26␤Fizz!␤28␤29␤FizzBuzz!␤31␤32␤Fizz!␤34␤Buzz!␤Fizz!␤37␤38␤Fizz!␤Buzz!␤41␤Fizz!␤43␤44␤FizzBuzz!␤46␤47␤Fizz!␤49␤Buzz!␤Fizz!␤52␤53␤Fizz!␤Buzz!␤56␤Fizz!␤58␤59␤Fiz…
masak a bit shorter.
21:16 wk joined
masak nom: say $_ %% 15 ?? "FizzBuzz!" !! $_ %% 3 ?? "Fizz!" !! $_ %% 5 ?? "Buzz!" !! $_ for 1..100 21:16
p6eval nom ebd495: OUTPUT«1␤2␤Fizz!␤4␤Buzz!␤Fizz!␤7␤8␤Fizz!␤Buzz!␤11␤Fizz!␤13␤14␤FizzBuzz!␤16␤17␤Fizz!␤19␤Buzz!␤Fizz!␤22␤23␤Fizz!␤Buzz!␤26␤Fizz!␤28␤29␤FizzBuzz!␤31␤32␤Fizz!␤34␤Buzz!␤Fizz!␤37␤38␤Fizz!␤Buzz!␤41␤Fizz!␤43␤44␤FizzBuzz!␤46␤47␤Fizz!␤49␤Buzz!␤Fizz!␤52␤53␤Fizz!␤Buzz!␤56␤Fizz!␤58␤59␤Fiz…
masak shorter still.
djanatyn Is this a round of perl 6 golf? :) 21:17
flussence nom: say $_ %% 15 ?? "FizzBuzz!" !! $_ %% 3 ?? "Fizz!" !! $_ %% 5 ?? "Buzz!" !! $_ for ^100 21:18
p6eval nom ebd495: OUTPUT«FizzBuzz!␤1␤2␤Fizz!␤4␤Buzz!␤Fizz!␤7␤8␤Fizz!␤Buzz!␤11␤Fizz!␤13␤14␤FizzBuzz!␤16␤17␤Fizz!␤19␤Buzz!␤Fizz!␤22␤23␤Fizz!␤Buzz!␤26␤Fizz!␤28␤29␤FizzBuzz!␤31␤32␤Fizz!␤34␤Buzz!␤Fizz!␤37␤38␤Fizz!␤Buzz!␤41␤Fizz!␤43␤44␤FizzBuzz!␤46␤47␤Fizz!␤49␤Buzz!␤Fizz!␤52␤53␤Fizz!␤Buzz!␤56␤Fizz!…
flussence hm, not the same...
TimToady rosettacode.org/wiki/FizzBuzz#Perl_6 <-- of course there's an RC entry for that, several in fact
masak rests his case 21:19
21:19 jaldhar joined, MayDaniel left
TimToady none of which use ??!! 21:19
djanatyn I have a friend who was going through the project euler questions using perl6, and pushed all his solutions to a git repo. 21:20
flussence nom: use FizzBuzz qw(100);
p6eval nom ebd495: OUTPUT«===SORRY!===␤arglist case of use not yet implemented at line 1, near ";"␤»
djanatyn I did a couple project euler problems in haskell, and learned a lot by comparing my solutions with the solutions given in the wiki.
masak TimToady: infix:<x>. that's cute, but Merlyn wouldn't like it. :P
djanatyn I learned to appreciate a lot of things in Perl after using Python for a bit.
for one, string interpolation in Python *SUCKS*. 21:21
21:21 y3llow left
masak TimToady: he once said Bool::True should be 43 or something, so that people wouldn't abuse it in exactly this way... 21:21
21:21 y3llow joined
flussence std: use Acme::FizzBuzz:from<perl5>; # there, that's as short as it gets in p6 for now. 21:22
p6eval std e3c970e: OUTPUT«===SORRY!===␤Can't locate module Acme::FizzBuzz at /tmp/vLZiCn3IKc line 1:␤------> use Acme::FizzBuzz:from<perl5>⏏; # there, that's as short as it gets in␤Check failed␤FAILED 00:01 118m␤»
TimToady mapping Yes/No to There/NoThere is perfectly natural
[Coke] well, if you're going to make it non-1, make it 42. sheesh. ;)
TimToady *Not
djanatyn Also, I was very saddened to discover there was no qw() in python.
