Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2015-10-14

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 n0tjack g'night all, and thanks again
00:05 softmoth m: my $x = 3;  (&sin, &cos X $x :with({&^a($^b)})).perl.say; # I thought there was something like this :with adverb for Z and X?
00:05 camelia rakudo-moar f3b535: OUTPUT«5===SORRY!5=== Error while compiling /tmp/hnbnlZB1PK␤You can't adverb $x␤at /tmp/hnbnlZB1PK:1␤------> 3= 3;  (&sin, &cos X $x :with({&^a($^b)})7⏏5).perl.say; # I thought there was someth␤»
00:11 softmoth m: cross((&sin, &cos), 3, :with({&^a($^b)}))
00:11 camelia rakudo-moar f3b535: OUTPUT«WARNINGS:␤Useless use of "cross((&sin, &cos), 3, :with({&^a($^b)}))" in expression "cross((&sin, &cos), 3, :with({&^a($^b)}))" in sink context (line 1)␤»
00:11 softmoth m: cross((&sin, &cos), 3, :with({&^a($^b)})).perl.say
00:11 camelia rakudo-moar f3b535: OUTPUT«(0.141120008059867e0, -0.989992496600445e0).Seq␤»
00:14 hohoho joined #perl6
00:14 [Coke] m: .say for .reals
00:14 camelia rakudo-moar f3b535: OUTPUT«Method 'reals' not found for invocant of class 'Any'␤  in block <unit> at /tmp/IaG_qS1uZH:1␤␤»
00:15 [Coke] m: given 42 { .say for .reals }
00:15 camelia rakudo-moar f3b535: OUTPUT«Method 'reals' not found for invocant of class 'Int'␤  in block <unit> at /tmp/kwojiobzrB:1␤␤»
00:16 laouji joined #perl6
00:17 softmoth sub infix:<-$->(&a, $b) { &a($b) }; (&sin, &cos X[-$-] 3).say.perl
00:17 softmoth m: sub infix:<-$->(&a, $b) { &a($b) }; (&sin, &cos X[-$-] 3).say
00:17 camelia rakudo-moar f3b535: OUTPUT«(0.141120008059867 -0.989992496600445)␤»
00:35 kid51 This bug ticket, https://rt.perl.org/Ticket/Display.html?id=126349, lacks a meaningful Subject.  It just says '[BUG]'.  Can someone suggest a better Subject for it?  If so, I or someone with rt privileges can add that Subject.
00:36 tokuhirom joined #perl6
00:38 telex joined #perl6
00:43 laouji joined #perl6
00:45 CurtisOvidPoe joined #perl6
00:47 Kogurr joined #perl6
00:48 BenGoldberg m: sub infix:<of> { &^x($^y) }; say &cos of 3; say &sin, &cos, &tan Xof 3;
00:48 camelia rakudo-moar f3b535: OUTPUT«-0.989992496600445␤(0.141120008059867 -0.989992496600445 -0.142546543074278)␤»
00:54 n0tjack joined #perl6
00:55 n0tjack last annoying newb question of the day: is there a pointfree way of expressing map {'0123456789'.index($_)}, "string".comb?  ?
00:55 n0tjack it just feels like overkill to introduce a new block context just to invoke a single function with a single unbound param
00:55 ofzyh_ joined #perl6
00:57 [Coke] kid51: perhaps you can ask the person who opened the ticket.
00:58 sQuEE joined #perl6
00:58 adhoc joined #perl6
00:58 gfldex kid51: "substitution of Array of Str fails to provide backreferences" and i can understand why he left the subject out :)
00:58 gfldex m: my @strs = ("11 22 33"); for @strs { $_ ~~ s/(\S+)\s(\S+)\s(\S+)/$0-$1-$2/; }; say @strs;
00:58 camelia rakudo-moar f3b535: OUTPUT«Use of Nil in string context  in code  at /tmp/pq2S60RUgD:1␤Use of Nil in string context  in code  at /tmp/pq2S60RUgD:1␤Use of Nil in string context  in code  at /tmp/pq2S60RUgD:1␤[--]␤»
00:58 [Coke] updating..
00:59 gfldex m: my @strs = ("11 22 33"); for @strs { $_ ~~ s/\s(\S+)\s(\S+)/$0-$1-$2/; }; say @strs;
00:59 camelia rakudo-moar f3b535: OUTPUT«Use of Nil in string context  in code  at /tmp/NG8WX_iNEp:1␤Use of Nil in string context  in code  at /tmp/NG8WX_iNEp:1␤Use of Nil in string context  in code  at /tmp/NG8WX_iNEp:1␤[11--]␤»
00:59 gfldex m: my @strs = ("11 22 33"); for @strs { say $_ ~~ s/\s(\S+)\s(\S+)/$0-$1-$2/; }; say @strs;
00:59 camelia rakudo-moar f3b535: OUTPUT«Use of Nil in string context  in code  at /tmp/MoBfYeEcUI:1␤Use of Nil in string context  in code  at /tmp/MoBfYeEcUI:1␤Use of Nil in string context  in code  at /tmp/MoBfYeEcUI:1␤「 22 33」␤ 0 => 「22」␤ 1 => 「33」␤[11--]␤»
00:59 BenGoldberg j: my @strs = ("11 22 33"); for @strs { $_ ~~ s/\s(\S+)\s(\S+)/$0-$1-$2/; }; say @strs;
00:59 ShimmerFairy gfldex: if that's what it can be called, I could swear there's an existing ticket about s/// in a loop :)
00:59 camelia rakudo-jvm f3b535: OUTPUT«Use of Nil in string context  in code  at /tmp/UtguVw6Bg0:1␤Use of Nil in string context  in code  at /tmp/UtguVw6Bg0:1␤Use of Nil in string context  in code  at /tmp/UtguVw6Bg0:1␤[11--]␤»
00:59 BenGoldberg n: my @strs = ("11 22 33"); for @strs { $_ ~~ s/\s(\S+)\s(\S+)/$0-$1-$2/; }; say @strs;
00:59 camelia niecza v24-109-g48a8de3: OUTPUT«sh: mono: command not found␤»
01:00 [Coke] updated.
01:02 timotimo n0tjack: time to build something like python's "itemgetter" or "methodcaller"? :P
01:02 gfldex #118705 is dupe of #123005 is dupe of #123005
01:02 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=118705
01:02 gfldex it's double duped twice! That's to much for me. Good night.
01:03 n0tjack timotimo: I feel there's a whatever-star solution lurking, that I just can't see
01:03 n0tjack maybe with .assuming?
01:04 sufrostico joined #perl6
01:04 timotimo putting a whatever star into the method list will not cause whatever currying
01:04 timotimo m: (ord *)('a')
01:04 camelia rakudo-moar f3b535: OUTPUT«5===SORRY!5=== Error while compiling /tmp/T6RywmNXwk␤Calling ord(Whatever) will never work with declared signature (Cool $s)␤at /tmp/T6RywmNXwk:1␤------> 3(7⏏5ord *)('a')␤»
01:04 timotimo right
01:05 timotimo not in that case, either
01:09 BenGoldberg m {ord $^x}('a')
01:09 BenGoldberg m: {ord $^x}('a').say
01:09 camelia rakudo-moar f3b535: OUTPUT«97␤»
01:09 n0tjack but it can't be order
01:09 n0tjack I mean ord
01:09 n0tjack it's gotta be a => 0, b => 1, etc
01:10 timotimo well, you can subtract ord('0') from the values
01:10 n0tjack and ord('a') - ord('0') won't work, because in the general case it won't be ASCII
01:10 n0tjack ha, beat me by a millisecond
01:10 timotimo you'll still have to check for out-of-bounds
01:11 n0tjack yeah, but I have plans for that
01:12 BenGoldberg m: my %lookup = ('0'..'9') X (0..9); my @whatever = %lookup{ "1234".comb }; @whatever.say;
01:12 camelia rakudo-moar f3b535: OUTPUT«[(Any) (Any) (Any) (Any)]␤»
01:13 BenGoldberg m: my %lookup = ('0'..'9') X (0..9); %lookup.say
01:13 camelia rakudo-moar f3b535: OUTPUT«0 0 => (0 1), 0 2 => (0 3), 0 4 => (0 5), 0 6 => (0 7), 0 8 => (0 9), 1 0 => (1 1), 1 2 => (1 3), 1 4 => (1 5), 1 6 => (1 7), 1 8 => (1 9), 2 0 => (2 1), 2 2 => (2 3), 2 4 => (2 5), 2 6 => (2 7), 2 8 => (2 9), 3 0 => (3 1), 3 2 => (3 3), 3 4 => (3 5), 3 6 …»
01:13 timotimo it's not getting any better :D
01:13 * BenGoldberg headscratches
01:13 timotimo X is a *bit* much :D
01:13 n0tjack X is cartesian, no?
01:13 timotimo yeah
01:13 BenGoldberg m: my %lookup; %lookup{'0'..'9'} = 0..9; %lookup.say
01:13 camelia rakudo-moar f3b535: OUTPUT«0 => 0, 1 => 1, 2 => 2, 3 => 3, 4 => 4, 5 => 5, 6 => 6, 7 => 7, 8 => 8, 9 => 9␤»
01:14 BenGoldberg m: my %lookup; %lookup{'0'..'9'} = 0..9;  my @whatever = %lookup{ "1234".comb }; @whatever.say;
01:14 camelia rakudo-moar f3b535: OUTPUT«[1 2 3 4]␤»
01:14 n0tjack BenGoldberg: Yeah, I did that too, but it forces me to create intermediate vars.
01:14 n0tjack I'm trying to stay as functional, pointfree, and clean as I can
01:15 n0tjack I mean, I can achieve my ends, but I'm trying to make it hard for others to maintain my code ;)
01:16 n0tjack If you want to see the kind of monsters I'm inflicting on the world: https://gist.github.com/ano​nymous/483ece62552981cb47dc
01:16 BenGoldberg m: my %lookup = map { $_ => $_ } 0..9;  my @whatever = %lookup{ "1234".comb }; @whatever.say;
01:16 camelia rakudo-moar f3b535: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ok5PJnAIZC␤Missing comma after block argument to map␤at /tmp/ok5PJnAIZC:1␤------> 3my %lookup = map { $_ => $_ }7⏏5 0..9;  my @whatever = %lookup{ "1234".c␤    expecting any of:␤        infix␤…»
01:16 BenGoldberg m: my %lookup = map { $_ => $_ }, 0..9;  my @whatever = %lookup{ "1234".comb }; @whatever.say;
01:16 camelia rakudo-moar f3b535: OUTPUT«[1 2 3 4]␤»
01:16 BenGoldberg m: my @whatever = ((map { $_ => $_ }, 0..9).hash){ "1234".comb }; @whatever.say;
01:16 camelia rakudo-moar f3b535: OUTPUT«[1 2 3 4]␤»
01:21 BenGoldberg n0tjack, What about doing: $/.make(  :10[$/.comb]  );
01:22 BenGoldberg On the one hand, it's not "doing it the hard way", but on the other, it restricts to base ten integers, just like your present grammar.
01:22 n0tjack BenGoldberg:  That does work, but I need a more general solution for other situations (also, for this particular grammar, I'm trying to "do the work myself", build up the values from fundamental principles).
01:24 n0tjack check out line 48, for example
01:34 vendethiel joined #perl6
01:37 softmoth joined #perl6
01:45 tokuhirom joined #perl6
01:45 rhr joined #perl6
01:49 adu joined #perl6
01:59 llfourn joined #perl6
02:05 uruwi joined #perl6
02:05 uruwi Ho!
02:06 awwaiid How do I get a module of mine listed in panda? Do I need to have the master branch be the stable branch?
02:07 uruwi I installed Rakudo* 2015.09, but can't find Panda, which is claimed to be packaged with the distribution, anywhere.
02:07 garu ilmari: follow up on the HTTP::UserAgent issue. According to tokuhirom, current version expects a \r\n, not just \n. Since RFC2616 also says so, I wonder if it should be implemented at all
02:07 uruwi Is there any kind of problem?
02:11 dayangkun joined #perl6
02:18 uruwi Never mind, found it!
02:19 noganex_ joined #perl6
02:19 awwaiid Yeah. me too.
02:25 colomon joined #perl6
02:26 dalek ecosystem: 7f59a4f | (Brock Wilcox)++ | META.list:
02:26 dalek ecosystem: Add LREP project
02:26 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/7f59a4f861
02:26 dalek ecosystem: a775816 | (Zoffix Znet)++ | META.list:
02:26 dalek ecosystem: Merge pull request #77 from awwaiid/patch-1
02:26 dalek ecosystem:
02:26 dalek ecosystem: Add LREP project https://github.com/awwaiid/p6-lrep/
02:26 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/a7758169e0
02:26 awwaiid In case anyone else doesn't know and wants to join the fun: to add a module to the ecosystem (ie panda) do a PR against https://github.com/perl6/ecosystem editing META.list
02:35 dalek ecosystem: 889c8e5 | (Zoffix Znet)++ | README.md:
02:35 dalek ecosystem: Suggest adding viewable URLs when adding modules to eco
02:35 dalek ecosystem:
02:35 dalek ecosystem: For ease of viewing during announcements. Suggested by Issue #76
02:35 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/889c8e5594
02:37 Zoffix awwaiid, you'll also need to run `panda update` for it to refresh its modules list, if you want your module to be installable
02:37 Zoffix (on your box)
02:41 dalek modules.perl6.org: e5ad02f | (Zoffix Znet)++ | web/build-project-list.pl:
02:41 dalek modules.perl6.org: Ensure sortable table "arrows" pics are copied to output dir
02:41 dalek modules.perl6.org: review: https://github.com/perl6/modul​es.perl6.org/commit/e5ad02f075
02:43 colomon joined #perl6
02:45 dalek specs: 8fb3cd9 | (Nova Patch)++ | / (13 files):
02:45 dalek specs: various minor casing/spacing/grammar/syntax fixes
02:45 dalek specs:
02:45 dalek specs: ◦ unicode                    → Unicode [except unicode pragma]
02:45 dalek specs: ◦ Perl6                      → Perl 6
02:45 dalek specs: ◦ Perl5                      → Perl 5 [except Perl5ish]
02:45 dalek specs: ◦ a                          → an [before vowel sounds]
02:45 dalek specs: ◦ c<…>                       → C<…>
02:45 dalek specs: ◦ $*ARGS                     → @*ARGS
02:45 dalek specs: ◦ :p5                        → :P5
02:45 dalek specs: ◦ Unicode database           → Unicode Character Database [http://unicode.org/glossary/​#unicode_character_database]
02:45 dalek specs: ◦ unicode normalisation form → Unicode normalization form [http://unicode.org/glossary/#normalization_form]
02:45 dalek specs:
02:45 dalek specs: also minor punctuation changes on some of the lines modified
02:45 dalek specs: review: https://github.com/perl6/specs/commit/8fb3cd9b33
02:46 timotimo .u
02:46 yoleaux Search for a Unicode character by codepoint, name, or raw character
02:46 timotimo .u l 6
02:46 yoleaux U+1974 TAI LE LETTER TONE-6 [Lo] (ᥴ)
02:46 yoleaux U+16F97 MIAO LETTER TONE-6 [Lm] (𖾗)
02:46 timotimo damn, not what i meant
02:49 aborazmeh joined #perl6
02:49 aborazmeh joined #perl6
02:50 ShimmerFairy timotimo: what did you mean, ooc?
02:50 ShimmerFairy m: say "◦".uniname
02:50 camelia rakudo-moar f3b535: OUTPUT«WHITE BULLET␤»
02:51 timotimo m: say "l 6".uniname
02:51 camelia rakudo-moar f3b535: OUTPUT«LATIN SMALL LETTER L␤»
02:51 timotimo m: say " ".uniname
02:51 camelia rakudo-moar f3b535: OUTPUT«SPACE␤»
02:52 timotimo that's not the right space to go between Perl and 6!
02:52 timotimo m: say " ".uniname
02:52 camelia rakudo-moar f3b535: OUTPUT«NARROW NO-BREAK SPACE␤»
02:52 timotimo ^- has to be this!
02:52 uruwi Just a question...
02:52 ShimmerFairy timotimo: that's the space I use, not NBSP :)
02:52 timotimo it's a NNBSP, though
02:52 uruwi how do I make perl6 recognize an internally used module?
02:53 timotimo uruwi: can you be a little more specific?
02:53 ShimmerFairy timotimo: I always saw NBSP in documents (a purple-ish underline on the space in emacs, which AFAIK is just for NBSP. maybe NNBSP is counted too)
02:53 uruwi Is it possible to use a module internally?
02:53 uruwi (within a project?)
02:54 Zoffix uruwi, sure
02:54 ShimmerFairy uruwi: yes it is. I've got quite a few projects where I may have a test file at the top level, module files in lib/, and I run the script with    perl6 -Ilib test.p6
02:54 timotimo you may have to "use lib" for things
02:54 timotimo or -I, yeah
02:54 Zoffix uruwi, you should be able to just `use` it if the paths are right. If they aren't. Just `use lib 'foobar';` or add them with -I
02:54 ShimmerFairy uruwi: you could also put  use lib "lib/";  in such a test script, as has been mentioned a couple times already :)
02:54 Zoffix Answer all the things \o/
02:55 * Zoffix goes to bed &
02:56 dalek doc: a8437e9 | (Tokuhiro Matsuno)++ | doc/Language/5to6-perlvar.pod:
02:56 dalek doc: $*PROGRAM_NAME was renamed to $*PROGRAM-NAME.
02:56 dalek doc: review: https://github.com/perl6/doc/commit/a8437e9b74
02:56 dalek doc: d4c4bfc | (Zoffix Znet)++ | doc/Language/5to6-perlvar.pod:
02:56 dalek doc: Merge pull request #161 from tokuhirom/program_name
02:56 dalek doc:
02:56 dalek doc: $*PROGRAM_NAME was renamed to $*PROGRAM-NAME.
02:56 dalek doc: review: https://github.com/perl6/doc/commit/d4c4bfcafa
02:57 uruwi So if I have a module with the source at lib/Terminal/Basic.pm6 then what would I type?
02:57 Zoffix uruwi, use lib 'lib'; or perl6 -Ilib
02:58 ShimmerFairy you'd use that as    use Terminal::Basic    , so long as you've told Perl 6 to look in that lib/ directory.
02:59 uruwi It doesn't work in the REPL, does it?
