🦋 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.
Xliff Why is this being emitted? "No exception handler located for warn" 00:01
tellable6 hey Xliff, you have a message: gist.github.com/146d57d7ff58c32400...179dc5411e
Xliff And how can I fix it?
tony-o: Can you show me what you did? I can't seem to get it to work on my updated gist. Feel free to fork it and send me the new URL. 00:07
guifa2 [Coke]++ codesections++ 00:18
Xliff Why is this being emitted? "No exception handler located for warn" 00:26
And how can I fix it? 00:27
berber44 Hello, I'm reading Apocalypses and have a question. In one place it is written: "we will make the rule that a left curly that has whitespace in front of it will never be interpreted as a subscript in Perl 6". Is this rule still valid in Raku? 00:36
guifa2 berber44: yes
berber44 Thanks, guifa2
guifa2 well, unless you count the unspace, but I don' think that counts since unspacing is literally removing spacing :-) 00:37
[Coke]: is that behavior documented anywhere? Like....it's awesome, but still… 00:38
moon-child guifa2: unspacing? 00:39
guifa2 moon-child: no, putting a constant as a type for multi dispatch
m: multi sub foo (1 :$a) { say "one" }; multi sub foo ('a' :$a) { say "ay" }; foo 'a'; foo 1 00:40
camelia 5===SORRY!5===
Calling foo(Str) will never work with any of these multi signatures:
(Int :$a where { ... })
(Str :$a where { ... })
at <tmp>:1
------> 3; multi sub foo ('a' :$a) { say "ay" }; 7⏏5foo 'a'; foo 1
Calling f…
guifa2 weird it literally just worked for me haha
err done 00:41
moon-child guifa2: no I mean, what's unspacing?
guifa2 moon-child: unspacing is using a backslash and then you can follow it as many spaces as you want, as if it weren't there
my @a = 1,2,3,4; say @a\ [1]
evalable6 2
moon-child m: my $abc = 5; say $a\ bc;
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$a' is not declared
at <tmp>:1
------> 3my $abc = 5; say 7⏏5$a\ bc;
guifa2 moon-child: it always breaks an identifier though IIRC 00:42
moon-child ahh ok
tony-o you can use a MAIN and accept an array 00:58
tellable6 2021-01-05T20:11:00Z #raku <melezhik> tony-o I've found one issue with fez - github.com/tony-o/raku-fez/issues/9
eax is there an equivalent text2html in raku? as in perl5. Or will I just have to inline it? 01:49
Xliff eax: text2html? What module is that from? 02:05
tony-o: Can you show me what you did with the regex? Thanks. 02:07
SmokeMachine .tell xinming it seems the `.grep: { $min < .col and .col <= $max }` is not related to the Optimization (but I'm not sure yet...). but it seems something like this works: `.say for Bla.^all.grep: { $min > .col >= $min }` maybe it could help while it's not fixed 02:09
tellable6 SmokeMachine, I'll pass your message to xinming
eax @Xliff HTML::FromText 02:11
Xliff eax: Nope. Looks like you will have to Inline it. 02:17
eax no worries, I was mentally prepped, just wanted to give it another check. Thank you Sir 02:19
tib . 06:57
tellable6 2021-01-05T23:51:34Z #raku <melezhik> tib I've create a simple Sparrow plugin tom list sparky builds, it's built on go and does not require any dependencies . It's very simple but illustrate the idea - sparrowhub.io/plugin/sparky-db/0.000001
Xliff Which is faster? .^can or .^lookup? 09:14
MasterDuke benchable6: compare HEAD my $a = "foo"; my $b; my $c; for ^100_000 { $b = $a.^can("index"); $c = $a.^can("bar") } ||| my $x = "foo"; my $y; my $z; for ^100_000 { $y = $x.^lookup("index"); $z = $x.^lookup("bar") } 10:23
benchable6 MasterDuke, starting to benchmark the 1 given commit
MasterDuke, ¦HEAD: «Benchmark: ␤«timed out after 10 seconds»»
MasterDuke benchable6: compare HEAD my $a = "foo"; my $b; my $c; for ^1_000 { $b = $a.^can("index"); $c = $a.^can("bar") } ||| my $x = "foo"; my $y; my $z; for ^1_000 { $y = $x.^lookup("index"); $z = $x.^lookup("bar") } 10:24
benchable6 MasterDuke, starting to benchmark the 1 given commit
MasterDuke, gist.github.com/2e12a32826aaddca44...c02e57f138
MasterDuke Xliff: gist.github.com/2e12a32826aaddca44...c02e57f138 seems `.^lookup` is faster
Xliff MasterDuke++: Thanks. I'll have to save that bit you did for my own evil purposes! :) 10:26
MasterDuke: Do you have the lowdown on how to use sourceable?
