🦋 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.
00:01 sena_kun joined 00:03 Altai-man_ left 00:28 sno left 00:32 Cabanossi left
raku-bridge <DataKinds> rypervenche: might be very overkill, but that's totally something that mongoDB or similar could make quick work of 00:39
<DataKinds> and you get cool things like very fast indexing and bloom filters if you go that route
<DataKinds> modules.raku.org/dist/MongoDB:cpan:MARTIMM looks very cute 00:41
00:41 xinming left 00:42 xinming joined, MilkmanDan left, Cabanossi joined 00:45 vike left 00:54 vike joined 00:55 Xliff joined 00:56 xinming left, xinming joined
elcaro The title on moarvm.org says "MoarVM - A VM for NQP and Rakudo Perl 6" 00:59
jdv79 what does it all mean?
elcaro what should this change to? "... NQP and Raku"? "... NQP and Rakudo Raku"?
"MoarVM - An elegant VM for a more civilized age" 01:01
01:10 oddp left, molaf left 01:22 molaf joined 01:30 softmoth joined
timotimo ah don't you just love it when a tiny usability feature feels good to use 01:37
even though you're working in a browser
softmoth :) 01:48
timotimo the heap snapshot explorer has two panes for one object each with all its outgoing or ingoing references, and each pane has a "copy over to the other side" button 01:51
but now holding shift while clicking a reference will put the object into the other pane instead of the one it's from 01:52
01:59 maggotbrain joined 02:01 Altai-man_ joined 02:03 sena_kun left 02:05 Manifest0 left, Manifest0 joined 02:11 rbt left, rbt joined
softmoth that's a nice touch 02:14
02:15 Cabanoss- joined 02:17 Cabanossi left, Cabanoss- is now known as Cabanossi 02:21 softmoth left 02:22 softmoth joined 02:24 Doc_Holliwould left
timotimo perhaps even necessary to make it remotely usable :) 02:27
02:30 cpan-raku left 02:31 cpan-raku joined, cpan-raku left, cpan-raku joined 02:45 zacts joined 03:26 sftp left 03:30 sftp joined 03:32 Cabanossi left 03:35 Cabanossi joined 03:55 rbt left, rbt joined 03:59 zacts left 04:02 sena_kun joined 04:03 Altai-man_ left
raku-bridge <DataKinds> hey all, I have a work in progress expansion upon that blog post that I posted earlier 04:21
<DataKinds> datakinds.github.io/2020/06/25/exp...-itertools
<DataKinds> don't post it anywhere but if anyone has any feedback I'd like to hear it 🙂
Xliff DataKinds: Nice job from what I've read so far. 04:26
One thing that twisted my head was the accumulate function...
raku-bridge <DataKinds> why's that?
04:26 epony left
Xliff "sub accumulate(@p, &func = * + *) { [\[&func]] @p }" 04:26
raku-bridge <DataKinds> ahhhh 04:27
Xliff I have never seen that in Raku, before. Can you explain why the nested square braces?
raku-bridge <DataKinds> docs.raku.org/language/operators#R...aoperators
04:27 epony joined
raku-bridge <DataKinds> > To wrap a function instead of an operator, provide an additional layer of square brackets 04:27
<DataKinds> > Prefix the wrapped operator with a , to return a lazy list of all intermediate values instead
<DataKinds> er. that's supposed to be a \, but discord ate it 04:28
Xliff Ah! OK. I must have missed that construct, first time around.