Is that just a perl thing?
flussence yep 21:23
masak [Coke]: I can't swear that he didn't actually write 42.
PerlJam djanatyn: doesn't python have %w() or something ?
flussence does it?
PerlJam or am I thinking of ruby?
plobsing djanatyn: shell somewhat has that by design
djanatyn PerlJam: Not sure.
masak Ruby definitely has it.
djanatyn Ruby has some interesting things, especially with string interpolation.
IIRC, you can embed a Ruby statement in a string. 21:24
TimToady we can do it easier
djanatyn Oh?
(v5 or v6?)
TimToady v6
plobsing v5 does it, but it ain't pretty
PerlJam djanatyn: v6 of course (check the channel you're in :) 21:25
djanatyn :D
masak it's almost too easy in Perl 6! :P
djanatyn how is it done? :)
TimToady perl6: say "The answer is { 7 * 3 * 2 }"
lichtkind mberends: im very grateful but excuse me had lot of other things to do
p6eval pugs, rakudo a55346, niecza v8-64-ga39a953: OUTPUT«The answer is 42␤»
flussence
.oO(works in PHP too, you just have to prefix your statements with a useless variable...)
 21:26
djanatyn hmm...
TimToady perl6: say "The answer is { $_ for 1..10 }"
p6eval pugs, niecza v8-64-ga39a953: OUTPUT«The answer is ␤»
..rakudo a55346: OUTPUT«The answer is 1 2 3 4 5 6 7 8 9 10␤»
masak nieczabug!
djanatyn perl6: $foo = 1; $string = "the number is { $foo }"; $foo += 1; say $string
p6eval rakudo a55346: OUTPUT«===SORRY!===␤Symbol '$foo' not predeclared in <anonymous> (/tmp/K9glA0Nkm8:22)␤»
..niecza v8-64-ga39a953: OUTPUT«===SORRY!===␤␤Variable $foo is not predeclared at /tmp/XAAGri65kD line 1:␤------> <BOL>⏏$foo = 1; $string = "the number is { $fo␤␤Variable $string is not predeclared at /tmp/XAAGri65kD line 1:␤------> $foo = 1; [33…
..pugs: OUTPUT«*** ␤ Unexpected " ="␤ expecting "::"␤ Variable "$foo" requires predeclaration or explicit package name␤ at /tmp/dZ7PQ_mgTh line 1, column 5␤»
djanatyn ...oh 21:27
I suppose I did that wrong
TimToady need my
djanatyn ah, use strict is predeclared :D
TimToady no, it's not that lazy
djanatyn perl6: my $foo =1; my $string = "the number is { $foo }"; $foo += 1; say $string
p6eval pugs, rakudo a55346, niecza v8-64-ga39a953: OUTPUT«the number is 1␤»
djanatyn So, it interpolates the string at the time of it's declaration? 21:28
TimToady assignment
21:28 jferrero left
TimToady when the string is eval'd 21:28
we don't automatically close over strings
you can always write a closure for htat 21:29
*that even
plobsing perl6: my $builder = "the number is { * }"; say $builder(2);
masak Groovy closes over strings, IIRC.