MasterDuke i'm not very good at using sourceable6. AlexDaniel` is much better. but the wiki does have some documentation 10:27
sourceable6: help
sourceable6 MasterDuke, Like this: sourceable6: 42.base(16) # See wiki for more examples: github.com/Raku/whateverable/wiki/Sourceable
MasterDuke what are you trying to find? 10:28
jmerelo Surprisingly (or not) I managed to compile up to NQP in FreeBSD... github.com/raku-community-modules/...focus=true 10:44
MasterDuke there are a couple people using rakudo on some of the bsds, don't know which ones though 10:48
istr tyil is on a bsd 10:49
jmerelo MasterDuke: Kaiepi IIRC
MasterDuke: not sure about tyil
tyil I have some BSD machines, but those aren't my "main" machines 10:50
jmerelo MasterDuke: anyway, I was trying to address an issue in LWP-Simple, and then fell into the rabbit hole. Then, one thing is knowing some people use it, and another setting up a workflow to compile and test stuff for you. 10:50
tyil: did you use Raku here?
tyil only for testing purposes 10:51
(also, only freebsd so far)
jmerelo tyil: so you managed to compile it there... 10:51
tyil yes 10:52
jmerelo I mean, it does not seem impossible, only you have to know all the tricks. Latest one is you can't use the supplied make, but gmake.
tyil I even have precompiled r* for freebsd
jmerelo tyil: oh. That would have saved me some trouble... Are they accessible somewhere?
jmerelo tyil: oh yes www.tyil.nl/post/2020/06/21/lately-in-raku/ 10:53
tyil dist.tyil.nl/tmp/
no organisation in there yet :p 10:54
jmerelo OK. Anyway, I'm reproducing it in a Github action workflow, and I think I can create artifacts for downloading there. When I've got it working (or if I got it working) I can upload it to the rakudo repo so that it can create... something for download. 10:55
tyil should be very possible
sadly sourcehut started limiting to 4 build jobs :( 10:56
(and their freebsd env is software emulation, so it takes about an hour instead of 15-20min)
jmerelo tyil: this will probably take around 30 minutes github.com/raku-community-modules/.../465826226 10:58
tyil: it's a FreeBSD vm running within vmbox in macosx. Just booting the vm takes around 5 minutes. 10:59
tyil: which is why it could be a nice thing to have binaries, as well as actual vm ci. I'll try to look up if there's something around.