raku-bridge <DataKinds> yeah, it just stuck out to me because i'm an APL guy 04:32
04:32 xinming left 04:33 xinming joined 04:53 sena_kun left 04:54 sena_kun joined, schlaftier left, schlaftier joined 05:13 abraxxa left 05:16 abraxxa joined 05:28 molaf left 05:33 aborazmeh joined, aborazmeh left, aborazmeh joined 05:36 bocaneri joined 05:44 aborazmeh left 05:45 bocaneri is now known as Boc 05:49 aborazmeh joined, aborazmeh left, aborazmeh joined 06:01 Altai-man_ joined 06:03 Black_Ribbon left 06:04 sena_kun left
Xliff m: my @a = <a b c d e>; @a .= splice(2, 0); @a.gist.say 06:04
camelia []
Xliff m: my @a = <a b c d e>; @a.splice(2, 0); @a.gist.say 06:05
camelia [a b c d e]
Xliff m: my @a = <a b c d e>; @a.splice(2, 1); @a.gist.say
camelia [a b d e]
06:06 aborazmeh left, zacts joined 06:36 clarjon1 left 06:49 Xliff left 06:52 Doc_Holliwould joined 06:53 oddp joined, antoniogamiz joined, zacts left, zacts joined 06:57 Kaeipi left 07:00 Kaiepi joined, wamba joined 07:10 Guest98635 left 07:11 rbt left, softmoth left, rbt joined 07:16 Guest98635 joined 07:19 nadim joined 07:20 nadim is now known as Guest14298 07:33 Guest14298 left 07:45 tejr left 07:58 rbt left, rbt joined 07:59 tejr joined 08:02 sena_kun joined 08:03 Altai-man_ left 08:06 xinming left 08:07 xinming joined 08:13 JJMerelo joined 08:33 Sgeo left, sarna joined 08:38 antoniogamiz left 08:44 aborazmeh joined 08:46 pecastro joined 08:48 aborazmeh left 08:53 zacts left 08:54 zacts joined 09:12 JJMerelo left, aborazmeh joined, aborazmeh left, aborazmeh joined 09:24 aborazmeh left 09:26 antoniogamiz joined 09:51 sno joined 10:01 Altai-man_ joined 10:02 libertas left 10:03 sena_kun left
El_Che lizmat: you're a busy twitter bee :) 10:12
10:12 antoniogamiz left
lizmat yeah.. never waste an opportunity for more exposure 10:13
El_Che It seems to annoucement was mostly positively received 10:14
on the other hand, if it was a physical conf we would have seen some lynching ;)
10:17 riatre left 10:20 riatre joined 10:23 JJMerelo joined 10:32 Doc_Holliwould left, libertas joined 10:51 zacts left, zacts joined
Geth ecosystem: e5dc5c2246 | (Richard Hainsworth)++ (committed using GitHub Web editor) | META.list
Adding Pod::From::Cache to ecosystem

Pod::From::Cache replaces the functionality of Pod::To::Cache, removes unnecessary functionality, and has no other non-core modules.
11:01 tejr left 11:02 rindolf joined 11:04 vike left 11:05 xinming left 11:06 xinming joined 11:14 atta joined 11:18 kensanata joined 11:21 tejr joined 11:24 tejr left 11:25 tejr joined 11:27 antoniogamiz joined 11:44 patrickb joined 11:46 vike joined 11:49 tejr left
abraxxa andrewshitov.com/2020/06/06/some-t...s-in-raku/ is sadly a perfect example what's wrong with Rakus type system: from four types in Perl 5 to too much types which even overlap with the sigils that as available for some types but not all and even change what they do ;( 11:53
11:54 tejr joined 11:56 lichtkind joined
ShimmerFairy It seems that blog post focuses a lot on "look at these weird corner cases that can trip you up", and not enough on "it's because the system works like this, which isn't that complex" 12:00
12:01 sena_kun joined 12:03 Altai-man_ left
ShimmerFairy (though I will say that having sigils beyond $ is way less useful than I presume they are in Perl 5) 12:03
12:11 rbt left, rbt joined 12:12 brtastic joined 12:14 antoniogamiz left
jnthn I'm not really sure what point that article is trying to make. It's pretty simple really: assignment is a copying operation into the target on the left, so the thing you end up with is always the type of the thing on the left. 12:17
I think the problems are largely that folks thing assignment is *replacing* the thing on the left with the thing on the right. And until that distinction is understood, all kinds will be confusing.