03:01 ShimmerFairy uruwi: I'm not sure, but you should be able to type  'use lib "lib/"' (replacing that string with the correct relative path to lib/), and then  use Terminal::Basic on the next line
03:01 ShimmerFairy uruwi: you should also be able to start the REPL as  perl6 -Ilib  instead of the 'use lib' statement
03:02 uruwi Thanks ShimmerFairy
03:12 uruwi Any different for source files?
03:18 atweiden joined #perl6
03:25 softmoth joined #perl6
03:43 tokuhirom joined #perl6
03:47 atweiden i'm testing this https://github.com/atweiden/nightscape/​blob/develop/t/grammar/01-header.t#L99
03:47 atweiden it passes in isolation here: https://github.com/atweiden/nightscape/​blob/develop/t/grammar/01-header.t#L118
03:47 atweiden and here: http://ix.io/lpw
03:47 atweiden but the original test fails
03:47 atweiden is this a rakudobug?
03:47 BenGoldberg joined #perl6
03:51 atweiden (running latest nom)
04:01 uruwi Is there any way to get a keypress without waiting for the user to press [Enter]?
04:01 uruwi Ideally I'd use the NCurses module but at this time it doesn't work with Windows.
04:17 WizJin joined #perl6
04:23 atweiden oops, figured it out
04:23 dalek ecosystem: b42c954 | lestrrat++ | META.list:
04:23 dalek ecosystem: Add Apache::LogFormat
04:23 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/b42c9540bc
04:23 dalek ecosystem: 2c8368d | (Tokuhiro Matsuno)++ | META.list:
04:23 dalek ecosystem: Merge pull request #78 from lestrrat/patch-1
04:23 dalek ecosystem:
04:23 dalek ecosystem: Add Apache::LogFormat
04:23 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/2c8368d6b5
04:34 WizJin joined #perl6
04:39 laouji joined #perl6
04:57 uruwi another question: where does perl6 search for dlls on Windows?
04:57 moritz in PATH I believe
05:02 FROGGS[mobile] joined #perl6
05:03 FROGGS[mobile] dlls are searched in CWD too I believe
05:07 uruwi I'm trying to get the NCurses module to work on Windows.
05:07 FROGGS[mobile] cool
05:07 uruwi What does "No appropriate parametric role variant available for 'Native'" mean?
05:07 uruwi I tried to search for it, but I got confused
05:09 FROGGS[mobile] hmmm
05:10 uruwi Does that mean that there's no function found with the appropriate signature in the dll?
05:10 FROGGS[mobile] m: use NativeCall; sub foo is native(<foo bar>) { *}
05:10 camelia rakudo-moar f3b535: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Ji4wHTAxFh␤No appropriate parametric role variant available for 'Native'␤at /tmp/Ji4wHTAxFh:1␤»
05:10 FROGGS[mobile] ha!
05:11 FROGGS[mobile] uruwi: pass a string to the is native trait, not a list or something else
05:12 uruwi Ah, it was because $NCURSES_LOC didn't exist.
05:14 softmoth joined #perl6
05:14 uruwi Perl still can't locate the dll though
05:14 FROGGS[mobile] ohh
05:14 uruwi even if I add the directory to the WINDOWS %Path% variable
05:14 FROGGS[mobile] yeah, traits run at compile time, which can be confusing
05:16 FROGGS[mobile] do you add it to PATH before starting your script?
05:17 uruwi Yes
05:17 uruwi Not every time, though
05:18 FROGGS[mobile] setting it in the script it too late though
05:20 FROGGS[mobile] but you can also provide an absolut path to is native, without setting Path at all I think
05:22 uruwi Still no luck.
05:24 uruwi Does it matter if I use a version from Cygwin?
05:30 uruwi Tried to use a Windows binary and still doesn't work.
05:31 uruwi oh, wrong file name
05:32 uruwi unfortunately Murphy's law strikes back "Error opening terminal: xterm."
05:35 uruwi This much work just to be able to read keypresses without pressing Enter.
05:36 uruwi And even less luck on cmd
05:38 uruwi On a break from all this madness, I've abandoned this project long long ago https://github.com/bluebear94/sun1p1
05:45 tokuhirom joined #perl6
05:45 bjz joined #perl6
05:47 jkva joined #perl6
05:50 uruwi It's a problem with initscr()
05:52 rindolf joined #perl6
05:53 uruwi It seems I'll have to link a Cygwin-aware DLL, but I can't find libncursesw6.dll anywhere
06:01 kjones_ joined #perl6
06:08 bjz joined #perl6
06:08 kmel joined #perl6
06:10 kmel m: say 'Hello'
06:10 camelia rakudo-moar f3b535: OUTPUT«Hello␤»
06:12 uruwi say 99 ** 99
06:14 uruwi m: say 99 ** 88
06:14 camelia rakudo-moar f3b535: OUTPUT«4129496711338884686279985453869161796875​23572093437777826575141669857000764249874724781​22770530747238801087245466136620792550517990555​186110294214867126284223076111559302271201␤»
06:14 uruwi m: say 99 ** 99
06:14 camelia rakudo-moar f3b535: OUTPUT«36972963764972677265718790562880544​059566876428174110243025997242355257045527​752342141065001012823272794097888954832654​011942999676949435945162157019364401441807​1060667659301384999779999159200499899␤»
06:15 TEttinger m: say 99 ** 99 ** 99
06:15 camelia rakudo-moar f3b535: OUTPUT«0␤»
06:16 TEttinger m: say 99 ** 99 * 99
06:16 camelia rakudo-moar f3b535: OUTPUT«36603234127322950493061602657251738​618971207663892369140595737269931704475072​474818719654351002695040066156910065284327​471823569680179941585710535449170757427389​035006098270837114978219916760849490001␤»
06:16 TEttinger m: say 99 * 99 ** 99
06:16 camelia rakudo-moar f3b535: OUTPUT«36603234127322950493061602657251738​618971207663892369140595737269931704475072​474818719654351002695040066156910065284327​471823569680179941585710535449170757427389​035006098270837114978219916760849490001␤»
06:16 TEttinger interesting
06:16 TEttinger so ** really does have the right priority for the exponentiation operator
06:18 FROGGS[mobile]2 joined #perl6
06:18 ShimmerFairy .tell jnthn just taking a quick glance at NQP's regex stuff, I notice case-insensitive stuff uses uc and lc variants. Now that we have nqp::fc, would it be beneficial to have regexes use fc for :i and :ii ? Or do those modifiers explicitly require just uc/lc checking in spec somewhere (nothing suggesting such in S05)?
06:18 yoleaux ShimmerFairy: I'll pass your message to jnthn.
06:20 bjz_ joined #perl6
06:25 Hotkeys I like how sequence generation looks when multiplying
06:25 Hotkeys m: my @test = 1, 2, 3, * * * * * ... *; say @test[^10];
06:25 camelia rakudo-moar f3b535: OUTPUT«(1 2 3 6 36 648 139968 3265173504 296148833645101056 135345882205792807436868315512832)␤»
06:26 uruwi So that was the product of the 3 previous terms right?
06:26 Hotkeys yeah
06:26 uruwi m: my @test = 1, 1, 2, * * * * * ... *; say @test[^10];
06:26 camelia rakudo-moar f3b535: OUTPUT«(1 1 2 2 4 16 128 8192 16777216 17592186044416)␤»
06:26 Hotkeys just a line of asterisks
06:26 ShimmerFairy Yeah, I think it's delightfully cryptic. I can't help but think it could also serve as a good tool for teaching the difference between "term" and "operator", at least for parser-writing things :)
06:26 Hotkeys beauty
06:27 uruwi m: my @test = 1, 2, 3 * ** * ** * ... *; say @test[^10];
06:27 camelia rakudo-moar f3b535: ( no output )
06:27 firstdayonthejob joined #perl6
06:27 uruwi m: my @test = 1, 2, 3 * ** * ** * ... *; say @test[^5];
06:27 camelia rakudo-moar f3b535: ( no output )
06:27 uruwi m: my @test = 1, 2, 3, * ** * ** * ... *; say @test[^10];
06:27 camelia rakudo-moar f3b535: OUTPUT«(1 2 3 1 8 3 1 512 3 1)␤»
06:27 Hotkeys that's neat
06:28 ShimmerFairy m: my @test = 1, 1, 2, {$^a * $^b * $^c} ... *; say @test[^10];    # the fun-ruining clearer version of that line of asterisks, fwiw
06:28 camelia rakudo-moar f3b535: OUTPUT«(1 1 2 2 4 16 128 8192 16777216 17592186044416)␤»
06:28 Hotkeys m: my @test = 1, 2, 3, * ** * ** * ... *; say @test[^20];
06:28 camelia rakudo-moar f3b535: OUTPUT«(1 2 3 1 8 3 1 512 3 1 134217728 3 1 2417851639229258349412352 3 1 1413477651822707463666638000594334812​6619871175004951664972849610340958208 3 1 28240139587082174969491088422046278633​51353911851577524683401930862693830361​1984999058739209952299969708978654…»
06:28 Hotkeys interesting that it goes to 3 1 every time
06:28 uruwi m: my @test = 2, 2, 2, * ** * ** * ... *; say @test[^10];
06:28 camelia rakudo-moar f3b535: OUTPUT«(2 2 2 16 20035299304068464649790723515602557504478254755697​51419265016973710894059556311453089506130880933348​10103823434290726318182294938211881266886950636476​15470291650418719163515879663472194429309279820843​0910485599057015931895963952486337236720300291…»
06:28 uruwi m: my @test = 2, 3, 4, * ** * ** * ... *; say @test[^10];
06:28 camelia rakudo-moar f3b535: OUTPUT«(2 3 4 2417851639229258349412352 1 0 2417851639229258349412352 1 0 2417851639229258349412352)␤»
06:28 Hotkeys I never knew perl 5, only started perl 6 a few days ago
06:28 Hotkeys this is fun
06:28 uruwi I broke Camelia
06:29 Hotkeys m: 3 ** 4 ** 99999999999999999999999999
06:29 camelia rakudo-moar f3b535: OUTPUT«WARNINGS:␤Useless use of "**" in expression "3 ** 4 **" in sink context (line 1)␤»
06:29 uruwi Still trying to get NCurses module to work with Windows
06:29 kmel joined #perl6
06:30 uruwi m: my @test = 2, 2, 2, * ** * ** * ... *; say @test[^10].map(&log);
06:30 camelia rakudo-moar f3b535: OUTPUT«Attempt to divide Inf by zero using /␤  in block <unit> at /tmp/v2TkLbcMfT:1␤␤Actually thrown at:␤  in block <unit> at /tmp/v2TkLbcMfT:1␤␤»
06:31 uruwi m: (1, 2).map(&log)
06:31 camelia rakudo-moar f3b535: ( no output )
06:31 uruwi m: say (1, 2).map(&log)
06:31 camelia rakudo-moar f3b535: OUTPUT«(0)␤»
06:31 uruwi m: say (1, 2, 3).map(&log)
06:31 camelia rakudo-moar f3b535: OUTPUT«(0 1.09861228866811)␤»
06:31 uruwi m: say (1, 2, 3, 4, 5).map(&log)
06:31 camelia rakudo-moar f3b535: OUTPUT«(0 0.792481250360578 1.6094379124341)␤»
06:32 uruwi say (1).map(&log)
06:32 uruwi m: say (1).map(&log)
06:32 camelia rakudo-moar f3b535: OUTPUT«(0)␤»
06:32 uruwi m: my @test = 2, 2, 2, * ** * ** * ... *; say @test[^10].map(&length);
06:32 camelia rakudo-moar f3b535: OUTPUT«5===SORRY!5=== Error while compiling /tmp/XBTugJvACT␤Undeclared routine:␤    &length used at line 1. Did you mean '&elems', '&chars', '&graphs', '&codes'?␤␤»
06:32 uruwi m: my @test = 2, 2, 2, * ** * ** * ... *; say @test[^10].map(&chars);
06:32 camelia rakudo-moar f3b535: OUTPUT«(1 1 1 2 19729 1 1 19729 1 1)␤»
06:33 uruwi m: my @test = 1, 3, 5, * * * * * ... *; say @test[^10].map(&chars);
06:33 camelia rakudo-moar f3b535: OUTPUT«(1 1 1 2 3 5 8 15 27 49)␤»
06:33 Hotkeys also lazy geometric sequences are nice too
06:33 Hotkeys m: my @geo = 1, 2, 4 ...^ * > 10000; say @geo
06:33 camelia rakudo-moar f3b535: OUTPUT«[1 2 4 8 16 32 64 128 256 512 1024 2048 4096 8192]␤»
06:33 uruwi m: my @test = 1, 3, 5, * * * * * ... *; say @test[^80].map(&chars);
06:33 camelia rakudo-moar f3b535: OUTPUT«(timeout)»
06:34 uruwi m: my @test = 1, 3, 5, * * * * * ... *; say @test[^40].map(&chars);
06:34 camelia rakudo-moar f3b535: OUTPUT«(timeout)»
06:34 TimToady m: my @geo = 1, 2, 4 ... 10000; say @geo
06:34 camelia rakudo-moar f3b535: OUTPUT«[1 2 4 8 16 32 64 128 256 512 1024 2048 4096 8192]␤»
06:34 uruwi m: my @test = 1, 3, 5, * * * * * ... *; say @test[^20].map(&chars);
06:34 camelia rakudo-moar f3b535: OUTPUT«(1 1 1 2 3 5 8 15 27 49 90 164 302 555 1020 1875 3448 6342 11664 21453)␤»
06:34 Hotkeys TimToady, oh
06:34 TimToady intuited sequences don't have to match exactly any more
06:34 Hotkeys neat
06:35 TimToady m: my @geo = 1, -2, 4 ... 10000; say @geo
06:35 camelia rakudo-moar f3b535: OUTPUT«[1 -2 4 -8 16 -32 64 -128 256 -512 1024 -2048 4096 -8192]␤»
06:35 TimToady even if they change sign
06:35 Hotkeys cool
06:35 Hotkeys m: my @geo = -1, 2, -4 ... 10000; say @geo
06:35 camelia rakudo-moar f3b535: OUTPUT«[-1 2 -4 8 -16 32 -64 128 -256 512 -1024 2048 -4096 8192]␤»
06:35 uruwi m: my @what = 1, 6, 8, ... 10000; say @what
06:35 camelia rakudo-moar f3b535: OUTPUT«5===SORRY!5=== Error while compiling /tmp/CRSEYRiGhm␤Comma found before apparent series operator; please remove comma (or put parens␤    around the ... listop, or use 'fail' instead of ...)␤at /tmp/CRSEYRiGhm:1␤------> 3my @what = 1, 6, 8,7…»
06:36 uruwi m: my @what = 1, 6, 8 ... 10000; say @what
06:36 camelia rakudo-moar f3b535: OUTPUT«Unable to deduce arithmetic or geometric sequence from 1,6,8 (or did you really mean '..'?)␤  in block <unit> at /tmp/rF_mweXWcH:1␤␤»
06:36 uruwi m: my @what = 1, 4, 9 ... 10000; say @what
06:36 camelia rakudo-moar f3b535: OUTPUT«Unable to deduce arithmetic or geometric sequence from 1,4,9 (or did you really mean '..'?)␤  in block <unit> at /tmp/7sXfX4HhmV:1␤␤»
06:36 Hotkeys so it ignores sign for the upper limit?
06:36 uruwi m: my @what = 1, 4, 16 ... 10000; say @what
06:36 camelia rakudo-moar f3b535: OUTPUT«[1 4 16 64 256 1024 4096]␤»
06:36 TimToady yes, it takes absolute value for that
06:37 ShimmerFairy That error message suddenly makes me think of what use something like    1, 2, 3, fail 10000    would have :P
06:37 TimToady that's more to tell people what the ... was interpreted as than a serious suggestion
06:37 FROGGS joined #perl6
06:38 ShimmerFairy TimToady: I figured it's one of the less common intentions in that context, but it's still a funny expression to consider :)
06:38 Hotkeys so if you wanted to take into consideration the sign,
06:38 Hotkeys m: my @geo = -1, 2, -4, ...^ * > 10000; say @geo
06:38 camelia rakudo-moar f3b535: OUTPUT«5===SORRY!5=== Error while compiling /tmp/eLM7NZ5w_8␤Comma found before apparent series operator; please remove comma (or put parens␤    around the ... listop, or use 'fail' instead of ...)␤at /tmp/eLM7NZ5w_8:1␤------> 3my @geo = -1, 2, -4,7…»
06:38 Hotkeys m: my @geo = -1, 2, -4 ...^ * > 10000; say @geo
06:38 camelia rakudo-moar f3b535: OUTPUT«[-1 2 -4 8 -16 32 -64 128 -256 512 -1024 2048 -4096 8192 -16384]␤»
06:39 uruwi Anyone managed to build and run the NCurses module on Windows?
06:39 FROGGS uruwi: I've never tried
06:39 uruwi wb
06:39 FROGGS uruwi: is there a Perl 5 version that works on windows?
06:39 FROGGS maybe we can look what they do
06:42 uruwi Well, not ncurses per se, but it has what I want https://metacpan.org/pod/Games::Roguelike::Console
06:43 Hotkeys m: my @alphabet = 'a', 'b', ... 'z'; say @alphabet
06:43 camelia rakudo-moar f3b535: OUTPUT«5===SORRY!5=== Error while compiling /tmp/tG32An3I0Q␤Comma found before apparent series operator; please remove comma (or put parens␤    around the ... listop, or use 'fail' instead of ...)␤at /tmp/tG32An3I0Q:1␤------> 3my @alphabet = 'a', 'b'…»
06:43 Hotkeys m: my @alphabet = 'a', 'b' ... 'z'; say @alphabet
06:43 camelia rakudo-moar f3b535: OUTPUT«[a b c d e f g h i j k l m n o p q r s t u v w x y z]␤»
06:44 Hotkeys neat
06:44 uruwi m: my @hiragana = 'あ', 'い' ... 'ん'; say @hiragana
06:45 FROGGS uruwi: so, port that instead?
06:45 camelia rakudo-moar f3b535: OUTPUT«(timeout)»
06:45 Hotkeys lol
06:45 uruwi You know what, the framebuffer functionality would be useful too
06:46 ShimmerFairy m: say "あ" ~~ /<rangechar>/
06:46 camelia rakudo-moar f3b535: OUTPUT«Method 'rangechar' not found for invocant of class 'Cursor'␤  in block <unit> at /tmp/1R_5FaG3d_:1␤␤»
06:46 FROGGS grr, I can't even copy+paste these properly in my chat client
06:46 tokuhirom joined #perl6
06:46 ShimmerFairy I could swear that was a rule name...
