»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
lizmat re packages, you might want to check out nxadm.github.io/rakudo-pkg/ 00:00
00:01 astronavt joined, silug joined 00:06 astronavt left 00:13 rindolf left 00:18 nadim_ left 00:43 patrickz_ joined 00:45 Cabanossi left
Zoffix "2017 Perl 6 Advent Day 2 – Perl 6: Sigils, Variables, and Containers": perl6advent.wordpress.com/2017/12/...ontainers/ 00:45
00:46 patrickz left 00:47 Cabanossi joined 00:49 margeas left
lizmat Zoffix: weird line ending at "implies (NL) a type-constraint" 00:50
Zoffix Thanks. Fixed. 00:54
00:59 evalable6 joined, ChanServ sets mode: +v evalable6
lizmat "$ = $ for $primes" ?? am I missing something ? 01:00
Zoffix: ^^^
Zoffix lizmat: dunno. Why do you think you're missing something? The idea is nqp::null for $primes
lizmat shouldn't that be "$_ = $_ for $primes" ? 01:02
01:02 astronavt joined
Zoffix lizmat: doesn't matter. The problematic bit is the `|$primes` iteration itself, not what the for loop does 01:02
I just don't know anything better than `$ = $` for a null operation that doesn't emit a warning 01:03
lizmat then maybe "Nil for #primes" ?
*$primes
Zoffix m: Nil for <a b c>
camelia ( no output )
Zoffix How come that doesn't warn?
lizmat special cased afaik
Zoffix lizmat++ that is indeed better 01:04
TimToady yes, that is special-cased
01:05 comborico1611 left, comborico1611 joined
squashable6 🍕 tbrowder++ wrote a comment on “Fix table caption RT #126740”: github.com/rakudo/rakudo/pull/1291...-348654459 01:05
synopsebot RT#126740 [new]: rt.perl.org/Ticket/Display.html?id=126740 [POD] Pod::Block::Table node caption property is not populated properly
Zoffix Updated. 01:06
01:06 astronavt left 01:08 BenGoldberg joined
jnthn "It's easy to remember: it looks like a squished box" <3 01:08
I never noticed :D
Zoffix :)
Zoffix calls it a night 01:11
\o
jnthn o/ 01:12
jnthn also, for that matter :)
01:23 troys is now known as troys_ 01:24 eliasr left 01:29 Cabanossi left 01:32 Cabanossi joined
timotimo m: my @stuff = (1, 2, (3, 4)), %(:42foo, :70bar); say flat @stuff»<> 01:32
camelia (1 2 3 4 bar => 70 foo => 42)
timotimo m: my @stuff = (1, 2, (3, 4)), %(:42foo, :70bar); say (flat @stuff»<>).perl 01:33
camelia (1, 2, 3, 4, :bar(70), :foo(42)).Seq
01:33 astronavt joined
timotimo Zoffix: it may be a good idea to recommend map alongside the hyper op and point out the difference with regards to nested structures? 01:33
01:50 wamba left 01:57 pilne joined
Zoffix timotimo: which differences? 02:02
timotimo m: my @stuff = (1, 2, (3, 4)), %(:42foo, :70bar); say (flat @stuff.map(*.<>)).perl 02:03
camelia (1, 2, 3, 4, :bar(70), :foo(42)).Seq
timotimo oh, huh.
i guess there wasn't one
m: my @stuff = (1, 2, $(3, 4)), %(:42foo, :70bar); say (flat @stuff.map(*.<>)).perl 02:04
camelia (1, 2, $(3, 4), :bar(70), :foo(42)).Seq
timotimo m: my @stuff = (1, 2, $(3, 4)), %(:42foo, :70bar); say (flat @stuff»<>).perl
Zoffix m: my @stuff = (1, 2, $(3, 4)), %(:42foo, :70bar); say [].^lookup("AT-KEY").can: "nodal"
camelia (nodal)
(1, 2, $(3, 4), :bar(70), :foo(42)).Seq
timotimo okay, never mind me!
Zoffix It's nodal, so it don't descend :)
timotimo that makes sense, yeah
Zoffix But I didn't check if it were when I was writing the article, so timotimo++ for noticing :)
timotimo going to bed, too
02:05 joy joined
timotimo Zoffix++ Zoffix++ # two good posts 02:05
Zoffix \o/
timotimo hm. i guess i could have also contributed a post to the advent by using dragon speech recognition to circumvent my rsi 02:06
Zoffix That would've been quite a commitment :) Rest. 02:07
Zoffix &
02:16 Cabanossi left 02:17 Cabanossi joined 02:18 rp-- joined
comborico1611 Goodnight, everybody! 02:21
02:21 comborico1611 left 02:40 kitsunenokenja left 02:45 ilbot3 left 02:51 sivoais joined 02:53 piojo left 03:10 pentashift joined 03:11 pentashift left 03:12 cdg joined 03:15 Cabanossi left 03:16 cdg left 03:17 Cabanossi joined 03:26 astronavt left 03:35 ilbot3 joined, ChanServ sets mode: +v ilbot3 03:43 ctilmes left 03:44 kitsunenokenja joined, Cabanossi left 03:47 Cabanossi joined 03:50 cdg joined 03:58 rp-- left 04:15 Cabanossi left 04:16 Cabanossi joined 04:38 mson left 04:47 cdg left, cdg joined 04:50 BenGoldberg left 04:52 cdg left 04:55 mson joined 05:07 Herby_ joined
Herby_ new advent of code problem up: adventofcode.com/2017/day/2 05:08
there are some fast people out there
teatime hrm, that seems awfully easy/straightforward 05:09
teatime wonders what he's missing
05:10 reportable6 left, reportable6 joined
Herby_ probably nothing. the first few problems are typically pretty simple 05:11
gives the commoners like me a chance :)
the difficulty does ramp up
teatime ah, it's only day 2 :) 05:12
gotcha
Herby_ .tell Zoffix: ++ on Day 2 advent article 05:14
yoleaux Herby_: What kind of a name is "Zoffix:"?!
Herby_ .tell Zoffix ++ on Day 2 advent article
yoleaux Herby_: I'll pass your message to Zoffix.
05:16 khw left
Zoffix Thanks. Glad you liked it. 05:16
yoleaux 05:14Z <Herby_> Zoffix: ++ on Day 2 advent article
05:37 wander joined
wander afternoon #perl6 05:38
yoleaux 1 Dec 2017 17:39Z <AlexDaniel> wander: OK rt.perl.org/Ticket/Display.html?id...xn-1511996
wander ok
05:39 aborazmeh joined, aborazmeh left, aborazmeh joined
Herby_ I'm lacking sleep and caffeine but I'm confused with this 05:49
m: my $line = "86 440 233 83 393 420 228 491 159 13 110 135 97 238 92 396"; say $line.split(/\s/).max()
camelia 97
geekosaur it found the lexical max
not the numeric one
Herby_ doh 05:50
geekosaur you use split, it produces Strs
Herby_ what would be the clean way to do that, given a string?
simple for loop?
teatime m: my $line = "86 440 233 83 393 420 228 491 159 13 110 135 97 238 92 396"; say $line.split(/\s/)».Int.max()
camelia 491 05:51
geekosaur that yes
teatime is one possibility
Herby_ teatime: ++
teatime yay I answered a question correctly
Herby_ :)
05:51 kitsunenokenja left
wander m: my $line = "86 440 233 83 393 420 228 491 159 13 110 135 97 238 92 396"; say $line.words.max(*.Int) 05:53
camelia 491
teatime alas, that is nicer. 05:54
wander wrap-handle we have :-)
teatime I might be tempted to split on [^0-9], but I guess you can trust your input
wander yes, it depends
Herby_ well, I solved Advent of Code Day 2 Part 1! adventofcode.com/2017/day/2 pastebin.com/mWqypbEr 05:56
not high achievement but I got tripped by that lexical max :(
part 2 looks a tad trickier 05:58
05:59 philomath_ joined 06:00 philomath_ left 06:06 Xliff_ joined 06:09 notbenh left, notbenh joined 06:11 riatre joined 06:15 philomath_ joined
Herby_ Ok, part 2 completed. and I cleaned up part 1 06:27
pastebin.com/jAzWKZgV
If anyone has feedback on improvements, I'm all ears
06:31 philomath_ left, philomath_ joined 06:36 mcmillhj joined
wander Heyby_: line 21, is it `elsif`? how about $a == $b? 06:36
Herby_ I did make a (luckily) correct assumption that they would be unique numbers for each line 06:39
are you saying to add another elsif to check if $a == $b?
wander replace `elsif` with `if` 06:40
06:40 pedh joined, mcmillhj left
Herby_ ahhh 06:41
thanks!