p6eval niecza v8-64-ga39a953: OUTPUT«Unhandled exception: Unable to resolve method postcircumfix:<( )> in class Str␤ at /tmp/LK5PIopNb8 line 1 (MAIN mainline @ 7) ␤ at /home/p6eval/niecza/lib/CORE.setting line 2045 (CORE C954_ANON @ 2) ␤ at /home/p6eval/niecza/lib/CORE.setting line 2046 (CO… 21:30
..rakudo a55346: OUTPUT«invoke() not implemented in class 'Perl6Str'␤ in main program body at line 22:/tmp/k4LhEhX_rQ␤»
..pugs: OUTPUT«*** Cannot cast from VStr "the number is Inf" to Pugs.AST.Types.VCode (VCode)␤ at /tmp/rd6zLdeC83 line 1, column 42-53␤»
masak it confuses the heck out of most Groovy developers :)
TimToady perl6: my $foo = 1; my $closure = {"the number is $foo"}; $foo++; say $closure();
p6eval pugs, rakudo a55346, niecza v8-64-ga39a953: OUTPUT«the number is 2␤»
TimToady perl6: my $foo =1; my $string = "the number is { lazy $foo }"; $foo++; say $string 21:31
p6eval niecza v8-64-ga39a953: OUTPUT«===SORRY!===␤␤Action method statement_prefix:lazy not yet implemented at /tmp/jxWSY1L7HI line 1:␤------> my $string = "the number is { lazy $foo ⏏}"; $foo++; say $string␤␤Unhandled exception: Unable to resolve method sta…
..pugs: OUTPUT«*** Cannot cast from VInt 1 to Pugs.AST.Types.VCode (VCode)␤ at /tmp/bypCtZcCi4 line 1, column 43-53␤»
..rakudo a55346: OUTPUT«Could not find sub &lazy␤ in main program body at line 22:/tmp/uWntd9_2YY␤»
plobsing so whatever-star in a string interpolation doesn't turn into a closure?
TimToady aw
no, isolated * is always just a Whatever 21:32
masak nom: say "This is just an asterisk: *"
p6eval nom ebd495: OUTPUT«This is just an asterisk: *␤»
TimToady you'd have to write it with concat
plobsing I mean, you can't do it with "{ * }"? Does "{ }" work different from concat? 21:33
TimToady it's more like ~ do { * } ~
plobsing ah 21:34
TimToady you could use ~ * ~ though
masak 'night, #perl6
21:34 masak left
TimToady perl6: my $closure = "foo " ~ * ~ " bar"; say $closure(42) 21:34
p6eval pugs: OUTPUT«*** Cannot cast from VStr "foo Inf bar" to Pugs.AST.Types.VCode (VCode)␤ at /tmp/1EMrRKdXdJ line 1, column 40 - line 2, column 1␤»
..rakudo a55346, niecza v8-64-ga39a953: OUTPUT«foo 42 bar␤»
21:37 bluescreen10 left 21:48 lue joined
lue hello planet o/ 21:48
TimToady I don't think we've cleared our orbit yet 21:49
21:52 bluescreen10 joined
sorear good * #perl6 22:03
im2ee Could I make something with TCP connection in perl6? :)
Using sockets e.g.? 22:04
22:04 wamba left
sorear flussence: mock up some code for me and I'll make it run 22:04
22:09 frhodes joined
flussence (ooc: are "return;" and falling off the end of a block treated differently by CONTROL?) 22:10
plobsing im2ee: you could. google suggests looking at 'IO::Socket::INET'
im2ee plobsing, p6 has it? 22:11
flussence (oh, maybe I should read more than the latter half of a sentence.)
(answer: yes)
sorear flussence: in niecza, they are differnet 22:12
flussence I wonder what rakudo does...
sorear rakudo doesn't have CONTROL, iirc.
