🦋 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.
rypervenche releasable6: status 00:07
releasable6 rypervenche, Next release will happen when it's ready. There are no known blockers. 0 out of 289 commits logged
rypervenche, Details: gist.github.com/7a45dda8cb9278510e...30137d78de
Tirifto Hello! 00:46
Side question: Could one say that the ‘&’ sigil marks the infinitive? 00:49
Juerd That's a nice observation. Seems to work, but there are probably enough edge cases :) 00:51
Tirifto Juerd: Gotcha. Then I'll keep it in mind and keep an eye out for edge cases; thanks! :) 00:55
tbrowder .tell chloekek user-format pod leading decl works on master branch and should be in upcoming release. until i v
tellable6 tbrowder, I'll pass your message to chloekek
tbrowder .tell chloekek until i can update docs, look at roast test: S26*/block-leading-user-format.t for an example 01:00
tellable6 tbrowder, I'll pass your message to chloekek
kalkin-- If I have Pod on a class and it attributes + methods, I would expect to get a Pod Structure as a tree, but currently I'm getting a list. Is it a bug or a feature? 09:58
Geth_ doc: 402e897c6a | (JJ Merelo)++ | doc/Language/statement-prefixes.pod6
Comment automatic serialization of *Seq in loops

And how to avoid it with statement prefixes #2632.
10:14
Kaiepi m: class Foo { has Int:D @!foo is Set; submethod BUILD(:@!foo) { } }; say Foo.new: :foo[1,2,3] 10:34
camelia Cannot modify an immutable Set[Int:D] (set())
in submethod BUILD at <tmp> line 1
in block <unit> at <tmp> line 1
Kaiepi m: $eval my class Foo { has Int:D @!foo is Set; submethod BUILD(:@foo) { @!foo := @!foo.new: @foo } }; Foo.new: :foo[1,2,3]
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$eval' is not declared. Did you mean any of these?
&EVAL
&val

at <tmp>:1
------> 3<BOL>7⏏5$eval my class Foo { has Int:D @!foo is
Kaiepi m: my class Foo { has Int:D @!foo is Set; submethod BUILD(:@foo) { @!foo := @!foo.new: @foo } }; say Foo.new: :foo[1,2,3]
camelia Foo.new
Kaiepi is the former supposed to happen? 10:35
or rather why does the former work the way it does? 10:37
oh 11:03
Kaiepi m: my Int:D @foo is Set; @foo = 1, 2, 3 11:04
camelia Cannot modify an immutable Set[Int:D] (set())
in block <unit> at <tmp> line 1
Kaiepi i was expecting it to work like assigning to `my Int:D @foo` would if i hadn't already assigned anything to it 11:05
kalkin- p6: say "asdasd\x00foo".split: "\x00"; 11:45
camelia (asdasdoo)
kalkin- p6: say "asdasd:foo".split: ":"; 11:46
camelia (asdasd foo)
kalkin- split a string by ASCII NULL doesn't work
bug or feature?
lizmat I'd say bug
kalkin- p6: say "asdasd\x0foo".split: "\x0"; 11:47
camelia (asdasdoo)
kalkin- p6: say "asdasd\0foo".split: "\0";
camelia (asdasd foo)
kalkin- p6: say "asdad\x3easd" 11:48
camelia asdadϪsd
kalkin- p6: say "asdad\x00asd"
camelia asdad
sd
kalkin- I thought i can enter any char via \xHEXHEX 11:48
looks like it doesn't work for 0 byte?