06:41 philomath_ left 06:49 pedh left 07:16 Todd joined
Todd HI all, A dumb question (but that never stopped me!) 07:16
Why does this work: $cmd = "/usr/bin/atop 1 1 > $GoodATop 2>&1"; $str = qqx{ $cmd };
but including the string on the qqx lie does not. Why do I need to use a variable? 07:17
symptom: /bin/sh: /usr/bin/atop 1 1 > /opt/Cimcor/CimTrak/CimTrakServer/logs/atop.txt: No such file or directory
$str = qqx{ "/usr/bin/atop 1 1 > $GoodATop 2>&1" }; doesn't work 07:18
teatime that's not getting shell quoted
or rather, I think it is, but you put it in "" which get passed to the shell :)
Todd Oh, so leave the quotes off?
teatime try: $str = qqx{ /usr/bin/atop 1 1 > $GoodATop 2>&1 }; 07:19
yeah. qqx{} is itself the quoting construct
Todd Bugger. Thank yoU!
worked perfectly Thank yoU! 07:20
teatime excellent.
Todd This one worked too! 07:22
my $str = qqx{ cd /opt/Cimcor/CimTrak/CimTrakServer; ./CimTrakServer.exe -Diagnostics \"Level=255\" > $GoodCimDiagErr 2>&1 };
07:22 wamba joined
Todd Will wonder ever cease! 07:23
geekosaur I think the backslashes are unnecessary there
teatime do you even need to escape
yeah
geekosaur in a "" string theyd be needed but here the quotes are { }
so they unescape to "
also I can't imagine $str contains much useful after that since you send it all to a file 07:24
Todd ordinarily no, but the program required it for some reason.
It is ported from that "other" operating system, whose name I can not mention. 07:25
geekosaur qqx does because it's an expression. you probably wanted
teatime Todd: VMS?
geekosaur shell " cd /opt/Cimcor/CimTrak/CimTrakServer; ./CimTrakServer.exe -Diagnostics \"Level=255\" > $GoodCimDiagErr 2>&1 "
and here you do need the escapes 07:26
teatime well only because you chose to use " " ;)
geekosaur also note that, as with perl 5's ``, using qqx when you don;t need the output is kinda expensive
yes, I could have used « » or something instead 07:27
also, uh, good luck running that on VMS :p
directories don't work the same way, redirection requires something that looks more like JCL than Unix 07:28
teatime looks at geekosaur.
geekosaur (yes, I know. ironically that'd actually stand soem chance of working as is on modern windows)
teatime oh, the command line
geekosaur simple Bourne-shell-style redirections have worked in cmd.exe since windows nt 3.5 and win98 (win95 had a few more limitations but I think that one did work) 07:29
ok, modulo the slashes needing to be backslashes in cmd.exe 07:30
teatime heh
Todd This is Fedora 27. I need the quotes and backslash when run from the (bash) command line too. I was warned by the program's tech support. It also has to run in its own directory too. Weird, but .. 07:31
geekosaur oh, one of those programs. bad win port 07:32
in that case I would actually be surprised that the qqx one worked because perl would strip the backslashes before it went to the shell
Todd I asked this last time, but has anyone come up with a better perl 6 for win 32? I really do not want to use Perl 5. 07:33
geekosaur last I heard the win32 build reliably ran out of memory
while building 07:34
Todd I got the above to work trial and error. I had the 2>&1 redirect wrong and you guys helped me figure it out.
geekosaur might be fixable but I don't think we have anyone working on perl 6 who has a win32 setup
(like, I have a windows vm always running but it's 64 bit) 07:35
07:35 anant joined
Todd that is what I heard. Any way to build a win32 msi on a 64 bit system? 07:35
anant Does perl6 support nested classes?
teatime yes.
geekosaur (and insufficient memory to run a second concurrently, and I use the main one often enough that shutting it down kinda messes things up)
it would have to be a cross-build and I don't think that's currently supported 07:36
Todd bugger.
teatime ... did I just lie
geekosaur (windows has odd restrictions. best bet might be to try to get a wow32 build env; not sure how painful that is.)
nested classes should work fine
Todd anyone know what the P5 equivalent of slurp and spurt is? 07:37
geekosaur but accesses might be funny
Todd When building Wine Staging, I have a "./configure --disable-win64" option. Nothing like that for Perl6? 07:39
geekosaur don't think there is a spurt. slurp is { my $o = select $fh; local $/ = undef; my $r = <$fh>; select $o; $r } assuming $fh is the filehandle (this can be simplified a bit in some circumstances)
anant I didn't an example of a nested class in docs.perl6.org/language/objects
geekosaur that tells wine to not build win64 emulation into it, not that it should build a win32 program
anant I didn't = I didn't see
Todd no sign of either over on perldoc.perl.org/index-functions.html#S
geekosaur it'll generate a 64-bit executable that only handles win32 emulation
m: class A { has $.a; class B { has $.b }}; my $x = A::B.new(b => 6); say 'alive' 07:40
camelia alive
geekosaur so it may depend on what you mean by nested 07:41
since it's mostly namespacing not e.g. need an A to make an A::B
heh. back in the old days FileHandle.pm had a slurp but it looks like current IO::Handle lacks it 07:45
07:45 aborazmeh left
geekosaur I recall it being somewhat discouraged 07:46
although there were two camps on the issue
anant Thanks geekosaur and teatime :)
teatime geekosaur++ 07:47
anant teatime++
geekosaur oh, if you wanted that nested class to only be accessible nside of class A, it's 'my class B { ... }' 07:51
07:53 darutoko joined
anant ok. Are nested classes discouraged in general? I searched many perl6 tutorials, but didn't see them mentioned. 07:53
buggable New CPAN upload: Inline-Go-0.0.4.tar.gz by AZAWAWI cpan.metacpan.org/authors/id/A/AZ/...0.4.tar.gz
geekosaur not so much discouraged as of dubious use 07:54
07:55 Todd left
anant Hmm. Does Perl6 have a "Zen" guide like Python (www.python.org/dev/peps/pep-0020/) 07:56
geekosaur nope.
anant It would be nice to have :) 07:57
I liked how the Python Zen said: Flat is better than nested.
geekosaur python is One Correct Way; perl is There Is More Than One Way To Do It
anant ah!
geekosaur++
geekosaur which is why perl 6 mixes OO, FP, procedural, etc. and lets you mix and match according to need or convenience 07:58
I imagine some conventions will eventually be agreed on by the community, but p6 is young enough that there aren't many such yet 08:00
wander still so many things TODO 08:03
anant Hmm. 08:04
wander that is why i am somewhat anxious about our Advent Calendar Day1. feature `but` has several known bugs 08:05
but, perl6 makes code happy 08:06
08:06 rindolf joined
geekosaur well, it *was* about ruining christmas... :p 08:06
wander that is the most important to me, my thoughts could be expressed freely 08:07
08:09 setty1 joined 08:12 anant left 08:21 nadim joined 08:23 sena_kun joined 08:28 Cabanossi left 08:31 Cabanossi joined 08:39 a3f joined 08:48 mson left 08:51 donaldh_ joined 08:52 donaldh_ is now known as donaldh 08:58 philomath_ joined 08:59 philomath_ left 09:00 philomath_ joined 09:09 troys_ is now known as troys 09:11 raiph left
Geth mu: 972e9bc5ce | (Moritz Lenz)++ | misc/perl6advent-2017/schedule
Per 6 advent: commit to day 8
 09:24
09:25 philomath_ left, philomath_ joined 09:26 Phi_mb joined, Phi_mb left 09:29 domidumont joined 09:35 philomath_ left, Phi_mb joined, Phi_mb left 09:36 domidumont left, domidumont joined 09:43 troys left 09:48 cdg joined 09:52 cdg left 09:58 Cabanossi left 10:01 jeek left, Cabanossi joined 10:17 pdcawley joined 10:18 stmuk left, stmuk joined 10:22 rindolf left 10:29 Cabanossi left 10:31 Cabanossi joined 10:34 pdcawley left 10:39 reportable6 left, reportable6 joined 10:40 someuser left 10:42 margeas joined 10:47 rindolf joined 10:48 reportable6 left 10:49 reportable6 joined, ChanServ sets mode: +v reportable6 10:50 reportable6 left, reportable6 joined
AlexDaniel squashable6: status 10:51
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in 1 day and ≈1 hour. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/db790f480c845f3b3e...c78e787ee0
AlexDaniel reportable6: 2017-12-01T07:33:00Z 2017-12-02T06:00:00Z 10:52
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
10:52 reportable6 left, reportable6 joined, ChanServ sets mode: +v reportable6
AlexDaniel reportable6: 2017-12-01T07:33:00Z 2017-12-02T06:00:00Z 10:52
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
bartolin kupopo: yeah, unfortunately the jvm backend was badly broken when Rakudo Star 2017.10 was released. it should look a bit better (and actually be installable) on HEAD. there are major problems, still (e.g. 'make' does not work before 'make test', nativecall is broken). i added some details to your gist: gist.github.com/shicks/27df7cf0d86...50942d4c62 10:53
yoleaux 1 Dec 2017 23:56Z <lizmat> bartolin: is gist.github.com/shicks/27df7cf0d86...50942d4c62 something you'd expect from a "make install" ?