s/thing/think/ 12:18
So assign into $ or & means store the thing on the right into the value of the scalar container, assignment into @ means iterate the thing on the right and store each value it produces into the array, and assignment into % means iterate the thing on the right and find pairs or take the things pair-wise 12:20
12:28 JJMerelo left
ShimmerFairy I still have quite a bit of trouble with arrays and hashes honestly, though I wonder how much of that comes from the fact that I used Raku pre-GLR. 12:29
12:34 chloekek joined 12:39 rindolf left 12:45 Kaiepi left 12:48 Kaiepi joined 12:50 chloekek left
abraxxa but it answers exactly those questions that come up in my mind every time I touch Raku code 12:56
even knowing the difference between a %hash variable holding an object of type Hash directly vs. a scalar container pointing to a Hash object doesn't help
I *think* I would have preferred to abandon the sigils at all 12:57
and you can code Raku without them, but it might have some drawbacks and the code style clashes with people that prefer them 12:58
jnthn = vs := is also hard to grasp 12:59
lizmat personally, I have been using the syntax "my \foo := bar"
where the := is currently not needed
I hope a future language version would allow you to drop the \ in that case
abraxxa I say \ in an article yesterday for the very first time
lizmat which would allow you to program sigilless if you want to 13:00
jnthn It won't because the \ is there to solve an ambiguity
lizmat which ambiguity ?
jnthn With a type name
El_Che well, if people found $@%& ugly, wait until they hear about \ 13:01
lizmat you mean, in the case "my int foo := bar" ?
El_Che ;)
jnthn lizmat: Well, just so far as `my foo` - is `foo` a type name or a symbol to be introduced?
lizmat if we would go for allowing that, it would mean a new symbol 13:02
13:02 Kaeipi joined
abraxxa why not use $foo? 13:02
lizmat agree you have some potential issue there
abraxxa because of the additional container memory usage? 13:03
and I thought the language design phase is over after 20 years...
to me this is bikeshedding
lizmat no, the language continues to evolve
but in a controlled way :-)
abraxxa the languages that have a character in front of variable names use $
jnthn < jnthn> It won't <== pretty clear, I think ;)
abraxxa end of discussion 13:04
lizmat [14:57:28] <abraxxa>I *think* I would have preferred to abandon the sigils at all
abraxxa what do they add but confusion and more ways to do basic things? 13:05
when everything is an object, $var holds an object, which one doesn't matter
lizmat ah, so you want to abandon @ and % ?
abraxxa if you want to know you call isa on it
lizmat I see... well, you can write your code like that, so go for it! 13:06
13:06 Kaiepi left
abraxxa that would at least stop the question which one you can use because both work on assignment but are different later one 13:06
that confuses starters completely
lizmat my $list = (1,2,3,4); for $list { } # how many times will this iterate ?
abraxxa i guess once 13:07
but this is where the problem starts
lizmat is that what you want ?
abraxxa either an array has a sigil and is NOT an object but something special, but not both
lizmat that's why we have the @ sigil :-)
in Raku, *everything* is an object
abraxxa yes, but Array and Hash have a sigil, the others don't
lizmat even native variables are, at some place 13:08
which others ?
abraxxa how do other languages solve this problem?
jnthn There's no direct connection between Array and Hash and sigils, other than they are the default container type for @ and %
Any other Positional or Associative type can be bound to an @ or %
my %h is BagHash # for example 13:09
lizmat m: my @a is List = 1,2,3; @a[0] = 42
camelia Cannot modify an immutable List ((1 2 3))
in block <unit> at <tmp> line 1
jnthn Or that
abraxxa the missing thing for me is the guideline when to use what and why
lizmat so we have a documentation issue?
abraxxa lizmat: what is a List? Did we *really* need another Array type?
and there are so many of them, and each call returns a different one! 13:10
synthmeat wasn't it for the laziness?
