🦋 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.
guifa hrm 00:34
p6: my $t = token { 'a' <?{ say self.pos }> 'r' }; 'bbbbbbbar' ~~ $t
camelia 0
guifa p6: my token t { 'a' <?{ say self.pos }> 'r' }; 'bbbbbbbar' ~~ /<t>/
camelia 7
guifa What’s the rationale for the difference? 00:35
It feels buggy to me
elcaro guifa: this works... 01:38
p6: my $t = token { 'a' <?{ say self.pos }> 'r' }; 'bbbbbbbar' ~~ /$t/
camelia 7
guifa elcaro: so it seems putting the token in a regex is required for it to function properly. 01:39
elcaro yeah, seems related to how $t is interpreted in smartmatch (.ACCEPTS)
guifa I couldn’t quite see what was causing it in the ACCEPTS method. But that’s also all written in NQP and I might be missing something. My guess is it has something to do with it being the top level and tokens aren’t generally expected to be used as a top level regex, but called by others. 01:43
guifa I filed an issue, but maybe it’s correct by design. If so, that’s a good gotcha to mention in the docs somewhere 01:44
elcaro i think it's by design. possibly something to do with token ratcheting behaviour 02:17
if you use the token itself (ie. not inside a regex) you get the same result as the lone scalar
p6: my $t := token { 'a' <?{ say self.pos }> 'r' }; 'bbbbbbbar' ~~ &$t
camelia 0
elcaro using the token as a callable in ~~ works as follows: 02:18
p6: my token t { 'ar' }; 'bbbbbbbar' ~~ &t
camelia ( no output )
elcaro p6: my token t { 'ar' }; say 'bbbbbbbar' ~~ &t
camelia ｢ar｣
elcaro i could be wrong, but maybe the ratcheting essentially ignores everything that failed, so when the token finally matches, the pos is 0. but inside a regex, that regex is keeping track of the pos, not the token itself. 02:19
i'm just guessing here, but it's plausible
eh, maybe not. `token` ratches, `regex` does not, but assigning a `regex` to a scalar and matching on that still produces `0` 02:25
regardless, you are not comparing apples to apples. `'bbar' ~~ $token` is eqv to `'bbar' ~~ &token` 02:27
moritz don't confuse ratcheting (not backtracking) with searching for a match by applying a top-level .pos
Geth doc: dd5b764fab | (JJ Merelo)++ | doc/Type/Pair.pod6
Adds note on synthetic numerics refs #2632
 06:38
Geth doc: 02c55490a2 | (JJ Merelo)++ | doc/Language/list.pod6
Adds enums as shape specifier refs #2632
 06:53
jmerelo 214 out of 296 new items in 6.d already documented github.com/Raku/doc/issues/2632 82 to go... Any help is appreciated... 06:55
chloekek p6: use 🙈; EVAL 'say "hi"' 13:26
camelia 5===SORRY!5=== Error while compiling <tmp>
Bogus term
at <tmp>:1
------> 3use7⏏5 🙈; EVAL 'say "hi"'
expecting any of:
argument list
infix
infix stopper
postfix
prefix
…
jmerelo chloekek: you added that as a synonym? Cool! 13:51
chloekek No. 13:52
chloekek It’s also wrong, because evil and eval are different words. 13:52
Geth ecosystem: 35adc2934a | (Jonathan Worthington)++ (committed using GitHub Web editor) | META.list
Remove OO::Monitors; it's now on CPAN
 17:22
cpan-raku New module released to CPAN! OO::Monitors (1.1.1) by 03JNTHN 17:27
lizmat weekly: andrewshitov.com/2020/01/08/callin...interface/ 17:57
notable6 lizmat, Noted! (weekly)
Tirifto Hello! 18:31
tellable6 hey Tirifto, you have a message: gist.github.com/6c006d356501e511e9...7f800893b9
Tirifto uzl[m]: Thanks for the note! For this case I'll go with ‘use lib’, since it's just for a single portable script. 18:36
[Coke] wonders if he can get raku on his raspberry pi 4 19:43
(without having to self compile)
ToddAndMargo HI All. Newbie question. How do I write (transfer just the bits) back and forth from an Integer (int16) and a Cardinal (unit16)? 20:11
my workaround is to do a bitwise OR on them($x = $y +| 0x0000), but I'd like to know the right way of doing it. 20:13
guifa2 [Coke]: self compiling is the worst with limited resources. I tried to get it on my shared hosting but it popped my memory limit =\ 20:15
ToddAndMargo: Hmm, I don't think there's a great way beyond what you're doing. I'd probably do it as a sub though 20:19