im2ee plobsing, I use IO::Socket::INET in perl 5. :)
plobsing rakudo star includes an LWP::Simple implementation, so I'd assume it would have that 22:13
flussence perl6: sub a($b); { return 1 if $b; CONTROL { say 'control'; return 0 } }; say a(True); say a(False);
p6eval rakudo a55346: OUTPUT«Could not find sub &a␤ in main program body at line 22:/tmp/61YvleS8rw␤»
..niecza v8-64-ga39a953: OUTPUT«===SORRY!===␤␤Malformed block at /tmp/35C74XzD0M line 1:␤------> sub a($b)⏏; { return 1 if $b; CONTROL { say 'contr␤␤Parse failed␤␤»
..pugs: OUTPUT«*** ␤ Unexpected ";"␤ expecting bare trait, trait or block␤ at /tmp/xV7sV5rhsZ line 1, column 10␤»
sorear Rakudo has an IO::Socket::INET, but it's horrendously feature-poor
plobsing im2ee: I would expect some slight API differences
flussence whoops
perl6: sub a($b) { return 1 if $b; CONTROL { say 'control'; return 0 } }; say a(True); say a(False); 22:14
p6eval rakudo a55346: OUTPUT«(timeout)control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤control␤…
..niecza v8-64-ga39a953: OUTPUT«1␤Nil␤»
..pugs: OUTPUT«1␤␤»
flussence niecza++, looks good to me
well, good-er than an infinite loop 22:15
sorear flussence: niecza generates a return handler inside &a which is slightly smaller than the entire block, so it takes precednce over the CONTROL, which does scope over the entire block
flussence (control exceptions are weird, I'm going to pretend they're not there for today.) 22:17
sorear lichtkind: slide 53 is slighty out of date 22:20
perl6: my @cities = <roma milano turino>; say @cities 22:21
nom: my @cities = <roma milano turino>; say @cities
p6eval pugs, rakudo a55346: OUTPUT«romamilanoturino␤»
..niecza v8-64-ga39a953: OUTPUT«roma milano turino␤»
nom ebd495: OUTPUT«roma milano turino␤»
lichtkind sorear: you mean say @cities.perl; ?
sorear lichtkind: no, look at the say output.
lichtkind: page 53/158 of your slideset gives 'say' output without spaces. 22:22
lichtkind sorear: yes as checked with my version of nom
sorear: is what niecza does really the specced? 22:23
sorear: its now 49/154
flussence interesting... perl6.vim highlights "throws_ok" as a special function but it's not in any Test.pm6 I know of... 22:24
sorear lichtkind: nom and niecza give the same answer
rakudo master is lagging
lichtkind: I just redownloaded, still 158 pages. 22:25
22:25 jevin left
lichtkind yes the last version is not up 22:26
i could swear nom did too
22:26 jevin joined
lichtkind but i checked and you right 22:26
i changed it and will be up soon
thank you
sorear: you will come also to turino or frankfurt? 22:27
sorear I have no plans at this time. 22:29
lichtkind new version is up, sorear++ 22:30
sorear perl6: sub mysub(*@arr) { say @arr.elems }; mysub(1, (2, [3, 4]), 5); 22:31
p6eval pugs, rakudo a55346, niecza v8-64-ga39a953: OUTPUT«4␤»
22:31 frhodes left
sorear perl6: my $number = 0 .. *; say $number; # lichtkind? 22:34
p6eval pugs: OUTPUT«Stack space overflow: current size 8388608 bytes.␤Use `+RTS -Ksize -RTS' to increase it.␤»
..niecza v8-64-ga39a953: OUTPUT«0..Infinity␤»
..rakudo a55346: OUTPUT«␤»
sorear nom: my $number = 0 .. *; say $number; # lichtkind? 22:35
p6eval nom ebd495: OUTPUT«0..Inf␤»
sorear lichtkind: page 145/158, "With eval" is slightly misleading since it doesn't use eval(). "With interpolation" may be better
lichtkind sorear: you mean on slide 143 22:40
where its about hash slices?
sorear yes 22:42
lichtkind sorear: so its fixed now too, thanks 22:45
sorear \o/ All tests successful. 22:49
lichtkind yeah
dalek ecza: 4e8d22a | sorear++ | / (2 files):
Remove deleted test files, and make lives_ok et al not intercept warnings
22:51 frhodes joined 22:52 [Coke] left 22:54 [Coke] joined 23:05 donri left 23:06 jevin left 23:09 frhodes left, frhodes joined
lichtkind new version is up 23:11
with yet 5% more fun
23:14 im2ee left 23:21 jevin joined, jaldhar left
flussence I'll leave this in the hands of people who know what they're talking about - some ramblings on what to do with control exceptions: raw.github.com/gist/1179681/5e80f3...l-tests.p6 23:23
23:24 jferrero joined 23:25 Helios joined 23:27 wk left 23:29 araujo left 23:34 Psyche^ joined 23:35 Patterner left, Psyche^ is now known as Patterner
flussence ^Zzz 23:39
23:48 tokuhirom joined
lichtkind good night you brave ones 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!