06:46 uruwi m: my @alphabet = 'α', 'β' ... 'ω'; say @alphabet
06:46 camelia rakudo-moar f3b535: OUTPUT«[α β γ δ ε ζ η θ ι κ λ μ ν ξ ο π ρ σ τ υ φ χ ψ ω]␤»
06:48 TimToady m: say "あ"..."ん"
06:48 camelia rakudo-moar f3b535: OUTPUT«(あ ぃ い ぅ う ぇ え ぉ お か が き ぎ く ぐ け げ こ ご さ ざ し じ す ず せ ぜ そ ぞ た だ ち ぢ っ つ づ て で と ど な に ぬ ね の は ば ぱ ひ び ぴ ふ ぶ ぷ へ べ ぺ ほ ぼ ぽ ま み む め…»
06:49 [Sno] joined #perl6
06:52 ecocode joined #perl6
06:52 uruwi FROGGS I probably should have used a frame buffer from the start
06:54 uruwi Anyway, it's way too late at night for me to port this today.
06:54 uruwi I'll probably begin tomorrow.
06:54 FROGGS ++uruwi :o)
06:55 bjz joined #perl6
06:55 uruwi Yikes, what's been done for Win32...
06:57 kmel m: say 'ا'...'ي'
06:57 camelia rakudo-moar f3b535: OUTPUT«(ا ب ة ت ث ج ح خ د ذ ر ز س ش ص ض ط ظ ع غ ػ ؼ ؽ ؾ ؿ ـ ف ق ك ل م ن ه و ى ي)␤»
06:58 uruwi m: say [+] 'ا'...'ي'
06:58 camelia rakudo-moar f3b535: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '7⏏5ا' (indicated by 7⏏)␤  in block <unit> at /tmp/Ue8ms31u19:1␤␤Actually thrown at:␤  in block <unit> at /tmp/Ue8ms31u19:1␤␤»
06:58 uruwi m: say [~] 'ا'...'ي'
06:58 camelia rakudo-moar f3b535: OUTPUT«ابةتثجحخدذرزسشصضطظعغػؼؽؾؿـفقكلمنهوىي␤»
06:59 kmel i see additional letters is this normal?
06:59 uruwi say chr(0) .. chr(65535)
06:59 uruwi m: say chr(0) .. chr(65535)
06:59 camelia rakudo-moar f3b535: OUTPUT«"\x[0]".."\x{FFFF}"␤»
06:59 uruwi m: say chr(0) .. chr(255)
06:59 camelia rakudo-moar f3b535: OUTPUT«"\x[0]".."ÿ"␤»
06:59 uruwi say chr(7)
07:00 uruwi m: say chr(7)
07:00 camelia rakudo-moar f3b535: OUTPUT«␤»
07:00 uruwi m: for 0 .. ^100 {say chr(7);}
07:00 camelia rakudo-moar f3b535: OUTPUT«Range objects are not valid endpoints for Ranges␤  in block <unit> at /tmp/AQvN6cHeEK:1␤␤»
07:00 uruwi m: for 0 .. 100 {say chr(7);}
07:00 camelia rakudo-moar f3b535: OUTPUT«␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤​␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤…»
07:00 uruwi m: for 0 .. 100 {say "\a";}
07:00 camelia rakudo-moar f3b535: OUTPUT«␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤​␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤␤…»
07:00 Hotkeys oh
07:00 Hotkeys I wondered where the bels were coming from
07:01 moritz .u ␤
07:01 yoleaux U+2424 SYMBOL FOR NEWLINE [So] (␤)
07:01 moritz not a BEL
07:02 Hotkeys oh
07:02 softmoth joined #perl6
07:02 Hotkeys well it's making me bel
07:10 CIAvash joined #perl6
07:20 ShimmerFairy Hotkeys: one reason I disabled my system beep :)   (the main reason being that, using headphones as my speakers, there's never a time where I could hear system noises that won't also interrupt me listening to something)
07:20 ShimmerFairy moritz: I'm sure it's the 0x07 being printed between those NL symbols, not the symbols themselves
07:21 ShimmerFairy m: say "\a".uniname
07:21 camelia rakudo-moar f3b535: OUTPUT«BELL␤»
07:21 moritz ShimmerFairy: ah
07:22 laouji joined #perl6
07:22 abraxxa joined #perl6
07:24 RabidGravy joined #perl6
07:31 Hotkeys night all
07:34 leont joined #perl6
07:35 masak good morning, #perl6
07:38 ely-se joined #perl6
07:43 adhoc joined #perl6
07:48 RabidGravy joined #perl6
07:50 [ptc] o/ masak
07:50 eliasr joined #perl6
07:55 xinming joined #perl6
07:57 ][Sno][ joined #perl6
08:01 zakharyas joined #perl6
08:02 FROGGS joined #perl6
08:08 WizJin joined #perl6
08:12 darutoko joined #perl6
08:12 kmel m: 'good morning'.say
08:12 camelia rakudo-moar f3b535: OUTPUT«good morning␤»
08:15 labster joined #perl6
08:25 nine FROGGS: Perl 5 for Windows is http://strawberryperl.com/
08:26 FROGGS nine: hi, I know strawberry... why are you telling about this?
08:26 jkva_ joined #perl6
08:26 FROGGS I'm a windows + strawberry + activeperl user after all :o)
08:27 Timbus he probably misinterpreted this: <FROGGS> uruwi: is there a Perl 5 version that works on windows?
08:28 FROGGS ohh
08:28 FROGGS nine: this was about an ncurses lib for windows
08:28 nine FROGGS: < FROGGS> uruwi: is there a Perl 5 version that works on windows?
08:28 nine ok :)
08:28 FROGGS :o)
08:29 Hor|zon joined #perl6
08:29 virtualsue joined #perl6
08:29 nine Probably been skimming the backlog too fast. It's reeeeeally hard to keep up when one is afk for an evening.
08:31 FROGGS aye
08:32 ely-se joined #perl6
08:34 dakkar joined #perl6
08:36 sivoais joined #perl6
08:42 adhoc joined #perl6
08:46 sivoais joined #perl6
08:47 tokuhirom joined #perl6
08:50 luiz_lha joined #perl6
08:50 softmoth joined #perl6
08:51 davido_ joined #perl6
08:57 sivoais joined #perl6
08:59 Ulti how do you set the max threads for the thread pool?
09:00 Ulti (and does that include GC threads or are they lower level)
09:00 moritz I don't think the GC runs in a separate thread
09:01 Ulti oh really, I thought it runs in parallel itself when it gets called
09:01 moritz maybe I'm wrong though
09:02 moritz m: say $*SCHEDULER
09:02 camelia rakudo-moar f3b535: OUTPUT«ThreadPoolScheduler.new(initial_threads => 0, max_threads => 16, uncaught_handler => Callable)␤»
09:03 moritz Ulti: you can assign a new ThreadPoolScheduler to $*SCHEDULER, and give it a max_threads
09:03 Ulti okedoke thanks
09:04 Ulti though sort of wonder what happens if there are already threads when you do that o___O
09:04 jsnathan joined #perl6
09:04 FireFly joined #perl6
09:06 rurban joined #perl6
09:06 kmel what is the correct way of referencing perl 6: Perl 6 or Perl6 or perl 6 or perl6 ?
09:06 rarara joined #perl6
09:06 moritz kmel: Perl 6, preferably with a non-breaking space between Perl and 6 :-)
09:07 sivoais joined #perl6
09:07 kmel thanks moritz
09:08 Ulti perl6 would be the actual compiler you call specifically rather than the language
09:12 chenryn_ joined #perl6
09:12 kmel So Perl 6 is the language specification
09:12 kmel and perl6 is the compiler
09:12 diana_olhovik_ joined #perl6
09:12 ShimmerFairy I personally don't think it matters how you write it ("the actual compiler" depends on the implementation, happens to be "perl6" for rakudo). I also don't believe NBSP is anything work worrying over :)
09:13 masak kmel: yes. well, `perl6` is the name of the executable.
09:13 masak kmel: people refer to the compiler+runtime as "Rakudo"
09:13 brrt joined #perl6
09:13 chenryn_ how to use varibles in regex/token/rule?
09:13 masak chenryn_: just use them
09:14 masak m: my $s = "oo"; say so "foo" ~~ / f $s /
09:14 camelia rakudo-moar f3b535: OUTPUT«True␤»
09:14 kmel thanks
09:14 ely-se joined #perl6
09:14 masak m: my $s = /oo/; say so "foo" ~~ / f $s /
09:14 camelia rakudo-moar f3b535: OUTPUT«True␤»
09:14 masak m: my $s = /o+/; say so "foo" ~~ / f $s /
09:14 camelia rakudo-moar f3b535: OUTPUT«True␤»
09:14 kmel m: say 'thanks everyone :)'
09:14 camelia rakudo-moar f3b535: OUTPUT«thanks everyone :)␤»
09:16 RabidGravy Ulti: also for convenience most of the methods that start new threads such as Promise.start take an optional scheduler argument
09:16 espadrine_ joined #perl6
09:17 sivoais joined #perl6
09:18 kmel say 'hello world'; and 'hello world'.say yield the same result. Is it a valid example of Tim Toady? If I was to explain what Tim Toady is.
09:18 Ulti RabidGravy does .race() ?
09:19 RabidGravy I somewhat doubt it
09:20 baest kmel: I would say so
09:20 brrt good * #perl6, too
09:20 masak kmel: sure
09:21 kmel thanks baest and masak
09:21 masak kmel: another example: `for @a -> $elem { ... }` vs `@a.map: -> $elem { ... }`
09:21 virtualsue joined #perl6
09:22 moritz constructing a hash from keys values: my %h; %h{@keys} = @values;  or   my %h = @keys Z=> @values;
09:22 masak kmel: and you probably want to write it as TIMTOWTDI (There Is More Than One Way To Do It)
09:22 llfourn m: say &say; say Str.^find_method('say').package;
09:22 camelia rakudo-moar f3b535: OUTPUT«sub say (| is raw) { #`(Sub|60192304) ... }␤(Mu)␤»
09:23 llfourn though they are not exactly the same thing under the hood :)
09:23 kmel masak: ok
09:23 rarara time perl6 -e 'say "Ciao"' : user    0m0.156s ; time perl6 <<< 'say "Ciao"' : user    0m2.070s. Wow
09:23 masak kmel: though, of course, there's more than one way to write it :P
09:24 chenryn_ p6: say "key=value" ~~ /$<key>=\N+ '=' $<value>=\N+/;
09:24 camelia rakudo-moar f3b535: OUTPUT«「key=value」␤ key => 「key」␤ value => 「value」␤»
09:24 chenryn_ p6: my rule field { $<key>=\N+ '=' $<value>=\N+ }; say "key=value" ~~ &field;
09:24 camelia rakudo-moar f3b535: OUTPUT«Nil␤»
09:24 chenryn_ can't write named capture directly in rule?
09:24 kmel masak :)
09:25 DrForr_ rarara: Oddly I get the exact opposite. sys 0m0.053s without <<<, 0m0.008s with.
09:25 DrForr_ But I might be more up-to-date, maybe rebuild your rakudo?
09:25 rarara I will
09:27 sivoais joined #perl6
09:30 mattprelude joined #perl6
09:30 ShimmerFairy m: my regex field { $<key>=\N+ '=' $<value>=\N+ }; say "key=value" ~~ &field;
09:30 camelia rakudo-moar f3b535: OUTPUT«「key=value」␤ key => 「key」␤ value => 「value」␤»
09:30 RabidGravy It would be amusing to find the thread from clpm from about 1998 wherein someone asked "how do I find the length of a string" and it went on for a week with people posting ever more ludicruous ways of doing it
09:30 ShimmerFairy chenryn_: 'rule' considers whitespace significant; 'regex' is equivalent to the ~~ / ... / example you used
09:30 RabidGravy and then redo it as Perl 6
09:30 ShimmerFairy (there's also 'token' , which is like 'regex' but disables backtracking behavior)
09:31 tadzik joined #perl6
09:31 * llfourn just learnt something new about token
09:32 M-tadzik1 left #perl6
09:33 chenryn_ so, can't ~~ a rule but only regex?
09:34 ShimmerFairy chenryn_: that's not it, the problem is that you had whitespace in that rule, but there's no whitespace in the text you were matching
09:37 tokuhirom joined #perl6
09:37 sivoais joined #perl6
09:38 ShimmerFairy chenryn_: oh, forgot to mention that 'rule' also doesn't do any backtracking (just like 'token'), so the first \N+ would've captured that entire string and left nothing for the rest of it to match.
09:40 margeas joined #perl6
09:43 chenryn_ p6: my $field_split = ' ';my $value_split = '=';my regex field { $<key>=\N+ $value_split $<value>=\N+ };my regex kv { <field>+ % $field_split };say 'key=value k2=value2' ~~ &kv;
09:43 camelia rakudo-moar f3b535: OUTPUT«「key=value k2=value2」␤ field => 「key=value k2=value2」␤  key => 「key=value k2」␤  value => 「value2」␤»
09:43 chenryn_ oh, yes...
09:48 sivoais joined #perl6
09:48 sivoais joined #perl6
09:48 spider-mario joined #perl6
09:52 jsnathan joined #perl6
09:54 rarara Is it possible to tell rakudobrew to download all the packages and then to rakudobrew to build withoud downloading?
09:54 rarara github.com[0: 192.30.252.128]: errno=Network is unreachable
09:54 rarara this is what i get on the cluster
09:58 sivoais joined #perl6
09:58 rarara because, btw, a simple rsync -av .rakudobrew me-there@where does also not work
10:01 carlin colomon: any idea why Crypt::Bcrypt isn't listed on smoke.perl6.org ?
10:02 jnthn afternoon, #perl6
10:02 yoleaux 06:18Z <ShimmerFairy> jnthn: just taking a quick glance at NQP's regex stuff, I notice case-insensitive stuff uses uc and lc variants. Now that we have nqp::fc, would it be beneficial to have regexes use fc for :i and :ii ? Or do those modifiers explicitly require just uc/lc checking in spec somewhere (nothing suggesting such in S05)?
10:05 jnthn .tell ShimmerFairy The regex stuff doesn't use fc 'cus we only just implemented fc. :) We should use it instead of lc.
10:05 yoleaux jnthn: I'll pass your message to ShimmerFairy.
10:06 ShimmerFairy ding!
10:06 yoleaux 10:05Z <jnthn> ShimmerFairy: The regex stuff doesn't use fc 'cus we only just implemented fc. :) We should use it instead of lc.
10:06 ShimmerFairy jnthn: I of course figured it's because fc is new, I was just wondering if that was something to be done, or if there was perhaps a specific reason to not change it over :)
10:07 jnthn m: grammar G { token TOP { <digit> <alpha> } }; class A { method TOP((:$digit, :$alpha)) { say $digit; say $alpha; } }; G.parse('1a', actions => A)
10:07 camelia rakudo-moar f3b535: OUTPUT«「1」␤「a」␤»
10:08 jnthn .tell n0tjack about your unpacking $/ question in the backlog that I don't think anyone answered: grammar G { token TOP { <digit> <alpha> } }; class A { method TOP((:$digit, :$alpha)) { say $digit; say $alpha; } }; G.parse('1a', actions => A)
10:08 yoleaux jnthn: I'll pass your message to n0tjack.
10:08 sivoais joined #perl6
10:09 vytas joined #perl6
10:11 risou_awy joined #perl6
10:14 hohoho joined #perl6
10:19 sivoais joined #perl6
10:23 ely-se joined #perl6
10:23 brrt afternoon jnthn
10:25 jnthn o/ brrt
10:27 g5 joined #perl6
10:29 cygx joined #perl6
10:29 cygx o/
10:29 sivoais joined #perl6
10:30 cygx any thoughts on adding a :$match flag to Str.split and have it retunr strings by default in case of :all?
10:31 cygx I found it rather surprising to get back a list of mixed strings and Match objects...
10:36 vytas joined #perl6
10:37 lizmat good *, #perl6!
10:37 dalek rakudo/nom: 5271fc7 | lizmat++ | src/core/ (4 files):
10:37 dalek rakudo/nom: Use private methods for better optimizations
10:37 dalek rakudo/nom:
10:37 dalek rakudo/nom: Well, at some point in the future.
10:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5271fc7d69
10:38 softmoth joined #perl6
10:39 jnthn lizmat: I think we already do something with 'em :)
10:39 lizmat well, I tested it yesterday, didn't see a difference in performance though  :-(
10:39 sivoais joined #perl6
10:40 lizmat class A { method a() {}; method b() { self.a } }; for ^1000000 { A.b }    # versus:
10:40 lizmat class A { method !a() {}; method b() { self!a } }; for ^1000000 { A.b }
10:41 jnthn It'll be hard to see
10:41 jnthn I suspect in those cases everything gets inlined and it's a wash
10:41 jnthn And non-privates get dynamic optimization
10:42 lizmat ok, in any case, I thought it would be better that way :-)
10:42 tokuhirom joined #perl6
10:42 * lizmat starts backlogging
10:42 jnthn So while the static code generation is perhaps a bit better for privates (a lookup by name less), once spesh gets its hands on the code it'll all be the same. :)
10:42 dalek rakudo/nom: 9db48e9 | cygx++ | src/core/Str.pm:
10:42 dalek rakudo/nom: Add missing Slip.new(...) to Str.split(:all)
10:42 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9db48e945a
10:42 dalek rakudo/nom: b16110d | lizmat++ | src/core/Str.pm:
10:42 dalek rakudo/nom: Merge pull request #551 from cygx/patch-3
10:42 dalek rakudo/nom:
10:42 dalek rakudo/nom: Add missing Slip.new(...) to Str.split(:all)
10:42 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/b16110d293
10:44 cygx lizmat++
10:49 |Tux| test             50000    36.887    36.778
10:49 |Tux| test-t           50000    37.560    37.452
10:49 sivoais joined #perl6
10:56 nine Regardless of performance considerations, making these methods private sounds like a good idea.