is this a bug or did I misunderstood the purpose/working of \xFF syntax? 11:49
p6: say "asdasd\0foo"; 11:51
camelia asdasd␀foo
kalkin- p6:"foo:ver<1.2>\0asd", "bar:ver<2.3>\0foo" ==> map *.split(':ver')[1] ==> map *.split: "\0" ==> my %h; say %h 11:53
evalable6 {<1.2> asd => (<2.3> foo)}
kalkin- not what i expected, hmm somehow \0 parsing is wrong, or I'm missing something obvious 11:54
p6: "foo:ver<1.2>\0asd", "bar:ver<2.3>\0foo" ==> map *.split(':ver')[1] ==> map *.split: "\0" ==> flat() ==> my %h 11:57
camelia ( no output )
kalkin- p6: "foo:ver<1.2>\0asd", "bar:ver<2.3>\0foo" ==> map *.split(':ver')[1] ==> map *.split: "\0" ==> flat() ==> my %h; say %h
camelia {<1.2> => asd, <2.3> => foo}
chloekek Inline::Perl5 gives me Prototype mismatch: sub Digest::SHA::DESTROY ($) vs none at (eval 10) line 5. if I import Digest::SHA with use Digest::SHA:from<Perl5> <sha256_hex>;. But everything works fine though. 12:55
tellable6 2019-12-22T00:55:58Z #raku <tbrowder> chloekek user-format pod leading decl works on master branch and should be in upcoming release. until i v
2019-12-22T01:00:19Z #raku <tbrowder> chloekek until i can update docs, look at roast test: S26*/block-leading-user-format.t for an example
chloekek tbrowder: neat! 12:57
kalkin- did p6doc Language/testing or p6doc Str ever worked for any one? 13:06
kalkin- ok I see. I neeed to install from source and then use `zef install --force .` 13:59
chloekek p6: sub g { $*x = 2; }; sub f { my $*x = 1; g; say $*x }; f 14:16
camelia 2
chloekek p6: sub g { $*x = 2; }; sub f { my $*x := 1; g; say $*x }; f
camelia Cannot modify an immutable Int (1)
in sub g at <tmp> line 1
in sub f at <tmp> line 1
in block <unit> at <tmp> line 1
chloekek p6: sub g { my $*x = 2; }; sub f { my $*x := 1; g; say $*x }; f
camelia 1
kalkin- Hmm, am I right that it's under specified if a leading/trailing pod can contain elements like L<> and B<> ? 14:35
Current rakudo does not parse formatting codes in leading/trailing pods. Bug or feature? 14:40
tbrowder .tell jmerelo i can't tell if my advent post is scheduled properly--the bloody calendar display seems to have changed on me! please help! 14:47
tellable6 tbrowder, I'll pass your message to jmerelo
tbrowder afk&
kalkin: formatting codes are not yet implemented in all pod objects, PRs welcome! 14:51
uzl[m] tbrowder: Will you latest work allow for full POD formatting in declarator blocks? Or only the comments won't be squashed into a single line as before? 14:54
kalkin- tbrowder: my issue isn't even that it's not implemented, skimming S26 i'm not sure if formatting codes should be implemented as part of leading/trailing pod 15:18
guifa kalkin-: I think ultimately it would make sense — after all theoretically it’s just a POD. But they get stored as Str ATM 15:49
cpan-raku New module released to CPAN! Getopt::Subcommands (0.0.1) by 03LEONT 15:53
kalkin- .tell guifa looking at Grammar.nqp there're two methods for attaching leading/trailing pod. I'm not sure why it's implemented this way 16:21
tellable6 kalkin-, I'll pass your message to guifa
kalkin- In general I find Grammar.nqp is very confusing and complicated. My understanding of it is very limited
.tell guifa never mind, it's for attaching it the actual code parts 16:25
tellable6 kalkin-, I'll pass your message to guifa
chloekek First time running Raku code on a server. :D 16:28
tbrowder kalkin: i agree it wasn't planned, 17:37
*planned for all pod objects, and as i read it that includes table cells. however, that doesn't mean we couldn't add that feature. but i think that would entail a new internal design. 17:39
likewise, S26 doesn't address formattingb 17:40
*formatting in pod declarator blocks as far as i can see. 17:41
kalkin- tbrowder: can you elaborate on what you mean by “new internal design”. What is wrong with the current, besides that it stringifies content for leading/trailing pods 17:46
tbrowder well, when you look at how pod parts are assembled, i think it would be easier to handle formatted code, as well as other objects, as classes, so that we could more easily create such things as tables within table cells, etc. 17:52
johnjohn101 hi raku people 18:49
Altai-man_ o/
tbrowder \o 18:53
kalkin- huhu 18:56
AlexDaniel Altai-man_: ahhhhhhhh 19:03
I have an idea
Altai-man_ yup?
cpan-raku New module released to CPAN! RakuAdvent::WordPress (0.0.2) by 03TBROWDER
AlexDaniel Altai-man_: can you try wiping both installed/ and output/ dirs before running blin?
or is it already the case?