reportable6 AlexDaniel, gist.github.com/7535fa27421221ed95...8ed234dd79
AlexDaniel wander: ↑ progress so far :)
bartolin AlexDaniel++ 10:54
AlexDaniel buggable: tag MOLD 10:55
buggable AlexDaniel, There are 191 tickets tagged with MOLD; See fail.rakudo.party/t/MOLD for details
wander AlexDaniel++ 10:58
hahainternet .min and .max on an array of strings gives back nonsense results 10:59
».Numeric solves it, but I wouldn't expect them to give back such weird outputs 11:00
AlexDaniel hahainternet: why so
hahainternet the strings don't get coerced into the numbers they represent though, i am far too perl6 ignorant to discover why
AlexDaniel m: say <abcdoeu sthiozeu uloeuhal>.max
camelia uloeuhal
AlexDaniel m: say <abcdoeu sthiozeu zzozzzt uloeuhal>.max
camelia zzozzzt
AlexDaniel m: say <abcdoeu sthiozeu zzozzzt uloeuhal>.min
camelia abcdoeu
hahainternet AlexDaniel: i understand
but this kinda violates the whole cool interoperability of strings/numbers in perl doesn't it? 11:01
perl6*
AlexDaniel I don't think so. You gave it a list of strings, and then asked for the max element
hahainternet i actually gave it just a list, never specifying strings 11:02
(semantics i know, but from a programmer's perspective it's important)
geekosaur you have it a list of things that don't look like numbers
hahainternet they certainly do look like numbers
geekosaur if they look like numbers then you get IntStr-s
if they don't then you get Str-s
AlexDaniel what's the code we're talking about?
geekosaur (this means you can actually get a mixture)
hahainternet AlexDaniel: it's just for AOC #2, for 'file'.IO.lines -> $l { my @n = $l.split(/\t/); } 11:03
presumably that produces some sort of hard String type
that won't coerce in a cool way 11:04
AlexDaniel well, you took a string and split it, so you got strings back
geekosaur ^
hahainternet ok, how would i recover the cool behaviour?
geekosaur we discussed thsi earlier, actually, split is a Str operation, it will always give you Strs
AlexDaniel hahainternet: $l.split(/\t/)».Cool ? :) 11:05
geekosaur $l.split(/\t/)>>.Int (or that)
AlexDaniel that doesn't work :(
hahainternet AlexDaniel: doesn't appear to exist?
geekosaur: doing ».Numeric, so that's fine
AlexDaniel .IntStr doesn't work too
hahainternet anyhow not a major complaint, but it felt a bit wart-y 11:06
AlexDaniel lizmat: any reason why we don't have Str.IntStr ?
geekosaur too much magic has its own problems
"every DWIM has a WAT"
hahainternet geekosaur: speak for your own wizardry ;)
geekosaur you say that now...
hahainternet no i know what you're saying 11:07
but it should be up to the programmer to opt in or out to the wizardry they want
and i got opt-outed by default :)
geekosaur because it's easy to opt back in with the hyperop
hahainternet geekosaur: right but it isn't, that only coerces, it doesn't restore the cool behaviour i mentioend
i mean my problem is solved, so ty
geekosaur and you have more control, which matters quite a lot in many cases
geekosaur keeps being reminded of nancy typing 11:08
AlexDaniel hahainternet: but you can also think about performance. That is, should it actually look into every single element returned from .split, and check if it's a number or not?
hahainternet AlexDaniel: if it's an IntStr sure? :)
geekosaur (which was p5's "cool" behavior gone wrong plus a NaN parsing issue)
hahainternet i don't mean the existing case should be changed
i would just like ».Cool or ».IntStr as you gave as an example 11:09
AlexDaniel hahainternet: not sure I understand, how would it know if it's an IntStr? :)
hahainternet AlexDaniel: methods are dispatched on types? i'm not suggesting .split or anything do the coercion 11:10
11:10 Aaronepower joined
geekosaur you seem to think a type is determined solely by looking at it. do you really want everything that touches it to have to compute its current type first 11:11
hahainternet geekosaur: where do i think that? I just would like a way to coerce them
11:09.07 < hahainternet> i would just like ».Cool or ».IntStr as you gave as an example
lizmat AlexDaniel: I don't see a reason not to have Str.IntStr 11:13
but I would make it slightly more involved than just IntStr.new(self.Int,self), because with that, any error in converting to Int becomes multi-level, and thus complicated 11:15
m: dd "foo".Int
camelia Failure.new(exception => X::Str::Numeric.new(source => "foo", pos => 0, reason => "base-10 number must begin with valid digits or '.'"), backtrace => Backtrace.new)
lizmat as Str.Int returns a Failure, that will explode when trying to bind
so if self.Int returns a Failure, you should throw that immediately, I think
wander m: my $x = IntStr.new(12, "42"); say $x + 4; say $x; 11:16
camelia 16
42
wander the evil way :)
AlexDaniel m: my $x = IntStr.new(12, "I'm not a Str, get off my lawn!"); say $x + 4; say $x; 11:21
camelia 16
I'm not a Str, get off my lawn!
lizmat not too dissimilar from dualvar() in P5 :-) 11:28
handy for things like errno(), where the Str version contains the error message 11:29
11:31 poohman joined 11:33 stmuk left
squashable6 🍕 W4anD0eR96++ opened issue “MOP ^add_method could be renamed as ^add-method”: github.com/rakudo/rakudo/issues/1292 11:34
poohman hello all, I have a question regarding grammar tokens - if the * is for zero or more, why does the following token work and not the next 11:36
token frame {\s*<top_left>\r\n\s*<date_or_time>'</TD>'\r\n}
the above works
wander mainly about consistency, not have to be kebab-case
poohman token frame {\s*<top_left>\s*\r\n\s*<date_or_time>'</TD>'\r\n}
the only difference is the \s* before \r 11:37
geekosaur because * is greedy and \r and \n both match \s, so you have \s*\s* which is kinda asking for problems
wander what is the text?
geekosaur or put otherwise, you've given that regex *way* too much rope 11:38
wander m: say so "\r\n" ~~ /\s*/
camelia True
wander m: say so "\r\n" ~~ /\s*?\r\n/
camelia True
11:39 espadrine left
wander also note that it's 'token' 11:39
so backtracking is disable 11:40
geekosaur (I oversimplified that a bit; the reality is that it's got far too many ways to match --- and token is ratcheting, so once the first \s* commits to matching the \r\n, it can;t backtrack when it sees the literal ones
poohman ok
geekosaur so (a) since it can;t backtrack, it can't "unmatch" the \r and \n it committed to the first \s* and will fail when it hits them in the pattern (b) and if you make it non-ratcheting, it will be sloooooooooooooooow 11:41
squashable6 🍕 AlexDaniel++ wrote a comment on “MOP ^add_method could be renamed as ^add-method”: github.com/rakudo/rakudo/issues/12...-348686524
🍕🍕🍕 First contribution by lizmat++! ♥ 11:42
poohman super - thanks 11:44
11:44 ZzZombo joined
squashable6 🍕 tbrowder++ wrote a comment on “Fix table caption RT #126740”: github.com/rakudo/rakudo/pull/1291...-348686656 11:44
synopsebot RT#126740 [new]: rt.perl.org/Ticket/Display.html?id=126740 [POD] Pod::Block::Table node caption property is not populated properly
squashable6 🍕 lizmat++ wrote a comment on “MOP ^add_method could be renamed as ^add-method”: github.com/rakudo/rakudo/issues/12...-348686720 11:45
11:49 Aaronepower left 11:50 pdcawley joined 11:52 ctilmes joined
squashable6 🍕 W4anD0eR96++ wrote a comment on “MOP ^add_method could be renamed as ^add-method”: github.com/rakudo/rakudo/issues/12...-348687536 12:01
AlexDaniel buggable: tag TESTNEEDED 12:07
buggable AlexDaniel, There are 49 tickets tagged with TESTNEEDED; See fail.rakudo.party/t/TESTNEEDED for details
12:11 ChoHag joined
squashable6 🍕 W4anD0eR96++ wrote a comment on “MOP ^add_method could be renamed as ^add-method”: github.com/rakudo/rakudo/issues/12...-348688003 12:11
12:19 Aaronepower joined
pdcawley jamesnvc: I'm not sure that Emacs has a concept of a <menu> key by default. However, you can pick something (say Key_Menu or whatever) and then use `key-translation-map` to map from whatever emacs thinks it is initially, to `(kbd "<menu>")`, and you should be able to bind to `(kbd "<menu>")` rather than the weird name. 12:20
AlexDaniel pdcawley: that's great to know, but is it the right channel? :D 12:27
pdcawley Arse!
And, it turns out, I'm wrong. Emacs _does_ know about the menu key.
AlexDaniel pdcawley: yeah, it defaults to M-x I think
12:29 kitsunenokenja joined, poohman left
pdcawley AlexDaniel: Keyboards are weird. 12:29
AlexDaniel just realized he has no menu key
(there's PrtSc on its place, wtf) 12:30
12:30 ChoHag left
AlexDaniel pdcawley: some keyboard *layouts* are weirder. I remapped my spacebar to AltGr :) 12:32
pdcawley AlexDaniel: Don't talk to me about weird layouts. I've got a KeyboardIO model 01, 64 keys in an an approximation of the Malt layout ('E' on the left thumb, ANIS/THOR on the home row. 12:33
AlexDaniel (or is it other way around? What I meant is that my spacebar is AltGr now) 12:34
pdcawley: oh, that one is cool. Have you seen X-Bows by the way? Another relatively similar keyboard, although it probably doesn't come with a crazy keyboard layout :) 12:35
pdcawley Nope. But I do have an original Maltron that I've cut down to a 64 key matrix, rewired and rebrained to do NKRO and other such stuff, and a Kinesis that I keep meaning to hack palm keys into (and possibly rebrain, but it's got more sophisticated hardware than the Maltron, which was entirely hand wired to begin with. 12:37
AlexDaniel maaaan… keyboardio should really come with perl 6 manual or something :D 12:38
12:38 epony joined
pdcawley AlexDaniel: Yeah, I'm sure obra will be up for that. Right after he's managed to get enough enclosures out of his wood suppliers to ship everything after the first 1000 kickstarter orders. 12:39
AlexDaniel: Not. (And not just because I just spotted obra lurking in here) 12:40
12:42 poohman joined, poohman left, poohman joined 12:44 Cabanossi left 12:46 Cabanossi joined 12:54 stmuk joined
AlexDaniel obra: imgur.com/a/J06Ve 12:54
12:56 ChoHag joined
El_Che wow, man, those keyboard are expensive 13:04
(on the other hand, I spend €92 today on a mouse) 13:05
"jesse vincent" <-- p5's Jesse vincent? 13:07
13:08 natrys joined 13:12 ChoHag left 13:19 ctilmes left, ctilmes joined 13:20 konsolebox left, konsolebox joined 13:21 xcm left 13:23 xcm joined 13:28 Cabanossi left 13:31 Cabanossi joined 13:35 poohman left 13:40 mson joined
tyil Zoffix: poke 13:42
Zoffix: in the advent article, under the "The One And Only" heading, it has a sentence "Since scalars, as the same suggests, contain just one thing… ", should "same" not be "name" there?
13:42 pmurias joined
pmurias in roast we test tan(pi/2) 13:43
is-approx tan(my num $ = π/2), (sin(π/2) / cos(π/2)) 13:44
is tan(pi/2) something we can expect sane values for
AlexDaniel pmurias: I dunno, but fwiw python returns something similar 13:47
13:47 cdg joined
AlexDaniel and ruby too 13:49
squashable6 🍕 jnthn++ wrote a comment on “MOP ^add_method could be renamed as ^add-method”: github.com/rakudo/rakudo/issues/12...-348693494 13:53
🍕🍕🍕 First contribution by jnthn++! ♥
pmurias jnthn: how should I get more than 14 digits of a floating point number in NQP? 13:54
13:54 piojo joined
jnthn pmurias: No idea, I thought the "normal" stringification of a float would include all of the available precision? 13:55
If not, maybe we should check why not.
lunch, bbiab
13:55 rindolf left
pmurias jnthn: it's set to 14 13:56
err 15 13:58
a double has ~15.9 decimal digits of precision 13:59
13:59 rindolf joined
dogbert17 pmurias: I seems to remember there being a few RT's about that. Perhaps AlexDaniel has the RT numbers in his head :) 14:00
14:03 poohman joined
pmurias the problem I have is that when emitting floating point literals I get wrong numbers 14:04
AlexDaniel down to 179 MOLD tickets \o/
pmurias what are MOLD tickets? 14:05
AlexDaniel dogbert17: no idea. RT#127184 ? 14:06
synopsebot RT#127184 [open]: rt.perl.org/Ticket/Display.html?id=127184 [BUG] One digit short to correctly stringify a double
AlexDaniel pmurias: tickets that were not updated in more than one year or so 14:07
fail.rakudo.party/t/MOLD
squashable6: status 14:08
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈21 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/15ec6884b94a7088aa...1c430be6a3
dogbert17 AlexDaniel: yes, and perhaps RT #127201 as well 14:14
synopsebot RT#127201 [new]: rt.perl.org/Ticket/Display.html?id=127201 sprintf() float precision cannot go beyond 15[efg]
AlexDaniel break & 14:23
14:24 philomath_ joined 14:28 domidumont left 14:29 ChoHag joined 14:30 domidumont joined 14:35 domidumont left 14:36 steve joined
AlexDaniel here is a low-hanging fruit that can close 3 tickets: RT#124624 14:36
synopsebot RT#124624 [open]: rt.perl.org/Ticket/Display.html?id=124624 [LHF] :./S02-magicals/KERNEL.t line:8 reason: 'unimpl 0KERNEL'
14:37 poohman left 14:38 poohman joined, poohman left 14:39 philomath_ left
AlexDaniel Actually 5, if you also unfudge $*DISTRO tests 14:39
14:40 philomath_ joined, poohman joined
AlexDaniel and 7 if you do the same with $*VM 14:41
pmurias m: say $?KERNEL
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$?KERNEL' is not declared. Did you mean 'Kernel'?
at <tmp>:1
------> 3say 7⏏5$?KERNEL
pmurias m: say $*KERNEL 14:42
camelia linux (4.4.70.18.9.default)
AlexDaniel actually 16, if you do the same with $*PERL… :)
pmurias AlexDaniel: $?KERNEL has to be actually implemented first before we unfudge it right?
AlexDaniel didn't we decide to have $*KERNEL instead of $?KERNEL? or is there supposed to be a difference? 14:43
14:44 Cabanossi left
pmurias isn't $?KERNEL the kernel we are compiling under as opposed to the $*KERNEL we are running under? 14:45
14:46 Cabanossi joined
AlexDaniel uhhhh… I see… 14:46
yeah, I guess I'm wrong then 14:47
pdcawley El_Che: And the first perl 6 project manager Jesse Vincent, yes. Or obra. 14:49
14:52 philomath_ left
AlexDaniel well ok, at least these tickets no longer have mold because I linked them all together 14:54
14:56 raiph joined
wander m: .say for [[1,2,3]][0] 15:04
camelia 1
wander m: .say for [[1,2,3]][0].Array
camelia 1
wander m: .say for [[1,2,3],][0] 15:05
camelia [1 2 3]
wander m: .say for [[1,2,3],][0].Array
camelia 1
2
3
wander most frequent asked question about 'container' 15:06
15:08 poohman left
wander m: say |[[1,2,3],[4,5,6]] 15:08
camelia [1 2 3][4 5 6]
wander m: say [|[1,2,3],|[4,5,6]]
camelia [1 2 3 4 5 6]
jnthn Well, single arg rule as well as containers :) 15:09
pmurias hmm, is there a way to exactly serialize a double to javascript? 15:13
wander m: say reduce { $^a.append: $^b.flat }, [[1,2,3],[4,5,6]] 15:17
camelia [1 2 3 4 5 6]
15:22 SHODAN joined 15:24 ctilmes left
wander m: say [1,2,3] + [4,5,6] 15:26
camelia 6
wander m: say +[4,5,6]
camelia 3
El_Che pdcawley: nice project. I saw something on twitter about it. 15:28
15:32 darutoko left 15:44 domidumont joined, Cabanossi left
wander m: say gather for [[1,2,3],[4,5,6]] { .take for .flat } 15:44
camelia (1 2 3 4 5 6)
15:45 khw joined 15:46 Cabanossi joined 15:50 mson left
wander m: [[1,2,3],[4,5,6]].join('#').split('#') # dirty 15:51
camelia ( no output )
15:58 rindolf left 16:02 rindolf joined
Herby_ o/ 16:10
Zoffix tyil: thanks. Fixed. 16:16
Herby_: you were looking for containers article the other day. It's now up: perl6advent.wordpress.com/2017/12/...ontainers/ 16:17
Herby_ Zoffix: o/\o gonna read it this morning, just waiting for my coffee to brew :) 16:18
16:33 perlpilot_ joined 16:34 Sgeo_ joined
squashable6 🍕 zoffixznet++ wrote a comment on “MOP ^add_method could be renamed as ^add-method”: github.com/rakudo/rakudo/issues/12...-348703641 16:40
16:43 Zoffix left 16:44 ZzZombo left 16:47 ZzZombo joined 16:54 ctilmes joined
El_Che developers.slashdot.org/story/17/1...edium=feed 16:56
First comment hasn't been claimed yet :)
timotimo didn't we have some kind of decision to have underscores as part of "internalsy" names? 17:08
17:08 poohman joined 17:11 someuser joined
teatime heh, (10!) seconds is exactly 6 weeks. neat. 17:12
17:13 sftp joined
timotimo hm, 3 * 4 is already 12, and 2 * 12 is 24, that's the "60 seconds per hour" bit 17:14
um, that's the 24 hours per day bit, of course 17:16
take 6 and 10 to get the 60 seconds per hour
17:16 sftp left
teatime that's a neat way of looking at it that did not occur to me. 17:16
timotimo you're left with 4, 5, 7, 8, 9; it's probably easy to figure out how to re-arrange them to get to 6 weeks, but i'm in a tiny hurry
teatime </nerdsniped> 17:17
timotimo m: (6 * 7 * 24 * 60) / (2 * 3 * 4 * 6 * 10)
camelia WARNINGS for <tmp>:
Useless use of "/" in expression "(6 * 7 * 24 * 60) / (2 * 3 * 4 * 6 * 10)" in sink context (line 1)
timotimo m: say (6 * 7 * 24 * 60) / (2 * 3 * 4 * 6 * 10)
camelia 42
timotimo m: say 4 * 5 * 7 * 8 * 9
camelia 10080
timotimo i made a mistake somewhere 17:18
m: say (6 * 7 * 24 * 60) / (2 * 3 * 4 * 5 * 6 * 7 * 8 * 9 * 10)
camelia 0.016667
17:18 perlpilot_ left
timotimo so .. not quite? 17:18
teatime wth 17:19
m: say (6 * 7 * 24 * 60 * 60) / (2 * 3 * 4 * 5 * 6 * 7 * 8 * 9 * 10)
camelia 1
timotimo oh, haha
i forgot to go from minutes to hours 17:20
or equivalently from seconds to minutes
17:22 sftp joined 17:26 nadim left 17:30 nadim joined 17:38 patrickz joined 17:39 patrickz_ left 17:42 patrickz left 17:43 patrickz joined 17:45 llfourn_ joined, llfourn left 17:49 patrickz_ joined 17:52 patrickz left 17:54 patrickz joined 17:55 patrickz_ left 17:58 patrickz left 17:59 patrickz joined 18:04 cdg left, patrickz left 18:05 cdg joined 18:08 Zoffix joined
Zoffix m: say flat [[1,2,3],[4,5,6]]».<> 18:08
camelia (1 2 3 4 5 6)
Zoffix wander: ^
18:09 patrickz joined
wander wow 18:09
Zoffix wander: you just need to decont two inner Arrays. perl6advent.wordpress.com/2017/12/...rs/#decont
18:09 cdg left
wander if only I don't put the stuff decont into a new container 18:10
Zoffix m: say flat ((1,2,3),(4,5,6)) # or not use Arrays in the first place
camelia (1 2 3 4 5 6)
wander thanks a lot, looks I didn't read the post attentively :P 18:12
Zoffix You can also write your gather/take version like this:
m: say flat gather [[1,2,3],[4,5,6]]».take
camelia (1 2 3 4 5 6)
Zoffix But note that that method destroys information from Hashes:
m: say flat gather [[1,2,3],[4,5,6], %(:42foo, :60meow)]».take
camelia (1 2 3 4 5 6 42 60)
Zoffix m: say flat gather [[1,2,3],[4,5,6], %(:42foo, :60meow)].map: *.take # this won't tho 18:13
camelia (1 2 3 4 5 6 foo => 42 meow => 60)
18:13 patrickz left 18:14 patrickz joined
wander "Hyper operators are defined recursively on nested arrays.", huh 18:14
Zoffix Where's that from?
wander under docs.perl6.org/language/operators#..._Operators 18:15
Zoffix m: say :foo.^roles
camelia ((Associative))
Zoffix m: say gather :42foo».take
camelia (42)
Zoffix wonders if that's actually a bug :/ 18:16
18:16 domidumont left
Zoffix m: say -« [[[1, .5], 2], 3]; 18:17
camelia [[[-1 -0.5] -2] -3]
wander m: say [[1,2,3],[4,5,6]].flatmap(*.self) 18:19
camelia (1 2 3 4 5 6)
wander interesting
Zoffix inconsistency; descends into itemized iterables yet method form doesn't
s: &METAOP_HYPER_PREFIX 18:20
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/4fca...ps.pm#L519
18:20 patrickz left
Zoffix s: [], 'dispatch:<hyper>' 18:21
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/4fca...Mu.pm#L855
Zoffix s: &HYPER
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/4fca...ps.pm#L529
Zoffix m: [[[1, [.5, 1]], 2], 3]».&[-].say 18:23
camelia [[[-1 [-0.5 -1]] -2] -3]
Zoffix m: [[[-1, [.5, -1]], -2], 3]».abs.say 18:24
camelia [[[1 [0.5 1]] 2] 3]
Zoffix
.oO( am I misremembering? :S )
18:24 patrickz joined
wander it reminds me of this guy 18:27
m: [[[1, [.5, 1]], 2], 3]»(&say)
camelia 1sub say (| is raw) { #`(Sub|51541408) ... }
0.5sub say (| is raw) { #`(Sub|51541408) ... }
1sub say (| is raw) { #`(Sub|51541408) ... }
2sub say (| is raw) { #`(Sub|51541408) ... }
3sub say (| is raw) { #`(Sub|51541408) ... }
Zoffix I guess I confused .flat not descending into containerized iterables with ». not descending with nodal callables
teatime tries to figure out what wander's last example is doing 18:28
18:29 pmurias left, kitsunenokenja left
Zoffix m: [[[1, [.5, 1]], 2], 3]»(&say) 18:30
camelia 1sub say (| is raw) { #`(Sub|47003008) ... }
0.5sub say (| is raw) { #`(Sub|47003008) ... }
1sub say (| is raw) { #`(Sub|47003008) ... }
2sub say (| is raw) { #`(Sub|47003008) ... }
3sub say (| is raw) { #`(Sub|47003008) ... }
Zoffix oops
m: sub meow ($x, $y) { dd [$x, $y] }; [[[1, [.5, 1]], 2], 3]»(&meow)
camelia [1, sub meow ($x, $y) { #`(Sub|65895176) ... }]
[0.5, sub meow ($x, $y) { #`(Sub|65895176) ... }]
[1, sub meow ($x, $y) { #`(Sub|65895176) ... }]
[2, sub meow ($x, $y) { #`(Sub|65895176) ... }]
[3, sub meow ($x, $y) { #`(Sub|65895176) ... }]
Zoffix m: say (1, 2, 3) »+« (42)
camelia Lists on either side of non-dwimmy hyperop of infix:<+> are not of the same length
left: 3 elements, right: 1 elements
in block <unit> at <tmp> line 1
Zoffix m: say (1, 2, 3) »+» (42)
camelia (43 44 45)
Zoffix I'm guessing like this, except meow() instead of `+` and &meow instead of 42 18:31
Zoffix &
18:31 Zoffix left
teatime I def. still don't get it. 18:32
buggable New CPAN upload: PDF-Font-0.1.1.tar.gz by WARRINGD cpan.metacpan.org/authors/id/W/WA/...1.1.tar.gz 18:33
18:35 pmurias joined
TimToady m: [[[1, [.5, 1]], 2], 3]».&say 18:36
camelia 1
0.5
1
2
3
18:36 patrickz_ joined
TimToady m: 1»(&say) 18:37
camelia 1sub say (| is raw) { #`(Sub|60519792) ... }
TimToady I have no idea what that thinks it's doing...
teatime welp I no longer feel bad about it :) 18:38
TimToady m: 1(&say)
camelia No such method 'CALL-ME' for invocant of type 'Int'
in block <unit> at <tmp> line 1
wander m: my @a = [1,2,3]; @a>>++; say @a # normal use 18:40
camelia [2 3 4]
18:40 patrickz left
wander note that 18:40
m: [1,2,3]>>&say 18:41
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed postfix
at <tmp>:1
------> 3[1,2,3]>>7⏏5&say
expecting any of:
postfix
wander m: [1,2,3]>>(&say)
camelia 1sub say (| is raw) { #`(Sub|60384624) ... }
2sub say (| is raw) { #`(Sub|60384624) ... }
3sub say (| is raw) { #`(Sub|60384624) ... }
TimToady I have no idea what that thinks it's doing...
wander so it escapes 18:42
TimToady that should be trying to call 1 as a subroutine, passing it &say as the first arg, which should fail
m: 1(&say)
camelia No such method 'CALL-ME' for invocant of type 'Int'
in block <unit> at <tmp> line 1
TimToady like that
say is both being called and used as data somehow
teatime yar 18:43
18:44 noganex left
wander maybe it trans to say(1, &say) 18:45
when >>++, postfix:<++>($a) is called(still 'maybe')
18:46 noganex joined, cdg joined
TimToady but (&say) is not a valid postfix, except insofar as it passes &say as an argument to something 18:46
and that something has to be Callable 18:47
this is definitely somewhere in the "undefined behavior" realm
wander yes, so you can see `[1,2,3]>>&say` throws "expecting any of:␤ postfix␤"
however the form (&say) escapes for some reasons 18:48
TimToady m: [1,2,3]».&say # it requires the .
camelia 1
2
3
TimToady not that you should be using say with hypers in any case
since order of evaluation is undefined 18:49
18:49 cdg left, patrickz_ left
TimToady errands & 18:49
18:49 cdg joined 18:50 patrickz joined
wander m: say [1,2,3]>>(sub {dd @_}) 18:51
camelia [1, sub (*@_) { #`(Sub|66679448) ... }]
[Nil Nil Nil]
[2, sub (*@_) { #`(Sub|66679448) ... }]
[3, sub (*@_) { #`(Sub|66679448) ... }]
18:54 cdg left
wander m: [&say, &say]>>(42) 18:54
camelia No such method 'CALL-ME' for invocant of type 'Int'
in block <unit> at <tmp> line 1
wander it should be &say(42), however looks like 42.(&say, 42) now 18:55
m: [&say, &say]>>.(42) # workaround
camelia 42
42
teatime wth
how does that work?! 18:56
18:56 cdg joined
AlexDaniel ma6ic 18:57
teatime oh, I forgot 42.say works heh 18:58
le whoops
AlexDaniel that's irrelevant :)
m: &say(42)
camelia 42
AlexDaniel m: &say.(42)
camelia 42
teatime but isn't it trying to do 42(&say) ?
er 18:59
ok I just give up
18:59 patrickz_ joined 19:01 patrickz left 19:04 patrickz joined 19:05 kitsunenokenja joined
wander m: sub f() { say 42 }; [&f, &f]>>(); 19:05
camelia ===SORRY!===
Unknown QAST node type NQPMu
19:06 patrickz_ left
wander s: Grammar 19:06
SourceBaby wander, Something's wrong: ␤ERR: Cannot resolve caller sourcery(Grammar); none of these signatures match:␤ ($thing, Str:D $method, Capture $c)␤ ($thing, Str:D $method)␤ (&code)␤ (&code, Capture $c)␤ in block <unit> at -e line 6␤␤
19:09 patrickz left 19:14 patrickz joined
AlexDaniel reportable6: list 19:20
reportable6 AlexDaniel, gist.github.com/22af8503523d183b34...b0f135a95f
AlexDaniel reportable6: 2017-12-01T07:33:00Z 2017-12-02T18:00:00Z
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
19:20 reportable6 left, reportable6 joined, ChanServ sets mode: +v reportable6
AlexDaniel oh gosh please someone fix GH#1278 :) 19:20
synopsebot GH#1278 [open]: github.com/rakudo/rakudo/issues/1278 [SEGV] Another SEGV in whateverable
AlexDaniel reportable6: 2017-12-01T07:33:00Z 2017-12-02T18:00:00Z 19:21
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
AlexDaniel, gist.github.com/b3b546a1ad9be8eb50...d83c199c14
bartolin for a second I thought, reportable6 responded to your "oh gosh please someone fix ..." with "OK, working on it! This may take up to 40 seconds" :-) 19:23
AlexDaniel ha, you wish… 19:24
.oO( jnthnable? )
bartolin lol
19:25 patrickz_ joined 19:26 patrickz left 19:28 Cabanossi left 19:29 comborico1611 joined 19:30 patrickz joined 19:31 patrickz_ left, Cabanossi joined
TreyHarris Is there a shorthand way for accepting <st nd rd th> numeric options and merging them so you can do :0th :1st :2nd :3rd fun? 19:33
19:35 releasable6 left, patrickz_ joined, releasable6 joined 19:36 patrickz left, unicodable6 left, unicodable6 joined, ChanServ sets mode: +v unicodable6, bisectable6 left, committable6 left, bisectable6 joined, ChanServ sets mode: +v bisectable6, committable6 joined, eliasr joined
AlexDaniel TreyHarris: I think the answer is no: github.com/rakudo/rakudo/blob/5929...1042-L1066 19:39
TreyHarris AlexDaniel: And I think it's yes? 19:42
one sec, checking
AlexDaniel m: sub foo(:st(:$nd)) { say $nd }; say foo :2nd 19:43
camelia 2
True
AlexDaniel m: sub foo(:st(:$nd)) { say $nd }; say foo :1st
camelia 1
True
AlexDaniel the syntax is backwards so you can't alias more than one it seems? 19:44
TreyHarris m: sub ordinated( :st(:nd(:rd(:th(:$num)))) ) { $num //= "Nothing"; say "I got $num." }; ordinated(:0th); ordinated(:1st); ordinated(:2nd); ordinated(:3rd); ordinated(:4th)
camelia I got 0.
I got 1.
I got 2.
I got 3.
I got 4.
TreyHarris You just can't variablize the last one unlize there are just two. 19:45
Uh.... optionize the variable
Er, mumblize the mumble
AlexDaniel :O
this is awesome 19:47
TreyHarris m: sub ordinated ( :st(:nd(:rd(:th(:$num = -1 )))) ) { say "I got $num." }; ordinated(:0th); ordinated(:1st); ordinated(:2nd); ordinated(:3rd); ordinated(:4th) 19:50
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse named parameter; couldn't find right parenthesis
at <tmp>:1
------> 3sub ordinated ( :st(:nd(:rd(:th(:$num7⏏5 = -1 )))) ) { say "I got $num." }; ord		 19:51
AlexDaniel m: sub ordinated ( :st(:nd(:rd(:th(:$num))))=-1 ) { say "I got $num." }; ordinated(:0th); ordinated(:1st); ordinated(:2nd); ordinated(:3rd); ordinated(:4th); ordinated()
camelia I got 0.
I got 1.
I got 2.
I got 3.
I got 4.
I got -1.
TreyHarris AlexDaniel++ 19:53
jnthn m: sub ordinated(:st(:nd(:rd(:th($num))))) { say $num }; ordinated(:1th)
camelia 1
TreyHarris I coulda sworn I tried that first
jnthn m: sub ordinated(:st(:nd(:rd(:th($num))))) { say $num }; ordinated(:1num)
camelia Unexpected named argument 'num' passed
in sub ordinated at <tmp> line 1
in block <unit> at <tmp> line 1
jnthn Note that if you omit the : before $num then it isn't one of the exposed names
You can in fact do this any time you want to have one external name for a named parameter but put it into a variable of a different name 19:54
AlexDaniel github.com/perl6/doc/issues/1696
TreyHarris oh, duh, I did try it, but I tried to name it "ordinated-2", which isn't a legal sub name. :-P 19:56
AlexDaniel m: sub ordinated ( :st(:rd(:th(:$nd)))=-1 ) { say "I got $nd." }; ordinated(:0th); ordinated(:1st); ordinated(:2nd); ordinated(:3rd); ordinated(:4th); ordinated()'
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3ated(:3rd); ordinated(:4th); ordinated()7⏏5'
expecting any of:
infix
infix stopper
statement end
statement …
AlexDaniel m: sub ordinated ( :st(:rd(:th(:$nd)))=-1 ) { say "I got $nd." }; ordinated(:0th); ordinated(:1st); ordinated(:2nd); ordinated(:3rd); ordinated(:4th); ordinated()
camelia I got 0.
I got 1.
I got 2.
I got 3.
I got 4.
I got -1.
jnthn That's another way to avoid an unexpected bonus name for it, yes :) 19:57
AlexDaniel maybe it'd make sense to support something like :st:nd:rd:th($num) 19:58
ctilmes Should :3st be an error?
TreyHarris jnthn: "bonus name"? what's the "bonus name"?
AlexDaniel TreyHarris: ordinated(:32num)
TreyHarris ctilmes: no, specifically for the benefit of non-English-speakers, th/st/rd/nd are interchangeable 19:59
jnthn TreyHarris: If you do sub foo(:nd(:st(:$num))) { } then you can pass it as :num($value) too
TreyHarris: Make it just sub foo(:nd(:st($num))) { } and the $num name is purely for internal use
wander m: say :6lang 20:00
camelia Unexpected named argument 'lang' passed
in block <unit> at <tmp> line 1
TreyHarris jnthn++ I think in an effort to be pithy, the docs leave out examples like that. Thanks
jnthn
.oO( Well, that's a pithy )
wander m: (:6lang).say
camelia lang => 6
TreyHarris ctilmes: I think that was specifically mentioned in Apocalypse <mumble> in 2002 20:01
I know this because I can remember the desk I was sitting in when I read that and thought, "ah, makes sense" 20:02
AlexDaniel aliases are not even documented I think. Maybe a little bit in docs.perl6.org/language/functions#sub_MAIN
wander m: (:6lang).say; (:perl6).say
camelia lang => 6
perl6 => True
TreyHarris AlexDaniel: re your issue, is that an faq, or for a useful example of aliasing in functions.pod6? I'll take the issue and write it up 20:03
AlexDaniel TreyHarris: I'd say aliasing should be documented, and this can be one of the examples 20:04
TreyHarris (Though as an FAQ it's an easy add, for functions we should add a bit about aliases.)
AlexDaniel: agreed.
AlexDaniel TreyHarris: also this docs.perl6.org/language/functions#sub_MAIN should probably link to that newly written section 20:05
20:05 tailgate joined
TreyHarris AlexDaniel: yup, saw that, thanks. 20:05
20:07 mson joined
TreyHarris Ack! 20:07
AlexDaniel: Meant to cancel my comment since you just added the same, but closed the issue by mistake
20:08 comborico1611 left 20:09 eliasr left
squashable6 🍕 AlexDaniel++ opened issue “Proposal: easier syntax for multiple param aliases”: github.com/rakudo/rakudo/issues/1293 20:10
🍕 AlexDaniel++ labeled issue “Proposal: easier syntax for multiple param aliases”: github.com/rakudo/rakudo/issues/1293
20:11 donaldh left, eliasr joined, donaldh_ joined
TreyHarris AlexDaniel: are lang-spec issues now going straight to rakudo? I was ill and dropped out of actively monitoring things for a few months 20:11
20:11 patrickz joined
TreyHarris (Referring to the rakudo issue mentioned just now, 1293) 20:12
AlexDaniel TreyHarris: previously they went to RT, so now that we're using rakudo/issues as an alternative to that, yeah.
TreyHarris: in an ideal world we'd want them to be on roast/issues
20:12 donaldh_ is now known as donaldh
TreyHarris AlexDaniel++ thanks. 20:12
Before I start integrating this, anyone written a POD6 editing mode for any common editor? Not that I need it, just always like to try such things, especially if I've been away awhile. 20:13
20:14 patrickz_ left
TreyHarris AlexDaniel: And while I'm trying to get catch up, is there a simple-summary place I can read to catch up to the ecosystem/CPAN deal now as compared to, say, the begininng of the year (when we'd create a package with mi6, push it up to you own repo, and submit a link to perl6/ecosystem)? 20:19
I was nearly done with a package that I need to finish but need to update it first 20:20
Can we get squashable6 to report an perl6/doc issues too? 20:21
Herby_ I think Zef is broke on my Windows 10 running 2017.10: gist.github.com/sylints/ec6dd9aece...3013efe5da
any ideas?
I can't install modules
20:22 wander left
TreyHarris Herby_: github.com/ugexe/zef/issues/213 ? 20:26
AlexDaniel TreyHarris: this particular squashathon is not about doc tickets, so it doesn't report stuff about the doc repo. github.com/rakudo/rakudo/wiki/Mont...Squash-Day and github.com/rakudo/rakudo/wiki/MOLD...QUASHathon 20:28
TreyHarris: as for the ecosystem, I don't think much changed, but see this: docs.perl6.org/language/modules.ht...le_to_CPAN 20:29
Herby_ TreyHarris: thats why they pay you the big bucks. That fixed the problem. Thanks!
TreyHarris AlexDaniel: Well, that section of modules.pod6 was written in the past month, so that counts :-) Thanks 20:30
Herby_: they do? Where do I collect my check? 20:31
:-D
20:31 pmurias left 20:32 yqt joined, benchable6 left 20:33 benchable6 joined, ChanServ sets mode: +v benchable6
Herby_ :) 20:33
squashable6 🍕 zoffixznet++ wrote a comment on “Proposal: easier syntax for multiple param aliases”: github.com/rakudo/rakudo/issues/12...-348718558 20:39
TreyHarris I thought the autogenerated html files like in routine/ contained the source pod6 file(s) in comments, but I don't see them anymore... is there a place they've migrated, or do you just have to figure it out yourself? 20:42
20:43 lizmat left 20:44 Cabanossi left 20:46 Cabanossi joined, kaare_ joined, pdcawley left
squashable6 🍕 treyharris++ wrote a comment on “Proposal: easier syntax for multiple param aliases”: github.com/rakudo/rakudo/issues/12...-348718976 20:46
🍕🍕🍕 First contribution by treyharris++! ♥
TreyHarris squashable6: 🤩 20:47
squashable6 TreyHarris, I cannot recognize this command. See wiki for some examples: github.com/perl6/whateverable/wiki/Squashable
TreyHarris heh
AlexDaniel buggable: pizza TreyHarris 20:49
buggable TreyHarris, enjoy this slice of Margherita pizza, my friend! Yummy 🍕
Herby_ is there a way to what functionality the bots in this channel have? 20:51
for instance, is there a bot that return perl 6 modules that fit a tag?
*is there a way to see
Morfent i think i might port net::dns to perl 6 20:52
TreyHarris Morfent: it's only 5 years old, from the CPAN history? That really surprises me, I feel like I've known about it forever 20:54
TreyHarris realizes that "only 5 years old" is a ridiculous statement for a software library, but for important CPAN modules, it's not 20:55
AlexDaniel Herby_: github.com/perl6/doc/issues/711#is...-235414744 20:59
Herby_: as far as I know only whateverables have reasonable documentation 21:00
Herby_ AlexDaniel: thanks 21:02
21:05 pdcawley joined
squashable6 🍕 zoffixznet++ opened issue “Passing named args witho…”: github.com/rakudo/rakudo/issues/1294 21:23
🍕 zoffixznet++ opened issue “LTA error with named arg…”: github.com/rakudo/rakudo/issues/1295 21:25
🍕 zoffixznet++ labeled issue “LTA error with named arg…”: github.com/rakudo/rakudo/issues/1295
21:31 someuser left
Herby_ greppable: ftp 21:37
greppable6: ftp
21:40 greppable6 joined, ChanServ sets mode: +v greppable6
AlexDaniel greppable: ftp 21:40
greppable6 AlexDaniel, gist.github.com/13dfebd0fe7798dac5...3b89a371d1
AlexDaniel Herby_: ↑ some bots are off due to github.com/rakudo/rakudo/issues/1280 21:41
Herby_ gotcha 21:42
21:44 rindolf left
AlexDaniel greppable: \bftp\b 21:48
greppable6 AlexDaniel, gist.github.com/4d1c6d7b0251dadc11...37d54086e5
squashable6 🍕 zoffixznet++ wrote a comment on “Proposal: easier syntax for multiple param aliases”: github.com/rakudo/rakudo/issues/12...-348723095 21:57
21:59 eliasr left
DrForr I've got to get to bed, but here's a suggestion for the Test module - Add 'multi subtest( Hash, Str ) { note "Subtests without a body are not recommended" }' as I keep accidentally writing subtest { #'() }, 'to be filled-in'; and the error is somewhat nonintuitie. 22:10
yoleaux 1 Dec 2017 22:30Z <AlexDaniel> DrForr: Please check RT#125299 and do something with it. I really don't understand what has to be done in order to close it. If it's just that the line number of the starter has to be mentioned, then it was done in github.com/rakudo/rakudo/commit/3e...965336147c
DrForr O hai. Will check in a second...
22:11 comborico1611 joined, comborico1611 left 22:13 sena_kun left
DrForr .tell AlexDaniel RT tels me I have no permissions to viiew that ticket, Please feel free to close it. 22:15
yoleaux DrForr: I'll pass your message to AlexDaniel.
AlexDaniel .
yoleaux 22:15Z <DrForr> AlexDaniel: RT tels me I have no permissions to viiew that ticket, Please feel free to close it.
AlexDaniel RT#125299
wtf where did the ticket go
DrForr I can view it in the public tracker, I can't view it when logged in. 22:16
AlexDaniel rt.perl.org/Ticket/Display.html?id=125299
hmmm
DrForr Likely because of how I entered it, IE. via email. If you have perms to edit it, feel free to close it and paste this log for explanation. 22:17
geekosaur that's kinda a thing with rt.perl.org, yeh 22:19
annoying af
22:20 dmaestro joined
DrForr Blasted keyboard. Must get to bed here, like I said you hav my blanket permission to do whatever you like with the bug. 22:23
El_Che funny, bed and blanket in same sentence
DrForr And the same room in my case. Brand new duvet too. 22:24
El_Che DrForr: cold in Prague? 22:25
DrForr It was snowing this afternoon; I have a loft space here, and the (upward-facing) windows were covered in frost. 22:26
Okay, bedtime it is. G'night, good luck with RT. 22:29
22:34 lizmat joined
AlexDaniel DrForr: marked as ｢testneeded｣ 22:39
22:46 wamba left 22:47 natrys left 22:53 kitsunenokenja left
squashable6 🍕 jnthn++ wrote a comment on “Proposal: easier syntax for multiple param aliases”: github.com/rakudo/rakudo/issues/12...-348726245 22:53
Herby_ rookie question incoming... 22:56
how should module naming conventions work? sometimes I see Net::FTP convention 22:57
other times I just see FTP
why one or the other?
If I wanted to create a module for the oxford dictionary api, should I name it Oxford::Dictionary? or just Dictionary
that seems to be something Python avoids, for one reason or another 22:58
ctilmes Or Dictionary::Oxford so later someone else makes a similar API to Dictionary::Foo
Herby_ ah
good point
22:59 Cabanossi left
Herby_ one more dumb question: does the file structure matter with the naming convention used? 23:00
23:01 donaldh left
Herby_ If I use Dictionary::Oxford, does my file structure need to be: ./lib/dictionary/oxford.pm6 23:01
ctilmes Perl5 makes a concerted effort to maintain organization of module hierarchies. It kind of failed at that. IMHO, it is much less important in Perl 6 world
Herby_ or should it be ./lib/oxford.pm6
23:01 Cabanossi joined
ctilmes Convention would be to match file tree to name scheme, but it actually doesn't have to. 23:02
Herby_ thats what I assumed but I never knew for sure.
thanks
23:04 dmaestro left
ctilmes You have to put it somewhere it can be found in the manner you cause it to be searched for. 23:07
jnthn If you want it to work with -I then you need to follow the directory structure
ctilmes docs.perl6.org/language/modules#Finding_Modules 23:08
jnthn If you don't care about that (though it'd make development a bit "fun" :P) then one doesn't *have* to follow the convention, however :)
Herby_ gotcha 23:09
thanks for that link ctilmes
jnthn As ctilmes said, though, trying to build some neat hierarchy isn't something we worry about so much in the Perl 6 world. Searching and tags have won as ways of people fidning stuff. 23:10
23:12 kitsunenokenja joined
Herby_ makes sense. as a beginner I just want to make sure I stick to any standard conventions if there are some 23:12
23:13 ShalokShalom joined
AlexDaniel .tell wander good news!!! I think your RT comment went through 23:13
yoleaux AlexDaniel: I'll pass your message to wander.
AlexDaniel squashable6: status
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈12 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel, Log and stats: gist.github.com/d14b219452916bb401...b4b84c73dc
23:21 Aaronepower left
timotimo so ... the official csv games use 100x more lines in their hello.csv than we do in our test-t 23:28
but almost all of them are faster than rakudo is at the task
geekosaur was going to say "they're also probably 100x faster" :p 23:29
timotimo yup.
geekosaur we... have a ways to go
timotimo the slowest seems to be ruby, which takes about 9.3 seconds 23:30
Herby_ and what is p6 at?
timotimo takes just below 3 seconds now 23:31
maybe we want to have a version based on libcsv, too :P 23:32
the luajit version uses that adn clocks in at just below a second
23:34 dmaestro joined 23:36 mson left
timotimo since the input file is 23 megs big, taking a second means you're doing 23 megs per second 23:40
at that point you might as well be bottlenecked by spinning rust and filesystem :P
our csv library spends barely any time doing GC, that is good 23:42
i'm refering to percentage-wise, though 23:43
14 seconds in total
the gc runs 9950 times
we're still spending 20.7% of total time doing splitting 23:46
3.46% of that is sorting the indices of different split-pieces found in the string (adds up to 13 secodns) 23:47
Herby_ This API requires headers in the get request. app_id = 123, app_key = xyz . How do I use those in an HTTP::UserAgent get request?
I'm poking around the documentation but I can't quite figure it out
23:50 yqt left
timotimo it looks like headers get put as named arguments into the call 23:51
i see an example with post rather than get that passes X-Foo => "foodle"
so it's probably the same way for get 23:52
23:52 kitsunenokenja left
Herby_ something like: my %response = $ua.get($url, {app_key => $app_key, app_id => $app_id}) ? 23:54
timotimo no, without the braces 23:55
i think what you pass as a hash there would be form data for post requests and such 23:56
squashable6 🍕 vendethiel++ wrote a comment on “Proposal: easier syntax for multiple param aliases”: github.com/rakudo/rakudo/issues/12...-348729378 23:58
🍕🍕🍕 First contribution by vendethiel++! ♥
Herby_ timotimo: that works. thank you
timotimo YW 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!