Hm, Azure Pipelines could be it... 11:00
jmerelo Or CirrusCI... cirrus-ci.org/guide/FreeBSD/ 11:03
notagoodidea Looking at Sparrow repo, what is the "glue" lib that I see in every file in the ressources dir? 11:09
tyil jmerelo: welp, let's see how this goes github.com/rakudo/star/runs/1656083082 11:45
jmerelo tyil: It failed here github.com/raku-community-modules/...focus=true 11:58
tyil device not configured doesn't tell me anything on first glance :< 11:59
Kaiepi jmerelo: my vps runs freebsd 12:19
El_Che "11:49 < MasterDuke> istr tyil is on a bsd" 12:20
I read: tyil is on lsd
tyil I mena 12:21
I could be
Kaiepi that's a bmake issue i think
tyil who knows ;)
Kaiepi if gmake is installed, that should be used
since nqp and rakudo's build system will generate a gnu makefile if that's the case 12:22
oh i should read more of chat 12:25
tyil does anyone know how I can remove this circle-ci check from the rakudo star repo? I don't use it, and I'm unsure why it's even there 12:39
Geth ¦ problem-solving: ash assigned to JJ Issue Subdomain for the Raku Course github.com/Raku/problem-solving/issues/257 12:58
tyil also, can a release on github be deleted after publishing, or is that only possible with drafts? 13:04
(I'm trying to make github actions build a r* on a release, but apparently release events only trigger on non-drafts, so I'll need to use an actual release to test and I don't want to clutter up with test releases) 13:05
Geth ¦ problem-solving: lizmat assigned to rba Issue Subdomain for the Raku Course github.com/Raku/problem-solving/issues/257 13:43
¦ problem-solving: lizmat unassigned from JJ Issue Subdomain for the Raku Course github.com/Raku/problem-solving/issues/257
notagoodidea For those than could be interested: a patch has been merged to KDESyntaxHighlight to consodilate the raku highlighting (used by Kate) and if everything goes well will be pushed to skylighting the syntax highlighting used by neuron and pandoc. 13:44
MasterDuke nice 13:45
xinming SmokeMachine: with Model.^all method which returns the ResultSeq, We should take snapshot of the table name. 13:52
tellable6 hey xinming, you have a message: gist.github.com/18914fae6d856161cc...3396e828d8
xinming SmokeMachine: otherwise, when we get ResultSeq, which go out of the scope of the block uses `temp Model.^table = "blabla"`, The ResultSeq will use the old table name. 13:53
SmokeMachine: so soemthing like this will work. my $rs-seq = do { temp Model.^table = "blabla"; Model.^all.grep({ }); }; 13:55
xinming SmokeMachine: also, For now, .grep({ ... }) is chained with "and", do we have methods to do chain with "or"? 13:58
guifa2 When doing a lot of string concating in perf-sensitive code, is there a known best approach? Switch to native str? Generate a list and then join at the end? I think I remember that string handling is done in chunks that are joined in the VM but I don't remember the details 14:03
(I'll do some tests to see if no one knows off hand, just curious if there was a standard approach)
SmokeMachine xinming: &&, and, || and or should work... if it isn’t, that’s a bug... 14:06
MasterDuke yeah, the vm has a concept of strands, so repetition (i.e., the `x` operator) and concatenation can just increment a strand count or create a new strand
SmokeMachine xinming: but that’ll really work well when I concert Red to use RakuAST... 14:07
MasterDuke but it does depend on how you're using the strings, so the vm may end up flattening all the strands anyway
so you really need to benchmark with the operations/code you will be doing for real 14:08
guifa2 MasterDuke: thanks. I'm having to create my own sprintf effectively and trying to come up with different ways to generate the digits in numbers 14:10
notagoodidea guifa2: For the integration of Intl::CLDR in a program, is it possible to only get a fixed set of words without having to take the all the files available? 14:19
guifa2 notagoodidea: depends, and there are several approaches. Each language has a strings and data file, and that's only opened on demand (in fact, the loading is even lazier than that). So long as you're working on a single language, that should be fairly efficient, but it does require the installation of all of Intl::CLDR's resource files (which is a hefty 75 MB or so) 14:22
xinming SmokeMachine: What I mean is, the .grep({ }).grep({ }) is a `and` op, Do we have chaining op to make the new chain `or`
guifa2 If you're wanting to reduce that weight (or allow for faster loading of a small subset of data across all languages), I'd recommend using a preprocessing script that directly accesses Intl::CLDR, and generates the specific data you need 14:23
xinming SmokeMachine: In DBIx::Class, We can do something like this, I just forgot the syntax. 14:24
notagoodidea mm ok:) My idea was : I am gonna need those words, say <hello bye foobar>, in all language available. Best to preprocess and back it somewhere. 14:25
guifa2 (e.g. a calendar app you might use `cldr{$lang}.dates.calendars.gregorian.months.stand-alone.wide[1..12] for @langs` and output that into a file and read it in more quickly
yeah, exactly. I imagine about half of people using it will have the `use` statement in the main script, and about the half using it for preprocessing, depending on use case 14:26
xinming SmokeMachine: BTW, the reason I ask that ResultSeq should remember the 'temped' table name is.