Altai-man_ AlexDaniel, I did that before running it the second time (with 12 nproc), but if you want to...
AlexDaniel hmm…
Altai-man_ AlexDaniel, as for "green light" from me: well, no regressions were spotted yet. what's the usual number for ZefFailure count? 19:04
AlexDaniel Altai-man_: 5
ok, there were a lot more, for the file I have at hand there were 10 19:05
rindolf AlexDaniel: hi, sup? ltns
Altai-man_ hmm, ~300 is too much compared to 10... 19:06
AlexDaniel Altai-man_: exactly
Altai-man_: many times there's only 1 module that breaks when we introduce a serious regression, so not testing 300 modules is pretty bad 19:07
Altai-man_ AlexDaniel, I probably have to read the blin code more carefully, how the situation with "already installed" can occur at all? a single zef instance is used? 19:08
AlexDaniel rindolf: I'm enjoying a little break from raku stuff and at the same time being busy on some of my other projects (though I use raku for all of them…) 19:09
Altai-man_: I don't get it too, yes
rindolf AlexDaniel: ah, i am stuck on a node-inspect issue
AlexDaniel Altai-man_: it installs modules into separate directories, and these directories are only included when the module is in the dependency chain 19:10
Altai-man_: so I have absolutely no idea how that can happen, unless installed/ dir was not purged 19:11
Altai-man_ oh
so `installed/` too...
I only purged output/, ouch.
AlexDaniel ok one issue down :)
there's no purge command because I thought the image with blin will be fresh every time you start it 19:12
and that's how it should be, actually…
like, scrap the whole vm or whatever there is and start from scratch for every run 19:13
Altai-man_ I am sure rm -r can delete some things...
AlexDaniel but there are also processes and other stuff that can be left behind
Altai-man_ AlexDaniel, there are 26 tasks right now and I don't see anything related except my ssh session, so... 19:15
AlexDaniel ok let's figure out the proper way of running it later :)
Altai-man_ so I am running nproc 12 again, with installed/ output/ contents purged, right? 19:16
AlexDaniel Altai-man_: right, you can also maybe purge /tmp/whateverable
Altai-man_: also, are you sure you have all dependencies installed? 19:17
===SORRY!===\nPlease install Inline::Perl5 for Perl 5 support. \n\n
ah
"output": "===> Searching for: Inline::Perl5\n«timed out after 600 seconds»"
either increase the timeout or run with less nproc
Altai-man_ AlexDaniel, even less than 12?
AlexDaniel try with 8 maybe 19:18
it's gonna take a while but… that's maybe ok…
I'm gonna go get some food
AlexDaniel & 19:19
kalkin- what does <.foo> mean in Grammer.nqp 20:33
is it just a call to a method? 20:34
leont No, means it matches the foo rule, but doesn't capture it into a foo element
kalkin- leont: but there is a attach_leading_docs method but no rule 20:36
or is it the case of as long as you return a Match it doesn't matter that it's a method? 20:37
like documented here: docs.perl6.org/language/grammars#M...n_grammars ?
I've been starring into Grammar.nqp for a few hours now, but i still have no idea how to allow to handle leading/trailing pods as proper pod and not Str 20:39
e.g in Grammar.nqp:2415 the leading pod is set. But at which point is $*DOC set, which is used by attach_leading_docs? 20:41
I suspect it habens in the token pblock, but i don't see where it is matched for a subroutine 20:42
Xliff o/ 20:47
Will raku work in a 32-bit environment? 20:48
AlexDaniel yes, without JIT 20:49
and possibly with some bugs
but yes
Xliff Thanks.
