🦋 Welcome to Raku! raku.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: colabti.org/irclogger/irclogger_log/raku
Set by ChanServ on 14 October 2019.
00:26 zacts left 00:29 finsternis joined 00:46 cpan-raku joined, cpan-raku left, cpan-raku joined 01:06 AlexDani` joined 01:08 AlexDaniel left 01:11 _jrjsmrtn joined 01:12 Roamer` joined 01:14 __jrjsmrtn__ left 01:16 molaf__ left 01:22 maggotbrain777 joined 01:25 maggotbrain left 01:29 molaf__ joined 01:40 a3f left 01:49 sjaveed joined 01:51 sjaveed left 01:59 sjaveed joined
sjaveed hello folks! 02:00
[Coke] hello
02:02 a3f joined
sjaveed so I'm just getting started with Raku - former perl monger that went to ruby but am now interested in falling in love with raku 02:04
i'm running into an issue when attempting to use a regex that's stored in a string - i'm sure i'm doing something wrong 02:05
i've got a gist with a testable script here: gist.github.com/sjaveed/061c743b70...46bcc5c000 02:06
wondering if experienced rakuist has thoughts on what i'm doing wrong
[Coke] if you want to "eval" a regex, I think it's /<$string>/ 02:07
m: my $regex-str = '"/greeting/" $<name> = (.+)'; my $path = "/greeting/bob"; say $path ~~ /<$regex-str>/
camelia 「/greeting/bob」
[Coke] which I only remember because I basically asked the same question this weekend. :)
sjaveed actually yeah my bad - that was an older version
i switched to that
but it's not picking up my capture groups with that syntax 02:08
compare that to
[Coke] m: my $regex-str = '"/greeting/" $<name> = (.+)'; my $path = "/greeting/bob"; $path ~~ /<$regex-str>/ ; say $0
camelia Nil
[Coke] m: my $regex-str = '"/greeting/" $<name> = <(.+)>'; my $path = "/greeting/bob"; $path ~~ /<$regex-str>/ ; say $/
camelia 「/greeting/bob」
sjaveed m: my $path = "/greeting/bob"; say $path ~~ /"/greeting/" $<name> = (.+)/; 02:09
camelia 「/greeting/bob」
name => 「bob」
sjaveed i'm certain it's something trivial and i'm not seeing it
also it's been 20 years since i've been on irc 02:10
i think i've missed the bots :-)
oh lol just saw your comment [Coke] - about having asked the same question recently 02:12
02:12 zacts joined
sjaveed do you have an example of where you've used string interpolation in your code? 02:12
wondering if seeing an example might help figure this out
[Coke] question: do you really need to store the regex as a string? 02:13
you could do my $regex = /..../ and then use the regex object.
(assuming you just wanted to pull it out into a separate object, not necessarily as a string)
sjaveed the's a good question - i *think* i need to store it as a string 02:14
let me explain what i'm trying to do and maybe a better alternative will appear
so this is essentially the toy functionality i figured i'd build - a simple routing engine for the web
02:14 zacts left
sjaveed something like that rails has 02:14
so you could provide it a route that looks like e.g. /greeting/:name and it will match against /greeting/bob and give me name as the param 02:15
my initial take on that was to take the route /greeting/:name and convert it into a regex
similar to the one in the gist
do that for all routes specified
and see which one matches any given path
i'm not sure i can do this with a static regex 02:16
02:17 MilkmanDan left
[Coke] ... have you seen Cro? 02:17
sjaveed what's that?
02:18 MilkmanDan joined
[Coke] (someone already built something similar to the thing you're looking for) - but I know that building one is very attractive to the folks that like perl & raku 02:18
sjaveed sorry really new to the ecosystem
oh hmmm
[Coke] cro.services/
sjaveed i'll take a look
02:18 maggotbrain777 left
sjaveed yeah this is also a bit of a personal challenge really 02:18
nothing like a project to learn a language i think
but maybe there are pointers to using regexes for this there 02:19
thanks for the pointer!