10:56 jnthn Yes, agree
10:56 diana_olhovik joined #perl6
10:59 sivoais joined #perl6
11:03 ely-se joined #perl6
11:04 cvxcv joined #perl6
11:11 |Tux| lizmat, your CSV6 commit causes a syntax error (you do not even need to run the test suite)
11:11 |Tux| ===SORRY!=== Error while compiling /pro/3gl/CPAN/Text-CSV6/lib/Text/CSV.pm
11:11 |Tux| Missing block
11:11 |Tux| at /pro/3gl/CPAN/Text-CSV6/lib/Text/CSV.pm:1680
11:11 lizmat grrr  :-)
11:15 dalek ecosystem: c7f50ff | azawawi++ | META.list:
11:15 dalek ecosystem: Add Selenium::WebDriver
11:15 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/c7f50ffa2b
11:16 larion joined #perl6
11:16 lizmat m: class A { method a(@a) { @a.grep(self) } }; say A.a(A xx 5)  # [Tux]  not sure why that fails in CSV
11:16 camelia rakudo-moar 5271fc: OUTPUT«((A) (A) (A) (A) (A))␤»
11:18 lizmat [Tux]: better ?
11:19 |Tux| yep
11:20 lizmat I do so many recompiles, and needing to reinstall everything... I hope we get a better fix for that soon
11:21 * jnthn too :)
11:21 jnthn Hairy problem is hairy
11:25 azawawi joined #perl6
11:27 ely-se Hairy Potter
11:27 azawawi hi
11:27 ely-se hi azawawi!
11:28 azawawi is there an http client library in Perl 6 that .post actually works?
11:30 RabidGravy HTTP::UserAgent will happily do a POST
11:31 RabidGravy though this may depend on what you want to actually post, lots of work going in at the moment
11:31 RabidGravy if it doesn't work file an issue ;-)
11:32 RabidGravy anyway, getting cabin fever - off out to the pub for a bit ;-)
11:32 diana_olhovik_ joined #perl6
11:32 ggoebel2 joined #perl6
11:33 hohoho joined #perl6
11:33 mayuresh joined #perl6
11:33 mayuresh p6: say 3;
11:33 camelia rakudo-moar b16110: OUTPUT«3␤»
11:34 mayuresh p6: say "Hello, world!";
11:34 camelia rakudo-moar b16110: OUTPUT«Hello, world!␤»
11:34 mayuresh ah, the camelia bot is rather cool ...
11:34 mayuresh :)
11:36 araujo joined #perl6
11:37 araujo joined #perl6
11:38 pmurias joined #perl6
11:39 lizmat m: Str.new(value => q/Hello, world!/ ~ "\n").print   # more complicated way
11:39 camelia rakudo-moar b16110: OUTPUT«Hello, world!␤»
11:39 |Tux| test             50000    36.255    36.146
11:39 |Tux| test-t           50000    37.664    37.555
11:39 lizmat .5 sec less ?
11:39 azawawi https://github.com/sergot/http-userage​nt/blob/master/lib/HTTP/UserAgent.pm6 # doesnt contain even a post method :)
11:41 pmurias hi
11:43 dalek nqp: 1764d85 | (Pawel Murias)++ | src/vm/js/nqp-runtime/io.js:
11:43 dalek nqp: [js] Fix bug in nqp::getenvhash.
11:43 dalek nqp: review: https://github.com/perl6/nqp/commit/1764d8599f
11:43 dalek nqp: 0945366 | (Pawel Murias)++ | src/vm/js/ (3 files):
11:43 dalek nqp: [js] Implement nqp::spawn using the nqp-js-io module (For now lives on github as pmurias/nqp-js-io).
11:43 dalek nqp: review: https://github.com/perl6/nqp/commit/0945366d7d
11:43 dalek nqp: 7330628 | (Pawel Murias)++ | / (7 files):
11:43 dalek nqp: [js] Load the setting in the bootstrapped nqp-js.
11:43 dalek nqp:
11:43 dalek nqp: node nqp-bootstrapped.js t/nqp/01-literals.t passes.
11:43 dalek nqp:
11:43 dalek nqp: It uses a bunch of hacks that need to be cleaned up.
11:43 dalek nqp: review: https://github.com/perl6/nqp/commit/7330628d94
11:43 jnthn pmurias++
11:44 doublec_ joined #perl6
11:44 CIAvash azawawi: https://github.com/sergot/http-userag​ent/blob/master/lib/HTTP/Request.pm6
11:44 kid51 joined #perl6
11:45 travis-ci joined #perl6
11:45 travis-ci NQP build failed. Pawel Murias '[js] Load the setting in the bootstrapped nqp-js.
11:45 travis-ci https://travis-ci.org/perl6/nqp/builds/85310473 https://github.com/perl6/nqp/com​pare/6ac0c44f92ad...7330628d94a6
11:45 travis-ci left #perl6
11:49 pmurias hmm, nqp-js-io fails to build on travis for some reason :(
11:51 loren joined #perl6
11:52 iv looks like it's using a very old version of node
11:53 loren good evening, perl6
11:54 dalek nqp: 16cd47d | (Pawel Murias)++ | src/vm/js/ (2 files):
11:54 dalek nqp: [js] Implement nqp::getmessage.
11:54 dalek nqp: review: https://github.com/perl6/nqp/commit/16cd47d1c2
11:54 dalek nqp: de79fcf | (Pawel Murias)++ | t/nqp/44-try-catch.t:
11:54 dalek nqp: Test nqp::getmessage.
11:54 dalek nqp: review: https://github.com/perl6/nqp/commit/de79fcf514
11:57 travis-ci joined #perl6
11:57 travis-ci NQP build failed. Pawel Murias 'Test nqp::getmessage.'
11:57 travis-ci https://travis-ci.org/perl6/nqp/builds/85312361 https://github.com/perl6/nqp/com​pare/7330628d94a6...de79fcf514f8
11:57 travis-ci left #perl6
11:59 nine pmurias++ # still my lonesome hero :)
12:12 ely-se joined #perl6
12:12 telex joined #perl6
12:15 pierrot joined #perl6
12:18 hohoho joined #perl6
12:19 Jaglor joined #perl6
12:19 Jaglor left #perl6
12:21 cognominal joined #perl6
12:22 reneeb joined #perl6
12:27 softmoth joined #perl6
12:27 RaTTuS|BIG joined #perl6
12:27 jkva joined #perl6
12:39 azawawi sergot: ping
12:39 sufrostico joined #perl6
12:42 kmel joined #perl6
12:44 tokuhirom joined #perl6
12:44 moritz ok, heretic question: python has inverted stack traces, and the error message at the bottom
12:44 moritz what do you think of that? something to adopt? or rather avoid?
12:45 arnsholt Confuses me endlessly, when it's been a while since I programmed in Python =)
12:45 arnsholt But potentially a good idea, yeah
12:45 [Coke] -1 from me; I like having the error be right next to the last output
12:46 arnsholt It's actually very ergonomic when the stack traces start getting long
12:46 lizmat simple to add with an env variable. fwiw
12:46 lizmat well, the non --ll-exception case
12:50 ZoffixW joined #perl6
12:50 azawawi sigh, hit another http redirection bug with HTTP::UserAgent :(
12:51 azawawi so far POST + HTTP redirection is not working. Nearly tried everything, LWP::Simple, HTTP::Client (hangs), HTTP::UserAgent(hangs)
12:51 ZoffixW Hey. Does a "clock sequence" have meaning in Perl 6? What is it and how do I obtain it? It's mentioned in 4.1.5 of RFC 4122, but I'm not 100% sure what it is (http://www.rfc-base.org/txt/rfc-4122.txt)
12:51 ZoffixW azawawi, how did HTTP::Tinyish fair?
12:52 azawawi let me check :)
12:55 [Coke] .see curtisovidpoe
12:55 [Coke] .seen curtisovidpoe
12:55 yoleaux I saw CurtisOvidPoe 13 Oct 2015 13:44Z in #perl6: <CurtisOvidPoe> Heh.
12:55 * CurtisOvidPoe looks for CurtisOvidPoe
12:55 [Coke] FOUND HIM!
12:56 dalek nqp: cc2b977 | (Pawel Murias)++ | / (2 files):
12:56 dalek nqp: [js] Implement goal and dba rxtypes.
12:56 dalek nqp:
12:56 dalek nqp: Pass rx_goal.
12:56 dalek nqp: review: https://github.com/perl6/nqp/commit/cc2b977852
12:56 dalek nqp: e5c012a | (Pawel Murias)++ | / (3 files):
12:56 dalek nqp: [js] Add a make js-test-bootstrapped target.
12:56 dalek nqp: review: https://github.com/perl6/nqp/commit/e5c012aeb1
12:59 eone joined #perl6
12:59 travis-ci joined #perl6
12:59 travis-ci NQP build failed. Pawel Murias '[js] Add a make js-test-bootstrapped target.'
12:59 travis-ci https://travis-ci.org/perl6/nqp/builds/85321737 https://github.com/perl6/nqp/com​pare/de79fcf514f8...e5c012aeb1d4
12:59 travis-ci left #perl6
13:01 ely-se joined #perl6
13:03 * [Coke] shakes his fist at travis.
13:03 azawawi ZoffixW: i tried it with HTTP_TINYISH_DEBUG=1, it uses curl and gets stuck
13:03 azawawi ZoffixW: This is working perfectly https://github.com/azawawi/perl6-selenium-webdrive​r/blob/master/examples/test_selenium_phantomjs.pl
13:03 azawawi ZoffixW: this gets stuck https://github.com/azawawi/perl6-selenium-webdrive​r/blob/master/examples/test_selenium_phantomjs.p6
13:04 azawawi hmmm even curl get stuck
13:05 ZoffixW azawawi, what HTTP method does it redirect to?
13:05 ZoffixW IIRC a POST should be only redirected to a GET
13:05 brrt joined #perl6
13:06 azawawi ZoffixW: https://gist.github.com/az​awawi/e30986625e31cae4f2e9
13:06 azawawi ZoffixW: it is basically against a `phantomjs --webdriver=5555`
13:07 azawawi ZoffixW: it is installed via 'sudo apt-get install phantomjs'  # 1.9
13:08 azawawi LWP::UserAgent from Perl 5 has no problem with it. i think it is doing a workaround.
13:09 ZoffixW I don't have that package. But I'm guessing "* Disables POST, goes with GET" and "* Empty reply from server" are clues. It's expecting the redirect to follow with a POST, not GET.
13:09 ZoffixW Oh, there's actually "> POST /wd/hub/session/185fea20-"... then maybe the other way around, it's expecting a GET, but receiving a POST :P
13:10 ZoffixW In other words... ¯\_(ツ)_/¯
13:12 azawawi well that's true because i added the following https://github.com/azawawi/perl6-se​lenium-webdriver/blob/master/exampl​es/test_selenium_phantomjs.pl#L47
13:13 colomon carlin: Crypt::Bcrypt got on my “skip” list because it was hanging too often.  I’ll try reinstating it.
13:14 llfourn joined #perl6
13:14 ZoffixW azawawi, aha! :)
13:15 sergot azawawi: pong
13:15 sergot it does not, we need to add it :)
13:15 sergot you can build your post request using HTTP::Request
13:16 azawawi sergot: i was trying to use HTTP::UserAgent and somehow hit a redirection bug
13:16 azawawi sergot: already did :)
13:16 sergot and then $useragen->request($your_request)
13:16 sergot azawawi++
13:17 azawawi the goal here is to run automate testing for websites over phantomjs as phase #1 and then go with firefox, ...
13:17 azawawi s/automate/automated/
13:18 azawawi im basically implementing https://code.google.com/p/se​lenium/wiki/JsonWireProtocol
13:18 kmel is there a function I can call on a scalar that returns its type (Int, Str ect.)?
13:18 cygx m: 42.WHAT
13:18 camelia rakudo-moar b16110: ( no output )
13:18 ely-se m: my $x = 42; say $x.WHAT
13:18 camelia rakudo-moar b16110: OUTPUT«(Int)␤»
13:19 psch m: say Int.^can('WHAT') # it's not quite a function though
13:19 camelia rakudo-moar b16110: OUTPUT«()␤»
13:19 sergot azawawi: what is the problem now?
13:19 sergot could you show me the code?
13:19 psch but i imagine that probably doesn't matter... :)
13:20 kmel thanks ely-se cygx psch
13:21 azawawi sergot: LWP::UserAgent is doing some magic with POST redirection. We're not. Same with curl.
13:21 azawawi sergot: the Perl 5 version that is
13:23 azawawi sergot: basically phantomjs 1.9 which is now in debian stable responds with an HTTP status 303 to redirect
13:23 azawawi sergot: 2.0 fixed that
13:23 azawawi sergot: let me update the script to use HTTP::UserAgent
13:23 pmurias joined #perl6
13:27 dalek modules.perl6.org: 05c783d | (Zoffix Znet)++ | web/js/main.js:
13:27 dalek modules.perl6.org: Coerce sorter plugin to treat all module names as text
13:27 dalek modules.perl6.org: review: https://github.com/perl6/modul​es.perl6.org/commit/05c783d171
13:29 azawawi sergot: https://github.com/azawawi/perl6-se​lenium-webdriver/blob/master/exampl​es/test_selenium_phantomjs.p6#L50
13:30 dalek modules.perl6.org: 31e9990 | (Zoffix Znet)++ | web/ (2 files):
13:30 dalek modules.perl6.org: Disable sort on columns that do not make sense
13:30 dalek modules.perl6.org:
13:30 dalek modules.perl6.org: Use sorter plugin's options instead of half-fixing this with CSS
13:30 dalek modules.perl6.org: review: https://github.com/perl6/modul​es.perl6.org/commit/31e9990811
13:33 * azawawi starts debugging HTTP::UserAgent
13:33 sergot azawawi: hnn
13:33 sergot hmm
13:35 sergot What magic actually do you mean? :)
13:35 literal spacebat: there's also perl6-mode, but it's not quite there yet in terms of highlighting
13:36 ZoffixW sergot, I'm guessing this: https://github.com/azawawi/perl6-se​lenium-webdriver/blob/master/exampl​es/test_selenium_phantomjs.pl#L47
13:40 sergot oh, ok.
13:41 skids joined #perl6
13:43 sufrostico joined #perl6
13:46 azawawi sergot: aha
13:46 azawawi sergot: headers are not preserved on redirection in HTTP::UserAgent?
13:46 bjz joined #perl6
13:46 sergot azawawi: let me check
13:47 azawawi sergot: they're not
13:47 sergot azawawi: right
13:49 bjz_ joined #perl6
13:49 sergot azawawi: please create an issue for that
13:49 azawawi sergot: im trying to fix it :)
13:49 sergot azawawi++ that's nice! thank! :)
13:50 dalek ecosystem: f30d76f | moznion++ | META.list:
13:50 dalek ecosystem: Add Stream::Buffered
13:50 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/f30d76f614
13:54 rurban joined #perl6
13:55 margeas joined #perl6
13:57 softmoth joined #perl6
14:04 khw joined #perl6
14:06 [Coke] .tell nine - please log out and log back in to RT.
14:06 yoleaux [Coke]: I'll pass your message to nine.
14:06 [Coke] Also - if folks need help with RT that isn't "make me a perl 6 bug admin", there's an email addy at the bottom of each page, the rt bugadmins seem very helpful.
14:08 dalek rakudo/nom: e3ab6cc | lizmat++ | src/core/Rakudo/Internals.pm:
14:08 dalek rakudo/nom: Add WeightedRoll class
14:08 dalek rakudo/nom:
14:08 dalek rakudo/nom: It might be of general usage, and was the only way to not have the code
14:08 dalek rakudo/nom: live twice: once for Mix and once for MixHash
14:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e3ab6ccb3c
14:08 dalek rakudo/nom: 254c1a8 | lizmat++ | src/core/ (6 files):
14:08 dalek rakudo/nom: Some more Bag/BagHash/Mix/MixHash reorganization
14:08 dalek rakudo/nom:
14:09 dalek rakudo/nom: Basically, backport my experiment of using 2 hashes from last Monday
14:09 dalek rakudo/nom: by taking only the good bits I did.
14:09 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/254c1a881e
14:09 rindolf joined #perl6
14:09 cygx_ joined #perl6
14:11 dalek roast: 08e8c9a | lizmat++ | S32-basics/xxKEY.t:
14:11 dalek roast: Fix test because of simplified code
14:11 dalek roast:
14:11 dalek roast: Previously, we had an ASSIGN-KEY for Bag/Mix, now we just let the
14:11 dalek roast: fact that there is no container, take care of throwing the error.
14:11 dalek roast: review: https://github.com/perl6/roast/commit/08e8c9adb0
14:11 lizmat cycling&
14:11 azawawi sergot: going to debug it @home. nearly there :)
14:11 azawawi sergot: thanks for your help
14:11 azawawi Zoffix: thanks :)
14:14 sergot azawawi: thank you too
14:14 [Coke] RT: 1029: nom: 8; glr: 4; weird: 11; LTA: 87; testneeded: 9; xmas: 75
14:20 ely-se joined #perl6
14:23 tadzik oh gods that Ovid's blog post is hilarous
14:23 tadzik and a bit sad as well
14:25 sufrostico joined #perl6
14:25 [Coke] I hire consultants here. That rings pretty true, and I get the ones that have -already gone through- a consultanting company screener before they get to me!
14:25 [Coke] *consulting
14:33 CJKinni joined #perl6
14:34 weihan joined #perl6
14:42 muraiki joined #perl6
14:42 zostay is react/whenever/supply documented somewhere? i'm not even finding them described in S17
14:43 RabidGravy joined #perl6
14:45 edehont joined #perl6
14:45 tokuhirom joined #perl6
14:45 ShimmerFairy joined #perl6
14:49 [Coke] jnthn++'s presentations are the best place to find that at the moment.
14:50 [Coke] try this: http://www.jnthn.net/paper​s/2015-spw-concurrency.pdf
14:50 jnthn Also there's a gist somewhere that describes it in ratehr more detail that is to be incorporated into S17 at some point.
14:50 ecocode left #perl6
14:50 [Coke] and a todo for docs to get it in there.