Xliff Is there a programmatic way to handle delegation, yet? 20:59
Answer: No. See src/core.c/traits.pm6:422 21:04
Actually, that's line 421. The reason it's "no" is because the role that does all the work is scope-limited
lizmat Xliff: I see no reason why that couldn't be publized / exposed 21:05
but how would you indicate you'd want it your way? 21:06
Xliff lizmat: Actually, couldn't you call it as trait_mod:<handles>() and manually pass the attribute and method name?
lizmat preferably in a BEGIN block, but yeah, that could work 21:07
Geth_ ecosystem/bbkr-patch-1: b4b5ede86c | (Pawel Pabian)++ (committed using GitHub Web editor) | META.list
Added HomoGlypher module - toolset for handling similarly looking characters in strings.

I agree to the usage of the META file as listed here.
   I have a license field listed in my META file that is one of spdx.org/licenses
lizmat alternately, expose this as an Atribute method ?
Geth_ ecosystem: bbkr++ created pull request #476:
Added HomoGlypher module.
21:09
leont has had to call trait_mod's directly before
Xliff lizmat: role Delegatable {}; for self.^attributes.grep(* ~~ Delegatable) { trait_mod<handles>($_, .name.substr(2)) } # Attributes have been pre-marked with Delegatable 21:11
lizmat: Definitely in a begin block.
lizmat: ClassHOW, I would think. 21:12
ClassHOW.add-deletation($attriubte, $thunk)
But Attribute could work, too. 21:13
Attribute.add-delegator($thunk)
Aaannnnd.... crap. 21:19
If this needs to be done in a BEGIN block, I can't do this from a role.
Xliff Please tell me I'm mistaken... ;S 21:19
lizmat is trying to make it an Attribute method
Xliff lizmat++ 21:20
lizmat Xliff: ok, this gets into bootstrap issues... I guess there is a reason why that role lives in the place that it does 21:35
Stage start : 0.000 21:36
Ambiguous call to trait_mod:<handles>
Xliff: gist.github.com/lizmat/31a39891dce...39846fa047 # if you wanna have a go at it yourself 21:39
Xliff lizmat: Yeah. There's this weird thing with roles and attributes. 22:01
At compile time I don't think they are completely there, yet. 22:02
Especially when I am trying to get the attributes of a CStruct.
Hmm... they are there if I strictly go from the CStruct, but not for the role that contains the CStruct 22:03
lizmat: This seems to work. 22:06
github.com/Xliff/p6-GIMP/blob/mast...g.pm6#L135 22:07
^^ Compiles, not yet tested, but I would expect I would get immediate feedback from the BEGIN block if there were problems. 22:08
Xliff Will this work? -> "multi submethod BUILD (:$color-config) { ... }; multi submethod BUILD (:$mode, :$properties, :$something-else) { ... }; " 22:14
jnthn Xliff: You'd need to make those required named arguments for it to work 22:16
Xliff So more like: multi submethod BUILD (:$color-config is required ) { ... }; multi submethod BUILD (:$mode, :$properties, :$something-else) { ... }; "
jnthn I'd just spell it ! but yes 22:17
chloekek p6: use NativeCall; say byte ~~ Int 22:45
camelia True
chloekek p6: use NativeCall; say $_ ~~ Int for (int8, int16, uint8, uint16); 22:46
camelia True
True
True
True
chloekek p6: use NativeCall; say $_ ~~ UInt for (int8, int16, uint8, uint16); 22:47
camelia True
True
True
True
chloekek p6: my int $x = 0; say $x.VAR 23:01
camelia 0
lizmat chloekek: the int gets upgraded to Int 23:03
m: my Int $x = 0; say $x.VAR 23:04
camelia 0
lizmat m: my Int $x := 0; say $x.VAR
camelia 0
lizmat hmmm
guess I'm in he wrong mindset 23:05
chloekek p6: my $x = 0; say $x.VAR 23:28
camelia 0
chloekek p6: my $x = 'a'; say $x.VAR
camelia "a"
chloekek Ah, should check VAR.WHAT
p6: my int $x = 0; say $x.VAR.WHAT
camelia (IntLexRef)