00:36 arkiuat left, zetaaaa joined 00:42 melezhik_ joined 00:44 melezhik_ left 00:45 arkiuat joined 00:49 arkiuat left 01:02 zetaaaa left 01:04 arkiuat joined 01:09 ds7832 left 01:27 zetaaaa joined 01:30 arkiuat left 01:31 arkiuat joined
aruniecrisps just dipped my toes into Air @librasteve , I'm loving it so far. I'm just using Air::Functional for right now because I'm currently using Daisy UI alongside tailwind 01:31
01:58 arkiuat left 02:10 arkiuat joined 02:18 arkiuat left 02:31 arkiuat joined
[Coke] m: say Capture.^mro 02:33
camelia ((Capture) (Any) (Mu))
02:35 arkiuat left 02:45 melezhik_ joined 02:47 melezhik_ left, arkiuat joined 02:51 zetaaaa left 02:52 arkiuat left 03:04 arkiuat joined 03:08 arkiuat left 03:15 arkiuat joined
aruniecrisps i do wonder if there is a way to add <!DOCTYPE html> using Air::Functional 03:18
03:25 vasko4535 left, vasko4535 joined 04:34 kjp joined 04:36 kjp left, kjp joined 05:29 Aedil joined 05:49 kylese joined, kylese left 05:50 kylese joined 06:42 melezhik joined 06:43 arkiuat left 07:00 mimosa joined 07:03 justache left 07:08 wayland76 joined 07:12 arkiuat joined 07:17 arkiuat left 07:24 justache joined 07:28 silug left, silug5 joined 07:29 Sgeo left 07:44 arkiuat joined 07:49 arkiuat left 07:50 Aedil left 07:55 Aedil joined
librasteve in Air that's spelled say (page main p 'yo baby!').HTML 08:11
the base library has things like page and site so you need the incantation use Air::Functional :BASE; use Air::Base; to start 08:12
@aruniecrisps ^^ 08:13
08:17 arkiuat joined 08:21 librasteve_ joined
librasteve_ . 08:21
www.irccloud.com/pastebin/YNt8mtWn 08:22
08:26 arkiuat left
librasteve ^^ this should work (this is an older pre Air::Base example code untested for a bit, but hopefully it'll be easy to get it to work) ... so arguably Air::Functional needs one small additional - a routine called err page (so have to be suppressed in :BASE) that goes sub page(HTML $html) is export { '<!doctype html>' ~ $html } 08:28
I made an issue for it github.com/librasteve/Air/issues/38 ... feel free to make a PR (sorry I am a bit preoccuied with bits of Air on the workbench all around me atm) 08:30
wayland76 For those interested in the Raku keycaps I've been working on, I have a screenshot here: wayland.github.io/table-oriented-p...enshot.png 08:35
Key -- Red = Set theory/Relational Algebra (including tables :p ). Blue = the operators I want to use for trees. Purple = matched brackets of various kinds, including the ones asked about (on 9/0). Green = various others, some of which we use. 08:37
I'm still missing function composition, but hopefully I have the rest. 08:38
(not all of the relational algebra ones are implemented in Raku, but I'm hoping to get to them with table-oriented programming at some point, for the joins, etc) 08:39
08:47 LainIwakura joined 08:51 dakkar joined, LainIwakura88 joined 08:53 LainIwakura left 08:58 arkiuat joined 09:02 LainIwakura joined 09:04 arkiuat left 09:23 arkiuat joined 09:28 arkiuat left 09:39 mimosa left 09:40 arkiuat joined 09:43 LainIwakura88 left, LainIwakura left 09:44 arkiuat left 09:50 melezhik_ joined, apa_c joined 09:55 melezhik_ left
wayland76 Looks like I killed the conversation :p 09:56
09:56 mimosa joined 09:57 mimosa left 09:59 mimosa joined 10:07 ds7832 joined 10:09 arkiuat joined 10:12 ano__ joined
ano__ hello~ 10:13
I just pushed an update to `rirc` that allows to switch channels (so have a simili tab for each channels)
its probably buggy, if I can get any feedback, would be great
10:14 arkiuat left
ano__ the code is probably not very pretty as well xD I try to do what I can with the time I have :pray: 10:14
10:14 apa_c left 10:15 apa_c joined, ano__ left 10:21 mimosa_ joined 10:22 apa_c left