14:54 llfourn joined #perl6
15:01 CJKinni_ joined #perl6
15:05 blackcat_ joined #perl6
15:07 hohoho joined #perl6
15:14 daxim who is responsible for <https://p6weekly.wordpress.com​/2015/10/12/2015-42-unveiling/>?   "# even though there are officially no words starting with ß" is a daxim bait if I've ever seen one
15:15 daxim that only needs one counter example to disprove
15:15 daxim "ßo"
15:15 daxim (that's sorbian)
15:16 jnthn m: "ßo".fc
15:16 camelia rakudo-moar 254c1a: ( no output )
15:16 jnthn m: say "ßo".fc
15:16 camelia rakudo-moar 254c1a: OUTPUT«sso␤»
15:16 jnthn m: say "ßo".uc
15:16 ely-se joined #perl6
15:16 camelia rakudo-moar 254c1a: OUTPUT«SSO␤»
15:16 jnthn m: say "ßo".tc
15:16 camelia rakudo-moar 254c1a: OUTPUT«Sso␤»
15:16 jnthn :)
15:17 * jnthn had to look up what sorbian is :)
15:17 daxim no smiley, because that particular german orthographic rule does not apply
15:17 gfldex Sso is most likely wrong in that case because ß is a legature of s and z (the so called round z)
15:18 daxim it's a letter, not a ligature for at least 120 years
15:18 jnthn Doesn't matter. Perl 6 implements the Unicode SpecialCasing/CaseFolding rules.
15:18 jnthn If you want anything other than those, that's for module space.
15:18 daxim and it combined ſ and s, not s and z
15:19 daxim yeah, I' know, it's a bug in unicode
15:19 daxim they cannot fix it because of reasons.  it's complicated
15:19 jnthn :)
15:19 ShimmerFairy daxim: I concur with you, it really should be ß <-> ẞ   :)
15:19 daxim I have a p5 module that attempts to fix it
15:19 jnthn Yeah, I started reading a little on it and was like...OK, I don't even have time for this, I'll just get on with doing what Unicode says.
15:20 s34n left #perl6
15:20 jnthn They seem to take compatability quite seriously... That's probably a good thing overall.
15:20 ShimmerFairy daxim: AFAIU, the best/only solution would be if they just added two new symbols that implemented the proper casing, and deprecated the existing ones. (Though just changing the properties would be much simpler)
15:20 jnthn Well, stability, not compatibility.
15:21 loren my $l = IO::Socket::INET.new(:listen, :localport(xxxx)); my $c = $l.accept; while my $buf = $c.recv(:bin) { $buf.decode('utf-8').say; }; $c.close;
15:21 loren The simple service will terminate when i press CTRL+C in the client side.
15:22 zostay thx [Coke]
15:22 ShimmerFairy jnthn: btw, I do think we should be doing what Unicode says in core (as you've done), even if they let an erroneous application of NFKC slip through :P
15:23 daxim at least now I have a strong motivation to write my first p6 module
15:23 weihan joined #perl6
15:24 uruwi joined #perl6
15:24 uruwi Back
15:25 weihan joined #perl6
15:26 uruwi Still going to port Games::Rougelike
15:27 liztormato joined #perl6
15:27 sufrostico joined #perl6
15:28 liztormato joined #perl6
15:29 loren '$buf.decode('utf-8')' will throw a exception when $buf is not a string. the document not mention that.
15:31 jnthn loren: It's still appropriate. You can't write robust code like that. What if you get a multi-byte UTF-8 sequence spread over a couple of packets?
15:31 s_kilk joined #perl6
15:32 jnthn The documentation for .decode should mention what happens if it can't decode though, so if it doesn't that deserves an doc issue
15:33 loren ok, thks very much, jnthn
15:33 jonathan__ joined #perl6
15:33 jnthn (At some point we should also support things like inserting a replacement char too)
15:33 loren i'll take more test
15:34 mrf jnthn: is there currently any way to decode streamed in bytes without waiting for the entire stream to complete?
15:34 jnthn mrf: Use the string-level rather than byte-level API on sockets.
15:34 jnthn mrf: I plan to expose a streaming API for it too
15:35 jkva joined #perl6
15:35 jnthn (Hopefully pre-xmas)
15:35 RabidGravy while we're on the subject of sockets, does the NQP socket api expose timeout for connect?
15:36 jnthn RabidGravy: Not afaik
15:36 * jnthn will take a good pass over I/O things in November
15:37 dalek nqp: 932548b | (Pawel Murias)++ | src/vm/js/ (2 files):
15:37 dalek nqp: [js] avoid emitting a needless warning
15:37 dalek nqp: review: https://github.com/perl6/nqp/commit/932548bcfe
15:37 dalek nqp: b3a4cbf | (Pawel Murias)++ | src/vm/js/nqp-runtime/core.js:
15:37 dalek nqp: [js] Stringify the key value things we get while iterating on hash to the key.
15:37 dalek nqp: review: https://github.com/perl6/nqp/commit/b3a4cbf834
15:37 dalek nqp: 25008c3 | (Pawel Murias)++ | src/vm/js/nqp-runtime/core.js:
15:37 dalek nqp: [js] Make nqp::istype work on the result of nqp::hash.
15:37 dalek nqp: review: https://github.com/perl6/nqp/commit/25008c3d31
15:37 dalek nqp: 7b61ed4 | (Pawel Murias)++ | t/nqp/89-istype.t:
15:37 dalek nqp: Test calling nqp::istype on nqp::list() and nqp::hash().
15:37 dalek nqp: review: https://github.com/perl6/nqp/commit/7b61ed434d
15:37 dalek nqp: a133692 | (Pawel Murias)++ | t/nqp/68-associative-for.t:
15:37 dalek nqp: Test stringifying the keyvalue pair to the key.
15:37 dalek nqp: review: https://github.com/perl6/nqp/commit/a13369226b
15:38 jnthn mrf: In more detail: MoarVM has a nice implementation of these things that can handle not only codepoints hanging over multi-byte sequences, but also graphemes hanging over multi-codepoint sequences, to get NFG stuff right. So I want to provide a Perl 6 level API for that stuff.
15:38 pmurias jnthn: it's funny that I have to port over code from MoarVM to support doing some IO things synchronously on node.js
15:40 mrf jnthn: that would be nice for Web Streaming stuff
15:40 n0tjack joined #perl6
15:40 travis-ci joined #perl6
15:40 travis-ci NQP build failed. Pawel Murias 'Test stringifying the keyvalue pair to the key.'
15:40 travis-ci https://travis-ci.org/perl6/nqp/builds/85354276 https://github.com/perl6/nqp/com​pare/e5c012aeb1d4...a13369226b28
15:40 travis-ci left #perl6
15:40 jnthn mrf: Indeed. I'll try to make it happen pre-xmas so everyone'll have an easier time implementing such things.
15:42 mrf jnthn: is the idea that the methods would just return Str rather than Buf?
15:42 jax joined #perl6
15:43 virtualsue joined #perl6
15:43 dalek nqp: a187d4c | (Pawel Murias)++ | .travis.yml:
15:43 dalek nqp: Disable testing the js backend on travis as the node.js version is too old.
15:43 dalek nqp: review: https://github.com/perl6/nqp/commit/a187d4c93f
15:43 liztormato_ joined #perl6
15:44 jnthn mrf: There's a char-based API there today; the use case here is more like when you need to start out working at byte level because you didn't work out what encoding the rest of the message has to be decoded in
15:44 jnthn mrf: But the API will basically be something you can throw Buf/Blob into, and then ask for a Str with as many chars as it can completely decode/normalize at that point
15:46 timotimo oh, neat
15:46 tokuhirom joined #perl6
15:47 travis-ci joined #perl6
15:47 travis-ci NQP build passed. Pawel Murias 'Disable testing the js backend on travis as the node.js version is too old.'
15:47 travis-ci https://travis-ci.org/perl6/nqp/builds/85355579 https://github.com/perl6/nqp/com​pare/a13369226b28...a187d4c93f69
15:47 travis-ci left #perl6
15:48 gtodd1 hmm the "naive parsing benchmark" made a great leap forward in speed with GLR  ... but seems to have slowed down again
15:48 gtodd1 http://www.perlmonks.org/?node_id=1098390
15:48 gtodd1 or perhaps I am imagining it all
15:49 n0tjack lear
15:49 yoleaux 10:08Z <jnthn> n0tjack: about your unpacking $/ question in the backlog that I don't think anyone answered: grammar G { token TOP { <digit> <alpha> } }; class A { method TOP((:$digit, :$alpha)) { say $digit; say $alpha; } }; G.parse('1a', actions => A)
15:49 n0tjack oh, neat
15:49 * n0tjack goes off to refactor his grammar
15:51 Ven joined #perl6
15:57 n0xff joined #perl6
16:08 hohoho joined #perl6
16:10 virtualsue joined #perl6
16:11 timotimo gtodd: what parts have you tried?
16:11 liztormato joined #perl6
16:12 Ven joined #perl6
16:12 diana_olhovik_ joined #perl6
16:13 timotimo like, which of these alternatives?
16:14 loren good night, perl6
16:14 loren also timotimo
16:15 timotimo good night loren
16:16 gtodd timotimo: well I am not doing anything very formal .... I have a bunch of different command lines and some time a few weeks back compiled a post GLR perl6 and was shocked at how much it had sped up
16:16 timotimo how did you reach the conclusion that it has slowed down again?
16:16 yakudza joined #perl6
16:16 timotimo just by compiling a newer rakudo and re-running?
16:17 gtodd timotimo: yeah ... something like perl6 -n -e 'say $/ if m/<<\w**5>>/' logs.txt
16:17 gtodd had dropped from 40 to < 10 seconds ... unless I made a mistake
16:18 timotimo can i has your logs.txt?
16:18 gtodd recompiling from yestyerday it was back to something in the order of 30 seconds ...
16:18 timotimo wow
16:18 gtodd timotimo: ok just a sec trying with a freshly built rakudo again
16:18 timotimo that'd be a big loss indeed
16:22 ^elyse^ joined #perl6
16:22 TimToady you might try it with 'say $/ if .simplematch(/<<\w**5>>/)' to see if it's the gather/take overhead or not
16:22 gtodd ok
16:22 gtodd timotimo: easiest explanation I must have made a mistake seeing the earlier speed up
16:23 timotimo oh
16:23 timotimo that's actually a bit sad :(
16:23 gtodd but
16:24 gtodd one thing I notice is:
16:24 gtodd time perl6 -e 'for lines(:eager) {  say ~$/ if  m/(<<\w**5>>)/}' 1k-lines.txt  0.67s
16:24 TimToady is the -n loop being lazy or batching?
16:24 virtualsue joined #perl6
16:25 timotimo :eager doesn't exist any more, though
16:25 gtodd ah ok
16:25 gtodd my script needs tweaking then :-)
16:25 jnthn gtodd: What about just for lines() { ... } on a modern Rakudo?
16:25 tony-o_ m: say $?ENC
16:25 camelia rakudo-moar 254c1a: OUTPUT«5===SORRY!5=== Error while compiling /tmp/VjdRC4pxWx␤Variable '$?ENC' is not declared. Did you mean '@?INC'?␤at /tmp/VjdRC4pxWx:1␤------> 3say 7⏏5$?ENC␤»
16:25 jnthn TimToady: I'm not sure -n has been looked at since GLR...
16:26 tony-o_ r: say $?ENC;
16:26 camelia rakudo-{moar,jvm} 254c1a: OUTPUT«5===SORRY!5=== Error while compiling /tmp/tmpfile␤Variable '$?ENC' is not declared. Did you mean '@?INC'?␤at /tmp/tmpfile:1␤------> 3say 7⏏5$?ENC;␤»
16:26 tony-o_ hm
16:26 gtodd timotimo:  trying  with a file that's bigger doesn't errm "scale" ...  perl6 -e 'for lines(:eager) {  say ~$/ if  m/(<<\w**5>>)/}' 100k-lines.txt  30.90s
16:27 timotimo what happens if you drop the :eager?
16:27 gtodd ie. it's less than 100 times the time for 100 times as many lines ... which I assume is good
16:28 timotimo japhb: i'm getting some trouble with perl6-bench; after a ./bench setup i get Don't know how to process component ''
16:28 TimToady how does it compare with slurp.lines?
16:28 timotimo after "Setup complete."
16:28 tony-o_ timotimo: ugexe fixed a bug in it last night, if you haven't reinstalled since then
16:28 gtodd same ... I now grok why :eager would not be needed post GLR
16:28 tony-o_ it being perl6-bench
16:28 edehont joined #perl6
16:29 timotimo i just git pull'd it i think?
16:29 timotimo let me see.
16:29 timotimo "already up to date"
16:30 timotimo but the latest commit i see is 3 weeks old
16:30 matt_ joined #perl6
16:30 gtodd slurp.lines is ~ 2 seconds faster :-) 28.75 vs 31.25
16:30 gtodd :-D
16:30 timotimo and there's no pull requests or anything
16:31 timotimo tony-o_: where did that commit disappear to?
16:31 ugexe are you sure its not assuming its up to date based on the version in the meta.info file?
16:31 tony-o_ https://github.com/tony-o/perl6-bench/pull/8
16:33 tony-o_ timotimo: i just merged in the other two PRs from ugexe
16:33 Ven joined #perl6
16:33 tony-o_ so a git pull should pull down some killer stuff
16:34 timotimo oh, ok
16:34 timotimo oh, you have your own fork
16:34 gtodd and the winner is Larry Wall (AKA TimToady) ....
16:34 timotimo that explains it i guess
16:35 captain-adequate joined #perl6
16:35 gtodd time perl6 -n -e ' say ~$/ if .simplematch(/<<\w**5>>/)' 100k-lines.txt  7.49s user 0.08s system 98% cpu 7.681 total
16:35 timotimo ... wait
16:36 timotimo oh, it's about setting the outer's $/?
16:36 timotimo no, hold on, that's in simplematch, too
16:36 TimToady no, simplematch does that
16:36 gtodd so 7.49 seconds versus 30 seonds ... cause of .simplematch .... maybe I will check to see if the output is the same :-P
16:36 TimToady it's just all the rigamarole around handling multiple matches
16:36 cygx_ joined #perl6
16:36 TimToady was using it the other day to do my timings on the tweet database
16:37 gtodd will m//   "know" when it is supposed to be .simplematch in some closer to Xmas variant of perl6 :-)
16:37 timotimo merge conflict in readme.md
16:37 TimToady where about ⅓ of the overhead was gather/take, and about ⅔ was :i
16:38 gtodd wow perl5 -->  perl -wnl -E 'say $1 if /\b(\w{5})\b/' 100k-lines.txt   0.52seconds ... so it's only ~ 15 times faster
16:38 cygx_ Now that people are around, I'll ask again: Any thoughts on letting split(..., :all) return strings and only intermix Match objects in case of split(..., :all, :match)?
16:38 gtodd wheee!! (it used to be 75-80 times faster)
16:38 cygx_ We're no supposed to be doing high-energy physics, where pulling things apart can spawn all kinds of crazily complicated things ;)
16:38 cygx_ *not
16:38 pRiVi_ joined #perl6
16:39 Shozan joined #perl6
16:39 TimToady HLL::Compiler is very confused about what is the language name/version vs compiler name/version, and this shows up in both -v and -V
16:40 Upasaka_ joined #perl6
16:40 intrigue_ joined #perl6
16:40 intrigue_ joined #perl6
16:40 iv_ joined #perl6
16:40 btyler_ joined #perl6
16:40 TimToady "This is perl6 version 2015.09-373-g254c1a8"
16:41 ZoffixW joined #perl6
16:41 ZoffixW Shouldn't this flatten?
16:41 ZoffixW m: my @a = <foo bar baz>; my @b; @b.append: @a, 'mer'; say @b.perl
16:41 camelia rakudo-moar 254c1a: OUTPUT«[["foo", "bar", "baz"], "mer"]␤»
16:41 timotimo tony-o_: dooooood
16:41 timotimo tony-o_: that's a totally different perl6-bench you have there
16:41 * [Coke] stares at htmlify
16:42 TimToady should be more like "This is rakudo version 2015.09-373-g254c1a8 implementing Perl v6b" or so
16:42 jnthn TimToady: I susepct the code you're looking at predates us having much clue how such things would pan out :)
16:44 apotheon_ joined #perl6
16:46 * TimToady RT'd it, since it's fruit that's hanging somehwere
16:46 * timotimo bonks tony-o_ for deceiving him so bad
16:46 * gtodd reads about simplematch() and m/ and m:i etc.  ... argh :-)
16:47 * ZoffixW doesn't follow the subleties between .push and .append:
16:47 TimToady ZoffixW: no, that's two arguments under the one-arg rule
16:47 ZoffixW m: my @a = <foo bar baz>; my @b; @b.push(@a, 'mer'); my @c; @c.append(@a, 'mer'); say [ @b, @c ].perl
16:47 camelia rakudo-moar 254c1a: OUTPUT«[[["foo", "bar", "baz"], "mer"], [["foo", "bar", "baz"], "mer"]]␤»
16:47 TimToady .push only pushes arguments, .append only pushes a list
16:47 Ven joined #perl6
16:48 ZoffixW Alright. I'll read up on the difference between the two. Thanks.
16:48 TimToady because .push only pushes arguments, it doesn't care if any of those arguments are lists
16:48 sftp_ joined #perl6
16:48 TimToady because .append only pushes a list, it doesn't care whether that list was produced by argument comma or something else
16:48 timotimo m: my $spec = "/home/oldhome/timo/perl6/fr​esh_bench/components/arane"; my $COMPONENTS_DIR = "/home/oldhome/timo/perl6/fresh_bench/components"; $spec .= subst(/^ $COMPONENTS_DIR '/' /, ''); say $spec.perl
16:48 camelia rakudo-moar 254c1a: OUTPUT«"arane"␤»
16:49 timotimo for some reason that doesn't remove the text locally
16:49 Kogurr joined #perl6
16:49 jkva joined #perl6
16:50 timotimo m: my @specs = "/home/oldhome/timo/perl6/fr​esh_bench/components/arane"; my $COMPONENTS_DIR = "/home/oldhome/timo/perl6/fresh_bench/components"; for @specs -> $spec is copy { $spec .= subst(/^ $COMPONENTS_DIR '/' /, ''); say $spec.perl }
16:50 ZoffixW TimToady++ # That makes sense now. Thanks.