ShimmerFairy The non-$ sigils are the kind of thing I'd suggest maybe removing in a hypothetical Raku 2, but it's not big enough to worry about changing now.
lizmat m: say (1,2,3).WHAT
camelia (List)
ShimmerFairy List is immutable, Array is mutable.
abraxxa sometimes you need to cast, sometimes not
lizmat ShimmerFairy: wish it were so simple
abraxxa ShimmerFairy: and the names don't tell that difference
I'd like to be able to flag each var or maybe type with a read-only bit
lizmat m: my ($a,$b) = 42, 66; # the LHS is a List, but not immutable :-) 13:11
camelia ( no output )
ShimmerFairy well, the List itself is immutable, the contents (annoyingly) aren't :)
jnthn lizmat: Wrong, the List *is* immutable, the things inside it are mutable :)
abraxxa the things outsiders critizise about Perl 5 shouldn't be amplified in Raku like List vs Array 13:12
lizmat to me, saying something is immutable, implies it's a Value Type. Lists are not
ShimmerFairy m: class A { has @.b }; my $x = A.new; $x.b.push(5); say $x.perl; # I really wish you needed 'is rw' like with scalars
camelia A.new(b => [5])
lizmat m: my @a; sub a(@b) { @b.push(42) }; dd @a 13:13
camelia Array @a = []
lizmat hmmm
13:13 Kaeipi left
El_Che I must confess that I have been bitten quite few times needing to cast stuff between array, list and seq 13:13
adding ".list" has kind of a cargo cult + trial and error vibe in my usage 13:14
ShimmerFairy I've never needed to "cast" list types as far as I can recall, they just convert as needed.
El_Che the sigils are not problematic for me though 13:15
ShimmerFairy Same, if there's anything wrong with with lists and hashes, it ain't the sigils :) 13:16
El_Che I must say that sigils feel natural to me (knowing Perl), the \ does not 13:17
13:18 soursBot joined
lizmat even worse, the syntax for creating a Capture 13:19
m: dd \(1,2,3, :foo)
camelia \(1, 2, 3, :foo)
lizmat m: dd (\(1,2,3, :foo)).WHAT
camelia Capture
El_Che in my head: escaping regex? something utf8? a reference? a pointer?
lizmat yeah, I would be in favour of disallowing that
and just use: 13:20
El_Che while I know what it does, but my mind hits the breaks when I skim code that uses it
lizmat m: dd \(1,2,3, :foo).Capture
camelia \(1, 2, 3, :foo)
lizmat perhaps I should start with fixing the .raku representation :-)
jnthn What's broken about it? \(...) is the capture literal syntax, just as :(...) is the signature literal syntax 13:24
And calling .Capture on a list ain't the same, 'cus of the pair handling being different 13:25
ShimmerFairy Maybe it's that a backslash usually means you're escaping something? 13:26
jnthn Not really. Is \w in a regex escaping the w? Nope. :) 13:28
El_Che jnthn: what I mean is that \ is already heavily overloaded
thus it makes it difficult to skim code for me
ShimmerFairy well, you're, um, ESCAPING the usual meaning of 'w' in that context.
El_Che mm, indeed 13:29
jnthn Sure, and \(...) is ESCAPING the usual meaning of parentheses being just a grouping ;)
El_Che mmm
jnthn And don't forget [\+], where you're escaping the reduction operator throwing away the intermediate results
ShimmerFairy Related question: when would I, an ordinary Raku user, ever need to create a Capture or Signature? 13:30
jnthn ShimmerFairy: I think about 90% of my doing that is when using Test::Mock, 13:31
ShimmerFairy that would make sense
jnthn Where it's really quite useful 13:32
lizmat jnthn: re different Pair handling, I'm not seeing it ?
m: dd \(1,2,3,:foo)
camelia \(1, 2, 3, :foo)
lizmat dd \(1,2,3, foo => True) 13:33
evalable6 \(1, 2, 3, :foo)
lizmat m: dd (1,2,3,:foo).Capture
camelia \(1, 2, 3, :foo)
lizmat m: dd (1,2,3, foo => True).Capture
camelia \(1, 2, 3, :foo)
jnthn m: sub foo($x) { dd $x }; my $c = \((a => 42)); foo(|$c) 13:34
camelia :a(42)
jnthn m: sub foo($x) { dd $x }; my $c = ((a => 42)).Capture; foo(|$c)
camelia Too few positionals passed; expected 1 argument but got 0
in sub foo at <tmp> line 1
in block <unit> at <tmp> line 1
jnthn Uff, JITty segfault...this'll be fun 13:35
lizmat jnthn: ok, gotcha 13:36
13:38 antoniogamiz joined
El_Che ok, now it looks you're all speaking in tongues :P 13:45
13:48 sarna left
[Coke] m: :P 13:56
camelia WARNINGS for <tmp>:
Useless use of ":P" in sink context (line 1)
[Coke] m: say :P
camelia Unexpected named argument 'P' passed
in block <unit> at <tmp> line 1
[Coke] m: dd :P 13:57
camelia :P
[Coke] there we go. :)
13:57 softmoth joined 13:58 lucasb joined 14:01 Altai-man_ joined, zacts left 14:03 sena_kun left 14:04 xinming left 14:05 xinming joined 14:09 Doc_Holliwould joined 14:15 dataangel left, dataangel joined 14:16 antoniogamiz left 14:21 b2gills left 14:23 Boc left, orinthe7 joined 14:25 orinthe left, orinthe7 is now known as orinthe
rypervenche I'm trying to get this .map to work, but because I've added an if/else statement, I'm having trouble assigning the output to a variable, since it gives "Useless use of "=>" in expression "$num1 => set @num2" in sink context". 14:33
If I add "say" to line 25 it outputs the desired output, but I need that to go into the variable, not to stdout. 14:34
14:34 orinthe9 joined
lizmat rypervenche: what to you want to have in the map? just :num(set @num2) ? 14:35
14:35 orinthe left, orinthe9 is now known as orinthe
rypervenche In the resulting hash, yes that's what I would like. 14:35
14:36 Boc joined
rypervenche Errr, an example would be: 9831 => Set(189112 2170497 2170512) 14:36
lizmat you need to store the result of $num1 => set @num2 into a variable 14:37
either explicitely or implicitely
that's also what the warning is telling you 14:38
so something like: my \result = $num1 => set @num2; .....; result
rypervenche Ahh, I'll give that a shot then. Thanks.
lizmat I guess in the "if" case, you don't want anything added to the map, right ? 14:39
14:39 rindolf joined
lizmat also, I don't you need the outer %( ... ) 14:40
14:40 xinming left
lizmat actuallly, hang on... 14:40
14:40 skids joined
lizmat do you want the %links to be a Set of Pairs ? 14:41
14:41 JJMerelo joined
rypervenche I want it to be a hash of sets. The source file looks like this: gist.github.com/rypervenche/a1dfb2...7cf3cc1f2d 14:42
lizmat ah, yeah, duh...
lizmat has done too much social media today :-)
14:42 xinming joined
rypervenche Looking to make the hash key the first number, and then the set all of the numbers that go along with that number. 14:42
lizmat gotcha 14:43
14:43 Doc_Holliwould left 14:47 antoniogamiz joined, antoniogamiz left 14:55 Xliff joined, Kaiepi joined
Xliff \o 14:55
.tell vrurg Was R#3075 ever fixed? Looks like both it and R#3284 are still open 14:56
linkable6 R#3075 [open]: github.com/rakudo/rakudo/issues/3075 [precompilation] Conflicting use statement does not produce an error, but fails dynamic compilation.
R#3284 [open]: github.com/rakudo/rakudo/pull/3284 [precompilation] Fix Stash'es clashing after deserialization
tellable6 Xliff, I'll pass your message to vrurg
14:56 Xliff left 14:57 zacts joined
vrurg Xliff: no. I'm having a lightning talk soon, will elaborate later. 14:57
tellable6 vrurg, I'll pass your message to Xliff
rypervenche lizmat: Still not sure how to do that. Assigning the output of what I want to a variable and then calling it inside the else loop still throws the sink context error. 15:03
lizmat I have a file called "ten" that contains 0..9 as words on separate lines 15:05
m: my %h = "ten".IO.lines.map: -> $line { a => $line if $line eq "zero" }; dd %h # does that explain ? 15:06
camelia Failed to open file /home/camelia/ten: No such file or directory
in block <unit> at <tmp> line 1
lizmat Hash %h = {:a("zero")} # output for me :-)
15:16 b2gills joined 15:17 sno left 15:21 MilkmanDan joined
JJMerelo Just tallyed up again the number of lines the documentation includes. Some time in the past weeks, we went over 100K. 15:22
Here's the breakdown per contributor github.com/JJ/TPF-Grant/blob/maste...ritten.csv
15:23 sno joined
lizmat JJMerelo++ 15:25
15:25 zacts left 15:27 Kaiepi left
[Coke] wonders if he is the 'will' on the list. 15:29
JJMerelo [Coke] yep
Parsing of contributors is not perfect.
It's done by "blame"ing every single file, and adding up the lines of code in every fragment. 15:31
It's not really what every one has written, but how many lines written (or otherwise touched) by someone are in the current state of the repo. 15:32
Here's the presentation, BTW. jj.github.io/doccing-raku/#/6 15:34
I'd say it's 99% ready. Might change something from here to presentation time in 90 minutes.
rypervenche lizmat: Hmmm, I altered your little one-liner to look a bit more like my hash, and I don't understand why yours works and mine has the sink issue: gist.github.com/rypervenche/5f5fc3...c92d64b95c 15:35
lizmat rypervenche: because it is about the *last* value evaluated in a block
so you'll always get $last-num into the hash 15:36
rypervenche Oh!
15:37 soursBot left 15:39 kensanata left
rypervenche Now I understand. Thanks so much :) 15:41
lizmat weekly: github.com/JJ/TPF-Grant/blob/maste...ritten.csv
notable6 lizmat, Noted! (weekly)
lizmat rypervenche: you're welcome!
15:45 __jrjsmrtn__ joined 15:46 _jrjsmrtn left 15:50 Kaiepi joined 15:54 chloekek joined, xinming left 15:55 xinming joined 15:56 Nile joined, molaf joined 15:57 Nile left 16:00 stoned75 joined 16:01 sena_kun joined 16:03 xinming left, xinming joined, Altai-man_ left, chloekek left
Geth ecosystem: 3e39a32626 | (Richard Hainsworth)++ (committed using GitHub Web editor) | META.list
removing Pod::From::Cache until tests pass
16:14 cpan-raku left 16:16 cpan-raku joined, cpan-raku left, cpan-raku joined 16:17 soursBot joined 16:22 dolmen joined 16:23 chloekek joined 16:27 antoniogamiz joined
antoniogamiz is there a way to detect is raku is being run from cmd or Powershell? 16:27
sena_kun antoniogamiz, you mean the platform or? 16:29
antoniogamiz no, I mean the terminal 16:30
16:30 xinming left
antoniogamiz I want to delete a file, but in powershell I can use 'rm', but in cmd I have to use 'rd' 16:30
16:30 guifa left, xinming joined
antoniogamiz So I need a way to detect what terminal is being used to run the script 16:30
16:32 rindolf left
sena_kun antoniogamiz, can't you call unlink? 16:34
16:34 Kaiepi left
sena_kun docs.raku.org/routine/unlink <- and it'll do the thing for you. 16:34
antoniogamiz oh great, raku always has what I need :) thanks! 16:35
16:37 sno left 16:41 Doc_Holliwood joined, orinthe5 joined, guifa joined 16:43 orinthe left, orinthe5 is now known as orinthe
JJMerelo antoniogamiz there's also a test for running in Windows, is-win 16:43
antoniogamiz JJMerelo, I know, but I needed to differentiate terminals
unlink makes the trick 16:44
JJMerelo antoniogamiz that's not so easy... but if unlink does the trick...
16:45 finanalyst joined 16:46 guifa left
finanalyst I'm getting a '::' error in name from Test::Meta. I can suppress the error using :relaxed-name. But why am I getting the error at all? The only name I have with :: is in the provides and refers to the module being provided 16:49
JJMerelo The name of the module needs to include :: 16:51
finanalyst JJMerelo, so why am I getting this error: name has a '::' rather than a hyphen (if this is intentional please pass :relaxed-name to meta-ok) 16:53
16:53 soursBot left
antoniogamiz is there a way to make compare paths in a os-independent way? Something like "a\b".IO eq "a/b".IO => True 16:55
lizmat m: "a\\b".IO eqv "a/b".IO 16:56
camelia WARNINGS for <tmp>:
Useless use of "eqv" in expression ".IO eqv \"a/b\".IO" in sink context (line 1)
lizmat m: dd "a\\b".IO eqv "a/b".IO
camelia Bool::False
16:56 kensanata joined
antoniogamiz lizmat: so it cannot be done? 16:58
lizmat not immediately 17:00
it's a good question :-) 17:01
antoniogamiz I will compare all parts then :D
or replace \\ by /
lizmat yeah, feels like a thing to do 17:02
17:10 sno joined 17:11 rbt left, rbt joined 17:15 OpenZen joined 17:19 MasterDuke left
patrickb wrt distinguishing powershell and cmd: just for reference: github.com/Raku/App-Rakubrew/blob/...ell.pm#L31 17:21
Geth ecosystem/finanalyst-patch-2: f3e0715386 | (Richard Hainsworth)++ (committed using GitHub Web editor) | META.list
Adding Pod::From::Cache to the Ecosystem

Test::META test passing with :relaxed-name
ecosystem: finanalyst++ created pull request #511:
Adding Pod::From::Cache to the Ecosystem
17:27 dolmen left 17:31 soursBot joined 17:34 chloekek left 17:40 reach_satori left, reach_satori joined, antoniogamiz_ joined 17:42 Doc_Holliwood left 17:43 antoniogamiz left 17:49 rindolf joined 17:50 Guest14298 joined 17:53 antoniogamiz_ left 17:59 JJMerelo left 18:00 Doc_Holliwood joined 18:01 Altai-man_ joined, zacts joined, zacts left 18:03 sena_kun left 18:05 rjbs left 18:06 rjbs joined, rjbs left, rjbs joined 18:17 JJMerelo joined 18:22 rindolf left, Boc left 18:28 JJMerelo left 18:32 dolmen joined 18:33 Cabanossi left, Cabanossi joined 18:34 Sgeo joined 18:46 Guest14298 left 18:52 chloekek joined
Geth ecosystem/no-volatile-source-url: 3eaa0302b2 | (Patrick Böker)++ (committed using GitHub Web editor) | PULL_REQUEST_TEMPLATE.md
Add a check for volatile source-urls

Volatile `source-url`s will result in an inconsistent module state the users computer as soon as the contents of that source-url change without a matching change in the version number. Setting `source-url` to a git master branch is the typical faux-pas.
  *Instead* one should change the `source-url`to link to a never changing target (a tag, revision or some zip file).
19:01 kensanata left 19:04 soursBot left 19:05 soursBot joined 19:08 molaf left 19:15 rindolf joined
stoned75 hum... github.com/moose/Moose/search?l=raku 19:20
Geth ecosystem: patrickbkr++ created pull request #512:
Add a check for volatile source-urls
19:37 molaf joined 19:57 dolmen left 20:01 sena_kun joined 20:03 Altai-man_ left 20:07 rbt left 20:08 rbt joined 20:16 wamba left 20:17 dolmen joined
softmoth stoned75, well spotted. I think a bug report at github/linguist is what's needed. I've never dug into their code: github.com/github/linguist 20:18
20:19 brtastic left 20:23 wamba joined 20:30 kensanata joined 20:32 soursBot left 20:35 finsternis joined 20:40 xinming left 20:41 xinming joined 20:52 rindolf left 20:58 wamba left
kensanata Returning after a long time I uninstalled all the Perl6 stuff I had and installed rakubrew. So far so good. Then I tried zef install --/test cro and got "Enabled fetching backends [git path curl wget] don't understand www.cpan.org/authors/id/J/JN/JNTHN/....3.tar.gz" and something about configuring "one of the following backends" [pswebrequest] – that does that mean? 21:02
timotimo pswebrequest is, i think, based on powershell's built-ins? 21:07
but i think the actual problem is you don't have a tar installed?
[Coke] kensanata: what platform are you on? 21:08
kensanata [Coke]: Some Debian variant (PureOS) 21:10
timotimo: tar --version says I have 1.30
[Coke]: That is, I did a sudo apt install rakubrew and went from there. 21:11
Let me check what I did.
No, I did cpanm App::Rakubrew because apt only had rakudo stuff from 2018 21:13
21:14 aborazmeh joined, aborazmeh left, aborazmeh joined 21:30 softmoth left 21:33 softmoth joined
kensanata There used to be a prove6 binary, which I still refer to in my Makefile. What's it called, now? 21:39
21:40 softmoth left 21:54 kensanata left 22:01 Altai-man_ joined 22:03 sena_kun left 22:11 rbt left, rbt joined 22:12 nebuchadnezzar left, Celelibi left
timotimo could it be t/harness6? 22:13
or installed with TAP::Harness from the ecosystem 22:14
leont App::prove6 22:19
(which uses TAP::Harness) 22:20
22:22 Celelibi joined
leont I think someone is looking for raku github.com/Perl/perl5/issues/17886 22:24
raku-bridge <DataKinds> just finished the post that was WIP yesterday datakinds.github.io/2020/06/25/exp...-itertools 22:25
leont Yeah, some of those really do need an explanation 22:36
But half of that is that non-python programmers have no idea what some of these functions are supposed to do 22:38
22:39 Black_Ribbon joined
AlexDaniel kensanata: if you're on a debian derivative, any reason you can't use rakudo that is packaged in the repos? 22:40
tellable6 AlexDaniel, I'll pass your message to kensanata
AlexDaniel kensanata: ah, I've read your last message. Rakudo 2018.12 is not bad at all, but you can easily have 2020.05.1 from unstable 22:42
tellable6 AlexDaniel, I'll pass your message to kensanata
22:48 lucasb left 22:49 aborazmeh left 22:54 finanalyst left 22:58 skids left 23:04 dolmen left
Geth whateverable: 152bd39da6 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 21 files
Fix %*BOT-ENV situation

By using PROCESS variables instead. Dynamic variables shouldn't be used in code where start blocks are common. See
There's probably a better way to achieve this, but for now this will do.
whateverable: 4fea789613 | (Aleks-Daniel Jakimenko-Aleksejev)++ | xbin/Statisfiable.p6
Rework Statisfiable

It's an old bot so the code was a bit dated too. This should get it back on track.
23:10 pecastro left 23:16 chloekek left 23:30 Maylay left 23:32 finanalyst joined 23:33 Cabanossi left, maggotbrain left 23:34 Maylay joined
Geth whateverable: cf9bcd03b1 | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/Whateverable/Configurable.pm6
Can't use binding here (oops)
23:35 nebuchadnezzar joined 23:42 Cabanossi joined 23:47 patrickb left 23:54 MilkmanDan left 23:58 MilkmanDan joined