simon_sibl looks like this so far 10:23
cdn.discordapp.com/attachments/633...d85cf&
10:23 mimosa left
librasteve @wayland76 - that look's err interesting 0 guess I had something like this in mind www.crowdsupply.com/summa-cogni/mathpad that I can jack into my macbook usd c 10:27
10:27 apa_c joined
hmmm that's $190 US for a keypad - bit too pricey for me - maybe I can just get a standard keypad and some custom keycaps? 10:31
10:37 ds7832 left 10:40 arkiuat joined 10:45 arkiuat left 10:58 LainIwakura joined, LainIwakura3 joined 11:13 arkiuat joined 11:14 ds7832 joined 11:18 arkiuat left
hi folks - does RAKUDO_RAKUAST=1 engage the AST parser even if I go use 6.d? (or do I need to change to use 6.e.* to actually engage it) 11:22
11:25 arkiuat joined
lizmat RAKUDO_RAKUAST=1 will engage the new grammar, for all language versions 11:29
11:30 arkiuat left
librasteve cool - works with Air ;-) ... mebbe I shoukd turn on all the time 11:30
lizmat adding use 6.e.PREVIEW or use 6.* will only make the RakuAST:: classes visible (without needing to use "use experimental :rakuast")
librasteve =b 11:31
lizmat librasteve making it default, will sure help finding any still hidden issues :-)
11:36 apa_c left 11:47 LainIwakura left, LainIwakura3 left
wayland76 librasteve_: Well, these custom keycaps are looking to be $99 for the set. I'm thinking I get a second keyboard just for typing these. Oh, I also forgot to mention -- the Greek keyboard layout is the standard Greek layout from xkb, which will make it easier for me to design a layout 11:47
But the plan is definitely that I have this as a second keyboard, and only reach across to it when I need it. Anything I use regularly I'll incorporate into my regular setup. 11:48
Their two-tone keys are interesting, but I'm not sure I want to do that for now. 11:51
librasteve since I use Intellij+RIP, I have installed the Spec & Math Symbols plugin go I get autocomplete is I start typing eg symbol...union 11:52
wayland76 Interesting! I'd rather not be tied to one editor :) 11:53
librasteve just found Greek Symbol plugin also --- well these are pretty generic issues, so I guess that eg vi has code completion plugin has symbol lib (or you dould write it) 11:55
etc
btw did you notice the recent request on the stalled doc issue ? 11:56
11:58 arkiuat joined
github.com/Raku/doc-website/issues/456 11:58
12:03 arkiuat left
(hmm maybe this is the wrong issue - dang can't find the one - ahh - here it is github.com/Raku/doc/pull/4570) 12:06
actually a stalled PR, sorry
12:08 apa_c joined 12:12 melezhik left 12:18 anoano___ joined, anoano___ is now known as rirc-test 12:20 arkiuat joined, rirc-test left 12:25 arkiuat left 12:26 arkiuat joined 12:31 arkiuat left 12:41 melezhik_ joined 12:43 LainIwakura joined, LainIwakura58 joined
antononcube wayland76 "I'd rather not be tied to one editor :)" -- I am glad to hear that you are not necessarily a vi addict. 12:45
12:46 melezhik joined 12:47 melezhik_ left 12:58 melezhik_ joined 13:01 arkiuat joined 13:04 melezhik_ left 13:06 arkiuat left 13:10 melezhik_ joined, melezhik_ is now known as melezhik_3 13:12 melezhik_ joined 13:16 melezhik_3 left 13:17 melezhik_ left 13:18 melezhik_ joined, melezhik_ is now known as melezhik___1 13:20 melezhik_ joined 13:23 melezhik_ left, melezhik___1 left 13:25 apa_c left, apac joined 13:27 arkiuat joined, Aedil left 13:32 arkiuat left 13:43 ds7832 left 13:58 ds7832 joined 13:59 arkiuat joined 14:00 Rakuists joined 14:02 Rakuists left 14:03 Rakuists joined 14:04 arkiuat left 14:07 Rakuists left 14:08 melezhik_ joined 14:09 apac left