16:50 ZoffixW Weeeeeeeeeeeeeeeeee \o/
16:50 camelia rakudo-moar 254c1a: OUTPUT«"arane"␤»
16:51 TimToady m: say 'foo' ~~ S/foo/bar/
16:51 camelia rakudo-moar 254c1a: OUTPUT«Potential difficulties:␤    Smartmatch with S/// can never succeed because the subsequent string match will fail␤    at /tmp/TCGa4pn2jf:1␤    ------> 3say 'foo' ~~ S/foo/bar/7⏏5<EOL>␤False␤»
16:51 TimToady which is not strictly true, but close enough
16:51 TimToady m: say 'foo' ~~ S/foo/foo/
16:51 camelia rakudo-moar 254c1a: OUTPUT«Potential difficulties:␤    Smartmatch with S/// can never succeed because the subsequent string match will fail␤    at /tmp/Ffuqn5y8DN:1␤    ------> 3say 'foo' ~~ S/foo/foo/7⏏5<EOL>␤True␤»
16:52 timotimo damn
16:52 timotimo we have a strange bug going on here
16:52 jkva_ joined #perl6
16:52 timotimo $COMPONENTS_DIR gets imported from a module "Bench::Globals"
16:53 timotimo by puting a "my $COMPONENTS_DIR = OUTER::('$COMPONENTS_DIR')" at the beginning of this sub, i can get the regex to work
16:54 timotimo wizards, figure this out. i'm too tired and would prefer to not move a single inch today (but still have to do grocery shopping)
16:54 k2490klm joined #perl6
16:56 TimToady does it also fail with $spec.subst-mutate instead of .=subst?
16:56 timotimo i believe this is about the regex not matching because the $COMPONENTS_DIR doesn't get interpolated
16:57 TimToady I'm just wondering about different code paths in the compiler
16:57 TimToady I doubt the actual problem would be in .subst
16:58 patrickz joined #perl6
16:59 TimToady and does it fail with $spec ~~ s///?
16:59 bjz_ joined #perl6
16:59 adhoc joined #perl6
16:59 Woodi joined #perl6
16:59 jevin_ joined #perl6
16:59 |Tux| joined #perl6
16:59 woodruffw joined #perl6
16:59 Util joined #perl6
16:59 bbarker`` joined #perl6
16:59 tony-o_ joined #perl6
16:59 leedo joined #perl6
16:59 agentzh joined #perl6
16:59 mprelude joined #perl6
16:59 breinbaas joined #perl6
16:59 ssflying joined #perl6
16:59 revdiablo joined #perl6
16:59 sergot joined #perl6
16:59 atta joined #perl6
16:59 ingy joined #perl6
17:00 * TimToady looks around for the backlog he dropped a few minutes ago...
17:00 timotimo hold on
17:01 timotimo i'm running ./bench build at the moment, so can't check immediately
17:01 TimToady np
17:02 firstdayonthejob joined #perl6
17:06 softmoth joined #perl6
17:07 boggard joined #perl6
17:07 mrf joined #perl6
17:07 freeze joined #perl6
17:07 Grrrr joined #perl6
17:07 danstoner joined #perl6
17:07 lucs joined #perl6
17:07 yoleaux joined #perl6
17:07 mst joined #perl6
17:07 PerlJam joined #perl6
17:07 f3ew joined #perl6
17:07 oha joined #perl6
17:07 jantore joined #perl6
17:07 colomon joined #perl6
17:07 margeas joined #perl6
17:07 krakan joined #perl6
17:07 spacebat joined #perl6
17:08 yeltzooo joined #perl6
17:08 burnersk joined #perl6
17:08 sjohnson joined #perl6
17:08 go|dfish joined #perl6
17:08 skids joined #perl6
17:09 Humbedooh joined #perl6
17:09 timotimo i'll be doing a few nqp benchmarks now
17:09 timotimo moar vs jvm
17:10 Humbedooh joined #perl6
17:10 timotimo nine: need a bit more help with building the p6forloop desugar?
17:10 jdv79 joined #perl6
17:10 bjz_ joined #perl6
17:10 adhoc joined #perl6
17:10 Woodi joined #perl6
17:10 jevin_ joined #perl6
17:10 |Tux| joined #perl6
17:10 woodruffw joined #perl6
17:10 Util joined #perl6
17:10 bbarker`` joined #perl6
17:10 tony-o_ joined #perl6
17:10 leedo joined #perl6
17:10 agentzh joined #perl6
17:10 mprelude joined #perl6
17:10 breinbaas joined #perl6
17:10 ssflying joined #perl6
17:10 revdiablo joined #perl6
17:10 sergot joined #perl6
17:10 atta joined #perl6
17:10 ingy joined #perl6
17:11 timotimo as soon as that happens, i'll also do rakudo benchmarks
17:12 wtw joined #perl6
17:12 confuseAcat joined #perl6
17:12 avar joined #perl6
17:12 avar joined #perl6
17:12 literal joined #perl6
17:12 crux joined #perl6
17:13 timotimo TimToady: with only ~~ it says Nil
17:14 timotimo hold on, i mistaked
17:14 FROGGS joined #perl6
17:14 timotimo i get a match IFF i put in the OUTER:: thingie
17:16 * FROGGS tried yesterday to unbust s/// in for loops
17:16 * FROGGS did this after banging the head against jvm for hours
17:17 dha joined #perl6
17:18 Alina-malina joined #perl6
17:21 dalek doc/makefile: 6478396 | coke++ | t/pod-convenience.t:
17:21 dalek doc/makefile: fix test that was passing for the wrong reason
17:21 dalek doc/makefile: review: https://github.com/perl6/doc/commit/6478396ac5
17:25 Peter_R joined #perl6
17:31 TimToady oh yeah, that's the old subst in loop bug, dur
17:34 apotheon joined #perl6
17:34 FROGGS m: my @a = <a b c>; for @a { s[(.)] = "$0" }; say @a # it is as easy as this
17:34 camelia rakudo-moar 254c1a: OUTPUT«Use of Nil in string context  in code  at /tmp/PD5OPjMW0F:1␤[ a b]␤»
17:34 FROGGS m: my @a = <a b c>; for @a { my $/; s[(.)] = "$0" }; say @a
17:34 camelia rakudo-moar 254c1a: OUTPUT«[a b c]␤»
17:37 nine [Coke]: wooohooo...many thanks :)
17:37 yoleaux 14:06Z <[Coke]> nine: - please log out and log back in to RT.
17:37 ab5tract_ .tell uruwi regarding Roguelike: I can recommend my own https://github.com/ab5tract/Terminal-Print
17:37 yoleaux ab5tract_: I'll pass your message to uruwi.
17:38 dalek doc: 1a0ecc6 | (David H. Adler)++ | doc/Language/5to6-perlvar.pod:
17:38 dalek doc: Changed C<$*EXECUTABLE_NAME> to C<$*EXECUTABLE-NAME> in 5to6-perlvar.pod
17:38 dalek doc: review: https://github.com/perl6/doc/commit/1a0ecc69f4
17:40 ab5tract_ .tell uruwi it doesn't yet do the 'capture character without printing it to screen' via any API, but you have access to all of the tput commands that you could need, and a peg to hang more from if desired: https://github.com/ab5tract/Terminal-Print/​blob/master/lib/Terminal/Print/Commands.pm6
17:40 yoleaux ab5tract_: I'll pass your message to uruwi.
17:41 ab5tract_ .tell uruwi keep in mind though, that I'm currently working on a rewrite of the internals to use a multidimensional array and a react/whenever loop so that we can have multiple threads modifying the screen independently
17:41 yoleaux ab5tract_: I'll pass your message to uruwi.
17:42 dalek doc: 9413f6c | (Zoffix Znet)++ | doc/Language/5to6-perlfunc.pod:
17:42 dalek doc: Mention different flattening behaviour
17:42 dalek doc: review: https://github.com/perl6/doc/commit/9413f6c222
17:43 ab5tract_ speaking of -- can anyone clarify the syntax for a multi-dimensional ASSIGN-POS ? I tried emulating core, but it didn't work. Unless it needs to be declared as a multi? But I only want to support a three arg form (x,y,v).
17:43 kmel joined #perl6
17:44 kmel joined #perl6
17:45 nine timotimo: a look at Actions.nqp has made quite clear what you want :) What I have not yet figured out is what exactly the $qast argument of the block contains.
17:46 nine timotimo: currently trying: method statement_control:sym<for>($/) { my $xblock := $<xblock>.ast; make QAST::Op.new( :op<p6for>, :node($/), $xblock,); }
17:46 FROGGS we've got a p6for op?
17:46 dha Zoffix - re that commit: how recent is that change? I may need to rebuild rakudo, but I'm not seeing that behavior.
17:47 ZoffixW dha, pretty recent.
17:47 dha Ah, I'll rebuild and see what happens.
17:48 ZoffixW m: my @a = <foo bar>;  my @b.push: @a; my @c.append: @a; say [ @b, @c ].perl
17:48 camelia rakudo-moar 254c1a: OUTPUT«[[["foo", "bar"],], ["foo", "bar"]]␤»
17:48 tokuhirom joined #perl6
17:49 nine FROGGS: I'm creating a p6for op_desugar
17:49 FROGGS aha
17:51 ^elyse^ joined #perl6
17:51 timotimo nine: the thing you get in $qast is the qast node that you've put into the tree with the :op('p6forloop')
17:51 dha Would @a be the first element of @b if @b already had elements
17:52 dha ?
17:52 FROGGS no, push goes at the end
17:52 ZoffixW Nope
17:52 ZoffixW m: my @a = <foo bar>;  my @b = <moo mar mer>; @b.push: @a; say @b.perl
17:52 camelia rakudo-moar 254c1a: OUTPUT«["moo", "mar", "mer", ["foo", "bar"]]␤»
17:52 dha In that case, I find the wording there confusing.
17:53 xfix joined #perl6
17:53 jkva joined #perl6
17:53 dalek doc: 3aa60c6 | (Zoffix Znet)++ | doc/Language/5to6-perlfunc.pod:
17:53 dalek doc: Use clearer wording for flattening behaviour
17:53 dalek doc: review: https://github.com/perl6/doc/commit/3aa60c688b
17:54 timotimo nine: i can't help but notice the p6for seems to be lacking the thing to iterate about
17:55 El_Che is it me or is the modules.perl6.org growing fast?
17:55 ZoffixW El_Che, I notice that as well
17:55 TimToady maybe you're both growing fast
17:55 timotimo it's exploding
17:56 ab5tract_ m: class A { has @!a; method ASSIGN-POS(**@b) { my $v = @b.pop; @!a[ @b[0],@b[1] ] = $v }; method AT-POS(**@b) { @!a[ @b[0], @b[1] ] } }; my $t = A.new; $t[6;7] = "fooo"; dd $t[6;7]
17:56 camelia rakudo-moar 254c1a: OUTPUT«("fooo", Any)␤»
17:56 ZoffixW 15 new modules in the last 4 days
17:57 ab5tract_ TimToady: what's going on with that Any? am I wielding this right?
17:57 FROGGS m: "abc".match(/./); say $/
17:57 camelia rakudo-moar 254c1a: OUTPUT«「a」␤»
17:57 ab5tract_ also, happy belated 6.Birthday! :)
17:57 FROGGS I'm considering breaking that...
17:58 nine timotimo: isn't that $xblock?
17:58 nine timotimo: oh
17:59 timotimo :)
17:59 tony-o_ timotimo: did you get perl6-bench working?
18:00 timotimo tony-o_: i wasn't interested in *that* perl6-bench, i was interested in japhb's :)
18:00 nine timotimo: in the original code $xblock[0] is the thing iterated over and $xblock[1] is the loop body
18:00 timotimo but yeah, i was able to get it working
18:00 nine timotimo: so my assumption was that having $xblock as p6for's child will give me both
18:00 lizmat FROGGS: not returning match objects by default?
18:00 timotimo nine: ah, right.
18:01 timotimo in the desugar, you'll ge texactly that QAST::Op node back
18:01 timotimo and you can work from that
18:01 FROGGS lizmat: not setting $/ for method calls
18:01 FROGGS the method returns the match anyway
18:01 FROGGS and methods should not have side effects usually
18:01 lizmat ah, ok
18:01 lizmat true
18:02 kmel joined #perl6
18:02 FROGGS so, if the refactor I am attempting solves the bugs...
18:02 FROGGS we'll see
18:02 timotimo oof
18:03 nine timotimo: ok, then it will be interesting to find out why exactly it's failing
18:04 timotimo put more debug statements in :D
18:05 timotimo would you like to push it to a branch so i can have a look?
18:05 timotimo though, i'll be fooding soon
18:05 nine What's the NQP equivalent of $foo.WHAT.say?
18:05 arnsholt nqp::say(nqp::what($foo))
18:05 FROGGS nqp::say($foo.HOW.name($foo))
18:06 nine I definitely need to beef up my nqp debugging abilities. I want to contribute more in this area :)
18:06 azawawi joined #perl6
18:06 arnsholt Yeah, what FROGGS said is probably what you actually want
18:06 FROGGS nqp-m: my $foo := nqp::list(); nqp::say(nqp::what($foo)); nqp::say($foo.HOW.name($foo))
18:06 camelia nqp-moarvm: OUTPUT«␤NQPArray␤»
18:06 timotimo debugging nqp stuff is really hard, aye
18:07 FROGGS you get used to it :o)
18:07 dha speak for yourself.
18:08 azawawi hi
18:08 azawawi sergot: ping
18:10 luiz_lha joined #perl6
18:12 azawawi .tell sergot just submitted a PR for the 303 redirection logic + preserve headers for HTTP::UserAgent. Thanks for your time.
18:12 yoleaux azawawi: I'll pass your message to sergot.
18:13 uruwi joined #perl6
18:13 uruwi Ho!!!
18:13 yoleaux 17:37Z <ab5tract_> uruwi: regarding Roguelike: I can recommend my own https://github.com/ab5tract/Terminal-Print
18:13 yoleaux 17:40Z <ab5tract_> uruwi: it doesn't yet do the 'capture character without printing it to screen' via any API, but you have access to all of the tput commands that you could need, and a peg to hang more from if desired: https://github.com/ab5tract/Terminal-Print/​blob/master/lib/Terminal/Print/Commands.pm6
18:13 yoleaux 17:41Z <ab5tract_> uruwi: keep in mind though, that I'm currently working on a rewrite of the internals to use a multidimensional array and a react/whenever loop so that we can have multiple threads modifying the screen independently
18:14 uruwi ありがとう
18:14 uruwi 見ますよ
18:14 uruwi (Thanks, I'll be watching it)
18:15 azawawi .tell Zoffix it was the HTTP code 303 after all that caused the hanging. Now it is working :)
18:15 yoleaux azawawi: I'll pass your message to Zoffix.
18:15 uruwi Wait, I don't know you, do I?
18:16 ZoffixW botsnack
18:16 synbot6 om nom nom
18:16 uruwi I can definitely recall several eaux people on a community I used to frequent long long ago
18:16 nine How can a QAST::Op object have no op method?
18:17 n0tjack joined #perl6
18:17 FROGGS nine: what are you doing? :o)
18:17 FROGGS nine: can you paste it?
18:18 vendethiel joined #perl6
18:18 nine FROGGS: oh, debug error. It's really an NQPMu
18:18 FROGGS ahh
18:20 leont joined #perl6
18:20 s_kilk joined #perl6
18:21 s_kilk hi, I'm new to perl, just started looking at perl6. One thing I can't figure out: what's the difference between an Array and a List? some of the documentation seems to use the terms interchangably
18:22 [Coke] http://docs.perl6.org/type/Array : An Array is a List which forces all its elements to be scalar containers, which means you can assign to array elements.
18:22 FROGGS and you can push/pop/unshift/shift an Array, but not a List
18:23 [Coke] hurm. list says "Indexes into lists and arrays start at 0 by default." ... there is no way to override that, is there?
18:23 s_kilk ah, so a List is immutable? in that the elements of the list are not containers you can assign to?
18:23 punter joined #perl6
18:23 FROGGS I think there was meant to be a way to customize the starting index...
18:24 punter Is Ubuntu 16.04 going to include the Christmas version of Rakudo?
18:24 sergot azawawi: looks awesome but also makes some tests failing
18:24 yoleaux 18:12Z <azawawi> sergot: just submitted a PR for the 303 redirection logic + preserve headers for HTTP::UserAgent. Thanks for your time.
18:24 FROGGS s_kilk: if you put a literal value into a list, you cant assign to it.... if you put a (scalar) variable into a list, you probably can assign to that variable
18:24 sergot .seen azawawi
18:24 yoleaux I saw azawawi 18:15Z in #perl6: <azawawi> .tell Zoffix it was the HTTP code 303 after all that caused the hanging. Now it is working :)
18:24 [Coke] punter: doubtful.
18:25 FROGGS punter: we hope so... looks good so far
18:25 lizmat s_kilk: same for Map vs Hash
18:25 azawawi joined #perl6
18:25 [Coke] But, that's really up to ubuntu.
18:25 azawawi sergot: hi :)
18:25 s_kilk thanks lizmat, not sure I understand containers yet, but that does help
18:25 * [Coke] defers to FROGGS.
18:26 FROGGS [Coke]: hmmm, well, let's get it into debian first and then I will hunt some ubuntu packagers down :o)
18:26 sergot azawawi: hi :))
18:26 FROGGS (with cake or beer or something)
18:26 lizmat m: my $p = a => 42; $p.value = 666   # 42 is not a container
18:26 camelia rakudo-moar 254c1a: OUTPUT«Cannot modify an immutable Int␤  in block <unit> at /tmp/NdR_kY50e2:1␤␤»
18:27 lizmat m: my $a = 42; my $p = a => $a; $p.value = 666   # $a *is* a container
18:27 camelia rakudo-moar 254c1a: ( no output )
18:27 vendethiel ab5tract_: omg, your Terminal::Print looks amazing!
18:27 vendethiel I was looking for something like this at the SPW
18:27 sergot azawawi: looks like your PR makes some tests failing
18:27 sergot azawawi: could you please check that?
18:28 azawawi sergot: yeah im looking at them
18:29 sergot azawawi++
18:29 azawawi sergot: the funny thing is that they are random
18:30 sergot azawawi: hmm, that's weird
18:30 azawawi SSL?
18:32 sergot I've just rerun travis build to see if it happens again
18:32 sergot and it does :(
18:32 nine It works!
18:32 lizmat nine++
18:32 * [Coke] tries to run the terminal print examples and ends up in panda hell.
18:32 vendethiel nine++ lizmat++
18:32 sergot ua-simple.t and ua.t are failing, so this is not SSL problem I think
18:32 sergot brb azawawi
18:33 kmel It Works! => localhost :)
18:33 kmel ;)
18:33 azawawi sergot: sure i will take a look at it
18:34 kmel wasn't this the apache welcome page
18:34 diana_olhovik_ joined #perl6
18:35 azawawi sergot: found it... cloning headers is bad in redirection... my bad... fixing
18:38 dha Ugh. installing Task::Star got shot down by File::Temp failure. Is that something I should be reporting somewhere?