ToddAndMargo guifa2: that you for the confirmation. That is what I will do. 20:21
guifa2 I actually just looked up how to do it in C and it's not very straightforward there either so at least Raku's not trailing too far beyond the lowerlevel languages :-) 20:22
p6: sub into-uint16(int16 $n --> uint16) { $n +| 0 }; my int16 $a = -1; my uint16 $b = into-uint16 $a; say "Signed $a is unsigned $b" 20:25
camelia Signed -1 is unsigned 65535
guifa2 You could also maybe think about doing a circumfix operator instead of a sub. 20:27
p6: sub circumfix:<u( )>(int16 $n --> uint16) { $n +| 0 }; my int16 $a = -1; my uint16 $b = u($a); say "Signed $a is unsigned $b"
camelia Signed -1 is unsigned 65535
guifa2 Then you can use u( foo ) to convert to unsigned. Or use a different bracking like u< > or u[ ], etc dependin gon your taste. 20:29
ToddAndMargo Thank you! 20:34
ToddAndMargo forgive a dumb question. Are standard Raku Str strings UTF8 by default? 21:12
sena_kun yes 21:13
or, to be precise, ahem...
tadzik they're unicode strings, and they may become utf8 if you choose to encode them that way :)
ToddAndMargo I am converting from a UTF16 string coming back from a WinAPI call, so I wanted to make sure I knew exactly what a Raku Str was..Thank you! 21:21
ABC: "A", nul,"B", nul, "C", nul, nul, nul 21:22
guifa Note that internally there is some grapheme caching going on, but ATM I forget exactly what the details of it are 21:34
lizmat and you don't get back a UTF16 string from NativeCall, I would think you get back a Blob? 21:39
ToddAndMargo You get back byte wise a UTF16 string 21:46
ToddAndMargo Win api calls. I don't know about others 21:54
lizmat Blob.decode("utf-16") 21:56
ToddAndMargo cool. how do you decode a little endian? 21:58
lizmat Blob.decode("utf-16le") 21:59
ToddAndMargo very cool. Thank you! is a `CArray[WCHAR]` also a "Blob"? 22:01
Tirifto I’ve another question! If I want to make an object’s attributes read/write, but have them work read-only when passed to certain routines (specially the ones loaded from some modules), is wrapping it in a different object which has the same interface but only implements the read accessors and passing that to the said routines instead a decent way to do this in Raku? 22:06
lizmat Tirifto: you could make the attributes in question read-only, and create custom mutators 22:07
that would make clear that accessing is always read-only 22:08
or do you want the mutators also to not be available ?
Tirifto lizmat: Ideally the loaded routine visiting the object would have no way at all to alter it. (I reckon they would if you mean custom methods of different names by mutators.) 22:11
lizmat m: class A { has $.a; method set_a(\value) { $!a = value } }; my $a = A.new(a => 42); $a.set_a(666); dd $a # something like that 22:13
camelia A $a = A.new(a => 666)
Tirifto If access to the object shouldn't include the access to the means of modification, the latter just can not be contained within the object as a method, right? 22:16
lizmat if I parse your sentence correctly, yes 22:17
Tirifto (Sorry if my phrasing is odd! :P) 22:18
Then I might try the wrapper object… thank you for your hints, lizmat! 22:19
lizmat hmmm... I was going to suggest the "handles" trait, but I guess that will just pass on the containerness 22:21
Tirifto Reading up on ‘handles’ now! 22:25
Tirifto Okay, so if an object’s attribute ‘handles’ a method (name), and that method is called on the object, it will be called on whatever the attribute is holding instead, right? 22:34
lizmat m: class A { has $.a handles "foo" }; class B { has $.foo = "foo" }; my $a = A.new(a => B.new); dd $a.foo 22:36
camelia "foo"
lizmat hope that answers your question :-)
lizmat is tired and is off to bed&
Altai-man_ lizmat, good night. o/ 22:39
Tirifto lizmat: Yeah, it assured me. It doesn't provide a new solution, but might be helpful for the wrapper. Thank you and may you sleep well!

Powered by Cro and the Raku® Programming Language.

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

Thank you!