ds7832 regarding keyboards: personally I find the special key layers of the Neo2 layout extremely convenient for Raku. Neo2 is optimized for German, but there's also a QWERTZ and (on Linux) a QWERTY version available, which gives one the special key layers without the need to adapt to different letter positions. Chart available here, at the bottom (see esp. layers 3 and 6): neo-layout.org/Layouts/neoqwertz/ 14:09
14:11 hulk joined 14:12 kylese left 14:14 melezhik_ left, LainIwakura left 14:16 arkiuat joined
jubilatious1_98524 @wayland76 I've often wondered why--over the years--no one has seemed to come up with a Perl keyboard layout. Something like Vi/Vim but with option-HJKL mapped to $ @ % & 14:17
Here's a related Github-repo for "Perl6": github.com/donaldh/Perl6-macOS-keyboard 14:18
14:21 arkiuat left
timo is also a big fan of neo2 14:22
ds7832 jubilatious1_98524: good point with the $ @ % & (and I would add #). 14:24
I've just added some remappings to my helix config to make $ @ % easier to type there. Shift-1 is now %, Shift-2 is now $, and Shift-3 is now @. (The default symbols on those keys (Ā° Ā§ ā„“) are barely ever needed when programming.) 14:27
14:29 arkiuat joined
[Coke] (keyboards) anyone want a keyboardio keyboard? I got one off the kickstarter and never use it 14:32
jubilatious1_98524 @ds7832 Thanks! FYI I find myself using `*
14:34 arkiuat left 14:38 arkiuat joined 14:43 arkiuat left 14:57 arkiuat joined
Voldenet I'm a fan of en-US layout everywhere, because a lot of special characters are usable without Shift even 14:57
14:57 oodani left
Voldenet though () are a lot more useful than [] and they do require that additional combo, which is unusual 14:58
{} are more useful too, I wonder why square brackets were preferred 14:59
because I can't imagine [] being used more than () in regular text either 15:00
15:00 oodani joined
Voldenet in Poland hardly anyone uses polish layout anymore, it used to be default in windows 20 years ago, but everyone used US keycaps, so defaults were changed quickly 15:04
wait, it used to be default 30 years ago uh
15:10 apac joined
lizmat remembers Dutch IBM keyboard layout and shudders 15:12
if I remember correctly {} where hidden under an shoft-alt combination key 15:13
*shift
15:15 Aedil joined 15:26 LainIwakura joined
Voldenet it could be even worse - some variants of japanese keyboards type ļ½›ļ½instead of {} as default 15:31
15:33 Rakuists joined
Voldenet (of course, there is regular mode which uses US layout, but it's still very weird) 15:34
15:34 Rakuists left 15:35 Rakuists joined 15:36 Rakuists left 15:41 LainIwakura58 left 15:52 apac left 16:04 LainIwakura12 joined 16:23 ds7832 left
[Coke] m: my %a; %aļ½›"a"ļ½= 3; dd %a; 16:23
camelia ===SORRY!=== Error while compiling <tmp>
Bogus postfix
at <tmp>:1
------> my %a; %a<HERE>ļ½›"a"ļ½= 3; dd %a;
expecting any of:
infix
infix stopper
statement end
statement modifier
statemenā€¦
[Coke] I'm half-surprised that didn't already work. :)
16:46 LainIwakura12 left, LainIwakura left 16:51 ds7832 joined
librasteve use the dot [Coke] 17:26
m: `my %a; %a .{"a"}= 3; dd %a; 17:27
evalable6 (exit code 1) ===SORRY!=== Error while compilinā€¦
librasteve, Full output: gist.github.com/55c920e3c404d36c94...06a3417425
librasteve m: my %a; %a .{"a"}= 3; dd %a;
evalable6 (exit code 1) ===SORRY!=== Error while compiling /tmp/_E2Y3e67XA
Malformed postfix call (only basic method calls that exclusively use a dot can be detached)
at /tmp/_E2Y3e67XA:1
------> my %a; %a .<HERE>{"a"}= 3; dd %a;
librasteve oh
lizmat basically the same reason why: 17:28
Voldenet nah, special brackets don't work either way, but that's fine :D
lizmat m: my %a; %a <foo> = 42 # doesn't work
camelia ===SORRY!=== Error while compiling <tmp>
Preceding context expects a term, but found infix = instead.