18:39 dalek rakudo/nom: 59d4c40 | ugexe++ | src/core/CompUnitRepo/Local/Installation.pm:
18:39 dalek rakudo/nom: Supress CURLI warnings
18:39 dalek rakudo/nom:
18:39 dalek rakudo/nom: If the manifest is missing precomp info for a module it shouldn't warn each time one is loaded/used in a script, or complain about trying to use a value from an uninitialized hash key.
18:39 dalek rakudo/nom:
18:39 dalek rakudo/nom: This carries over the mistaken `:has_precomp` argument
18:39 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/59d4c40acc
18:39 moritz dha: in the File::Temp repo's issue track
18:39 dalek rakudo/nom: 3f3e032 | lizmat++ | src/core/CompUnitRepo/Local/Installation.pm:
18:39 dalek rakudo/nom: Merge pull request #552 from ugexe/patch-4
18:39 dalek rakudo/nom:
18:39 dalek rakudo/nom: Supress CURLI warnings
18:39 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/3f3e0328d7
18:39 [Coke] panda is broken.
18:39 [Coke] ------>     my $panda = Panda.new(:ecosystem(⏏make-d​efault-ecosystem($prefix)));
18:39 [Coke] Calling make-default-ecosystem(Str) will never work with declared signature ()
18:39 [Coke] ^^ few dozen of those.
18:40 uruwi Can't install any modules with Panda on cygwin. http://pastebin.com/XiNY0yqs
18:40 dha Oh. so is this Panda's fault or File::Temp's fault?
18:41 uruwi I have no idea.
18:42 * [ptc] has seen that in association with File::Temp more than once
18:42 sergot azawawi: why is it bad?
18:43 dha Huh. also, the list of modules in Task::Star's META.info file on modules.perl6.org does not include File::Temp...
18:43 uruwi ptc elaboration?
18:43 dha Ah. It's installing it in order to install p6doc.
18:44 [ptc] uruwi: I wish I had the tracebacks, but unfortunately I don't.  Upon running e.g. ./rebootstrap.pl or simply installing the modules for perl6-examples, I'd noticed File::Temp failing
18:44 azawawi sergot: it needs to exclude some headers as it broke some tests. I will revisit it.
18:44 azawawi sergot: the main problem was the 303 redirection
18:44 uruwi Is it the same error message?
18:44 [ptc] uruwi: then when I tried again later, it seemed to work (usually I'd also updated the full rakudo stack, panda etc., thus it's very hard to nail down where the problem was
18:45 [ptc] uruwi: sorry mate, I can't remember
18:45 uruwi What version of Rakudo are you running?
18:45 uruwi 2015.09?
18:45 ugexe 303 redirection used to work correctly
18:45 [ptc] uruwi: This is perl6 version 2015.09-363-g70eea19 built on MoarVM version 2015.09-81-g41002db
18:46 telex joined #perl6
18:46 uruwi Strangely enough, mine doesn't provide that much detail
18:47 uruwi same problem in cmd.exe
18:47 sergot azawawi: well, travis shows it still fails :(
18:47 uruwi So it happens with Windows, period.
18:47 [ptc] one thing I have had reasonably consistently is "Could not find file 'Build.pm' for module Build.pm" when installing Linenoise
18:48 [ptc] this is also something which is a bit flappy
18:51 uruwi Thinking of giving up on Cygwin and using full-on Linux instead...
18:52 uruwi How does no one else have this error, by the way?
18:53 uruwi Some person *has* to be running Perl6 on Windows
18:53 azawawi sergot: it sometimes hangs on 08-ua.t . Is that normal?
18:53 kmel m: say 'What is windows?'
18:53 camelia rakudo-moar 254c1a: OUTPUT«What is windows?␤»
18:54 El_Che uruwi: you got rakudo built on cygwin?
18:54 uruwi No, I'm using the regular windows build
18:54 uruwi Panda doesn't work from cmd either so this probably doesn't have anything to do with the shell
18:54 El_Che ah ok (I tried and failed :) )
18:55 [ptc] uruwi: afaik FROGGS and jnthn both use Windows
18:56 ugexe i run perl6 on windows
18:56 FROGGS and panda does work with latest star release (2015.09)
18:56 azawawi sergot: i think it is now a problem with https://github.com/perlpilot/p6-File-Temp/
18:56 uruwi Hmm
18:56 FROGGS I've not checked a newer version of rakudo
18:57 El_Che uruwi: starting a vm with windows 7, cygwin and rakudo
18:57 uruwi Oh
18:57 FROGGS why cygwin?
18:57 uruwi I use Win10 :P
18:57 ugexe im pretty sure its file::temp or file::directory::tree, and star is bundled what a specific version that still works
18:58 azawawi sergot: File::Temp is actually failing its tests with the latest batch of commits
18:58 jkva joined #perl6
18:58 * azawawi wishes for panda module version dependency
19:01 El_Che uruwi: same result in the Win VM
19:01 El_Che uruwi: same error message
19:01 El_Che uruwi: latest rakudo installed
19:02 uruwi Wait the modules in Rakudo* don't auto update do they.
19:02 uruwi I infer not.
19:03 El_Che however panda install Task::Star completed without errors (although it probably did nothing)
19:04 uruwi Succeeds on my machine too
19:06 sergot azawawi: looks so.. hmm.
19:06 El_Che uruwi: skipping tests (--notests) builds and installs the modules succesfully
19:06 jkva joined #perl6
19:06 ugexe zef works fwiw
19:06 ugexe task::star probably isnt trying to install a new File::Temp if you already have panda installed
19:07 dalek rakudo/nom: 17987c6 | lizmat++ | src/core/Exception.pm:
19:07 dalek rakudo/nom: Add 'type' to X::Adverb::Slice
19:07 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/17987c618d
19:07 dalek rakudo/nom: 3cdaf9a | lizmat++ | src/core/Any.pm:
19:07 dalek rakudo/nom: Handle unhandled named parameters to zen slice
19:07 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/3cdaf9a2be
19:07 uruwi Panda still tries to test even with --notests
19:07 azawawi sergot: so basically my headers fix was ok :)
19:07 El_Che panda --notests install ...
19:08 El_Che the params matching is a little basic it seems
19:08 dalek roast: 29f5477 | lizmat++ | S32- (2 files):
19:08 ugexe the thing is, we were cloning http::response like a week ago and it was removed
19:08 dalek roast: Adapt test for zen slice with unknown adverb
19:08 dalek roast: review: https://github.com/perl6/roast/commit/29f547712d
19:08 uruwi oh --notests comes before install
19:08 ugexe and its starting to get ugly with http::request generating other requests and modifying attributes instead of the transport/ua handling that
19:09 ugexe whereas with the cloned history the transport/ua can handle that stuff
19:09 punter left #perl6
19:09 rotcpy joined #perl6
19:10 llfourn joined #perl6
19:10 sergot ugexe: we're talking about travis, it should install the latest File::Temp, right?
19:11 jkva joined #perl6
19:13 ugexe yes. but its not testing it so you cant see that its broke from your travis output
19:14 azawawi m: $*SPEC.catpath('A', 'B');
19:14 camelia rakudo-moar 254c1a: OUTPUT«Too few positionals passed; expected 4 arguments but got 3␤  in block <unit> at /tmp/Q5pTBq7vrM:1␤␤»
19:14 timotimo a few push replaced with append and bench is a bit happier
19:15 azawawi m: $*SPEC.catpath('', 'A', 'B');
19:15 camelia rakudo-moar 254c1a: ( no output )
19:15 nine timotimo: the desugar op breaks loop control statements and for's return value. Will need some restructuring to get that working
19:15 azawawi this commit broke File::Temp, https://github.com/perlpilot/p6-File-Temp/com​mit/626af3f740a7e7059bb650c05dd3116ed6056429
19:16 sergot ugexe: oh, ok, right
19:18 uruwi Different error message though
19:18 timotimo nine: damn :(
19:18 timotimo jnthn: do you have a hunch why a desugar for for loops in rakudo would break things like nine just described?
19:19 nine timotimo: I guess it breaks the return value because the QAST::Want must be the outermost op
19:20 nine No idea where the "control exception without handler" comes from though
19:21 timotimo so are you going to push it up so i can have a quick look?
19:23 nine timotimo: current work in progress of trying to push the QAST::Want to the top: https://gist.github.com/niner/c08db40160de90a7c46b
19:23 rindolf joined #perl6
19:25 timotimo something's wrong with the perl5 portion of perl6-bench, it seems like o_O
19:26 azawawi sergot: anyway, https://github.com/azawawi/perl6-selenium-webdrive​r/blob/master/examples/test_selenium_phantomjs.p6 is now working with the fix and it is controlling phantomjs :)
19:27 azawawi sergot: thanks
19:28 timotimo oh!
19:28 timotimo could have something to do with the timings files for nqp-jvm being 260 megabytes big?
19:29 timotimo huh. the ones in nqp-moar are just half that size
19:30 timotimo japhb: anything we can do to make this less of a problem?
19:30 [Coke] ooh, p6 selenium
19:31 azawawi [Coke]: yup :)
19:34 sergot azawawi++ : awesome! thank you! :)
19:34 * azawawi &
19:34 johnjohn101 joined #perl6
19:34 johnjohn101 hi perl six peeps
19:34 sergot johnjohn101 hi \o
19:35 PerlJam johnjohn101: greetings
19:35 johnjohn101 what ever happened to the perl 6 on .net project?  seems like it hasn't had an update in a while.
19:36 PerlJam johnjohn101: champions are hard to come by.
19:36 johnjohn101 i understand
19:37 moritz in the past, I've wondered if writing a .NET backend for NQP was a good way to learn C#
19:37 moritz I came to the conclusion that it likely isn't :-)
19:37 dalek rakudo/nom: ee96fc1 | lizmat++ | src/core/Any-iterable-methods.pm:
19:37 dalek rakudo/nom: Make List.first/first-index 2x as fast
19:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ee96fc1190
19:38 pmurias_ joined #perl6
19:40 * [Coke] pokes the one guy he knows that works at microsoft about it.
19:42 masak m: class A { method m { "A" } }; class B is A { method m { "B"; } }; class C is A { method m { "C" } }; class D is B is C {}; say D.new.m
19:42 camelia rakudo-moar 3cdaf9: OUTPUT«B␤»
19:49 masak http://spec.perl6.org/ doesn't take me to http://spec.perl6.org/specification/ :(
19:49 masak it takes me to the main page
19:49 tokuhirom joined #perl6
19:50 timotimo http://t.h8.lv/p6bench/201​5-10-14-nqp_shootout.html - this was also supposed to have nqp-jvm in it, but it already took super long to do only the nqp-moar bits
19:50 timotimo i'm going to re-run the analyze command with also jvm parts
19:59 dalek rakudo/nom: dd15d70 | lizmat++ | src/core/Supply.pm:
19:59 dalek rakudo/nom: Give Supply.squish same semantics as List.squish
19:59 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/dd15d70464
20:00 PerlJam isn't http://spec.perl6.org/specification/ kind of a misnomer anyway?
20:00 dalek roast: 7ed4852 | lizmat++ | S17-supply/squish.t:
20:00 dalek roast: Add tests for 1 elem semantics of Supply.squish
20:00 dalek roast: review: https://github.com/perl6/roast/commit/7ed48521e1
20:03 larion joined #perl6
20:03 timotimo well, it does link to the test suite
20:04 ^elyse^ joined #perl6
20:07 timotimo http://t.h8.lv/p6bench/2015-10​-14-nqp_shootout_with_jvm.html
20:07 s_kilk joined #perl6
20:08 hohoho joined #perl6
20:11 uruwi Ho!
20:11 timotimo Hey!
20:12 lizmat timotimo: that's pretty bad, no ?
20:12 [Coke] have a 750 line or so perl6 script. add a multi MAIN("TEST") near a sub I'm trying to debug so I can quickly invoke a test without changing the rest of the code.
20:12 timotimo what is?
20:12 lizmat that shootout
20:12 lizmat the while empty ones
20:13 timotimo why do you think so?
20:13 PerlJam lizmat: bad for moar compared to jvm?
20:13 lizmat well, the nqp::jvm does 4x more ?
20:13 FROGGS head seems identical to 2015.07 and 2015.09
20:13 timotimo when it warms up, yeah
20:13 FROGGS lizmat: they payed more ppl to optimize that vm
20:13 FROGGS paid*
20:14 lizmat hehe: I bet there's some dieselgate in there as well  :-)
20:14 FROGGS one could say that the JVM is optimized to do nothing :o)
20:14 FROGGS hehe
20:14 gugod or collecting garbage
20:15 dalek roast: 364b771 | usev6++ | S (6 files):
20:15 dalek roast: Unfudge some tests for rakudo.jvm
20:15 dalek roast: review: https://github.com/perl6/roast/commit/364b7717ac
20:15 timotimo i'm not sure those benchmarks are worth anything, tbh :)
20:15 timotimo in a few hours i'll have some rakudo benchmarks
20:15 lizmat looking forward to that :-)
20:16 TEttinger joined #perl6
20:18 jnthn Such benchmarks are mostly a good test of how good your optimizer is at getting rid of "junk" and spitting out concise machine code. So, some value. :)
20:18 [Coke] if I have a parallel pipeline (e.g. something with .race.map.***) - what can I use to stop the pipeline? jnthn's example uses map...
20:18 timotimo right
20:18 jnthn 'night, #perl6
20:18 timotimo gnite jnthn!
20:19 lizmat gnight jnthn
20:24 [Coke] er, uses 'max'
20:25 dalek roast: 4a8cdeb | usev6++ | S02-literals/numeric.t:
20:25 dalek roast: Add another test for RT #124559
20:25 dalek roast: review: https://github.com/perl6/roast/commit/4a8cdebd44
20:25 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=124559
20:33 dha List.eager doc says "Evaluates all elements in the list eagerly, and returns them as a list." What does that mean?
20:34 dha specifically, what does it mean to evaluate something "eagerly"?
20:34 colomon joined #perl6
20:37 PerlJam some elements may be a "thunk" that generates list elements rather than actually be elements.  Eager evaluation will cause it to generate all of the values it can.
20:38 pink_mist that may lead to out of memory problems with infinite lists =)
20:38 dha ok. The docs should probably explain that. I'm not convinced I'm the person to do that, however.
20:39 psch "Eager evaluation" is a wikipedia article
20:39 psch that's not to say we don't need it in the docs, though
20:46 dha Ok, I'm confused. What is going on here: say (1, 2, 3).reduce: * - *;
20:46 dha This results in -4, but I'm not clear why.
20:46 psch m: say 1 - 2 - 3
20:46 camelia rakudo-moar dd15d7: OUTPUT«-4␤»
20:46 psch m: [-] 1, 2, 3
20:46 camelia rakudo-moar dd15d7: ( no output )
20:46 psch m: say [-] 1, 2, 3
20:46 camelia rakudo-moar dd15d7: OUTPUT«-4␤»
20:47 dha ok. I get that. What are the stars doing?
20:47 psch just a different form of specifiying the operator or function
20:47 psch '* - *' is a two-ary function that substracts the second argument from the first
20:47 dha and where is that documented?
20:48 dha Oh, placeholders. damn that looks weird.
20:48 psch just like &infix:<-> is a two-ary function that substracts the second argument from the first
20:48 dha C<*> means too many things, dammit.
20:48 PerlJam aye, there is a little of that.
20:48 psch m: say *.WHAT
20:49 camelia rakudo-moar dd15d7: OUTPUT«(Whatever)␤»
20:49 PerlJam same with { }
20:49 psch Whatever should be somewhat documented, iirc
20:49 psch yeah, pretty thoroughly even, it looks like
20:49 RabidGravy http://doc.perl6.org/type/Whatever
20:52 [Sno] joined #perl6
20:53 dha Having to read a section of an entirely different doc page seems problematic for an example. :-/
20:53 skids .oO(maybe .last-index should be renamed to .hind-index so we can have .hind which would not conflict with "last" and kinda rhymes with "find" anyway.)
20:54 psch dha: you're doc'ing .reduce?
20:54 * PerlJam would rather all of the *-index routines go away first
20:54 dha I'm looking through List.pod for where I can put in examples, and reading the ones that have examples already as I go.
20:54 dha That one caught my eye.
20:55 psch ah, i see
20:55 lizmat PerlJam: how would they be called then ?
20:57 PerlJam lizmat: something like   grep :index, { ... }, @stuff;
20:57 skids .first(:k)?
20:58 * masak .oO( .grep :kv, .grep :v )
20:58 PerlJam in any case, some modifier that say we want the ordinals rather than the values.
20:59 lizmat well, I'm fine with that: that's how I originally implemented them, I think
20:59 masak PerlJam: also, some other modifier that says we want the mutable container rather than the value in the container
20:59 lizmat but then the whole thing about named parameters are a smell came about
20:59 PerlJam masak: that would be nice too
21:00 lizmat masak: you get the container already if they are containers
21:00 kmel joined #perl6
21:00 niceperl joined #perl6
21:01 kmel i am reading http://doc.perl6.org/language/control
21:01 kmel can someone explain what does 'with' actually do?
21:02 PerlJam kmel: with $thing { ... }   is like if defined($thing) { ... }
21:02 masak m: my @a = 1, 2, 3; @a.grep(2)[0] = 7; say @a
21:02 camelia rakudo-moar dd15d7: OUTPUT«Cannot modify an immutable Int␤  in block <unit> at /tmp/lTm_2l8Fws:1␤␤»
21:03 masak lizmat: don't look like it
21:03 lizmat masak PerlJam : I will let TimToady rule on that, and make changes accordingly
21:03 kmel oh ok. thanks PerlJam
21:03 skids kmel: like if defined($thing) { ... }($thing) actually.
21:04 skids or well if defined($thing) {{ ... }($thing)}
21:04 PerlJam lizmat: yeah, he's the best person to ferret out any underlying commonality and unify it into a syntax that makes sense
21:04 lizmat masak: oops, looks like they're some "is raw"'s missing
21:04 kmel thanks skids
21:05 masak lizmat: tbh, I think "just the values" is an excellent default for .grep
21:05 masak lizmat: 99% of the time I use .grep, that's what I expect
21:05 PerlJam lizmat: and I wouldn't hold up 6.christmas for it in any case.  :)
21:05 masak probably great for optimization, too
21:05 lizmat masak: what is ?