02:19 maggotbrain joined
elcaro_ sjaveed: maybe look at the source of HTTP::Server::Router and see how it does it 02:19
sjaveed oooh 02:20
a pure router
[Coke] elcaro_++
sjaveed ok this might be more interesting
the Cro site's first example seemed to indicate they're solving the problem differently
02:20 elcaro_ is now known as elcaro
sjaveed get "greet", $name -> sub {} 02:20
[Coke] anyway, I don't know why your attempt doesn't preserve the captures. :(
sjaveed they're bypassing the problem it looks like
thanks elcaro_! this is promising 02:22
they're also using regexes but differently 02:23
elcaro No probs... you can search for modules at modules.raku.org
even if you're trying to roll your own solution for learning, it's often helpful to look at how others tackled it
sjaveed agreed
the thing i know i'll struggle with for a bit is finding the idiomatic way to raku 02:24
if nothing else, existing modules should help point me in the right direction for that
02:27 maggotbrain left 02:28 maggotbrain joined 02:36 stoned75 left 02:38 a3f left 02:39 a3f joined 02:42 maggotbrain left 02:43 maggotbrain joined 02:47 maggotbrain777 joined, maggotbrain left 02:48 stoned75 joined
sjaveed fwiw elcaro and [Coke], this gist works using EVAL gist.github.com/sjaveed/061c743b70...46bcc5c000 02:51
i'd rather not use EVAL of course 02:52
only the first and last examples in there work - the middle example continues to fail
03:11 MasterDuke left 03:13 maggotbrain777 left 03:14 maggotbrain joined
sjaveed hmmm so it looks like github.com/tony-o/perl6-http-server-router is bypassing a regex check as well 03:15
there's a regex check in there which caught my eye but it looks like that's used when the route is defined in terms of a regex 03:16
looks like github.com/moznion/p6-Router-Boost solves it using the EVAL approach 03:21
03:22 maggotbrain left
[Coke] maybe when moritz gets back in the morning (EU) he can figure it out. 03:23
sjaveed nice - yeah this is just weird - i'm glad i have a workaround but a proper solution or a definitive "that's not possible" would be good! 03:27
03:28 aborazmeh joined, aborazmeh left, aborazmeh joined
sjaveed oh nice - moritz as in Moritz Lenz? 03:28
i might come back in EU daylight hours then - thanks! 03:30
03:34 maggotbrain joined
elcaro 03:38
[Coke] aye, that one. 03:39
elcaro sjaveed: EVAL might be the only option. another example is jnthn's ECMA262Regex module, which converts a ECMA262 regex to a Raku regex using EVAL 03:40
jnthn has stated that this is something that RakuAST should be able to solve when it lands
sjaveed nice 03:45
ok this is all good info - looks like i have something to look forward to 03:46
i'll take a look at the ECMA262Regex module though - that might be the cleanest approach to this without using EVAL myself
03:50 maggotbrain777 joined 03:52 maggotbrain left 03:55 [Coke]_ joined, [Coke]_ left, [Coke]_ joined 03:57 Geth left, Geth joined 03:58 [Coke] left, sjn left, mendel left, mendel_ joined, sjn joined, zacts joined 04:22 aborazmeh left 04:24 stoned75 left, stoned75 joined 04:25 BenGoldberg left 04:26 [Coke]_ left 04:35 [Coke] joined, [Coke] left, [Coke] joined 04:37 ensamvarg3 joined 04:39 abraxxa left 04:40 abraxxa joined 04:43 abraxxa left 04:45 maettu joined, abraxxa joined 05:07 zacts left 05:09 zacts joined 05:20 molaf__ left 05:27 bocaneri joined 05:28 squashable6 left 05:31 squashable6 joined 05:32 maettu left 05:34 andrzejku joined, hungrydonkey joined 05:35 tejr left 05:45 aborazmeh joined, aborazmeh left, aborazmeh joined 05:48 zacts left, zacts joined 05:51 jmcgnh left 05:54 aluaces joined 05:58 jmcgnh_ joined 06:07 zacts left 06:09 andrzejku left 06:21 skids left 06:24 aluaces left 06:31 andrzejku joined 06:44 domidumont joined 06:47 aluaces joined 06:49 Sgeo left 06:53 Altai-man joined 06:59 earenndil joined 07:01 earenndil left, moon-child joined, rindolf joined 07:06 maettu joined
Geth ecosystem: 09cba83206 | Julio++ | META.list
Add BigRoot to ecosystem

See github.com/juliodcs/BigRoot
ecosystem: 2966a5f591 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | META.list
Merge pull request #541 from juliodcs/juliodcs-BigRoot

Add BigRoot to ecosystem Thanks!
07:09 dyske joined, dyske left 07:10 dyske joined 07:15 aborazmeh left 07:16 BenGoldberg joined 07:19 sena_kun joined 07:21 Altai-man left 07:27 tejr joined 07:32 aluaces left 07:34 dakkar joined 07:35 maettu left 07:38 dyske left, tejr left 07:49 maggotbrain777 left 07:50 BenGoldberg left 07:52 maggotbrain joined 07:53 tejr joined 08:01 ensamvarg joined 08:03 sno left 08:05 sno joined 08:12 domidumont left 08:16 domidumont joined 08:25 pecastro joined 08:47 Black_Ribbon left 08:48 Doc_Holliwould left 09:03 ensamvarg3 left 09:18 tejr left 09:25 maettu joined 09:28 stoned75 left 09:29 stoned75 joined 09:36 aluaces joined 09:42 natrys joined 09:48 maettu left 09:56 markoong joined 10:07 Kaiepi left
timotimo m: my $rexstr = "$<first>=<[a..z]> $<rest>=<[a..z]>+"; "hello world" ~~ / \s <blorp=$rexstr> /; say $<blorp> 10:07
camelia Use of Nil in string context
in block <unit> at <tmp> line 1
Use of Nil in string context
in block <unit> at <tmp> line 1
Unrecognized regex metacharacter = (must be quoted to match literally)
at /home/camelia/EVAL_0…
timotimo m: my $rexstr = "$<first>=<[a..z]> $<rest>=<[a..z]>+"; "hello world" ~~ / \s $<blorp>=<$rexstr> /; say $<blorp> 10:08
camelia Use of Nil in string context
in block <unit> at <tmp> line 1
Use of Nil in string context
in block <unit> at <tmp> line 1
Unrecognized regex metacharacter = (must be quoted to match literally)
at /home/camelia/EVAL_0…
timotimo m: my $rexstr = '$<first>=<[a..z]> $<rest>=<[a..z]>+'; "hello world" ~~ / \s $<blorp>=<$rexstr> /; say $<blorp>
camelia 「world」
first => 「w」
rest => 「orld」
timotimo sjaveed: i'm not sure if this will help; subcaptures of a regex-as-string interpolated bit can be received like this, by giving the result a name explicitly in the interpolating regex 10:09
10:09 Kaiepi joined
Xliff timotimo++ -- I was looking for that! 10:10
It uses EVAL under the hood, yes?
timotimo that's right 10:11
Xliff Thanks. 10:15
10:30 aluaces left 10:41 BenGoldberg joined 10:56 Doc_Holliwould joined 11:15 BenGoldberg left 11:18 Kaiepi left, Altai-man joined 11:21 sena_kun left 11:25 ensamvarg3 joined 11:26 Kaiepi joined 11:28 molaf__ joined 11:33 Noisytoot left 11:34 Noisytoot joined, molaf__ left 11:37 cpan-raku left, Kaiepi left 11:38 cpan-raku joined, cpan-raku left, cpan-raku joined 11:39 leont joined 11:49 hungrydonkey left 11:55 tejr joined 12:05 aluaces joined
lizmat .seen lizmat 12:06
tellable6 lizmat, I saw lizmat 2020-09-09T11:53:15Z in #raku-dev: <lizmat> ugexe: I tried
12:13 tejr left 12:14 tejr joined
Roamer` ...were you trying to get the bot to do a double-take?... I don't think it worked... 12:23
12:31 Archenoth joined 12:48 dogbert17 left 12:55 El_Che left
Xliff Is there any way to programmatically invoke need statements? 13:01
I would like to run them from a loop.
13:01 El_Che joined
rypervenche sjaveed: Any reason why you're not just using a named regex though? 13:01
m: my regex blah { '/greeting/' (.+) }; my $path = '/greeting/bob/'; $path ~~ &blah; 13:02
camelia ( no output )
rypervenche m: my regex blah { '/greeting/' (.+) }; my $path = '/greeting/bob/'; say $path ~~ &blah;
camelia 「/greeting/bob/」
0 => 「bob/」
Xliff m: my $a = 'NativeCall'; need $a; 13:10
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
need used at line 1 (in Raku please use "v" prefix and "use" for pragma (e.g., "use v6;", "use v6.c;") instead)
Xliff m: use v6; my $a = 'NativeCall'; need $a;
camelia Don't know setting RESTRICTED
at gen/moar/ModuleLoader.nqp:249 (/home/camelia/rakudo-m-inst-2/share/perl6/lib/Perl6/ModuleLoader.moarvm:previous_setting_name)
from gen/moar/ModuleLoader.nqp:254 (/home/camelia/rakudo-m-inst-2/share/perl6/lib/…
Xliff HUnh?
That definitely looks LTA
13:31 andrzejku left 13:33 tejr left 13:41 mst joined 13:45 aluaces left 13:57 tejr joined 14:04 cusion joined 14:07 mowotter left, BenGoldberg joined, hungrydonkey joined
timotimo c: HEAD my $a = 'NativeCall'; need $a 14:08
committable6 timotimo, ¦HEAD(e65466f): «04===SORRY!04=== Error while compiling /tmp/hNU22Qt_gA␤Undeclared routine:␤ need used at line 1 (in Raku please use "v" prefix and "use" for pragma (e.g., "use v6;", "use v6.c;") instead)␤␤ «exit code = 1»»
timotimo yeah, odd. 14:09
14:17 Sgeo joined
sjaveed Thanks for the additional comments timotimo, rypervenche 14:19
I'll take a look as soon as i get a moment to breathe here :-)
but tl;dr on why i can't use a named regex rypervenche: i'm generating these regexes from other strings 14:20
rypervenche Ahh ok 14:23
guifa Xliff: yes 14:29
need ::($a); if I remember correctly
14:29 orinthe left
guifa At least, it works with require 14:29
14:30 orinthe joined
guifa Xliff: actually, need says it runs at compile time, so it probably won’t work. I’d recommend going with require and then using the fully qualified name as a workaround (that’s what I’m using for Intl::X) 14:32
Altreus > Private multi-methods are not supported 14:34
why :)
I was sort of relying on it working tbh 14:36
guifa Altres: that’s a good question. Though there’s quite a few workarounds
Altreus I'm using the multi nature of it to halt recursion - I only made it private just now 14:37
doesn't have to be, but erm etc
timotimo private methods something something optimization? 14:39
one important thing about multis is that new candidates can be brought in in different contexts
like, with an inner scope that defines a multi candidate, that is only in there 14:40
Altreus can I define a method like that?
Altreus TIAS
timotimo but private methods are per exact class, so a little more like a sub
actually, with a multi sub you may be better served here?
Altreus shit it worked 14:42
14:42 BenGoldberg left, rbt joined
Altreus actually maybe it did, maybe it didn't 14:42
I got excited, hang on
guifa Altreus take a look at this 14:43
rbt Hello. Can anyone explain the difference between Buf derived from WHAT and Buf in the hash lookup?
my %H{Mu:U};
%H{Buf} = method (Buf $val) {'some buf'};
sub lookup(Mu $datum) {
my \type = $datum.WHAT;
dd $datum.WHAT, %H{$datum.WHAT};
dd type, %H{type};
dd Buf, %H{Buf};
my Buf $b = Buf.new(^256);
##### Output
# Buf
# Any element of %H = Any
# Buf
# Any element of %H = Any
# Buf
# Method %H = method <anon> (Mu: Buf $val, *%_) { #`(Method|93826602735792) ... }
moritz rbt: please don't past more than single line of code here; use a pastebin for that 14:44
guifa m: class A { multi sub b(\self,Int $) {‘int’}; multi sub b(\self,Str $) {‘int’}; method a (|c) { self.&bar(|c) } }; say A.new.test(1); say A.new.test(‘a’)
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
bar used at line 1. Did you mean 'bag', 'VAR'?
guifa erp 14:45
rbt moritz: Okay. Will do in the future.
guifa m: class A { multi sub b(\self,Int $) {‘int’}; multi sub b(\self,Str $) {‘int’}; method a (|c) { self.&b(|c) } }; say A.new.test(1); say A.new.test(‘a’)
camelia No such method 'test' for invocant of type 'A'. Did you mean any of
these: 'gist', 'List', 'list', 'Set'?
in block <unit> at <tmp> line 1
guifa okay I really need to stop renaming variables for golfing in irc
Altreus ok so it looks like the methods aren't constrained to the scope so they're still public even though I defined them elsewhere
so it looked like it worked but it was just the original with weird shapes :D 14:46
moritz rbt: I don't have time to dive into it right now, but maybe it's related to Buf being a (parametric) role; if you try to instantiate it, a class will be automatically created for you
Altreus but yeah I can use my multi sub
moritz so Buf could refer to eitehr of these
Altreus or similar
guifa Altreus: 14:47
m: class A { multi sub b(\self,Int) {‘int’}; multi sub b(\self,Str) {‘str’}; method a (|c) { self.&b(|c) } }; say A.new.a(1); say A.new.a('a')
camelia int
Altreus ah right
guifa will only add one character when you use (basically .&foo instead of !foo)
Altreus but if it's so easy why not just make private multis do that :)
guifa Could just be a NYI thing
Altreus yeah, pesky 14:48
get to it, unpaid labourers!
also thank you
guifa Also… grrr at whoever made the TWEAK method for X::Method::NotFound
It and three other classes in Exceptions.pm6 require special code for localization =/ 14:49
Speaking of weird things methods too btw 14:50
m: my &why-no-work-grrrrr = method { $:placeholder } 14:51
camelia 5===SORRY!5=== Error while compiling <tmp>
Placeholder variables (eg. $:placeholder) cannot be used in a method.
Please specify an explicit signature, like method <anon> (:$placeholder) { ... }
at <tmp>:1
------> 3my &why-no-work-grrrrr =…
Altreus what the heck is a placeholder
guifa whaaaaa? you don’t know the joys of placeholders??!?!?!??!! 14:52
guifa is only surprised because of how deep of a dive you’ve taken in Raku :-)
Placeholders occupy the happy place between a full block with signature and a whatever code
Altreus deep but not wide
I was going to say aren't they things like $^a, $^b
guifa yes they are :-)
Altreus but this does not match that pattern! 14:53
guifa $^positional, $:named
Altreus oh right hmm
I've not come across a block that uses named placeholders since I've not come across a procedure that's going to pass them
guifa (acutally in all honesty I had completely skimed over $:named until only a few monhs ago )
Altreus is it $: or :$ 14:54
you did $: before and it said please do :$ in a signature and I got confused
helpfully that's recognisably the "embarrased" emoji from before we had proper ones 14:55
timotimo $:blah is one way to declare that your block takes a named argument called "blah"
:$blah is the way to declare it in the signature
Altreus TIL about named placeholders, yay!
14:56 Maylay left
guifa It is weird that methods won’t allow them though. You’d think that they would, since self would be implied first positional, and all the rest would follow along from that 14:57
moritz if you can be bothered to writes classes/roles and methods, you can also be bothered to write signatures :D 14:59
14:59 Maylay joined
guifa moritz: if only you could see the nastiness I’m dealing with trying to create a semi passable localization system for Raku hahaha 14:59
moritz it's mostly meant for really short subs, often anonymous ones; for long ones, signatures help immensly with readabilty
guifa moritz (and timotimo for that matter), y’all are native German speakers yeah? 15:00
moritz ja
guifa iche habe eine Bitte O:-) 15:01
moritz shoot :D
guifa I wrote a script that parses out the .message for all the exceptions 15:02
I figured out a way to hack them in so that exceptions print out in non-English languages, even for compile time stuff
so…. naturlich, wanna translate ‘em to German? I’m doing Spanish/Asturian/Euro Portuguese, and rypervenche is going to help do Chinese 15:04
moritz I can translate some; probably don't have enough time/motivation to do all 15:05
guifa yay. (and don’t need all of them, that’s the joy of github and pull requests)
moritz so, where do I do that?
timotimo have you considered using the mechanism that also powers getting exceptions as json? 15:06
moritz SCOPE CREEP! :D 15:07
guifa timotimo I hadn’t seen that. Does it return a JSON object like {type: “foo”, message: “really long message here”} ? 15:09
Altreus je peux essayer de faire le français si tu veux
timotimo env RAKU_EXCEPTIONS_HANDLER=JSON perl6 -e 'die "oh no"' 15:10
Altreus «les traduire dans francais», I suppose
timotimo {"X::AdHoc":{"payload":"oh no","message":"oh no"}}???
it's not "a mechanism to get it as json" it's "a mechanism to get the exception and print it however you want" 15:11
guifa timotimo: did not know that that existed. But the messages are unfortunately generated in code, and sometimes through TWEAK methods (looking at you X::Method::NotFound)
guifa: as tu une^W^W^WDo you have a list of error strings somewhere? 15:14
codesections points for using 3 ^Ws instead of 9 ^Hs :D 15:15
guifa timotimo: I’m looking to actually adjust the message string, but it’s programmatically set — fairly simply in most cases, but quite complex in others (so localizers need some basic knowledge of Raku esp if their language does dual/paucal/etc numbers)
Altreus: I should be tonight. The only thing tripping me up from being done this morning was the no placeholders in methods. Debating my best workaround to keep my updater scripts simple 15:16
codesections guifa: If I could submit one minor patch for English, I think it'd be adding dual/paucal numbers 15:17
guifa I think I’ll probs just do __PLACEHOLDER__ and do a subst, since it’s just two of them and they have to have some special handling anyways
Altreus what are these numbers
codesections (a few years ago, I would have said "a gender neutral pronoun" but 'they' seems to have caught on well enough these days)
guifa English has singular and plural (the latter is technically “non-plural” because one has “zero things”)
Arabic, for instance, has singular, DUAL and plural, so that the number 2 gets handled specially 15:18
Altreus aha
I did wonder about that weird feature of Western languages
why is 1 so special
guifa paucal is a weird one, normally for “a few” (with languages determining that definition specially) 15:19
15:19 sena_kun joined
guifa It’s a localizer’s nightmare come true. Determining the correct number is … complex :-) 15:20
codesections imo, it infects thinking by inclining people towards binary divisions (1 vs not-1) in a way that would be less pronounced with a trinary division
but that's total speculation
15:21 Altai-man left 15:24 MilkmanDan left 15:28 MilkmanDan joined
Altreus we just group everything so it can be 1 again :D 15:29
guifa So there’s still one or two bugs in the generation of this file but …  15:31
But it gives a decent enough overview of how an Intl::X::en file might look 15:32
15:33 grumble joined
guifa Some of them have a “Note: “ (see X::Adhoc, line 133) which I’m creating bit by bit as I do my own localization 15:34
Altreus 'SORRY!' 15:35
:D 15:36
guifa I gave serious thought to just leaving that one as is haha
and then adding in a Canadian flag next to it ;-)
Altreus DÉSOLÉ! qqc a fait tort
not enough exclamation marks 15:37
The EVAL one has three, a sure sign of a diseased mind
guifa Oh but the Spanish and Asturian get six! 15:38
Altreus oy!
guifa errr
¡¡¡EVAL ye una función perpeligrosa!!!
Altreus I'm sort of surprised they still use those upside-downey ones
do they even have special names? 15:39
guifa You have to have them.. “It’s easy” and “Is it easy” use the same words, in the same order, but different intonation
Altreus I'm mildly upset that they render differently
Well that's true of French too, but the one at the end is enough
C'est facile. C'est facile? 15:40
guifa For a long question, it might not be obvious that something is a question until you get to the end
so it’s kinda nice
Altreus Well that's OK, the Germans put all their useful words at the end and cope perfectly well
It's a language of suspense
guifa But they just call them opening and closing marks, same model they use for ( ) or « »
Altreus oya
is fun 15:41
I was just chatting on twitch today about the way we English use the exact same words for like 5 different meanings depending on how you say it
But we completely refuse to use any sort of punctuation or accents to explain which one it is
guifa yeah 15:42
guifa . o O ( indeed | for real? | go English! | meh )
Altreus we need punctuation that means this wasn't enough of a sentence to deserve a full stop 15:46
sjaveed how about elipses...
Altreus that's three! 15:47
How did we get to the point where repeating punctuation makes it less effective? :|
[Coke] 15:48
[Coke] m: say 2÷6
camelia 0.333333
15:50 skids joined 15:56 Kaiepi joined 16:04 Kaiepi left 16:09 andrzejku joined
guifa` oh, I also totally forgot 16:14
Altreus: lizmat had made a proposal a while back to make ¿? and ¡! valid syntax in Raku. IIRC she was thinking of them as quotation marks, but if they could be used as a substitue for ?( ... ) and !( ... ) that would be kinda cool :-) 16:15
16:20 stoned75 left 16:22 stoned75 joined
Altreus ah and I can type them easily so bonus 16:29
does `my sub` close over `self` (not $self)? 16:30
16:30 billN1VUX joined
[Coke] does my sub even have a self? 16:30
Altreus as in method foo() { my sub bar { ... } } 16:31
I've used self.&bar for now
plus the relevant signature 16:32
16:34 xelxebar left 16:35 aluaces joined 16:36 xelxebar joined
guifa` subs explicitly reject the use of self 16:38
although maybe in a method block's lexical scope? 16:39
m: class A { method B { self #`(works okay here); my &c = sub { self #`(here though?) } } } 16:40
camelia WARNINGS for <tmp>:
Useless use of self symbol in sink context (line 1)
guifa` m: class A { method B { self.say #`(works okay here); my &c = sub { self.say #`(here though?) } } }
camelia ( no output )
guifa` seems to allow it, if you've got it in method's scope 16:41
16:44 jmcgnh_ left
Altreus I finally got to the position where I can TIAS :) 16:45
now I can TIAS
16:46 dakkar left
Altreus what function does try have over just using a block? 16:52
[Coke] legibility? 16:55
it also turns on "use fatal" 16:56
Altreus ah, implicit catch! of course 16:57
Is there a way to tell installations that a file has been deleted? That was a common issue with cpan 16:58
And I've just done it again :D
16:59 aluaces left 17:00 molaf joined 17:03 MasterDuke joined 17:10 rbt left 17:11 rbt joined
[Coke] Please consider casting a ballot for the RSC if you're eligible and haven't voted yet 17:19
17:23 markoong left 17:26 aluaces joined 17:32 zacts joined 17:33 BenGoldberg joined 17:37 Maylay left 17:42 jmcgnh joined 17:56 Maylay joined 18:01 vgrato_ left 18:07 BenGoldberg left 18:16 stoned75 left 18:17 stoned75 joined 18:22 Black_Ribbon joined 18:27 bocaneri left 18:30 Kaiepi joined 18:39 cgfbee left, cgfbee joined
El_Che until when can one vote? 18:47
[Coke] 20th 18:53
but please don't wait until the end to vote, as I have to click buttons to process each vote. 18:54
(and would rather not get 100 votes at the end of the run)
Juerd [Coke]: Do people who vote get a confirmation that it was received?
rypervenche guifa`: Does the ?? in your script allow for variants? 18:55
18:57 zacts left, zacts joined, zacts left
lizmat Juerd: they don't, afaik, but you can ask [Coke] here whether they received it 18:59
Juerd lizmat: It's fine, if the assumption is that email works then I'll assume the same :) 19:01
El_Che and to think that voting by mail is somewhat controversial in some countries :)
19:01 xelxebar left
lizmat well, we'll see whatever the RSC will come up with for the next election 19:02
19:02 xelxebar joined 19:03 wamba joined
[Coke] Juerd: I'm not replying to the emails, no. 19:03
Juerd [Coke]: ok, thanks :) 19:04
[Coke] I will post a list of email addressess that voted as part of the results.
(but not who they voted for)
Juerd Please obfuscate those a little because of address harvesting bots
[Coke] Sure, but in general that ship has sailed. 19:06
Juerd: patches welcome to github.com/Raku/Raku-Steering-Coun...tally.raku 19:07
Juerd In case of my own address, certainly, and it might be true for most. But you don't want to be responsible for someone's first, I think :)
[Coke] or just tell me what you want done: how would you want "Will Coleda [email@hidden.address] (@coke)" to appear?
mine's been out the mid 90s. :) 19:08
just s/@/ -at-/ ? 19:09
Juerd Simple substitution of the '@' for something uncommon goes a long way, such as s/'@'/ nospam~at /
guifa` rypervenche: that's actually one fo the advantages of messages being methods 19:10
Juerd Just the word 'at', with or without punctuation, will work for many but not most bots, while anything else will fool most (but still not all of course)
El_Che $whatbeforemail+domain@owndomain.com
guifa` rypervenche: oh, actually, you mean in terms of language codes. Yes, absolutely. Although it's not smart about it yet
Juerd [Coke]: Thanks :) 19:11
guifa` My Portuguese translation wil be Intl::X::pt-PT for instances, so that SmokeMachine can do Intl::X::pt-BR ;-)
Geth Raku-Steering-Council/main: 7e6800a260 | Coke++ | scripts/tally.raku
Add simplistic ranking output
Raku-Steering-Council/main: 0049ee5653 | Coke++ | scripts/tally.raku
Obfuscated reported email address

[Coke] Done. 19:13
SmokeMachine guifa`: I'm in! How should I start?! :)
[Coke] btw, I am not doing a great job with mime encoding of non-ascii. "Christian =?utf-8?Q?Bartolom=C3=A4us?=" 19:14
Juerd That's rfc2047 but I haven't done anything with in it Raku 19:16
It's in Perl's Encode.pm as MIME-Header 19:17
guifa` SmokeMachine: I should have the finalized template sometime tonight. Still have a small bug in one of my scripts to fix before it's usable 19:18
19:18 Altai-man joined 19:20 xinming left, xinming joined 19:21 sena_kun left 19:25 Altai-man left
bartolin [Coke]: oops. sorry for causing trouble there 19:28
[Coke] bartolin: :) 19:32
bartolin has (marginally) messed up the commit history in nfs-utils recently as well: git.linux-nfs.org/?p=steved/nfs-uti...09;hb=HEAD But then someone else managed to get an Umlaut in there ... 19:37
maybe I should stop doing that ;) 19:38
[Coke] wonders where he should report bartolin's email address as a bug. Email::MIME? 19:42
19:45 Archenoth left 19:47 Archenoth joined
lizmat remembers the day a new guy named Ævar started working at $work, and broke quite a few of the internal systems :-) 19:55
19:56 andrzejku left, Black_Ribbon left
guifa` Ævar was the first name or was he one of those mononymic types? 19:56
(or both, and if so, which broke the system?) 19:57
19:58 Black_Ribbon joined, andrzejku joined, zacts joined
lizmat the fact that there was no extended ASCII representation for Æ (if I remember correctly :-) 19:59
codesections Ah, good old #9 on the list: www.kalzumeus.com/2010/06/17/false...out-names/ 20:00