Did you make a mistake in Pod syntax?
at <tmp>:1
------> my %a; %a <foo> =<HERE> 42 # doesn't work
[Coke] That isn't a space before the bracket, it's a wide bracket from the previous line.
lizmat some constructs in Raku are whitespace sensitive
librasteve yeah I get it (now) 17:29
[Coke] what's the module that has a bunch of units builtin and (I think) a CLI for playing with ? 17:37
17:38 dakkar left 17:39 apac joined
jubilatious1_98524 @ds7832 Thanks! FYI I find myself using * much more than &, althought both could easily fit on the home row . [repeated post] 17:43
17:45 ds7832 left
m: my %a; %a.{"a"}= 3; dd %a; 17:48
evalable6 {:a(3)}
jubilatious1_98524 m: my %a; %a .{"a"}= 3; dd %a; 17:49
evalable6 (exit code 1) ===SORRY!=== Error while compilinā€¦
jubilatious1_98524, Full output: gist.github.com/f4fccf12b8d3d14e05...5363b58535
librasteve I thought that that would work too ;-( 17:50
17:51 mimosa_ left
lizmat m: my %a; %a\ .{"a"}= 3; dd %a; 17:52
camelia {:a(3)}
lizmat docs.raku.org/language/syntax#Unspace 17:53
jubilatious1_98524 m: my %a; for %a {say .{"a"} };
evalable6
jubilatious1_98524 m: my %a; %a{"a"} = 1; say .{"a"} for %a; 17:56
evalable6 1
jubilatious1_98524 ^^ maybe this.
18:10 arkiuat left 18:12 apac left 18:18 melezhik_ joined 18:23 arkiuat joined
Voldenet you're missing the point that it was not about spaces, in fact the example didn't have spaces before brackets ;P 18:26
I bet discord uses the same font for both characters which makes it extra confusing 18:27
18:28 arkiuat left
[Coke] ahhh 18:28
didn't even consider that
18:30 melezhik_ left, kylese joined, arkiuat joined, justache- joined 18:33 kjp_ joined 18:37 hulk left, justache left, kjp left, simcop2387 left, camelia left 18:39 simcop2387 joined 18:47 justache- is now known as justache, justache is now known as justache_ 18:48 MoC joined, justache_ is now known as justache
jubilatious1_98524 @Voldenet how could it "not (be) about spaces" if @lizmat 's unspace example corrects? 18:52
[Coke] My example had no spaces. 18:55
my example still didn't work
m: my %a; %aļ½›"a"ļ½= 3; dd %a;
evalable6 (exit code 1) ===SORRY!=== Error while compilinā€¦
[Coke], Full output: gist.github.com/48b1da030648dc6a75...5e3a1c4607
[Coke] m: "ļ½›".chars.say 18:56
evalable6 1
[Coke] m: "ļ½›".uninames.say
evalable6 (FULLWIDTH LEFT CURLY BRACKET)
[Coke] m: "{".uninames.say
evalable6 (exit code 1) ===SORRY!=== Error while compilinā€¦
[Coke], Full output: gist.github.com/0527a8f571483bca7d...71ebc53ed9
[Coke] m: '{'.uninames.say
evalable6 (LEFT CURLY BRACKET)
[Coke] So, that was my example. someone had mentioned the fullwidth as an example and I was just seeing if it already happened to work in the same context as the ascii one (and it doesn't) 18:57
cokebot9000 Yah, in discord, it looks like a space and a LEFT CURLY BRACKET - but you can highlight it and you'll see it's one character. on my irc font it's clearly a different character. 18:59
timo it looks like the exact same character in my terminal :| 19:00
[Coke] unicode hard. 19:04
Sorry about the confusion!
jubilatious1_98524 Okay, when I cut-and-paste @cokebot9000 's example into iTerm2 it errors. And adding a "." does not fix. I can clearly see the FULLWIDTH curlie there.
Voldenet m: sub a ļ½›"unicode is awesome" ļ½;a # :) 19:05
evalable6 (exit code 1) ===SORRY!=== Error while compiling /tmp/BC5Rp8nMOt
Missing block
at /tmp/BC5Rp8nMOt:1
------> sub a<HERE> ļ½›"unicode is awesome" ļ½;a # :)
expecting any of:
new name to be defined
timo you could use q with an opening regular curly and then put a fullwidth closing curly somewhere to make someone think that the string literal ends where it doesn't actually, or to have { } inside a qq string that are or are not actually code 19:14
Voldenet there are use cases for it, but I stand away from unicode that looks like ascii 19:17
m: ;
evalable6 19:18
Voldenet m: ;
evalable6
librasteve_ notable6: weekly
notable6 librasteve_, 2 notes: 2025-10-23T14:17:44Z <librasteve>: www.londonperlworkshop.com/ ; 2025-10-26T17:14:08Z <antononcube>: github.com/podlite/podlite-desktop...tag/v0.6.1
Voldenet huh, I didn't expect this to be working
m: ";".uniname.say
evalable6 SEMICOLON
jubilatious1_98524 Confusables: raku.land/zef:jjmerelo/Unicode::Security
Voldenet m: ; 19:19
evalable6
[Coke] if that's the greek one, yes, it's part of the encoding.
so by the time the code sees it, it's already a semicolon
Voldenet hm, that could be potentially security issue if you use raku to validate it
it'd be enormously shaky contraption though 19:20
[Coke] www.compart.com/en/unicode/U+037E - has a decomposition to U+003B, which I think we are respecting in our NFG 19:21
I assume anything with a decomposition would work, but haven't checked.
m: "Ļ²".uninames.say 19:22
evalable6 (GREEK LUNATE SIGMA SYMBOL)
[Coke] (nope, that one didn't get translated) 19:23
jubilatious1_98524 m: "(ā«·^Ļ¾ę„½Ļæ^ā«ø)".comb.map: *.uninames.say; 19:26
evalable6 (LEFT PARENTHESIS)
(TRIPLE NESTED LESS-THAN)
(Cā€¦
jubilatious1_98524, Full output: gist.github.com/87ffe8ec751ee76fe6...c48d6150b1
jubilatious1_98524 (above contains: GREEK CAPITAL DOTTED LUNATE SIGMA SYMBOL ) 19:27
19:28 camelia joined
[Coke] guessing this is in MVM_string_chr - turn codepoint to string 19:28
Voldenet hmā€¦ so if you try to to use grammars to parse binary string (I remember seeing such code in bencode), 0x037E would get turned into 0x003B 19:32
[Coke] m: "\c[GREEK QUESTION MARK]".uninames 19:34
is it binary or is it string? 19:35
if string, yes.
Raku has Blobs and Bufs, not binstrings.
m: "\c[GREEK QUESTION MARK]".uninames.say
did we lose camelia? 19:36
Voldenet but you probably can't use grammars on Buf/Blob
[Coke] (that prints (SEMICOLON), FYI) 19:37
timo indeed, though you can do the terrible hack of just stringifying the buf to hexadecimals separated by spaces (so you don't accidentally match the upper and lower nybble of two adjacent bytes)
camelia ( no output ) 19:38
(SEMICOLON)
lizmat m: say "hi" 19:39
camelia hi
librasteve_ rakudoweekly.blog/2025/10/27/2025-...-or-treat/
lizmat whee! librasteve___
whee! librasteve_++ rather :-)
19:42 abraxxa-home joined
Voldenet ah, right, matching hexadecimals would work - not elegant but would work 19:44
19:48 abraxxa-home left 19:49 abraxxa-home joined 20:03 apac joined 20:14 dmaestro joined
dmaestro I've been using the reduce function with a two-argument code block, and a single (init) value followed by a possibly empty list ($init, |@list) - this used to work with an empty list, but now, after upgrading to rakudo-star 2025.5 and later, does not. The documentation suggests a single value list should be accepted and return the value directly. 20:18
This works with built-in operators. Why did it stop working with user-supplied code blocks or functions?
Also, if someone could remind me how to run a code snippet here, it would be appreciated!Ā :b 20:27
Voldenet show code example so maybe bisect bot will reveal something
dmaestro Yes, I haven't been on IRC in ages, how do I quote or run code snippets? 20:28
Voldenet m: say "hi"
camelia hi
Voldenet but it's potentially this rakudoweekly.blog/2025/02/03/2025-...ie-awaits/
something about core development changes
dmaestro m: say <hi> 20:29
camelia hi
dmaestro m: say reduce &infix:<*>, 10, 20
camelia 200
dmaestro m: say reduce &infix:<*>, 10 20:30
camelia 10
dmaestro m: say reduce (* * *), 10, 20
lizmat m: say reduce &infix:<*>
camelia 200
1
dmaestro m: say reduce (* * *), 10
camelia Too few positionals passed; expected 2 arguments but got 1
in block <unit> at <tmp> line 1
lizmat dmaestro: the whatevercode block * * * expects 2 arguments 20:31
you're providing only one
bisectable6: old=2024.01 say reduce (* * *), 10
bisectable6 lizmat, Bisecting by exit code (old=2024.01 new=7117060). Old exit code: 0
lizmat the infix:<*> sub has special cases for 0 and 1 arg calls 20:33
* * * does not
dmaestro Right, so the "magic" that used to work for a single argument call no longer does? 20:34
lizmat yes, because there is no way that you can reliably guess what the correct behaviour is for user supplied Callables in that case 20:35
bisectable6 lizmat, bisect log: gist.github.com/894612d43e8e9ffc6f...7f292e5dc3 20:36
lizmat, There are 40 candidates for the first ā€œnewā€ revision. See the log for more details
lizmat if the documentation still suggest that a single value list should produce the single value, then that feels like a doc issue
dmaestro Sounds reasonable, just broke my code (which used to be pretty intuitive). Are there any simpler options (allowing inline Callable?) than creating a multi sub?
lizmat -> $a = 0, $b = 1 { $a * $b }
dmaestro Excellent! Thanks. 20:37
lizmat -> $a, $b? { $b.defined ?? $a * $b !! $a } 20:38
in case you just want to handle the 1 arg case
20:38 abraxxa-home left
dmaestro (y) 20:39
20:41 dmaestro left 20:42 apac left 21:11 apac joined 21:20 ds7832 joined 21:22 melezhik left 21:50 librasteve_ left, Aedil left 21:53 librasteve_ joined 22:06 wayland76 left 22:08 apac left
librasteve_ from the TPRC folksā€¦ www.irccloud.com/pastebin/XCRy15Sn 22:13
big thanks to all the rakuteers who helped to assemble our part of the fosdem devroom submission 22:14
please bear in mind that individual CFP is still open
weekly: www.irccloud.com/pastebin/XCRy15Sn 22:15
notable6 librasteve_, Noted! (weekly)
librasteve lol ... what about use v6.e.*; -> $a, $b? { // $b ?? $a * $b !! $a } 22:21
arkiuat so the zef troubles I was complaining about recently were my own fault: I had stupidly left RAKULIB set to . somewhere. ugexe has explained to me how this breaks stuff 22:24
22:25 LainIwakura joined 22:29 Sgeo joined 22:32 sftp left 22:35 MoC left
ds7832 wonders if that's also the reason for issues he's been having with zef not finding dependencies it previously installed 22:46
22:57 arkiuat left 22:59 arkiuat joined
arkiuat ds7832, it certainly could be 23:01
ugexe unlikely 23:03
that is almost certainly an issue with their environment / rakudos, which one is being used to invoke zef (and which rakudo zef would install those modules to), and which rakudo actually gets run when invoking raku directly
if zef doesnt find dependencies that were previous installed than your method of determining if they are installed is flawed (see previous comment) or you uninstalled them / nuked the repo 23:04
s/than/then/
ds7832 I should mention this is on a NixOS machine, which I don't really understand yet. My Rakudo build being compiled from source. 23:06
(the environment vars are indeed not set, though)
ugexe why do you think it isn't finding dependencies it previously installed? 23:09
ds7832 iirc, during the installation of some modules with many dependencies, zef would say it's installing the dependencies and report success of that, but then in the next step say it couldn't find that very module. I'm seeing now if I can reproduce this to give a log 23:12
ugexe it helps to be precise here... when you say during install do you mean during staging? installation is the last phase of `zef install .` which occurs after testing 23:13
staging happens before testing
also, are you sure it isn't zef that is finding it and not rakudo failing to find it when e.g. loading it for a test? 23:14
but yeah a log will be useful
23:16 LainIwakura left 23:18 sftp joined, sftp left, sftp joined
ds7832 this is a log illustrating it: pastebin.com/hZiF0DWX So a bit different from what I wrote above 23:20
yes, could really be the error is coming from Rakudo trying to load it 23:21
and it's definitely before the actual installation, apparently during staging. 23:33
ugexe what happens when you run `raku -e 'use File::Temp;' 23:38
ds7832 goes through without error 23:42
ugexe well, it is installed then. and the error you shown was indeed during staging, which is when raku precompiles it so it is a precompilation issue
not sure why that would happen though. the only thing i can guess would be because you are using rakudo star, and File::Temp is part of rakudo star, and somehow those things are causing this 23:43
if you wanted to experiment you could try `zef uninstall File::Temp` to get rid of whatever raku star installed, and then try to install again 23:46
ds7832 Ah! I think I remember suspecting that too at some point
and force installs of the precise "missing" module always resolved that issue
so doing that one by one, I eventually succeeded in installing e.g. Cro :) 23:47
23:48 guifa joined
aruniecrisps i'm wondering if there was a way to auto add files to the provides field in META6.json 23:49
ds7832 the problematic modules were always very "standard" ones, so I would suspect them to be ones included in Star.

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!