21:05 masak returning just the values, not their containers
21:06 masak it's one less level of indirection to worry about
21:06 PerlJam since you already have the containers, just returning them seems already "optimized"
21:06 lizmat m: my @a = 1, 2, 3; $_ = 7 for @a.grep(2); say @a
21:06 camelia rakudo-moar dd15d7: OUTPUT«[1 7 3]␤»
21:07 skids Depends entirely on use case I would say.
21:07 lizmat masak: I think this is really a problem of the iterator interface
21:07 lizmat your example did a push-all, which pushes values into a buffer
21:08 lizmat the for loop does pull-one, which gets the container
21:08 lizmat not sure how to solve that
21:09 dalek doc: 90ebc3c | (David H. Adler)++ | doc/Type/List.pod:
21:09 dalek doc: Added examples for grep-index, first-index, and last-index
21:09 dalek doc: review: https://github.com/perl6/doc/commit/90ebc3c9e3
21:09 dalek doc: 1c87ed7 | RabidGravy++ | doc/Type/List.pod:
21:09 dalek doc: Merge pull request #163 from dha/master
21:09 dalek doc:
21:09 dalek doc: Added examples for grep-index, first-index, and last-index
21:09 dalek doc: review: https://github.com/perl6/doc/commit/1c87ed7038
21:09 dha I assume that there is a difference between Range.list and Range.flat, but they are documented identically. Or are they in fact  just synonyms?
21:09 skids m: my @a = 1, 2, 2, 3; @a.grep(2)[0..1] = 7,8; say @a;
21:09 camelia rakudo-moar dd15d7: OUTPUT«Unexpected named parameter 'eagerize' passed␤  in block <unit> at /tmp/fGpowvDynL:1␤␤»
21:10 diana_olhovik___ joined #perl6
21:10 * skids decommute
21:10 llfourn joined #perl6
21:10 PerlJam dha: I think .list gives you an actual List, but .flat gives you a Seq
21:11 dha According to the docs, they both return a List.
21:11 dha and they both just say "Generates the list of elements that the range represents."
21:12 lizmat .tell jnthn couldn't we just not put a $no-sink in IterationBuffer.push ?
21:12 yoleaux lizmat: I'll pass your message to jnthn.
21:12 masak lizmat: I see. hm.
21:15 RabidGravy dha, it's entirely possible the docs weren't updated post-GLR
21:15 dha That's what it looks like, yes.
21:17 dha So that should be changed to "returns Seq:D"? And the doc perhaps changed to "Generates a sequence of elements..."?
21:18 lizmat well, not sure we want to expose Seq so clearly ?
21:19 PerlJam m: (^10).flat.WHAT.say
21:19 camelia rakudo-moar dd15d7: OUTPUT«(Seq)␤»
21:19 PerlJam It's already exposed to some extent
21:20 dha It certainly seems odd to have two methods that do the same thing. Also, since the function documentation has signatures, they should probably be correct...
21:21 margeas joined #perl6
21:22 dha I suppose you could just omit the return type from .flat, but that strikes me as inconsistent and weird. But maybe that's just me.
21:22 s_kilk The SCGI module appears to be broken currently, because its dependency Netstring doesn't build: http://pastebin.com/KaKrK7ku
21:23 s_kilk Which means that Web::App::Dispatch is broken too, I think, or at least the examples are. will dig in a bit further
21:24 PerlJam s_kilk: ping supernovus harder ... there's already a PR to fix it from back in May
21:24 dalek rakudo/nom: d578a56 | lizmat++ | src/core/Supply.pm:
21:24 dalek rakudo/nom: Introduce Supply.throttle
21:24 dalek rakudo/nom:
21:24 dalek rakudo/nom: So you can throttle your supplies by number of elems per time-unit
21:24 dalek rakudo/nom: (in seconds, or parts thereof)
21:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/d578a56c61
21:24 PerlJam ptc++  (I'm pretty sure he did bunches of PRs to add unit declarations to most modules in the ecosystem)
21:25 s_kilk Thanks PerlJam, I'll open an issue on the SCGI repo. I'm new to perl6 (new to perl actually), and have been digging around for a place to start contributing, looks like I found it :)
21:26 PerlJam s_kilk: alternatively, if no one can contact supernovus, someone can fork his repo, add the fix, then replace his repo in the ecosystem with their own.
21:27 PerlJam .seen supernovus
21:27 yoleaux I saw supernovus 30 Sep 2015 21:08Z in #perl6: <supernovus> I changed my build script so instead of using bootstrap.pl it uses rebootstrap.pl and it appears to work properly now.
21:28 PerlJam He was here a couple of weeks ago apparently.
21:28 dha Can I assume the examples for .floor, .ceiling, etc. from Cool.pod should be appropriate for, for instance, Real.pod?
21:28 s_kilk I'm tempted to do that actually, what's the current state of HTTP/Web stuff in Perl6? As a newcomer it seems that supernovus is responsible for most of the web projects in the ecosystem, is he pulling the cart alone? Is there a group of contributors I could speak to about me pitching in? I'd be keen on even just doing doc work
21:29 dha (Cool has lots of examples, Real, not so much)
21:30 RabidGravy s_kilk, there's a few people sparked up recently - I think that stuff is going to grow a bit
21:30 s_kilk Cool, I bet a handful of people could make some real progress with the right co-ordination
21:31 RabidGravy Crust came in just recently and those guys have been patching a load of other stuff which is cool
21:31 s_kilk The Web::App::Dispatch framework looks really promising, but it's currently broke.
21:32 s_kilk Ah, thanks RabidGravy, I hadn't seen Crust before.
21:36 RabidGravy I think in the fullness of time something really good will pop up, someone clever will have the urge to write the best web framework ever and actually have the time to try
21:37 colomon joined #perl6
21:37 ZoffixMobile joined #perl6
21:37 virtualsue joined #perl6
21:38 ZoffixMobile I really wish sri of the Perl 5's Mojolicious fame would finally decide to do a Perl 6 port. So far, they're unsure/need funding.
21:39 AlexDaniel joined #perl6
21:39 RabidGravy I don't, I want someone new to write something new
21:39 ZoffixMobile Even aside from the web framework, I miss the Mojo Toolkit in P6 (Mojo::DOM in particular)
21:40 ZoffixMobile RabidGravy, how come? sri knows the specs like the back of their hand and has creating two very successful web frameworks on their CV.
21:41 masak ZoffixMobile: in the meantime, I'm pretty sure you can use Mojo from Inline::Perl5
21:41 ZoffixMobile masak, yeah, you can (I tried).
21:42 sufrostico joined #perl6
21:42 sufrosti1o joined #perl6
21:43 RabidGravy ZoffixMobile, that doesn't mean someone else can't make something better and more Perl 6 like :) It would be really nice if someone who wasn't an existing Perl 5 author did that :)
21:44 ZoffixMobile But I wouldn't make a P6 module that has Inline::Perl5 and a Perl 5 module as a prereq; which is my more accurate feeling about missing Mojo :p
21:45 pierrot joined #perl6
21:47 ZoffixMobile RabidGravy, that's definitely true. :)
21:47 tony-o_ ZoffixMobile: i'd be interested in feedback on https://github.com/tony-o/perl6-hiker
21:47 ggoebel joined #perl6
21:48 s_kilk so, what would it take to replace the Netstring module in the ecosystem with a forked version?
21:48 ZoffixMobile .tell Zoffix check out https://github.com/tony-o/perl6-hiker when you're sober
21:48 yoleaux ZoffixMobile: I'll pass your message to Zoffix.
21:48 ZoffixMobile :p
21:48 tony-o_ lol ZoffixMobile++
21:48 leont integrating perl6's async power into a framework is where it starts being truly interesting to me, copying existing things is so-so
21:49 RabidGravy I'm with leont :)
21:49 tony-o_ hiker uses the async stuff
21:50 ZoffixMobile s_kilk, I suggest editing META.list in the https://github.com/perl6/ecosystem, while doing an @ mention of the original author of the intent to overtake. Wait 24 hours. Merge your overtake
21:51 tokuhirom joined #perl6
21:51 s_kilk Thanks ZoffixMobile, I've forked the project and patched it here: https://github.com/ShaneKilkelly/perl6-netstring . it seems to work, i think.
21:52 RabidGravy I'm taking a more relaxed approach to Cache::Memcached - I sent a PR to make it actually work in May, emailed him yesterday and will wait a week or two
21:53 ZoffixMobile Oh sure
21:53 ZoffixMobile Sorry, I assumed s_kilk attempted to contact the original author with no avail.
21:54 hohoho joined #perl6
21:55 s_kilk yeah, I think Netstring may be more broke than I realized, it installs now but crashes the server with "Invalid netstring stream data.". Not sure which side of the Web::App / SCGI / Netstring boundary is at fault there.
21:56 s_kilk sorry, I thought if it was an easy fix that it might be good to take some pressure off of supernovus. I've just wandered in off the street really, I'm quite taken with perl6, and was starting to play around with some of the web modules.
21:57 RabidGravy well, sounds like Netstring may or may not be properly tested
21:58 RabidGravy I'll put this on my TODO list for tomorrow
21:58 RabidGravy (it's actually more of a stack than a list nowadays but hey)
22:00 kid51 joined #perl6
22:00 dalek roast: ef6d4ee | lizmat++ | S17-supply/throttle.t:
22:00 dalek roast: Add basic Supply.throttle tests
22:00 dalek roast: review: https://github.com/perl6/roast/commit/ef6d4ee9a1
22:01 RabidGravy right bed time
22:01 lizmat gnight RabidGravy
22:01 ZoffixMobile s_kilk, yeah, Perl 6 is pretty cool :) If you haven't already, submit a PR to supernovus that fixes the issues. If they don't merge it soon enough, try to contact them about the merge, and if that fail, do the ecosystem editing stuff I suggested :) Sorry, my fault for not being clearer earlier :)
22:01 ZoffixMobile gnight
22:03 s_kilk Thanks ZoffixMobile, it looks like someone has had the same PR open for the last few months. I've added a comment to the PR too.
22:04 * dha sees the Document All Modules message and looks for modules in need of documentation.
22:06 dalek rakudo/nom: 271e84c | lizmat++ | t/spectest.data:
22:06 dalek rakudo/nom: Run the Supply.throttle tests
22:06 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/271e84c731
22:06 ZoffixMobile Thinking more of this, we probably need a more official guideline for module overtakes. I can imagine people getting touchy about someone just taking over their work willy-nilly
22:07 ZoffixMobile dha, make a list! :)
22:07 dha On what basis?
22:07 dha That is, one what basis would I add something to this hypothetical list? :-)
22:07 bjz joined #perl6
22:08 _sri ZoffixMobile: thanks for the vote of confidence :)
22:09 ZoffixMobile _sri, :) Join us! We have cookies! :p
22:10 _sri RabidGravy: it's not like perl6 is full of new programming paradigms that could result in radically new ways of developing web apps
22:10 _sri (and that doesn't mean it's bad at all!)
22:10 dha We do? cookies++
22:10 ZoffixMobile dha, does the documentation included with the module tell everything about how to use that module? The "this is a Port of Perl 5 blah" doesn't cut it. :)
22:10 dha *nod* ok.
22:11 dha Ooh. But now I have to get going. Something for tomorrow, I guess.
22:11 dha later, all.
22:11 ZoffixMobile \o
22:12 dalek doc: 913b7b8 | lizmat++ | doc/Type/Supply.pod:
22:12 dalek doc: Document Supply.throttle
22:12 dalek doc: review: https://github.com/perl6/doc/commit/913b7b8371
22:12 dalek perl6-most-wanted: 8849c04 | (Nova Patch)++ | most-wanted/ (2 files):
22:12 dalek perl6-most-wanted: add wanted i18n/NLP
22:12 dalek perl6-most-wanted: review: https://github.com/perl6/perl6​-most-wanted/commit/8849c04091
22:12 lizmat good night, #perl6!
22:13 ZoffixMobile night, lizmat
22:15 masak 'night, lizmat
22:16 lizmat .tell daxim if you can give me a paragraph about why my statement re ß in the P6W was wrong, I will add it as an erratum next week
22:16 yoleaux lizmat: I'll pass your message to daxim.
22:16 lizmat really afk&
22:17 dalek perl6-most-wanted: e1c1feb | (Nova Patch)++ | most-wanted/ (2 files):
22:17 dalek perl6-most-wanted: i18n and NLP headers
22:17 dalek perl6-most-wanted: review: https://github.com/perl6/perl6​-most-wanted/commit/e1c1feb9ee
22:19 leont _sri: I've seen few frameworks/languages handle concurrency and async well at the same time
22:19 leont few meaning "I haven't seen it, though it might exist, it must have not broken through"
22:20 [ptc] PerlJam: thanks :-)  Yes, I did.  I also wanted to fix kebab-case issues as well, but time sort of ran out on me
22:20 _sri leont: phoenix is pretty phenomenal
22:21 _sri i don't think anything in perl6 will ever be able to compete with that honestly
22:21 _sri (which again, is not bad! ;p)
22:21 _sri due to the erlang vm and otp
22:23 _sri play framework should be closer to what a perl6 framework will ultimately look like i guess, it's pretty good at concurrency/parallelism
22:24 leont I should look into that
22:26 _sri but yea, if you wanted to make concurrency/parallelism your big selling point you'll have to compete with phoenix/elixir, which would be an uphill battle
22:26 _sri it's not even such a radically innovative framework, it just takes advantage of all the erlang infrastructure
22:27 leont Only worked with erlang once, but yeah that was impressive
22:28 johnjohn101 erlang is ok but seems to me specialized usage
22:29 * _sri would focus more on expressiveness, like in perl5
22:32 _sri totally take advantage of the awesome concurrency features (love await and whenever), but have perl-ish stuff like one-liner web apps
22:33 leont Fair point
22:35 ugexe m: use trace; my $x = { :ok(1), :nok(0) }; my @a; @a.append($x); say @a.perl
22:35 camelia rakudo-moar 271e84: OUTPUT«2 (/tmp/WXFbbkbxjB:1)␤my $x = { :ok(1), :nok(0) }␤9 (/tmp/WXFbbkbxjB:1)␤my @a␤10 (/tmp/WXFbbkbxjB:1)␤@a.append($x)␤11 (/tmp/WXFbbkbxjB:1)␤say @a.perl␤[-> ;; $_? is raw { #`(Block|63458872) ... }]␤»
22:35 ugexe m: my $x = { :ok(1), :nok(0) }; my @a; @a.append($x); say @a.perl
22:35 camelia rakudo-moar 271e84: OUTPUT«[{:nok(0), :ok(1)},]␤»
22:37 uruwi joined #perl6
22:38 uruwi stupid mouse broke :(
22:52 AlexDaniel say '㎉'
22:52 AlexDaniel m: say '㎉'
22:52 camelia rakudo-moar 271e84: OUTPUT«㎉␤»
22:52 tokuhirom joined #perl6
22:52 AlexDaniel does anybody know any symbol with more than 4 letters?
22:52 AlexDaniel ㌲ I'm not sure if japanese counts
22:54 AlexDaniel but if yes, then I'm searching for a thing with more than 5… ㌮
22:56 AlexDaniel ㌖ or 6…
22:58 masak pretty sure those Japanese things you pasted don't contain any letters... :)
23:00 AlexDaniel masak: well, the last one is “kiromeetoru” written with katakana, if I'm not mistaken
23:01 TimToady funny they have that as well as ㎞
23:01 skids joined #perl6
23:05 timotimo http://t.h8.lv/p6bench/2015-10-15-r​akudo_moar_since_last_release.html - i don't really trust these measurements for some reason ...
23:05 TimToady by some reckoning there's nine letters in 𒐎
23:05 cognominal joined #perl6
23:07 AlexDaniel TimToady: I wonder if there is any font that supports it
23:08 kmel joined #perl6
23:08 kmel left #perl6
23:09 kmel joined #perl6
23:10 masak 'night, #perl6
23:10 AlexDaniel split_string_constant, split_string_regex – ouch!
23:12 * timotimo tries to get some rest and get well
23:12 TimToady AlexDaniel: shows up on my screen just fine :)
23:13 TimToady timotimo++
23:13 llfourn joined #perl6
23:14 TimToady AlexDaniel: but maybe that's because:
23:14 TimToady > ls .fonts/
23:14 TimToady Aegean.ttf      Aegyptus_R.ttf  Analecta.ttf   Gardiner.ttf   HAN NOM B.ttf  Musica.ttf   Unidings.ttf
23:14 TimToady Aegyptus_B.ttf  Akkadian.ttf    Anatolian.ttf  HAN NOM A.ttf  Maya.ttf       Symbola.ttf
23:23 EricMobiel joined #perl6
23:30 ShimmerFairy m: say "𒐎".uniname
23:30 camelia rakudo-moar 271e84: OUTPUT«CUNEIFORM NUMERIC SIGN NINE DISH␤»
23:32 AlexDaniel m: say 'ﯹ'.uniname
23:32 camelia rakudo-moar 271e84: OUTPUT«ARABIC LIGATURE UIGHUR KIRGHIZ YEH WITH HAMZA ABOVE WITH ALEF MAKSURA ISOLATED FORM␤»
23:34 AlexDaniel OK, it seems like this one is 18 characters or so
23:34 AlexDaniel m: say 'ﷺ'.uniname
23:34 camelia rakudo-moar 271e84: OUTPUT«ARABIC LIGATURE SALLALLAHOU ALAYHE WASALLAM␤»
23:46 tokuhirom joined #perl6
23:51 gfldex m: my constant @a := map { .chr.chars }, 1..33500; say [max] @a;
23:51 camelia rakudo-moar 271e84: OUTPUT«(signal SEGV)»
23:51 gfldex i think that's my first segfault rakudobug :)
23:53 gfldex sadly i cant think of a version that doesnt use map
23:54 gfldex r: my constant @a := map { .chr.chars }, 1..33500; say [max] @a;
23:54 camelia rakudo-moar 271e84: OUTPUT«(signal SEGV)»
23:54 camelia ..rakudo-jvm 271e84: OUTPUT«1␤»
23:56 ugexe m: my constant @a := (1..33500)>>.chr>>.chars; say [max] @a;
23:56 camelia rakudo-moar 271e84: OUTPUT«(signal SEGV)»
23:57 bigrattus joined #perl6
23:58 luiz_lha joined #perl6
23:58 [particle]1 joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo