»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.
MadcapJake I haven't made a slideshow in years, should I be doing 16:9 or 4:3? 00:15
MadcapJake 16:9 probably, (weird that LibreOffice still defaults to 4:3 at this point...) 00:16
b2gills sortiz: If you want to see what I was talking about earlier github.com/perl6/roast/pull/104 my be able to shine a light on it. 01:14
sortiz b2gills, thank you. I'll comment there. 01:22
dalek c: ff69616 | (Shlomi Fish)++ | doc/Language/5to6-nutshell.pod:
Add the equivalent of backticks - capture stdout.
 02:22
c: e0c27d6 | (Shlomi Fish)++ | doc/Language/5to6-nutshell.pod:
Add equivalent to each
c: 18da27a | (Shlomi Fish)++ | doc/Language/5to6-nutshell.pod:
Revise the qqx section.

Thanks to @AlexDaniel 's comment on the pull-req.
c: 81552ba | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Language/5to6-nutshell.pod:
Merge pull request #387 from shlomif/shlomif--enhance-5to6

Enhance the 5-to-6 documents .
Hotkeys is there something similar to smartmatch for regex that will return a boolean rather than me doing so $x ~~ $pattern' 02:58
er
quote before 'so' too
llfourn Hotkeys: don't think so 02:59
Hotkeys darn 02:59
llfourn m: $_ = 'abc'; say /ab/.so # well there's this 03:00
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
llfourn Hotkeys: ^ 03:01
Hotkeys weird
AlexDaniel m: $_ = ‘abc’; say ?m/ab/ 03:05
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: ↑
golfing again?
Hotkeys yeah
AlexDaniel 1 symbol is probably as short as it could get
Hotkeys I need it for two positionals though 03:06
so $_ is a no-go
Juerd m: "foo" ~~ /oo/ . Bool
camelia ( no output )
Hotkeys I'll have to stick with so
Juerd m: say("foo" ~~ /oo/ . Bool)
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
llfourn Hotkeys: what's wrong with match objects in this case?
Juerd Hotkeys: ? is shorter than so :)
AlexDaniel m: say ?(‘abc’~~?m/ab/)
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Juerd ?() is longer than so :) 03:07
AlexDaniel m: say so ‘abc’~~?m/ab/
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel same length
ahhh
no
m: say so‘abc’~~?m/ab/
Juerd What's ~~? :)
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel didn't know you can omit space after so…
Juerd: smartmatch
llfourn m: say &infix:<~~?> # I don't think it's a thing
camelia rakudo-moar 72dbb3: OUTPUT«5===SORRY!5=== Error while compiling /tmp/eKLy48ucXx␤Undeclared routine:␤ infix:<~~?> used at line 1. Did you mean 'infix:<~|>', 'infix:<∖>', 'infix:<lt>'?␤␤»
Juerd m: say "misleading" ~~? "nope"
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel whoah 03:08
I didn't notice it
Juerd AlexDaniel: No, you're actually doing FOO ~~ ?BAR
AlexDaniel yeah didn't notice it
Juerd $anything ~~ True is true.
AlexDaniel hey, but it works
m: say ‘abc’~~?m/abc/
Juerd "works" :)
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel m: say ‘abc’~~?m/abd/
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
llfourn it does?
AlexDaniel yup
llfourn so ~~ sets $_?
AlexDaniel m: say ‘abc’~~?m/abd{say $_}/ 03:09
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
konobi are there any "large" projects using perl6 yet at all?
AlexDaniel m: say ‘abc’~~?m/{say $_}abd/
camelia rakudo-moar 72dbb3: OUTPUT«abc␤abc␤abc␤abc␤False␤»
llfourn AlexDaniel: cool
Juerd konobi: That's not very likely
AlexDaniel konobi: depends on what you mean by “large”
llfourn I am building a large project in Perl 6. Not public yet though.
Hotkeys this the golf fn I have
m: my &f = {so$^a~$^b~~m:i/^[2|t0|t[w|o]?o][b[ee?|3]?]$/}; say f("2", "bee") 03:10
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys etc.
AlexDaniel Holy sh!
Juerd konobi: Or, well, since much of Perl 6 is written in Perl 6, and since Perl 6 is a large project: yep, there is! :) 03:10
konobi AlexDaniel: a good hierarchy
Hotkeys codegolf.stackexchange.com/question...1793#71793
AlexDaniel ok, let's see :)
konobi Juerd: =0P
Hotkeys that's why I can't use $_
I don't know if my regex is as minimal as can be there 03:11
oh
I can get rid of the group on the right
konobi AlexDaniel: i think i have a project idea that would really show off perl6 features... but trying to see if there's another that's a good example already
Juerd What's with this "bytes" thing? 03:12
Seems horribly antique
AlexDaniel konobi: I'd say just go with it if it makes sense
Hotkeys blame stack exchange
llfourn konobi: what's it do?
Hotkeys or actually
shoot
this doesn't work
AlexDaniel konobi: who knows, maybe somebody else will come with the same question and we will show off your project :)
konobi AlexDaniel: thinking a nix replacement
Hotkeys I'm a goof
llfourn what's a nix?
konobi mst and I have been throwing it back and forth for a while
AlexDaniel Hotkeys: yeah, just replace so with ~~? 03:13
AlexDaniel Hotkeys: and you'll save one character 03:13
konobi llfourn: nixos.org/nix/
Hotkeys hm?
AlexDaniel m: my &f = {$^a~$^b~~?m:i/^[2|t0|t[w|o]?o][b[ee?|3]?]$/}; say f("2", "beez")
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel m: my &f = {$^a~$^b~~?m:i/^[2|t0|t[w|o]?o][b[ee?|3]?]$/}; say f("2", "bee")
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys This regex is bad
I need to change it
llfourn checks it out
AlexDaniel I have no idea why that works but don't ask
Hotkeys m: my &f = {so$^a~$^b~~m:i/^[2|t0|t[w|o]?o][b[ee?|3]?]$/}; say f("2b", "ee")
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys see
I need to separate them 03:14
comma or something
Juerd t[w|o]?o is just obfuscation... too?|two is just as long.
Hotkeys hm
well this is golf
so it can stay
konobi llfourn: make sense? 03:15
Juerd Obfuscation isn't golf :)
Hotkeys m: {"$^a,$^b"~~?m:i/^[2|t0|t[w|o]?o]\,b[ee?|3]?$/}("2","Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Juerd Making shorter is
Hotkeys m: {"$^a,$^b"~~?m:i/^[2|t0|t[w|o]?o]\,b[ee?|3]?$/}("2B","ee").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys there
llfourn konobi: still absorbing gimmie a min :)
Hotkeys m: {"$^ax$^b"~~?m:i/^[2|t0|t[w|o]?o]xb[ee?|3]?$/}("2B","ee").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys m: {"$^ax$^b"~~?m:i/^[2|t0|t[w|o]?o]xb[ee?|3]?$/}("2","Bee").say 03:16
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
konobi Juerd: could be up your street too
Hotkeys m: {"$^a=$^b"~~?m:i/^[2|t0|t[w|o]?o]=b[ee?|3]?$/}("2","Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«5===SORRY!5===␤Unrecognized regex metacharacter = (must be quoted to match literally)␤at /tmp/EJLlcgHD2E:1␤------> 3{"$^a=$^b"~~?m:i/^[2|t0|t[w|o]?o]7⏏5=b[ee?|3]?$/}("2","Bee").say␤Couldn't find terminator / (corresponding / was at line 1…»
Hotkeys hmm
trying to not have to use an escape in there
Juerd konobi: I don't understand.
Hotkeys: _ 03:17
Hotkeys m: {"$^a,$^b"~~?m:i/^[2|t0|t[w|o]?o]\,b[ee?|3]?$/}("2B","ee").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys oops
didn't mean to do that
m: {"$^a_$^b"~~?m:i/^[2|t0|t[w|o]?o]_b[ee?|3]?$/}("2B","ee").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
konobi Juerd: it's a fully declarative package manager
Hotkeys m: {"$^a_$^b"~~?m:i/^[2|t0|t[w|o]?o]_b[ee?|3]?$/}("2","Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Juerd _ is the only non-alphanumeric ascii that doesn't need escaping, iirc
konobi with guarantees of versioning, etc. 03:18
AlexDaniel what about - and '?
Hotkeys I think it is being interpolated as part of the var name
Juerd konobi: What is?
Hotkeys in that string
konobi Juerd: nix
Juerd That's confusing. "Nix" sounds like "niks", which is Dutch for "nothing".
konobi but it has it's issues... but i think perl6 would excel at fixing em
llfourn konobi: it looks interesting. I am actually building something not similar but in the same realm. I'm making a cfg language.
AlexDaniel Hotkeys: once you smashed it into one string I don't think that there is a way out 03:19
konobi llfourn: oh nice! yeah... there's an aspect of that too
Hotkeys true
AlexDaniel Hotkeys: you need this 03:20
m: say so (‘hello’, ‘world’) ~~ (/‘hello’/, /‘world’/)
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel m: say so (‘hello’, ‘world’) ~~ (/‘hallo’/, /‘world’/)
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
llfourn konobi: I should have something up in the next week or two will appreciate your feedback :)
Hotkeys hmm
konobi llfourn: mst would have an interest too i think
llfourn Awesome. It needs some input from ppl with a lot of unix knowledge. 03:21
Juerd Hotkeys: "@_[]" and \s
llfourn it's essentially a glorified shell script transpiler designed for curl | sh usage
llfourn so you write shell scripts in Perl 6 basically 03:22
konobi Juerd: nix gives you a verifiabily reproduceable system based on declarative configs
Juerd konobi: That's nice.
konobi and they have a nixos linux distro 03:23
Hotkeys m: {($^a,$^b) ~~ (/^[2|t0|t[w|o]?o]$/,/^b[ee?|3]?$/)}("2", "bee")
camelia ( no output )
Juerd Hotkeys: Ooh, ~@_ even
Hotkeys m: {($^a,$^b) ~~ (/^[2|t0|t[w|o]?o]$/,/^b[ee?|3]?$/)}("2", "bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys Juerd: fancy
AlexDaniel konobi: what I can say is that if you're building something big then expect to report a few bugs. It's not too bad, with a 50% chance your bug will be fixed very quickly :) but just don't expect it to go too smoothly 03:24
Hotkeys m: {~@_ ~~ ?m:i/^[2|t0|t[w|o]?o]\sb[ee?|3]?$/}("2", "bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys m: {~@_ ~~ ?m:i/^[2|t0|t[w|o]?o]\sb[ee?|3]?$/}("2", "Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys m: {~@_ ~~ ?m:i/^[2|t0|t[w|o]?o]\sb[ee?|3]?$/}("2B", "Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys Juerd++
m: {~@_~~?m:i/^[2|t0|t[w|o]?o]\sb[ee?|3]?$/}("2B", "Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel Hotkeys: so what about making it right??
Hotkeys m: {~@_~~?m:i/^[2|t0|t[w|o]?o]\sb[ee?|3]?$/}("2", "Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys AlexDaniel: hm?
m: {~@_~~?m:i/^[2|t0|too|two?]\sb[ee?|3]?$/}("2", "Bee").say # less obfuscated 03:25
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
konobi AlexDaniel: heh... perl6 doesn't scare me... more interested in a way to really use the new functionality... roles feature heavily in my head
Juerd Hotkeys: too|two? looks wrong. ITYM too?|two
Hotkeys oops you're right 03:26
m: {~@_~~?m:i/^[2|t0|too?|two]\sb[ee?|3]?$/}("2", "Bee").say # less obfuscated
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys I inserted the |two in the wrong place
Juerd Darn, t[0|oo?|wo] isn't shorter :)
Hotkeys shh
er
darn
AlexDaniel m: {so@_~~(/:i^[2|t0|t[w|o]?o]$/,/:i^b[ee?|3]?$/)}("2", "Bee").say 03:27
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: ↑
Hotkeys oh
AlexDaniel and yeah, I just shove off 2 chars just by getting :i in
Hotkeys you see I couldn't get the adverbs to work when I tried it
when I did m:i with that
AlexDaniel yup because that's not how it works :)
Hotkeys well it should 03:28
AlexDaniel nope? Here we are smartmatching on a list of regexes
Hotkeys you actually don't need so with that
btw
AlexDaniel ooh right
Juerd Oh, heh, the ~ in front of @_ isn't even needed :)
Hotkeys m: {@_~~(/:i^[2|t0|t[w|o]?o]$/,/:i^b[ee?|3]?$/)}("2", "Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Juerd: ~ makes it wrong.
Hotkeys what do you mean wrong
Juerd AlexDaniel: Not if you match one thing, and use \s
AlexDaniel Juerd: ("2 Bee", "")
Hotkeys: ↑ 03:29
Juerd Hm. True.
AlexDaniel just use a list of regexes. It works this way
Hotkeys m: {~@_~~?m:i/^[2|t0|too?|two]\sb[ee?|3]?$/}("2 Bee". "").say # less obfuscated
camelia rakudo-moar 72dbb3: OUTPUT«5===SORRY!5=== Error while compiling /tmp/78iKHh3FVn␤Unsupported use of . to concatenate strings; in Perl 6 please use ~␤at /tmp/78iKHh3FVn:1␤------> 3/^[2|t0|too?|two]\sb[ee?|3]?$/}("2 Bee".7⏏5 "").say # less obfuscated␤»
Hotkeys beat
neat
konobi AlexDaniel: it's also a good excuse to get up to date on perl6
AlexDaniel konobi: :) 03:29
konobi has been quite a while since i was added to pugs =0) 03:30
Juerd AlexDaniel: @_~~/foo/,/bar/ ends up with ($something, /bar/), though. 03:31
AlexDaniel Juerd: yup, you need parens 03:31
Juerd Meh
I like the wrong version better :)
AlexDaniel it's wrong >:( 03:32
and honestly, the right version is not very longer
Hotkeys nope
it's actually very competitive
even with the golf langs
thank regex for that
Juerd But parens :( 03:33
AlexDaniel Juerd: yeah yeah, I know!
Hotkeys darn 03:34
1 byte more than the javascript answer
konobi AlexDaniel, llfourn: i think it's also an interesting thought exercise... what's the best way to model "packaging" and "software"
AlexDaniel Hotkeys: I think I can fix that 03:35
Hotkeys er actually
the javascript one doesn't look correct
"(a,b)=>/^(2|t[ow]?o|t0)b(3|e?e)?$/.test(a+b)"
that would have the same problem as my string mashing one no? 03:36
Juerd Worse
Hotkeys I don't know how to test a JS lambda
AlexDaniel m: {@_~~(/:i^[2|t0|two|to?o]$/,/:i^be?e?|3$/)}("2", "Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: I think that this is 2 chars shorter
Hotkeys: please verify
Hotkeys yep 03:37
Juerd Hotkeys: Test with (() => ...)(args)
llfourn konobi: I think you don't. Just start with a blank VM/container each time and write tests to make sure it got where it needs to be. 03:37
AlexDaniel OH
Juerd Hotkeys: Or var x = () => ...; x(args)
AlexDaniel m: {@_~~(/:i^2|t0|two|to?o$/,/:i^be?e?|3$/)}("2", "Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel ok we're getting closer to 0 bytes 03:38
Hotkeys are you sure that works
AlexDaniel Hotkeys: test it
Hotkeys my first try was without that group
Juerd AlexDaniel: It should test for b3, not 3
AlexDaniel Hotkeys: sure enough, but you had everything smashed together
Hotkeys m: {@_~~(/:i^2|t0|two|to?o$/,/:i^be?e?|3$/)}("2to", "Bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys it lumps the $ in the | bit 03:39
AlexDaniel m: {@_~~(/:i^[2|t0|two|to?o]$/,/:i^b[e?e?3]$/)}("too", "b3").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: oh really? Dammit! 03:40
indeed
llfourn m: say 'a','b' ~~ (/a/,/b/) # didn't know this worked
camelia rakudo-moar 72dbb3: OUTPUT«aFalse␤»
AlexDaniel m: {@_~~(/:i^[2|t0|two|to?o]$/,/:i^b[e?e?3]$/)}("too", "bee3").say # WRONG!
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys it needs the ? on the be group too
llfourn m: say ('a','b') ~~ (/a/,/b/) # didn't know this worked
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys cause it can test for b
AlexDaniel Hotkeys: ok then I don't know
konobi llfourn: well no... there's a fair amount of thought that needs to go into it
Hotkeys yeah
looks like 45 is where it stays 03:41
45 is good enough
that JS answer is wrong anyway
I think
I still haven't tested
Juerd Most answers are wrong because of string concat.
AlexDaniel ↑ point it out in your answer
Juerd Or reply to the wrong one 03:42
Off to bed, ciao!
Hotkeys I replied to the JS one 03:43
AlexDaniel xD 03:44
Hotkeys: I have a gut feeling that this could be shorter
let's see…
Hotkeys maybe
m: {**~~(/:i^[2|t0|two|to?o]$/,/:i^b[e?e?3]$/)}("too", "bee").say 03:45
camelia rakudo-moar 72dbb3: OUTPUT«Too many positionals passed; expected 0 or 1 arguments but got 2␤ in block <unit> at /tmp/_nLh0dl4V0 line 1␤␤»
Hotkeys darn
m: {** ~~(/:i^[2|t0|two|to?o]$/,/:i^b[e?e?3]$/)}("too", "bee").say
camelia rakudo-moar 72dbb3: OUTPUT«Too many positionals passed; expected 0 or 1 arguments but got 2␤ in block <unit> at /tmp/QmskbJCkLj line 1␤␤»
Hotkeys thought I'd be cheeky
flussence m: {@_~~(~2|/:i^t[0|wo|o?o]$/,/:i^b[e?e?3]$/)}("too", "b3").say 03:46
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
konobi llfourn: i mean, perl6 took a while to get into shape =0) 03:46
llfourn konobi: What I meant was I try to avoid issues with package versions by as much as possible by locking them down in docker and writing tests. 03:47
I never do apt-get updgrade etc 03:48
or try and model anything
but I am mostly developing software for myself no idea what it's like in enterprise :P
AlexDaniel Hotkeys: why are we doing this with regexes???
Hotkeys idk 03:49
everyone else was
I got tunnel vision
>_>
konobi llfourn: docker doesn't give you that.
llfourn give what?
atweiden konobi: yes on nix replacemen 03:50
something more like guix except in p6? 03:51
Hotkeys regex seems like it'd be the shortest, but then again I'm a bit tired
AlexDaniel m: {@_~~(any(<to too two t0 2>), any(<b be bee b3>))}("too", "b3").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel honestly that is not much longer
konobi atweiden: well... still trying to work it out... mst too
AlexDaniel Hotkeys: perhaps list that as an alternative non-golfy variant 03:52
Hotkeys: because hey, that's just some regular code
konobi atweiden: things like being able to apply a role to a package like "build::cmake" or "build::automake" or "build::gem" or "build::npm", etc. 03:53
atweiden sounds like guix
ugexe see %?RESOURCE build-time in s22
atweiden i really like the idea of replacing guile scheme with p6 though 03:54
AlexDaniel Hotkeys: -1 character if you remove parens in the second ｢any｣
Juerd m: {@_~~any(<t0 to too two 2>X,<b be bee b3>)}("2", "b3")
camelia ( no output )
Juerd m: {@_~~any(<t0 to too two 2>X,<b be bee b3>)}("2", "b3").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Juerd Couldn't resist.
mspo Juerd: case insensitive 03:55
Juerd It throws an exception instead of returning false, when invalid values are given, though
Oh, case :(
Regexes then :P
Hotkeys m: {@_~~(any(<to too two t0 2>), any(<b be bee b3>))}("too", "B3").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys AlexDaniel:
konobi atweiden: not too familiar with guix... that's a fsf thing, right?
AlexDaniel nooooooo case sensitivity!!
Juerd @_>>.lc is *long*
Hotkeys I was just about to add it to my post too 03:56
Juerd (Don't go for @_».lc because that's more bytes :P)
AlexDaniel m: {@_».lc~~(any(<to too two t0 2>), any <b be bee b3>)}("too", "B3").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys it's the same amount of bytes actually
Juerd Is it?
Oh, it is
flussence m: '»'.encode('windows-1252').bytes.say # :)
camelia rakudo-moar 72dbb3: OUTPUT«1␤»
Hotkeys AlexDaniel: I'll include that one
AlexDaniel flussence: oh come on
Hotkeys flussence: heh
konobi oh! 03:57
AlexDaniel flussence: don't bring your cave things here please :D
Hotkeys m: '»'.encode('C-1252').bytes.say # :)
camelia rakudo-moar 72dbb3: OUTPUT«Unknown string encoding: 'c-1252'␤ in block <unit> at /tmp/7jdaiot_4c line 1␤␤»
Hotkeys er
there's one that this other golf lang uses
flussence well if html5 says that's the default, why not...
konobi anyone know of any good articles on using perl6 for data analytics?
Juerd {@_».lc~~any(<t0 to too two 2>X,<b be bee b3>)} is shorter but I don't get why it errors on invalid values.
atweiden konobi: afaik guix wraps the nix pkg manager gnu.org/software/guix/
konobi ah yeah... i stop at gnu/fsf i'm afraid 03:58
AlexDaniel but I've never liked how all these golf tasks say “hey, in the fewest bytes possible”… bytes of what??
Juerd And with '?try ' in front it's 1 char longer than the any,any 03:58
AlexDaniel if I give them compiled C file is that going to count?
Hotkeys There's actually a guy that posts with some lang 03:59
AlexDaniel but then oh god, which architecture?
Hotkeys and specifies "Using blah encoding"
Juerd Oh, miscounted. It's the same length.
m: {?try @_».lc~~any(<t0 to too two 2>X,<b be bee b3>)}("twa", "ba").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Juerd m: {?try @_».lc~~any(<t0 to too two 2>X,<b be bee b3>)}("two", "be").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Juerd Oh, it doesn't need the ()! 04:00
Hotkeys I'll stick with {@_».lc~~(any(<to too two t0 2>), any(<b be bee b3>))} as the fairly readable non-competing alternative
Juerd m: {?try @_».lc~~any <t0 to too two 2>X,<b be bee b3>}("two", "be").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Juerd Shaved one off :)
AlexDaniel Hotkeys: don't you want to get it to 53 by removing a space and parens? :)
Hotkeys Nah
AlexDaniel fine
Hotkeys It's non-competing
it looks nicer with those in
:p
Juerd Hotkeys: ^
Hotkeys actually I'll remove a space 04:01
there it's at 53
er
54
AlexDaniel do we even need “any”? 04:02
Juerd Yes.
Hotkeys m: {@_».lc~~(<to too two t0 2>,<b be bee b3>)}("too", "B3").say 04:03
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Juerd {?try @_>>.lc~~any <t0 to too two 2>X,<b be bee b3>} is 52, isn't it?
Hotkeys hmm
AlexDaniel m: {@_».lc~~(<to too two t0 2>.any,<b be bee b3>.any)}("too", "b3").say 04:05
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: ↑ no parens
Hotkeys yeah Juerd
nice AlexDaniel
Juerd I already had a parenless one :)
Somehow, a golf without parens feels like earning bonus points 04:06
Anyhow, I should really get some sleep. Golf is addictive.
AlexDaniel yeah it is
procrastination
AlexDaniel I just ran some script which is going to take forever to finish… so, I guess I can play with golf for quite some time :) 04:07
Hotkeys: I am not satisfied. I want it shorter
Herby_ Evening, everyone! 04:07
Hotkeys AlexDaniel: I'm working on something 04:08
hold on
m: {?all @_ Z∈<to too two t0 2>».Str,<b be bee b3>}("2","bee") 04:10
camelia ( no output )
Hotkeys m: {?all @_ Z∈<to too two t0 2>».Str,<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys if you can find a way for me to not have to string the whole first match list
I would love you forever
because
AlexDaniel ooooooooh
Hotkeys m: <2 to>.perl.say 04:10
camelia rakudo-moar 72dbb3: OUTPUT«(IntStr.new(2, "2"), "to")␤»
Hotkeys it does that
which doesn't element
AlexDaniel m: <to 2>.perl.say 04:11
camelia rakudo-moar 72dbb3: OUTPUT«("to", IntStr.new(2, "2"))␤»
Hotkeys quite annoying
AlexDaniel ~« ?
Hotkeys I did try that
I think
Hotkeys m: say ~«<to too two 2 t0> 04:11
camelia rakudo-moar 72dbb3: OUTPUT«(to too two 2 t0)␤»
Hotkeys m: say ~«<to too two 2 t0>.perl
camelia rakudo-moar 72dbb3: OUTPUT«(("to", "too", "two", IntStr.new(2, "2"), "t0"))␤»
Hotkeys m: say (~«<to too two 2 t0>).perl
camelia rakudo-moar 72dbb3: OUTPUT«("to", "too", "two", IntStr.new(2, "2"), "t0")␤»
Hotkeys nope
azawawi hi
AlexDaniel seems like intstr is still a str 04:12
azawawi pasteboard.co/1uqPo9c8.png # Camelia The gray attacks Atom :)
Hotkeys ∈ is very strict though
llfourn m: say IntStr ~~ Str
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys m: "2" ∈ <2 to>
camelia rakudo-moar 72dbb3: OUTPUT«WARNINGS for /tmp/kwWL8IoifH:␤Useless use of "∈" in expression "\"2\" ∈ <2 to>" in sink context (line 1)␤»
Hotkeys m: say "2" ∈ <2 to>
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel m: say "2" ∈ <to 2>
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel m: say 2 ∈ <to 2>
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel m: say "to" ∈ <to 2> 04:13
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel isn't it a bug?…
Hotkeys could be
llfourn m: say "to" ∈ set(<to 2>)
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
llfourn m: say "2" ∈ set(<to 2>)
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
llfourn m: say "2".Int ∈ set(<to 2>) 04:14
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel “Returns True if $a is an element of $b.”
so it's not an element so fuck off? :/
Hotkeys idk
m: say <2 to> ∋ "2"
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys darn
llfourn m: <to 2>.map: *.^name..say
camelia rakudo-moar 72dbb3: OUTPUT«5===SORRY!5===␤Argument to "say" seems to be malformed␤at /tmp/f_hSe984uD:1␤------> 3<to 2>.map: *.^name..say7⏏5<EOL>␤Other potential difficulties:␤ Unsupported use of bare "say"; in Perl 6 please use .say if you meant $_, or use an …»
llfourn m: <to 2>.map: *.^name.say
camelia rakudo-moar 72dbb3: OUTPUT«Str␤IntStr␤»
llfourn m: say "2".IntStr ∈ set(<to 2>) 04:15
camelia rakudo-moar 72dbb3: OUTPUT«Method 'IntStr' not found for invocant of class 'Str'␤ in block <unit> at /tmp/uZuWao4YxI line 1␤␤»
masak ah, the joys of hybrid types.
llfourn m: <2>.map: { say "2" ~~ $_ } 04:16
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys why doesn't qw just do it as a string
llfourn m: <2>.map: { say .^name; say "2" ~~ $_ }
camelia rakudo-moar 72dbb3: OUTPUT«IntStr␤True␤»
Hotkeys m: say qw|2 to| ∋ "2"
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys OH
that's shorter than ».Str
AlexDaniel Hotkeys: so what's the final variant? 04:17
Hotkeys m: {?all @_ Z∈qw|to too two t0 2|,<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys I don't think there's any more whitespace I can remove
AlexDaniel remove something else then
Hotkeys heh
we're down to 47 bytes 04:18
if I can somehow lose 2 more I'll include it as an alternative
Juerd Nope. .lc
Hotkeys oh 04:19
right
darn
way to be a buzzkill Juerd
Juerd Closing laptop now :)
AlexDaniel Juerd: go get some rest…
llfourn (Juered still not in bed)
Hotkeys m: {?all @_».lc Z∈qw|to too two t0 2|,<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys that's another 5
back to 52
AlexDaniel ok well we should get rid of ‘all’, it gets in the way 04:20
Hotkeys m: {? @_».lc Z∈qw|to too two t0 2|,<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«(False)␤»
Hotkeys m: {so @_».lc Z∈qw|to too two t0 2|,<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«(False)␤»
Hotkeys m: {@_».lc Z∈qw|to too two t0 2|,<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«(True True)␤»
Hotkeys that's why I had ?all 04:21
hmm
m: {[&&] @_».lc Z∈qw|to too two t0 2|,<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys same number of bytes though
m: {[&&]@_».lc Z∈qw|to too two t0 2|,<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Sb_3aOeuQV␤Two terms in a row␤at /tmp/Sb_3aOeuQV:1␤------> 3{[&&]7⏏5@_».lc Z∈qw|to too two t0 2|,<b be bee b␤ expecting any of:␤ infix␤ infix stopper␤ statement en…»
AlexDaniel hmm 04:22
AlexDaniel hey look 04:22
AlexDaniel m: {@_».lc~~([&]<to too two t0 2>,[&]<b be bee b3>)}("too", "b3").say 04:23
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: that's two bytes shorter than <>.any
Hotkeys crazy
AlexDaniel how long is that by the way
Hotkeys m: say [&]<to too two t0 2>
camelia rakudo-moar 72dbb3: OUTPUT«all()␤»
Hotkeys ...
m: {[&&]@_».lc Z∈qw|to too two t0 2|,<b be bee b3>}("2","farts").say
camelia rakudo-moar 72dbb3: OUTPUT«5===SORRY!5=== Error while compiling /tmp/xiZFjk6Aja␤Two terms in a row␤at /tmp/xiZFjk6Aja:1␤------> 3{[&&]7⏏5@_».lc Z∈qw|to too two t0 2|,<b be bee b␤ expecting any of:␤ infix␤ infix stopper␤ statement en…»
Hotkeys m: {@_».lc~~([&]<to too two t0 2>,[&]<b be bee b3>)}("too", "farts").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel ah geez, it doesn't work
Hotkeys :p 04:24
AlexDaniel why?
Hotkeys sec
m: {@_».lc~~([|]<to too two t0 2>,[|]<b be bee b3>)}("too", "farts").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys m: {@_».lc~~([|]<to too two t0 2>,[|]<b be bee b3>)}("too", "b3").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys darn
AlexDaniel it needs parens 04:25
m: {@_».lc~~([|](<to too two t0 2>),[|](<b be bee b3>))}("too", "b3").say 04:26
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel m: {@_».lc~~([|](<to too two t0 2>),[|](<b be bee b3>))}("too", "b4").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel meh!
too long
m: {?[*] @_».lc Z∈qw{to too two t0 2},<b be bee b3>}("2","bee").say 04:32
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel m: {?[*] @_».lc Z∈qw{to too two t0 2},<b be bee b3>}("2","beez").say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel Hotkeys: now that's crazy, but not shorter
Hotkeys heh
m: {?[&] @_».lc Z∈qw{to too two t0 2},<b be bee b3>}("2","beez").say 04:33
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
Hotkeys m: {?[&] @_».lc Z∈qw{to too two t0 2},<b be bee b3>}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys might as well
although that brings us back to ?all
AlexDaniel Hotkeys: it is funny how pyth solution is 39 bytes 04:34
and perl6 is 45 :D
AlexDaniel that's really huge readability per byte ratio :) 04:35
Hotkeys yeah 04:36
AlexDaniel Hotkeys: you wouldn't believe it!
ah no…
Hotkeys he didn't even do a breakdown
like the other golf langs do
s/he/they/
the page he links is just a list of char -> instruction 04:37
s/he/they/
AlexDaniel m: {$^a.lc∈qw{to too two t0 2}&&$^b.lc∈<b be bee b3>}("2","bee").say 04:38
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel I thought that this is shorter but then I realized there is no .lc
AlexDaniel Hotkeys: well 04:49
-1 char 04:50
m: {?+(@_».lc Z∈<to too two t0 2>,<b be bee b3>)}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel oops
m: {?+(@_».lc Z∈qw{to too two t0 2},<b be bee b3>)}("2","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys wait 04:51
but that worked
without qw
m: {?+(@_».lc Z∈<to too two t0 2>,<b be bee b3>)}("2b","bee").say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys oh
nvm
er
oh 04:52
that won't work
AlexDaniel hahaha
right…
my bad
Hotkeys cause truthy is anything > 0
well actually that'll always be 2
so
AlexDaniel yup I see
but maybe we can do that with slip somehow? 04:53
actually, I've just noticed 04:56
Hotkeys: that this alternative has to use qw as well
gfldex .oO( use v6; to beat them at their own game. )
AlexDaniel m: say {@_».lc~~((qw<to too two t0 2>,<b be bee b3>)».any)}(‘2’,‘bee’) 04:59
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel same amount of bytes
Hotkeys smells longer
AlexDaniel Hotkeys: exactly the same 05:00
oooooh
m: say {@_».lc~~(qw<to too two t0 2>,<b be bee b3>)».any}(‘2’,‘bee’)
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: now that is two characters shorter! 05:01
Hotkeys same length in bytes actually
{@_».lc~~(qw<to too two t0 2>,<b be bee b3>)».any}
{?all @_».lc Z∈qw|to too two t0 2|,<b be bee b3>}
AlexDaniel I didn't try to get it to compete, but… 05:02
Hotkeys neither of those are competing
AlexDaniel there's one thing I often find crazy in code golf
gfldex m: -> Int $a, Str $b {1}(1,"2") 05:03
camelia ( no output )
AlexDaniel just look at this example! it is as short as it could be
Hotkeys what do you find crazy?
AlexDaniel Hotkeys: that it is not going to be shorter ever 05:04
Hotkeys oh
AlexDaniel and yet comes somebody who says “hey look, here it is, twice shorter than that” 05:04
and I'm like “whyyyy”
Hotkeys lol 05:05
like how the regex example is still 7 bytes shorter than the above
gfldex m: say sub f($a){$a}(1);
camelia rakudo-moar 72dbb3: OUTPUT«1␤»
gfldex calling functions (and blocks) in place seams not to be docked 05:07
AlexDaniel m: say {@_».lc~~qw<to too two t0 2 b be bee b3>[^5,5..*]».any}(‘2’,‘bee’) # nope
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
Hotkeys lol 05:08
azawawi about %?RESOURCES, does on have to specify every folder/file to be copied in META.info "resources" key? (e.g. resources/sample_dir) 05:10
AlexDaniel Hotkeys: hang on, it looks like regex example is -2 bytes 05:11
AlexDaniel m: {@_~~(/:i^t[0|oo?|wo]$/|2,/:i^b[ee?|3]?$/)}(‘to’,‘bee’).say 05:16
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: the problem is that it errors out if you do this 05:17
m: {@_~~(/:i^t[0|oo?|wo]$/|2,/:i^b[ee?|3]?$/)}(‘HAHA’,‘bee’).say
camelia rakudo-moar 72dbb3: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5HAHA' (indicated by ⏏)␤ in block <unit> at /tmp/MKPkUUV3Gi line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/MKPkUUV3Gi line 1␤␤»
Hotkeys lol
AlexDaniel but for valid inputs it works, yeah
:/
Hotkeys: you can put '2' in quotes but then it ruins everything 05:18
oh, but wait!
no, nothing 05:19
Hotkeys lol
AlexDaniel Hotkeys: oh, YEAH! 05:21
m: {@_~~(~2|/:i^t[0|oo?|wo]$/,/:i^b[ee?|3]?$/)}(‘MWAHAHAHA’,‘bee’).say
camelia rakudo-moar 72dbb3: OUTPUT«False␤»
AlexDaniel m: {@_~~(~2|/:i^t[0|oo?|wo]$/,/:i^b[ee?|3]?$/)}(‘2’,‘bee’).say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel m: {@_~~(~2|/:i^t[0|oo?|wo]$/,/:i^b[ee?|3]?$/)}(2,‘bee’).say
camelia rakudo-moar 72dbb3: OUTPUT«True␤»
AlexDaniel Hotkeys: that's -1 byte as far as I can see 05:22
Hotkeys cheeky
I like it
Hotkeys didn't mean to italics there 05:28
azawawi where can i find Distribution.install(...) source code? 05:29
jdv79 grep src/core ? 05:30
azawawi seems to me it does not handle recursive directories copying in %resources 05:30
jdv79 i bet you'll find it in src/core/Distribution.pm. wild guess. 05:31
AlexDaniel Hotkeys: HA-HA! 05:36
ah,no…
xD
Hotkeys lol
azawawi jdv79: nearly correct :) github.com/rakudo/rakudo/blob/nom/...on.pm#L112 05:37
AlexDaniel Hotkeys: Shakespeare answer is really cool 05:39
Hotkeys yeash 05:40
yeah 05:41
jdv79 cool
AlexDaniel Hotkeys: so will you change it to 44 byte version? 05:42
I don't think there is any way to make it shorter. That's it
Hotkeys oh 05:46
yeah
sec
Hotkeys AlexDaniel: there we go 05:52
tied with JS now
AlexDaniel Hotkeys: kay. It seems like my edit did not make it through, please fix <to too two t0 2> 05:53
Hotkeys: it has to be qw<>
Hotkeys oh
AlexDaniel you can change any() to .any to keep the byte count 05:54
Hotkeys good catch 05:56
Why does <> do IntStr but qw<> (s/<>/etc./) does Str 05:58
AlexDaniel m: say {("0."~.flip for ^$_)}(25) 06:03
camelia rakudo-moar 72dbb3: OUTPUT«(0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.01 0.11 0.21 0.31 0.41 0.51 0.61 0.71 0.81 0.91 0.02 0.12 0.22 0.32 0.42)␤»
AlexDaniel m: say {(^$_)».&{"0."~.flip}(25)
camelia rakudo-moar 72dbb3: OUTPUT«5===SORRY!5===␤Expression needs parens to avoid gobbling block␤at /tmp/NQ1yQR5tUH:1␤------> 3say {(^$_)».&{"0."~.flip}7⏏5(25)␤Missing block (apparently claimed by expression)␤at /tmp/NQ1yQR5tUH:1␤------> 3say 7⏏5{(^$_)».&{"0."~…»
AlexDaniel m: say {(^$_)».&{"0."~.flip}}(25)
camelia rakudo-moar 72dbb3: OUTPUT«(0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.01 0.11 0.21 0.31 0.41 0.51 0.61 0.71 0.81 0.91 0.02 0.12 0.22 0.32 0.42)␤»
AlexDaniel m: say {map ^$_: "0."~*.flip}(25) 06:05
camelia rakudo-moar 72dbb3: OUTPUT«(0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.01 0.11 0.21 0.31 0.41 0.51 0.61 0.71 0.81 0.91 0.02 0.12 0.22 0.32 0.42)␤»
AlexDaniel Hotkeys: come on… same character count
Hotkeys lol 06:07
AlexDaniel Hotkeys: perhaps you can also include the last one :) 06:07
Hotkeys Sure 06:08
Hotkeys I always instinctively go for the 'for' version 06:10
plus it usually ends up being about the same length
m: {map ^$_: "0."~.flip}(25).say
camelia rakudo-moar 72dbb3: OUTPUT«Cannot call map(Range: Str); none of these signatures match:␤ ($: Hash \h, *%_)␤ (\SELF: &block;; :$label, :$item, *%_)␤ (HyperIterable:D $: &block;; :$label, *%_)␤ in block <unit> at /tmp/dnORgOcmgb line 1␤␤»
Hotkeys aww
m: {map ^$_:"0."~*.flip}(25).say 06:11
camelia rakudo-moar 72dbb3: OUTPUT«5===SORRY!5=== Error while compiling /tmp/xCQtBYTYT1␤Confused␤at /tmp/xCQtBYTYT1:1␤------> 3{map ^$_:7⏏5"0."~*.flip}(25).say␤ expecting any of:␤ colon pair␤»
AlexDaniel m: say {"0."xx$_ Z~(^$_)».flip}(25) # slightly longer
camelia rakudo-moar 72dbb3: OUTPUT«(0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.01 0.11 0.21 0.31 0.41 0.51 0.61 0.71 0.81 0.91 0.02 0.12 0.22 0.32 0.42)␤»
AlexDaniel m: say {"0."X~(^$_)».flip}(25) 06:13
camelia rakudo-moar 72dbb3: OUTPUT«(0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.01 0.11 0.21 0.31 0.41 0.51 0.61 0.71 0.81 0.91 0.02 0.12 0.22 0.32 0.42)␤»
AlexDaniel Hotkeys: HA
Hotkeys damn
_Gustaf_ o/ 06:14
AlexDaniel Hotkeys: I liked this one 06:15
Hotkeys: that's one of these moments. Look how readable it was
and I only made it much uglier
19 bytes, right? 06:16
Hotkeys: perhaps keep previous attempts. People seem to like it 06:17
Hotkeys 20
AlexDaniel right. Forgot about » 06:18
Hotkeys What's your stackoverflow name 06:19
I'll credit you
AlexDaniel Hotkeys: codegolf.stackexchange.com/users/49...akimenko-a
goddamn thing can't accept my name properly
stupid length limits 06:20
Hotkeys lol 06:21
AlexDaniel Hotkeys: this one is easy 06:42
Hotkeys: codegolf.stackexchange.com/question...1393#71393
Hotkeys: replace ｢$_ <1｣ with ｢1>$_｣
Hotkeys: replace ｢by(*.chars)｣ with ｢by(&chars)｣ 06:43
Hotkeys: ｢* <10｣ with ｢10>*｣
Hotkeys can you chain 1 <= 10 > * the same as I have it? 06:44
er
do you mean 10 > * <= 1 06:45
AlexDaniel no idea actually
I didn't notice the chaining
Hotkeys: but yeah reverse the whole thing 06:46
1 <= * < 10 → 10 > * >= 1
Hotkeys: actually I still can't read it so I don't know :) 06:47
Hotkeys {my \s=($_,*×(1>$_??10!!.1)…10>*>=1);min(s[*-1]~"x10^"~s-(1>$_??s*2!!1),$_,:by(&chars))}
is where we're at
that post was just me attempting while half awake 06:48
AlexDaniel Hotkeys: by=>&chars 06:51
AlexDaniel Hotkeys: it is interesting that @a[*-1] is equal in length to @a.tail 06:58
Hotkeys they don't do quite the same thing though 06:59
AlexDaniel @a.tail(2) and @a[*-2..*] are same too
Hotkeys m: my @a = ^5; @a.tail.WHAT.say; @a[*-1].WHAT.say 07:00
camelia rakudo-moar 72dbb3: OUTPUT«(Seq)␤(Int)␤»
AlexDaniel Hotkeys: I'd say that tail is cooler because it returns a seq
for one element it is not relevant though :)
Hotkeys right
m: my @a = ^5; @a.tail(3) # more interesting
camelia ( no output )
Hotkeys m: my @a = ^5; @a.tail(3).say # more interesting
camelia rakudo-moar 72dbb3: OUTPUT«(2 3 4)␤»
Hotkeys I'll go to bed, if you want AlexDaniel leave your shortened version of the golf as a comment and I'll update the post tomorrow 07:11
(the codegolf board generally doesn't like edits for improvements not sure why)
AlexDaniel Hotkeys: I wont look at it anymore
Hotkeys: it is boring
Hotkeys lol 07:12
fair enough
'night
AlexDaniel so just get the latest version there, I guess
dalek osystem: 3f4fd5f | (Sam Gillespie)++ | META.list:
Update META.list

Adding the Net::OSC module to the ecosystem :)		 07:20
osystem: c1d4465 | sylvarant++ | META.list:
Merge pull request #152 from samgwise/patch-1

Update META.list
[Tux] test 22.228 07:36
test-t 12.130
csv-parser 51.622
lizmat good *, #perl6! 08:40
I hope the test-t > 12 is noise, and not cause by the Failure.handled changes
*caused
dalek kudo/nom: aaf5c26 | (Salvador Ortiz)++ | src/core/Setty.pm:
Fix for RT#127402: Set.hash stringifies its objects
 08:51
kudo/nom: 691b39b | lizmat++ | src/core/Setty.pm:
Merge pull request #711 from salortiz/setty_hash

Fix for RT#127402: Set.hash stringifies its objects
RabidGravy wh'appen peoples! 09:46
DrForr Nada mucho. 09:51
dalek ar: 4b44be8 | FROGGS++ | tools/star/mac-dmg.pod:
fix typo
 10:15
brrt fwiw, the perlsecco wine i got at FOSDEM was really nice :-) 11:25
fireartist just wrote a little script to pretty-print resursive directory listing - wow! - perl6 IO is *nice* 11:30
brrt perl6 is nice, in general :-) 11:36
i'm wondering if there is a perl6 equivalent of this
brrt my %hash = $string =~ m/(\w+) = (\w+)/g; 11:36
RabidGravy something with a split and map I would go for 11:38
got something very similar in the code I'm working on right now
brrt i think the perl5 idiom is really nice 11:41
RabidGravy m: say "foo=bar".match(/$<key>=(\w+) "=" $<value>=(\w+)/, :g).map({ ~$_<key> => ~$_<value>}) 11:42
camelia rakudo-moar 691b39: OUTPUT«(foo => bar)␤»
timotimo personally, i'd .comb 11:47
timotimo m: say "foo=bar baz=quux what=that".comb(/(\w+) "=" (\w+)/, :match)>>.list.perl 11:48
camelia rakudo-moar 691b39: OUTPUT«((Match.new(ast => Any, list => (), hash => Map.new(()), orig => "foo=bar baz=quux what=that", to => 3, from => 0), Match.new(ast => Any, list => (), hash => Map.new(()), orig => "foo=bar baz=quux what=that", to => 7, from => 4)), (Match.new(ast => Any, li…»
timotimo m: say "foo=bar baz=quux what=that".comb(/(\w+) "=" (\w+)/, :match)>>.list>>.Str.perl
camelia rakudo-moar 691b39: OUTPUT«($("foo", "bar"), $("baz", "quux"), $("what", "that"))␤»
timotimo m: say "foo=bar baz=quux what=that".comb(/(\w+) "=" (\w+)/, :match)>>.list>>.Str.flat.perl
camelia rakudo-moar 691b39: OUTPUT«($("foo", "bar"), $("baz", "quux"), $("what", "that")).Seq␤»
timotimo m: say "foo=bar baz=quux what=that".comb(/(\w+) "=" (\w+)/, :match)>>.list>>.Str>>.flat.perl
camelia rakudo-moar 691b39: OUTPUT«($(("foo",).Seq, ("bar",).Seq), $(("baz",).Seq, ("quux",).Seq), $(("what",).Seq, ("that",).Seq))␤»
timotimo that's terrible %)
m: say "foo=bar baz=quux what=that".comb(/(\w+) "=" (\w+)/, :match)>>.list>>.Slip>>.Str.perl 11:49
camelia rakudo-moar 691b39: OUTPUT«("foo", "bar", "baz", "quux", "what", "that")␤»
timotimo m: my %result = "foo=bar baz=quux what=that".comb(/(\w+) "=" (\w+)/, :match)>>.list>>.Slip>>.Str; say %result.perl
camelia rakudo-moar 691b39: OUTPUT«{:baz("quux"), :foo("bar"), :what("that")}␤»
timotimo m: my %result = "foo=bar baz=quux what=that".comb(/(\w+) "=" (\w+)/, :match)>>.Slip>>.Str; say %result.perl
camelia rakudo-moar 691b39: OUTPUT«{:baz("quux"), :foo("bar"), :what("that")}␤»
timotimo cool, i can drop the .list because .Slip does that already
moritz Slip is a subtype of List anyway 11:50
timotimo aye
moritz m: say Slip ~~ List
camelia rakudo-moar 691b39: OUTPUT«True␤»
timotimo is my solution acceptable? 11:50
i must admit, the .match one looks a bit nicer
though ... 11:51
m: say "foo=bar yours=mine".match(/$<key>=(\w+) "=" $<value>=(\w+) { make ~$<key> => ~$<value> /, :g)>>.ast
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5=== Error while compiling /tmp/h6oolbsQP7␤Preceding context expects a term, but found infix , instead␤at /tmp/h6oolbsQP7:1␤------> 3ue>=(\w+) { make ~$<key> => ~$<value> /,7⏏5 :g)>>.ast␤»
timotimo m: say "foo=bar yours=mine".match(/$<key>=(\w+) "=" $<value>=(\w+) { make ~$<key> => ~$<value> } /, :g)>>.ast
camelia rakudo-moar 691b39: OUTPUT«(foo => bar yours => mine)␤»
moritz m: my %result = flat "foo=bar baz=quux what=that".words>>.split('='); say %result.perl
camelia rakudo-moar 691b39: OUTPUT«{:baz("quux"), :foo("bar"), :what("that")}␤»
timotimo m: say "foo=bar yours=mine".match(/$<key>=(\w+) "=" $<value>=(\w+) { make ~$<key> => ~$<value> } /, :g)>>.ast.perl
camelia rakudo-moar 691b39: OUTPUT«(:foo("bar"), :yours("mine"))␤»
brrt acceptable, just not as well-worn-down, i'd say :-) 11:56
timotimo thing is, if we did it more like your idiom, we'd end up with a hash filled with match objects 11:57
so there has to be at least a .Str in there somewhere
or a ~*
brrt hmmm......
here's an idea though
way, way to late, but here it is
what if we give the hash a Cool method that converts all its keys and all its values to strings 11:58
timotimo .Coolize
brrt or, Cool objects at any rate
Coler
.Cooler
timotimo .refridgerate
brrt anyway, lunch & 11:59
timotimo lunch well
brrt ok, naive question 12:22
timotimo if it's a native question, you spell it with a lower-case q :) 12:23
brrt is a Match a descendant of Str, or does it at least does Stringy
why ask i
timotimo m: say Str ~~ Stringy
camelia rakudo-moar 691b39: OUTPUT«True␤»
timotimo er
m: say Match ~~ Stringy
camelia rakudo-moar 691b39: OUTPUT«False␤»
timotimo it does not, but it has .Str
brrt because.. this would make sense
a Match is always a part-of-a-string 12:24
hence, it *is* a string
moritz no
brrt why not
masak because a Match is not a kind of Str 12:24
moritz right
brrt why not
masak it's pointing into a Str
brrt well, yes, but it never incorrect to refer to it as a Str 12:25
moritz it's information about how a regex matched a string 12:25
masak it's not a Str
brrt it is Stringy, at least
masak no
you are misinformed about what it is
brrt yes
i said it was a naive question
moritz lots of things can be coerced to Str, but they don't do the Stringy role
masak right. exactly the same argument could be made about Bool 12:26
brrt no, but a Match always (refers to) a part of a Str 12:26
hmmmm
masak both True and False have Str-y representations
brrt that is actually a decent enough point
masak pause for breathing.
brrt thinks
masak :)
brrt but still, always a subset-of-matched string
well, reference is not identity, of course 12:27
hmm
masak brrt: designing type relationships is bloody hard. you're experiencing a *desire* for the model to be in a certain way. it isn't.
brrt aye 12:27
well, i was thinking about it, because if it *were* Stringy, then timotimo's sample would be simpler 12:28
masak in a sense, you're akin to the people who want Int to be a kind of Rat and Rat to be a kind of Num. 12:28
moritz or you always talk about a customer by name, which is a string. That doesn't make Customer a Str
masak I don't feel a great compulsion to make Match ~~ Stringy. it feels like conflating two things.
brrt in fact, i experience a revulsion to the idea :-) 12:29
(from you)
masak sometimes I do +$0 and coerce my Match to a Numeric type. doesn't mean I want Match ~~ Numeric
brrt but my counterargument remains. a match *is* always a subset-of-a-Str, it has certain string-like-behaviour; it also has very un-string-like behaviour, like being a decent target for assignment, whereas a substring - i think - is not 12:30
but yeah, it is hard
masak brrt: where you see an is-a, I see a has-a. a Match object has many properties, one of which is referring into part of a Str. 12:31
moritz there's also the representational aspect to consider 12:31
Match actually stores a references to the original string, and start and end indexes of the match 12:32
Juerd I think one of the most important distinctions between a Match's substring and a real string, is that a Match can "be" (have) an empty string, but it will still be true.
brrt well, that's not a terribly strong argument; you could have javastrings, and then match being a subset of String would be the most logical thing in the world
moritz if you make it inherit from Str, you *have* to store the matched string again
masak Juerd: good point. which spoils the monoidal aspect of Str, quite a strong deviation. 12:33
Juerd What does monoidal mean?
brrt (javastring as a representation)
masak thinks
Juerd: monoidal means having an associative operation with a unit. like (Int, +, 0) or (Bool, ||, False) or (Str, ~, "") 12:34
Juerd Also, a Match thingy can have multiple strings 12:35
brrt hmm.. can it
ok, well, i can't continue this now, will be back later :-)
Juerd m: say "foo" ~~ m:g/o/ 12:35
camelia rakudo-moar 691b39: OUTPUT«(｢o｣ ｢o｣)␤»
masak m: say ("foo" ~~ m:g/o/).^name 12:36
camelia rakudo-moar 691b39: OUTPUT«List␤»
Juerd Oh, that's a list of Matches
Never mind this argument then :)
masak yeah.
but the fact that this was even a *question* means that Match is not an obvious Str.
Juerd m: say "foo" ~~ /(f)(o+)/
camelia rakudo-moar 691b39: OUTPUT«｢foo｣␤ 0 => ｢f｣␤ 1 => ｢oo｣␤»
bioexpress Hello! "sub one { some_func( $_[0] ) }": was is the most similirar construct (with the smalles overhead regarding passing the parameter) in Perl 6 ? 13:08
timotimo what exactly do you want the sub to accept? 13:09
masak m: sub one { some_func(@_[0]) }; sub some_func($s) { say $s }; one("OH HAI")
camelia rakudo-moar 691b39: OUTPUT«OH HAI␤»
masak bioexpress: like that?
m: sub one { some_func(@_[0]) }; sub some_func($s) { say $s }; one("OH HAI", "I can send more stuff", "but it doesn't get printed") 13:10
camelia rakudo-moar 691b39: OUTPUT«OH HAI␤»
timotimo masak: it looks to me like they want a sub that extracts the first item in the first parameter passed
but what do i know
masak yes, I'm not sure either. :)
bioexpress timotimo: I think that is what I am searching for 13:11
timotimo okay, then the code is literally the same you already had :) 13:12
m: sub first_of_first { some_func($_[0]) }; sub some_func { say "i got ", $_ }; first_of_first([10, 20, 30]);
camelia rakudo-moar 691b39: OUTPUT«Too many positionals passed; expected 0 arguments but got 1␤ in sub first_of_first at /tmp/3hWl12_ERk line 1␤ in block <unit> at /tmp/3hWl12_ERk line 1␤␤»
timotimo whoops.
m: sub first_of_first($_) { some_func($_[0]) }; sub some_func { say "i got ", $_ }; first_of_first([10, 20, 30]); 13:13
camelia rakudo-moar 691b39: OUTPUT«Too many positionals passed; expected 0 arguments but got 1␤ in sub some_func at /tmp/rTwdAsJ51m line 1␤ in sub first_of_first at /tmp/rTwdAsJ51m line 1␤ in block <unit> at /tmp/rTwdAsJ51m line 1␤␤»
timotimo oh, *in* the sub
m: sub first_of_first($_) { some_func($_[0]) }; sub some_func($_) { say "i got ", $_ }; first_of_first([10, 20, 30]);
camelia rakudo-moar 691b39: OUTPUT«i got 10␤»
timotimo m: sub first_of_first { some_func($^thing[0]) }; sub some_func { say "i got ", $^thing }; first_of_first([10, 20, 30]);
camelia rakudo-moar 691b39: OUTPUT«i got 10␤»
timotimo m: sub first_of_first { some_func($^thing[0]) }; sub some_func { say "i got ", $^thing }; first_of_first([10, 20, 30], <foo bar baz>);
camelia rakudo-moar 691b39: OUTPUT«Too many positionals passed; expected 1 argument but got 2␤ in sub first_of_first at /tmp/uGfhwZysPy line 1␤ in block <unit> at /tmp/uGfhwZysPy line 1␤␤»
garu_ is doc.perl6.org/type/Signature the ultimate-goto-if-its-not-there-it-doesnt-exist reference on sub/method signatures in P6? Or should I look elsewhere as well? 13:14
timotimo if you want it to accept as many positional arguments, you'd add a "*@" in the signature
bioexpress I need only one argument
timotimo OK
bioexpress Thx!
timotimo what can we do to make /downloads/star prettier? 13:28
FROGGS garu_: it seems pretty complete
timotimo like, perhaps putting all files that belong to the same month into columns?
or friendlier icons?
FROGGS garu_: though from skimming it, constant expressions are just mentioned on return types, which is unfortunate I think 13:29
mspo you mean rakudo.org/downloads/star/ ?
awwaiid mmm. I think you could do just about anything and it would be prettier than it is
timotimo yes, that's the one
FROGGS garu_: like: sub foo("bar", $baz) { ... }
mspo Apache/2.0.63 (Unix) mod_ssl/2.0.63 OpenSSL/0.9.8e-fips-rhel5 mod_auth_passthrough/2.1 mod_bwlimited/1.4 Server at rakudo.org Port 80
timotimo without forcing people to spend more effort on this 13:30
mspo is that real?
awwaiid ideally big buttons at the top for the current version is all you want, and then only show the others if people go out of their way to get an outdated version
timotimo is that an OpenSSL version susceptible to heartbleed or something?
and yeah, it is real.
mspo old apache :) 13:30
awwaiid oh, zero-effort. hmm. maybe there are some like apache themes or something
timotimo rakudo.org isn't well maintained. the owner doesn't have time to work on it
FROGGS mspo: could be worse
(could be IIS)
mspo FROGGS: that's always true 13:31
timotimo i think the wordpress is also quite old
mspo ugh wordpress
moritz pmichaud runs rakudo.org; somebody should contact him 13:31
timotimo it says "update is available! please update now", but i'm *not* going to be the one doing the update click.
awwaiid mostly I think if you move /downloads/star to the main site as a regular page that links to the current versions, and then make this full-list page something almost nobody goes to
moritz if he doesn't want to maintain the server, we could host it on www.p6c.org (same as perl6.org) 13:32
awwaiid yeah, that sort of update is scary
moritz you only do that sort of update if you have shell access to the host *and* a recent backup 13:32
mspo yeah probably best to just have a button and hide that dirlist
timotimo: the lowest effort would be to fix the icons in the mod_dirlist :) 13:33
timotimo we could build a little something something that takes the dirlist apache generates (can it also generate it as .json or .xml or .txt?) and generate a nice little html page from it?
perhaps then we could remote-upload that to rakudo.org
awwaiid timotimo: is the goal that you drop a new version in the dir and don't do anything else? 13:34
timotimo aye
*perhaps* trigger the rebuild of the page and check if it doesn't crap itself due to unexpected filenames or something
awwaiid hmm. I'm still leaning toward having another wordpress download page being the best, but that would mean editing it to point to the lastest version 13:35
timotimo yeah
awwaiid well gimme edit access to it (wordpress side) and let's see if we can work it out? We don't do releases THAT often :)
timotimo i'm not sure i'd be allowed to extend trust from rakudo.org to you. also no clue if i have the rights to do that 13:36
let's see
awwaiid yeah, no worries
timotimo wow, i can! 13:37
would you be an author? editor? administrator?
contributor, maybe?
awwaiid for initial setup of a new page I think I'd need more (editor or admin, can't remember) -- thereafter less 13:38
timotimo i can easily set up a page for you
msg me your wanted e-mail address and username please?
awwaiid ideally, imo, right next to "Home" would be "Download" -- that page would have the most up-to-date / popular downloads and then link to the directories
fireartist need to be an editor to edit pages - contributor/author are blog-post only 13:39
codex.wordpress.org/Roles_and_Capabilities
timotimo thank you, fireartist, that helps :) 13:40
[Coke] . 13:51
[Coke] m: say sqrt(-1+0i) * 2**3 * sum pi 14:15
camelia rakudo-moar 691b39: OUTPUT«0+25.1327412287183i␤»
timotimo "sum pi"? %) 14:16
moritz m: say pi == sum pi
camelia rakudo-moar 691b39: OUTPUT«True␤»
timotimo is that supposed to read like a poem or something? 14:16
[Coke] i 8 sum pi 14:18
moritz groans 14:19
[Coke] usually written as √-1 2³ ∑ π
m: say sqrt(-1+0i) * 2³ * sum pi # slightly better 14:20
camelia rakudo-moar 691b39: OUTPUT«0+25.1327412287183i␤»
timotimo oh lord 14:21
dalek kudo/experiment-multi-line-repl: 398a92c | hoelzro++ | src/Perl6/Compiler.nqp:
Multi-line input: Don't segfault if an exception has no payload

We extract BOOTException's payloads via nqp::getpayload to detect X::Syntax::Missing errors; sometimes, an exception has no payload, which (until now) causes a segmentation fault		 14:33
garu_ FROGGS: that looks interesting. can you show me an example of where it would be useful? I might try and append to that doc 14:34
timotimo hoelzro: i'm on that segfault
timotimo pushed, please test :) 14:37
FROGGS garu_: github.com/tadzik/panda/blob/maste.../panda#L14 14:39
FROGGS m: #| do something sensible␤multi MAIN("doit") { ... };␤#| more verbose help␤multi MAIN("WAT") { ... } 14:42
camelia rakudo-moar 691b39: OUTPUT«Usage:␤ /tmp/3p3fJNo0eD doit -- do something sensible␤ /tmp/3p3fJNo0eD WAT -- more verbose help␤»
garu_ FROGGS: thanks!
FROGGS garu_: so one usage is subcommand style things#
-#
garu_ FROGGS: yeah, really interesting! 14:46
timotimo hoelzro: well? :) 14:50
rje_ Is +camelia written in Perl6? 14:55
moritz rje_: no
I guess it could be now 14:56
but back then, Perl 6 wasn't stable enough
timotimo only if enough hours in the day
rje_ Gotcha. Hey, due to the bad rep Perl5 has from the millennials here at work, maybe I should refer to Perl6 as "Camelia". 14:57
[Coke] rje_: the main implementation is Rakudo - just lead with that.
timotimo moritz: i think i already asked this, but i don't know if there was an answer yet. should we get collectd onto more than just hack.p6c.org?
moritz timotimo: you mean on the webserver too? 14:58
rje_ Code: I was thinking that last week, but it's weirder-sounding.
moritz timotimo: +1 from me
rje_ er, Coke
timotimo or on the VM host?
moritz timotimo: maybe that too
stmuk the most simple solution for rakudo.org/downloads/ and similar might be to create an 'archive' directory and move anything preXmas into it 14:59
moritz or maybe start to introduce subdirs for the years
rje_ Coke: your server is in TX. Are you in Texas?
moritz downloads/rakudo/2016/
rje_ Coke: I'm in Plano 15:00
timotimo stmuk: awwaiid is already working on a pretty little page for downloadingus
[Coke] rje_: I am not logged in from texas. :)
stmuk ah good 15:01
[Coke] Coincidentally, I lived in Houston for 2 years, worked for Enron.
timotimo moritz: strange. my reading of collectd.org/wiki/index.php/Plugin:CPU makes it seem like we're running some other cpu plugin on our collectd ... but how is that possible? 15:02
dalek href="https://perl6.org:">perl6.org: 9d04254 | (Steve Mynott)++ | source/downloads/index.html:
refer to Mac binary .dmg
rje_ Coke: unfortunately, our young over here are an opinionated lot. 15:03
moritz timotimo: dunno; I did a pretty minimal installation and stopped working on at as soon as it works 15:07
timotimo hum. interesting 15:07
why don't i have my password on my desktop?
timotimo grabs laptop
timotimo better now 15:09
hoelzro timotimo: sorry, was commuting; which segfault? 15:21
timotimo hoelzro: getexpayload on unitialized exceptions 15:24
hoelzro ahhh 15:26
getpayload is fine; it just fails when I do my $inner := nqp::getpayload($_); my $ex-type := nqp::what($inner).HOW.name(nqp::what($inner)) 15:27
status update on the multi-line input branch: it now passes roast 15:29
timotimo damn, i misread your commit in that case
so my patch is useless? :<
hoelzro timotimo: that just means I miswrote it =/
sorry =(
jnthn Fixing Moar SEGVs isn't really useless :)
timotimo moritz: how do we feel about having a notification from collectd when we have a process that's been eating too much cpu time? perhaps something like that is possible 15:30
timotimo (otherwise, it could probably be written in perl6 ;) ) 15:30
hoelzro jnthn: it certainly doesn't hurt =)
timotimo jnthn: i didn't actually check if you could get a segfault from that ... :|
i didn't know what would be required to reproduce hoelzro's segfault from getexpayload 15:31
hoelzro it wasn't getexpayload that was segfaulting 15:31
but nqp::what(nqp::getpayload($ex)) should probably never segfault
timotimo well, if it was returning c-level null, it could be causing the segfault
hoelzro mhmm
timotimo i put in a line that makes sure when it's null it gives VMNull instead
which can be WHAT'd
hoelzro oh, then it wasn't for nothing! 15:32
timotimo++ # fixing other's bug
*bugs
timotimo great! 15:34
timotimo i activated the "aggregation" plugin for collectd. it should now show a "all cpu usages added" graph on the cpu page, too 15:38
geekosaur misreads "aggravation"...
timotimo but i don't really know when that'll kick in. i've "systemctl restart collectd"
MadcapJake timotimo, wrt rakudo.org/downloads: adamwhitcroft.com/apaxy/ looks really nice 15:48
timotimo pretty! 15:49
we'd have to have someone with shell access to rakudo.org, though, i imagine
or we just use one of the probably millions of applicable wordpress exploits on it ... 15:50
MadcapJake lol
hoelzro oh, I forgot to finish my status report! the multi-line stuff works pretty well, except for class definitions
it really doesn't like those =/
MadcapJake anyone been following @perlhex? Curious if people are enjoying the content. Could use some more retweets to get the account more followers too! 15:52
timotimo is following it
the last few tweets i saw, i enjoyed 15:53
the ~= i found a bit strange, though
MadcapJake lol, i was trying to evoke "this was recently added to the docs" and thought I'd do something perl6ish :D
timotimo heh 15:54
okay, that's cute :)
ugexe should i be concerned if --profile always crashes a program? or is it just --profile is kinda fragile? 15:59
timotimo --profile might immediately die if you use anything multi-threaded or asynchronous I/O
ugexe ah
timotimo HA! 16:00
the aggregation plugin works now and has the cpu sum
moritz /w 10 16:07
kaare_ In Perl 5 it's rather common to pass a scalar to a sub, and then in there decide what to do based on the type [if (ref $parm eq 'ARRAY') {...}] 16:10
hoelzro multi subs?
kaare_ is there a common idiom to handle that in Perl 6?
That's not always really efficient.
hoelzro multi sub handle-it(Int $n) { ... } multi sub handle-it(Str $s) { ... }
[Coke] hoelzro++ is correct.
kaare_ But I guess that's the canonical answer :-) 16:11
[Coke] What makes you say it's not efficient?
Have you profiled it?
kaare_ Ah, no. Don't mean performance. 16:12
Just that it leads to too much clutter in some cases. 16:14
kaare_ I guess I'll try to rewire it one step further up. 16:15
alpha123 How's that cluttered? IMO it's *way* nicer than a bunch of if ... { ... } 16:17
hoelzro I think (personally) that multi subs are a lot less cluttered than if($type_check_1) { ... } elsif($type_check2) { ... }
kaare_ I agree that the perl 5 way is ugly too. Or rather that the construct probably never should have existed in the old code. 16:19
RabidGravy I actually remade a large given/when into a bunch of multis with literal parameters yesterday, much prettier 16:23
hankache hola #perl6 16:30
MadcapJake kaare_, as RabidGravy pointed out, given/when is another possible solution
fireartist m: Str.WHAT.say
camelia rakudo-moar 691b39: OUTPUT«(Str)␤»
fireartist say( "X: " ~ Str.WHAT ) 16:31
why doesn't .WHAT want to concat?
hankache m: say( "X: " ~ Str.WHAT ) 16:32
camelia rakudo-moar 691b39: OUTPUT«Use of uninitialized value of type Str in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in block <unit> at /tmp/UWxEQCR6OI line 1␤X: ␤»
fireartist ok, right, thanks :-)
timotimo kaare_: probably better than having a bunch of ifs is to have a given/when
hankache m: say( "X: " ~ .WHAT )
camelia rakudo-moar 691b39: OUTPUT«Use of uninitialized value of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in block <unit> at /tmp/8NJRRE5veT line 1␤X: ␤»
hankache m: say( "X: " "X"~ .WHAT )
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5=== Error while compiling /tmp/DkY8hUlwTu␤Unable to parse expression in argument list; couldn't find final ')' ␤at /tmp/DkY8hUlwTu:1␤------> 3say( "X: "7⏏5 "X"~ .WHAT )␤ expecting any of:␤ infix␤ infix stop…»
timotimo hankache: you want .WHAT.^name
or .WHAT.gist
hankache timotimo i was helping fireartist 16:33
timotimo oh
OK
hankache fireartist what are you trying to do?
MadcapJake fireartist, or Str.WHICH
timotimo .WHICH has a quite different use-case
m: say "hello how are you".WHICH 16:34
camelia rakudo-moar 691b39: OUTPUT«Str|hello how are you␤»
fireartist just printing some debugging - was surprised it didn't do as I expected
MadcapJake m: say( "X: " ~ Str.WHICH )
camelia rakudo-moar 691b39: OUTPUT«X: Str␤»
kaare_ thanks hankache/timotimo but I'm stepping back a little to see if I can avoid the ugliness completely.
MadcapJake timotimo, on a type, WHICH seems to just concatenate the type name 16:35
MadcapJake *when used with ~ 16:35
timotimo kaare_: you can also just try to call a method on something and catch exceptions and stuff like that 16:36
or try to coerce to a different type and catch exceptions when it doesn't work
RabidGravy this diff shows refactoring a given/when to multi with literal parameters github.com/jonathanstowe/Audio-Liq...fe4de3b687 16:38
MadcapJake with the lovely github showing it's syntax highlighting muscles... 16:40
s/it's/its/
timotimo yeah 16:42
timotimo i'm not sure how to remove the aggregation/ stuff, since apparently collectd-web keeps that around 17:00
even though since i moved the output into the cpu plugin's tab there'll not be new data there
hoelzro should I merge in the multi-line input work, in spite of its problem with classes? 17:05
I think it's better than nothing, but at the same time I think it might irritate users
geekosaur can you make it switchable? 17:10
skids Rabidgravy: you shouldn't need the else { Empty } these days. 17:11
m: say (if (0) { }); say (Empty)
camelia rakudo-moar 691b39: OUTPUT«()␤()␤»
hoelzro geekosaur: switchable? so one can turn off the feature? 17:13
RabidGravy yeah, I think it's superstitious after I found a case where it did matter
geekosaur yeh. $*MULTILINE-INPUT or some such so they can play with it and if it's too annoying push it out of the way 17:14
jnthn Maybe an env var 17:19
hoelzro ok 17:22
I'll add that later and merge it
ZoffixW m: say "㊤".uniname.words.first: :end 17:26
camelia rakudo-moar 691b39: OUTPUT«This Seq has already been iterated, and its values consumed␤(you might solve this by adding .cache on usages of the Seq, or␤by assigning the Seq into an array)␤ in block <unit> at /tmp/o5PFGbXALC line 1␤␤»
ZoffixW :/
ZoffixW m: say "㊤".uniname.words.first 17:26
camelia rakudo-moar 691b39: OUTPUT«CIRCLED␤»
ZoffixW m: say "㊤৫".comb».uniname».words».[*-1] 17:29
camelia rakudo-moar 691b39: OUTPUT«(HIGH FIVE)␤»
ZoffixW :D
TimToady m: say "㊤৫".uninames».words»[*-1] 17:31
camelia rakudo-moar 691b39: OUTPUT«(HIGH FIVE)␤»
timotimo oh hey TimToady 17:32
ZoffixW: i see your twitter exchange and ... oh well ...
ZoffixW heh 17:36
It's the equivalent of shouting "get a horse" at the dawn of automobile age :)
jnthn o/ TimToady 17:46
RabidGravy I might finally write Test::Kinda so I can express a "to be honest if only one or two fail then it's a pass" ;-) 17:54
xiaomiao RabidGravy: have a look at glibc test suite then 17:55
RabidGravy I think at this stage I'd rather not to be frank ;-) 17:55
RabidGravy testing against an external software that has its own view of deterministic scheduling with several distinct clocks is proving to be quite annoying 17:57
RabidGravy oh well, my back, my rod .... 18:00
Kermi Hi, I am reading perl6intro.com/#_what_is_perl_6 and I wonder Can we use perl6 with apache with mod_perl ? 18:04
[Coke] mod_perl is perl5 18:05
(so, no)
TimToady m: say "􏿽xEF􏿽xA9􏿽x9E" 18:06
camelia rakudo-moar 691b39: OUTPUT«艹␤»
TimToady I don't want to canonicalize that character
stmuk most perl 5 people have moved from mod_perl anyway
TimToady I want to be able to mention that variant in my file without it going away. 18:07
jnthn: ^^ 18:08
Kermi :) Thanks 18:08
What can I use for Apache and Perl6? 18:09
RabidGravy I'm not even clear in my mind how a mod_rakudo fir instance would work 18:10
RabidGravy Kermi, right now it's start a separate server and use Apache as a proxy 18:10
stmuk kermi: psgi
Kermi RabidGravy, where can read more about this? 18:11
RabidGravy about which part? 18:12
Kermi [20:10] <RabidGravy> Kermi, right now it's start a separate server and use Apache as a proxy
RabidGravy right, which part?
Kermi How to connect perl6 with apache
RabidGravy er
the Apache documentation describes how to use it as a frontend proxy for another application 18:13
(though I think all the cool kids are using nginx or something else for this purpose these days) 18:14
jnthn TimToady: That'd be part of NFC, no? 18:15
jnthn .u 􏿽xEF􏿽xA9􏿽x9E 18:15
yoleaux U+FA5E CJK COMPATIBILITY IDEOGRAPH-FA5E [Lo] (􏿽xEF􏿽xA9􏿽x9E)
jnthn m: say Uni.new(0xFA5E) 18:16
camelia rakudo-moar 691b39: OUTPUT«Uni:0x<fa5e>␤»
jnthn m: say Uni.new(0xFA5E).NFC
camelia rakudo-moar 691b39: OUTPUT«NFC:0x<8279>␤»
jnthn Yeah
Almost certainly a singleton decomposition.
There's not a good way yet, anyways. It's in Uni territory. 18:17
Kermi RabidGravy thanks 18:18
jnthn But we don't have uni literals yet.
And even when we do, we'll need to work out what happens if you try to concatenate them with Str.
.oO( NFG is a good default, but perhaps sometimes a bit too good... :P )
 18:19
RabidGravy Kermi, there are several things in the Perl 6 ecosystem that allow making a web application server, Bailador, Crust and so forth, you probably want to peruse a couple of those in the first instance 18:20
stmuk Kermi: examples.perl6.org/categories/cookb...cript.html 18:21
that's one miminal way of doing it and you would probably proxypass off apache or nginx to that port 18:22
Kermi I will read how to proxypass it. Thanks 18:24
timotimo i'm showing the "An Earthshaking Announcement" to a friend right now %)
jnthn noms & 18:29
TimToady m: say "􏿽xEF􏿽xAB􏿽x8F" 18:41
camelia rakudo-moar 691b39: OUTPUT«(signal SEGV)»
TimToady jnthn: ^^ there's another glitch :) 18:42
timotimo can you also make that with a \x literal, so that i can more easily copy-paste it? 18:43
TimToady m: say "\xfacf" 18:47
camelia rakudo-moar 691b39: OUTPUT«(signal SEGV)»
TimToady m: say "\xfad0"
camelia rakudo-moar 691b39: OUTPUT«(signal SEGV)»
[Coke] O_o 18:48
TimToady m: say "\xfad9"
camelia rakudo-moar 691b39: OUTPUT«龎␤»
TimToady m: say "\xfad8" 18:49
camelia rakudo-moar 691b39: OUTPUT«齃␤»
TimToady m: say "\xfad7"
camelia rakudo-moar 691b39: OUTPUT«(signal SEGV)»
TimToady there's the boundary
that trailing boundary, facf is the leading boundary 18:50
timotimo: thanks
timotimo er, thanks why exactly? 18:51
TimToady for copy-pasting it :) 18:52
timotimo damn, you expect me to fix that bug now? ;)
TimToady or at least for wanting to copy-paste it :)
timotimo 48543 const MVMint32 *pcs = comp_p[plane][upper][lower]; 18:53
the values of the indices are, respectively, <optimized out>, <optimized out>, and <optimized out>
:P
TimToady cool! 18:53
"Nobody will ever use those characters!" 18:54
timotimo 2, 126, and 211
(gdb) print comp_p[2][126] 18:55
$5 = (const MVMint32 **) 0x0
TimToady notes that those are, in fact, plane 0 characters 18:56
timotimo so ... do we actually expect there to be something?
TimToady I wouldn't expect to find something in plane 2 for them, no :) 18:57
timotimo i personally haven't a clue what that particular piece of code is precisely supposed to achieve; you said it's supposed to be plane 0, do you know why it'd get plane 2 instead? 19:01
given plane is just (l >> 16) & 0xF 19:02
TimToady maybe upstream, the same remapping thingy that turns 􏿽xEF􏿽xA9􏿽x9E into 艹 is turning those compat chars into something in plane 2? 19:03
but then they should find something
m: say "􏿽xEF􏿽xAB􏿽x8F".ord.base(16) 19:04
camelia rakudo-moar 691b39: OUTPUT«2284A␤»
TimToady yeah, looks like that's what's going on 19:04
m: say "􏿽xEF􏿽xAB􏿽x8F".ord.base(16).chr 19:05
camelia rakudo-moar 691b39: OUTPUT«Cannot convert string to number: trailing characters after number in '032284⏏5A' (indicated by ⏏)␤ in block <unit> at /tmp/fFGuCOiKMA line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/fFGuCOiKMA line 1␤␤»
TimToady m: say "􏿽xEF􏿽xAB􏿽x8F".ord..chr
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5===␤Argument to "chr" seems to be malformed␤at /tmp/EXvSTh8KTb:1␤------> 3say "𢡊".ord..chr7⏏5<EOL>␤Other potential difficulties:␤ Unsupported use of bare "chr"; in Perl 6 please use .chr if you meant $_, or use an expli…»
TimToady m: say "􏿽xEF􏿽xAB􏿽x8F".ord.chr
camelia rakudo-moar 691b39: OUTPUT«(signal SEGV)»
TimToady m: say "\x2284a"
camelia rakudo-moar 691b39: OUTPUT«(signal SEGV)»
timotimo gist.github.com/timo/0566aa167dbf0996bab9 19:06
that's the full backtrace of the thing you originally made copy-pastable
TimToady m: say "𢡊" 19:06
camelia rakudo-moar 691b39: OUTPUT«(signal SEGV)»
TimToady that's the actual plane 2 character that can't print
b2gills TimToady: should storing a fail in a variable stop it from going into $! ( unless you ask it the wrong question ) ｢{my $ = sub {fail 'e'}();Nil}; say $! === Nil｣ 19:32
arnsholt timotimo: IME, those <optimised out> values can frequently be recovered if you are one of those people who are able to remember which registers are used to pass arguments into a function 19:33
Sadly, I'm not one of those people =(
b2gills TimToady: I'm thinking it shouldn't fail at the end of the scope ｢{my @a = { Bool.pick ?? ++$ !! sub {fail}() } ... *;Nil}｣ Otherwise the only way to deal with this example would be to look at the internals of Array and handle the Failures that have already been generated. 19:37
skids b2gills: it is not that it is stored, it is that it is not sunk. 19:57
timotimo arnsholt: oh, that's a good idea actually. from the moar jit i know that a little bit 20:00
arnsholt Yeah, if you know it, it's useful 20:01
And there's a command to dump the contents of the registers as well, obv
timotimo print %foo should already do it, i think? 20:05
arnsholt Probably 20:06
moritz \o 20:08
timotimo moritz: do you happen to know if collectd-web has its own data store somewhere? that may need a tiny bi tof cleanup so that the aggregated cpu thingies that no longer get updates won't show any more 20:10
moritz timotimo: I don't think so; I think it just looks a the list of rrd files in some directories
timotimo: so just delete the old rrd files
timotimo ah. but i looked into the folder where it seemed like collectd was putting the rrd files and i didn't see the "old"/obsolete ones there 20:11
moritz hm
timotimo does collectd-web run on hack? if so, how should i transport the data from other hosts over to hack or make it reachable for collectd-web? 20:12
moritz yes to the first
and to the second: you can configure collectd on a second host to forward the data to hack 20:13
moritz collectd.org/wiki/index.php/Networ...troduction 20:14
dalek rl6-roast-data: 3d732a0 | (Heiko Jansen)++ | perl6_pass_rates.csv:
Remove incomplete entries

A bunch of rows for rakudo.moar-jit with an insufficient number of fields broke github CSV formatting. Since there are other entries from the same day and with the same sha1 hash I simply deleted the corrupted data sets.		 20:32
rl6-roast-data: 473e8d5 | (Will Coleda)++ | perl6_pass_rates.csv:
Merge pull request #12 from heikojansen/patch-1

Remove incomplete entries
[Coke] so, the daily tests I'm running on hack are getting stuck again. 20:34
no one else is seeing weird hangs running stresstest?
bartolin [Coke]: I've seen integration/advent2013-day14.t hanging lately (see irclog.perlgeek.de/perl6/2016-02-08#i_12007689) 20:37
[Coke]: but this week I saw no hangs
[Coke] argh, killed my screen session accidentally. 20:48
dalek rl6-roast-data: 4c3e1d0 | coke++ | / (7 files):
today (automated commit)
 20:49
rl6-roast-data: 197eac5 | coke++ | / (2 files):
today (automated commit)
timotimo video.fosdem.org/2016/STATUS.txt - has significantly changed 20:59
luckily we were in 2214, not in 2215
timotimo because in 2215, more than half the videos got lost 20:59
hoelzro was so confused reading that 21:05
hoelzro I was going to say "2215?! Perl 6 really *is* a one-hundred year language!" 21:05
jnthn TimToady: 'fraid I'm too tired to go SEGV hunting tonight on that one...please do RT it. 21:07
AlexDaniel I see that all bugs in RT have 0 priority. Is it OK use priority other than 0? 22:10
e.g. for segfaults? 22:16
moritz AlexDaniel: I don't think it'll accomplish anything 22:19
AlexDaniel: people who work on bugs have their own judgment about their relative priorities 22:20
[Coke] make sure the title has [SEGV] or Segmentation Fault in it. 22:30
AlexDaniel okay 22:31
pmurias AlexDaniel: are segfaults really higher priority? 22:46
AlexDaniel pmurias: imo yes
Hotkeys b2gills: are you around? 22:48
b2gills Hotkeys: nope ... wait a minute 22:48
that was meant as a joke 22:49
Hotkeys :p
I have a solution for "How even is a number" that is 23 bytes, but it's significantly different from yours
Should I post it as my own?
Or would it be better for you to add it to yours
I'm not sure what the protocol is here 22:50
Hotkeys b2gills: 22:51
b2gills If it is a vastly different approach, go ahead and post it if you want 22:52
really there is no hard and fast rules
Hotkeys mmkay
posted 22:54
m: my &f = {elems $_,*/2...^*!%%2}; say f(14); say f(20); say f(94208); say f(7); say f(-4)
camelia rakudo-moar 691b39: OUTPUT«1␤2␤12␤0␤2␤»
Hotkeys actually 22:55
Hotkeys m: my &f = {+($_,*/2...^*!%%2)}; say f(14); say f(20); say f(94208); say f(7); say f(-4) 22:55
camelia rakudo-moar 691b39: OUTPUT«1␤2␤12␤0␤2␤»
Hotkeys neat
b2gills {+($_,*/2...^*%2)} 22:56
Hotkeys :p
oh
I didn't realize that would work 22:57
Hotkeys oh 22:58
right
Hotkeys I forgot that it would be okay with a truthy value 22:59
Hotkeys and not just a True 23:00
weird
I can't get !% to work outside that example
b2gills In Perl 5 there is no True and False, but there are still if statements
Hotkeys m: say 3 !% 2
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Rd78q5E7l3␤Cannot negate % because multiplicative operators are not iffy enough␤at /tmp/Rd78q5E7l3:1␤------> 3say 3 !7⏏5% 2␤ expecting any of:␤ infix␤ infix stopper␤»
Hotkeys I mean I know that it would accept a truthy value logically 23:01
I just didn't think about it
because I'm used to doing like * > 10 or whatever in a sequence 23:02
so my mind was like "It's gotta be a boolean"
rje_ m: say "hello?"
camelia rakudo-moar 691b39: OUTPUT«hello?␤»
rje_ wow, cool
b2gills m: say "hello?".comb(/\w/).perl 23:03
camelia rakudo-moar 691b39: OUTPUT«("h", "e", "l", "l", "o").Seq␤»
rje_ comb! So that's what I should be using instead of split!
rje_ thanks for that 23:03
m: say "hello?".split('') 23:04
camelia rakudo-moar 691b39: OUTPUT«( h e l l o ? )␤»
b2gills m: say "hello?".split('').perl
camelia rakudo-moar 691b39: OUTPUT«("", "h", "e", "l", "l", "o", "?", "")␤»
rje_ it inserts undef thingies, see
unlike perl5
rje_ not the principle of least surprise to a Perl5 person :) 23:05
b2gills You can get Perl 5 to return them as well
Hotkeys sometimes split can be the same bytecount for a problem
rje_ eeek
m: say "hello?".split('').gist
camelia rakudo-moar 691b39: OUTPUT«( h e l l o ? )␤»
rje_ d'oh 23:06
b2gills I guess it's only the trailing empty string that you can get from Perl 5 ｢split "", "hello", -1｣ 23:07
Hotkeys m: say "11a1".comb(/\d+/).sum == "11a1".split(/\D/).sum
camelia rakudo-moar 691b39: OUTPUT«True␤»
Hotkeys :D 23:07
b2gills: perhaps you could add the split version as an alternate on your filter and add answer?
I know they're effectively the same 23:08
byte count and all
Hotkeys up to you 23:08
sortiz \o #perl6
Hotkeys \o
sortiz Still golfing Hotkeys? :) 23:09
Hotkeys of courwse 23:10
atweiden m: %h[0] = :zero(0); say %h.perl
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5=== Error while compiling /tmp/3Q5_xJEohQ␤Variable '%h' is not declared␤at /tmp/3Q5_xJEohQ:1␤------> 3<BOL>7⏏5%h[0] = :zero(0); say %h.perl␤»
atweiden m: my %h; %h[0] = :zero(0); say %h.perl
camelia rakudo-moar 691b39: OUTPUT«{:zero(0)}␤»
atweiden can anyone say what indexing a Hash type with a [0] accessor does?
Hotkeys m: my %h = a=>foo, b=>bar, c=>baz; say %h[0] 23:11
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5=== Error while compiling /tmp/z9ci6P85Yw␤Undeclared routines:␤ bar used at line 1. Did you mean 'VAR', 'bag'?␤ baz used at line 1. Did you mean 'bag'?␤ foo used at line 1␤␤»
Hotkeys m: my %h = a=>'foo', b=>'bar', c=>'baz'; say %h[0]
camelia rakudo-moar 691b39: OUTPUT«a => foo, b => bar, c => baz␤»
Hotkeys m: my %h = a=>'foo', b=>'bar', c=>'baz'; say %h[1] 23:12
camelia rakudo-moar 691b39: OUTPUT«Index out of range. Is: 1, should be in 0..0␤ in block <unit> at /tmp/GWFIZK16H9 line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/GWFIZK16H9 line 1␤␤»
b2gills Any non-positional will return itself when you use [0]
Hotkeys m: my %h = a=>'foo', b=>'bar', c=>'baz'; say %h.pairs[1]
camelia rakudo-moar 691b39: OUTPUT«c => baz␤»
Hotkeys m: my %h = a=>'foo', b=>'bar', c=>'baz'; say %h.pairs
camelia rakudo-moar 691b39: OUTPUT«(a => foo c => baz b => bar)␤»
Hotkeys oh hashes
ohhhh you
b2gills m: my %h; %h[0][0][0][0][0][0][0][0][0][0][0][0][0][0][0][0][0][0][0][0][0][0][0] = :zero(0); say %h.perl 23:13
camelia rakudo-moar 691b39: OUTPUT«{:zero(0)}␤»
dalek osystem: 809e5ff | (Claudio Ramirez)++ | META.list:
Move to META6.json
osystem: cb66171 | nxadm++ | META.list:
Merge pull request #153 from nxadm/master

Move to META6.json
Hotkeys that's a lot of zeroes
b2gills It's a lot of zeros that don't matter
sortiz atweiden, indexing any not Positional returns its type if index == 0, otherwise a Failure
timotimo i'd be glad if ecosystem commits were a bit more informative about what they do ... like, in the first line of the commit ... 23:14
and extra nice if they include a line that points at the github repository page of the module itself
b2gills some of them have had a line that pointed at the repository 23:15
timotimo yes 23:16
i was really glad about those when i wrote one of the weeklies
atweiden i'm having difficult seeing where indexing a non-positional with [0] would be practically useful. any hints? 23:17
why wouldn't it raise an exception?
timotimo that's the thing where single things pretend to be a list of a single thing 23:19
m: my $foo = 10; $foo[1] = "hi";
camelia rakudo-moar 691b39: OUTPUT«Index out of range. Is: 1, should be in 0..0␤ in block <unit> at /tmp/L_XsAgqY9h line 1␤␤Actually thrown at:␤ in block <unit> at /tmp/L_XsAgqY9h line 1␤␤»
timotimo that kind of thing still b0rks
sortiz m: 1.elems
camelia ( no output )
sortiz m: 1.elems.say
camelia rakudo-moar 691b39: OUTPUT«1␤»
sortiz m: "foo".elems.say 23:20
camelia rakudo-moar 691b39: OUTPUT«1␤»
sortiz atweiden, that way other common operations degrade gratefully 23:21
Hotkeys is .reverse the shortest way to reverse a list? 23:23
er nevermind
atweiden m: try { @a[-8]; CATCH { default { say $_.perl } } } 23:28
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5=== Error while compiling /tmp/oJWzT1pOHZ␤Unsupported use of a negative -8 subscript to index from the end; in Perl 6 please use a function such as *-8␤at /tmp/oJWzT1pOHZ:1␤------> 3try { @a[-8]7⏏5; CATCH { default { say $_.perl }…»
atweiden is there a way to catch that error?
m: try { my @a = 0, 1, 2; @a[-8]; CATCH { default { say $_.perl } } }
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ou_oO_IAUq␤Unsupported use of a negative -8 subscript to index from the end; in Perl 6 please use a function such as *-8␤at /tmp/ou_oO_IAUq:1␤------> 3try { my @a = 0, 1, 2; @a[-8]7⏏5; CATCH { defaul…»
sortiz A CATCH block
geekosaur atweiden, the ==SORRY== tells you it's compile time
hard to use runtime code to catch a compile time error 23:29
atweiden i see
sortiz .. can be used to catch runtime... early <ENTER> 23:31
Ups: 23:36
m: my @A[-1];
camelia rakudo-moar 691b39: OUTPUT«Memory allocation failed; could not allocate 18446744073709551608 bytes␤»
AlexDaniel m: my $x = -8; try { @a[$x]; CATCH { default { say $_.perl } } }
camelia rakudo-moar 691b39: OUTPUT«5===SORRY!5=== Error while compiling /tmp/hwg3DIh7YM␤Variable '@a' is not declared␤at /tmp/hwg3DIh7YM:1␤------> 3my $x = -8; try { 7⏏5@a[$x]; CATCH { default { say $_.perl } ␤»
AlexDaniel m: my @a; my $x = -8; try { @a[$x]; CATCH { default { say $_.perl } } }
camelia rakudo-moar 691b39: OUTPUT«X::OutOfRange.new(what => "Index", got => -8, range => "0..Inf", comment => Any)␤»
AlexDaniel ? 23:37
sortiz AlexDaniel, that's the runtime Failure 23:38
AlexDaniel sortiz: sure, why would you want to catch your typos?
RabidGravy right that's me done for the day. toodles.
AlexDaniel Hotkeys: did you figure it out? :)
m: my @a = <a b c>; say [R,] @a 23:39
camelia rakudo-moar 691b39: OUTPUT«(c b a)␤»
Hotkeys I just did something else
AlexDaniel Hotkeys: ↑ this is probably shorter
sortiz AlexDaniel, May be not _your_ typos, you are working with someone data.
AlexDaniel sortiz: you mean if I'm EVAL-ing someone's code? 23:40
Hotkeys m: my @a = <a b c d e f>; say [R,] @a
camelia rakudo-moar 691b39: OUTPUT«(f e d c b a)␤»
Hotkeys neat
sortiz AlexDaniel, your $x can be anything, not necessarily declared or assigned in the same scope. 23:41
sortiz AlexDaniel, That the general idea of a try { CATCH {} }, when your expectations can fail. 23:43
AlexDaniel sortiz: if $x is anything then you can use try { CATCH {} } without any problems 23:46
because it becomes a runtime problem
sortiz Exactly.
AlexDaniel the original snippet above had -8 hard coded, and there is no reason to try to catch that 23:47
that's what I was trying to say
sortiz There are other methods, but try/CATCH becomes a general way for handling Exceptions and Failures 23:48
sortiz Indeed, if the compiler can see a problem, it is reported at compiler time, otherwise is a runtime problem that can be handled with a try BLOCK 23:51
AlexDaniel Hotkeys: anything fun for today? :) 23:54

Powered by Cro and the Raku® Programming Language.

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

Thank you!