tyil notagoodidea: re: raku highlighting in KDESyntaxHighlight: nice! 14:27
notagoodidea guifa2: I was poking with the idea to find a way for localized CLI arguments but not right now.
xinming SmokeMachine: I just have some requirement which is like, my $rs1 = { temp Model.^table = "blabla"; Model.^all }; my $rs2 = { temp Model.^table = "bleble"; Model.^all };
guifa2 I'll have a localized number parser available again soon :-) 14:28
literally just use <local-number> in your regex and it will get you both the string and evaluate the value of the number as a mixin
notagoodidea tyil: Yep, it was already there but buggy due to Q String parsing. I just found the why and more knowledgeable people solved the issue :D 14:29
tyil notagoodidea++
something to add into next week's rakudo weekly :>
notagoodidea guifa2: It was more about the options/flag localized. That you could use --files (en) or --fichiers (fr) and the parser get the right options/flag modulo the existence of those words in the corpus. 14:30
SmokeMachine xinming: .grep (|) .grep? 14:31
notagoodidea tyil: :D If it is integrate to skylighting before end of the week, I will tell you. When it happends, raku will be supported by further version of pandoc/neuron.
guifa2 notagoodidea: Ah. You might find Fluent to have some better solutions there (that's my next target for rewriting after finishing up some CLDR work). It's Mozilla's localization framework. It's currently mostly usable in Raku, but I can do much better now 14:32
notagoodidea guifa2: Neat, even if right now, I am not confident to use a research project from Mozilla as they are dropped every six months (used to use Send, Notes, etc. and *poof* gone). 14:35
guifa2 Heh fair. But since fluent is an internal framework (really, at its core, it's just a file type description) it doesn't require any long-term support from them :-) 14:38
xinming SmokeMachine: Something like that, But the (|) union is different from a `where ... OR ... ` 14:48
SmokeMachine xinming: yes, it is...
xinming SmokeMachine: We need someway to add a top-or condition, instead of .grep only. 14:52
[Coke] yawns. 14:59
tyil .tell brass if you still want to look into mac builds of rakudo star, you might be able to get some automated builds working with github actions, I just made it work (I think?) for GNU+Linux tarballs 15:00
tellable6 tyil, I'll pass your message to brass
SmokeMachine xinming: With Red, I usually think the opposite... having this Raku's syntax, what should be the generated SQL to replicate Raku's behaviour 15:15
xinming: how would you do, with raku, grep a list by something OR something else... the best way I see would be: @list.grep: { .&cond1 || .&cond2 } 15:17
xinming: it's being hard to find another way... 15:18
cog Is there a raku routine which prints the printable glyphs for control chars ? 15:20
... in a string 15:21
SmokeMachine xinming: something that could be done is: for `$a.grep(&cond1) (|) $b.grep(&cond2)` when `$a eqv $b` optimise to `$a.grep: { .&cond1 || .&cond2 }` 15:22
xinming SmokeMachine: the problem with that approach is, we can't build search conditions based on "conditions" 15:25
SmokeMachine xinming: maybe, that way you could do something like: `my $a = Model.^all.grep: &cond1; ...; my $or = $a (|) $a.parent.grep: &cond2`
guifa2 cog: not a premade one, but you could do it fairly easy using Str.trans
cog Thx guifa2 15:26
SmokeMachine xinming: sorry, what do you mean by `we can't build search conditions based on "conditions"`
?
tony-o .tell patrickb merged the config stuff into fez 16:29
tellable6 tony-o, I'll pass your message to patrickb
jmerelo .tell tyil maybe you should post the instructions to build FreeBSD somewhere. I couldn't pull it off. 17:02
tellable6 jmerelo, I'll pass your message to tyil
tyil jmerelo: I just used `rstar fetch && rstar install`, though that's been a couple months ago 17:04
maybe one of the projects changed something that broke this
also, I dont seem to be installing gmake for freebsd in my build instructions
jmerelo The compilation I'm attempting breaks with "Device not configured" which apparently has nothing to do with compilation, but with using sshfs 17:05
SmokeMachine xinming: ???
jmerelo This is the log github.com/raku-community-modules/.../465898571
tyil interesting
I'm not well versed in github actions, but I'm guessing you can't use something else than sshfs then 17:06
jmerelo tyil: I really have no idea. Here's the action, it should be pretty straightforward to understand github.com/raku-community-modules/...eebsd.yaml 17:07
tyil: it sets up a freebsd vm, which is connected to the MacOSx port via sshfs. I have no idea why it breaks down after a while. Anyway, if you manage to package it for FreeBSD this could be use to run FreeBSD tests where they're needed. 17:09
tony-o jmerelo: sent you a pm
tellable6 2021-01-06T10:32:53Z #raku-dev <patrickb> tony-o Did you know JRaspass and JJAtria[m] are working on a pure-Raku modules.raku.org replacement? You should probably start a conversation on how to best integrate the fez ecosystem.
jmerelo Or maybe we can set up something in SourceHut, I have no idea how it works.
tony-o: gotcha
tony-o JRaspass: jjatria - i have some ideas about that if you guys want to chat. 17:10
JJAtria[m] tony-o Sure :D 17:11
JRaspass ditto
JJAtria[m] Just in case: I'm going to be afk for a bit in a bit, but I should be back in about an hour or so 17:13
tony-o what stage are you in? do you have a roadmap? 17:13
JRaspass for the site? there's an unordered list of things to do on the README.md called a roadmap :-P 17:14
tony-o what's the project link?
JRaspass gitlab.com/raku-land/raku-land 17:15
tony-o can we add the zef ecosystem to that roadmap? 17:16
JRaspass sure, patches welcome ;-) 17:17
tony-o just requested access as `tony-o` 17:19
tyil jmerelo: looking at my latest freebsd build (4 hours ago), it should still work, so I'm going to guess this is purely an issue with that freebsd vm setup (builds.sr.ht/~tyil/job/388507) 17:20
JRaspass approved
jmerelo tyil: how can that be used? 17:21
tyil jmerelo: the build itself?
tony-o thanks
jmerelo right
tyil jmerelo: download dist.tyil.nl/tmp/rakudo-star-amd64...f5.tar.gz, and extract it to /usr/local (or any other place you want to install it) 17:22
jmerelo tyil++ 17:29
notagoodidea tyil: The new raku definition have been merged in skylighting (github.com/jgm/skylighting/issues/114) :D 18:44
tyil notagoodidea: awesome!
jmerelo notagoodidea++ 18:53
Geth doc: 5b3407028c | (Stoned Elipot)++ | doc/Type/ValueObjAt.pod6
Typography tweak
 19:10
linkable6 Link: docs.raku.org/type/ValueObjAt
stoned75 m: releases my %m is Map = a => 42; say %m.WHICH, ' ', %m.WHICH.^name; 19:24
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
releases used at line 1. Did you mean 'reverse', 'repeated'?
stoned75 commit: releases my %m is Map = a => 42; say %m.WHICH, ' ', %m.WHICH.^name; 19:25
moon-child what's the difference between my %m is Map and my Map %m?
committable6 stoned75, gist.github.com/27dd1813076ed5f291...6571637487
stoned75 m: my Map %h; say %h.^name 19:26
camelia Hash[Map]
stoned75 m: my %m is Map; say %m.^name
camelia Map
stoned75 not the same type at all :) 19:27
stoned75 %h here is a Hash of Map while %m is a Map 19:30
hum... the evolution seen in gist.github.com/Whateverable/27dd1...6571637487 is ... interesting 19:32
moon-child stoned75: right 19:44
tony-o JJAtria[m]: what exactly is the bin/fetch mentioned in the README? i see the bin/(git|cpan) - are those being called via some cron? 20:03
.tell melezhik the issue on fez should be moot in the next couple of weeks
tellable6 tony-o, I'll pass your message to melezhik
JJAtria[m] tony-o I think that was renamed to 'parse' on gitlab.com/raku-land/raku-land/-/c...880eacb7bb 20:04
The documentation seems to be a bit out of date :P
JJAtria[m] The scripts in `bin` are mostly orchestrated by `bin/ingest`, which is run from the Makefile 20:06
tony-o gotcha - is that container just hotswapped upon being built then? 20:07
JJAtria[m] JRaspass would know the specifics about what exactly runs `make ingest`, which runs the process in the deployment box 20:08
I don't think this process requires it to be rebuilt, actually
It just re-creates the entire dataset in the DB in a transaction 20:09
JJAtria[m] But as far as I understand that's talking to the DB, not the container running the site 20:10
tony-o sounds good - i can work with that ty 20:16
i'll submit a patch for p6c and zef
[Coke] So, OT, but, as a US citizen... WTAF. 20:25
El_Che [Coke]: :/
tadzik yeah, surreal
El_Che [Coke]: I have seen the hand in the US supporting scenes like that in Latin America, but let me tell you, I have no schadenfreude at all. I hope democracy is not too badly damaged. 20:26
it reads like a bad scenario. Trump is even bad at coups 20:27
Seeing on how well armed are police, national guard, etc, I would expected rioters get into the senate floor 20:29
wow
melezhik I spin up a Sparky CI server for dog fooding , if someone wants to play with - 161.35.142.50/ cc tib 20:30
tellable6 2021-01-06T20:03:47Z #raku <tony-o> melezhik the issue on fez should be moot in the next couple of weeks
tony-o coup is kind of a stretch 20:36
not that i disagree with the rest of the WTAF sentiment 20:40
El_Che it doesn't have to be a succesful coup to be a coup 20:43
there was a similar failed coup in Venezuela last year 20:44
pretty identical scenario
timotimo it's not a coup if the military isn't involved 20:45
El_Che it's a coup if the military stays mostly out 20:46
that's was the original scenario in Bolivia
El_Che (the military had to step in later because the coup leaders weren't popular at all) 20:47
timotimo oh?
i guess that's a good point
El_Che it's was very innovative :)
timotimo if the military just shrugs from the sideline, you can't really say they are "not involved"
El_Che well, sone of that is maybe here in play
where is the National Guard? 20:48
how does a smallish number of protesters get into the Senate?
probably the best secured place next to the White House
let's hope it's just a farce of an attempt of Trump to poison the well while he leaves 20:49
sortiz El_Che, The request by DC officials to deploy de NG was denied by the Defense Department. 20:54
El_Che yeah, pretty WTF
[Coke] WaPo has an update on that, they are deploying. 20:55
there's a lot of churn right now in the news 20:56
El_Che I hate to see boots on the streets, but this looks warranted
[Coke] (washington post)
El_Che Remember the Senate elections in Georgia? Good times. 20:57
It sounds like ages ago :/
sortiz They are deploying at Trump request?! 20:58
[Coke] www.washingtonpost.com/dc-md-va/20...HC357IIUEM
El_Che if you wonder how the protestor got so fast inside the Senate: twitter.com/bubbaprog/status/13469...19520?s=20
twitter.com/FridaGhitis/status/134...14624?s=20 21:00
Improvised explosive device found in the Capitol: NBC
Xliff How can I get Cro::WebApps::Template to use a literal string as its template? 22:10
moritz Xliff: github.com/croservices/cro-webapp/...ry.pm6#L63 seems to be the answer 22:20
$source is the template string 22:21
Xliff Yeah. I am working through that now. Was hoping I wouldn't have to roll my own.
Xliff When I test this, I may have to formalize it for a non-file and submit a PR. 22:21
timotimo "D.C. Police Lose Control Of Rioting Trump Supporters After Hundreds Of Officers Called Away To Deal With Black Jaywalker" 22:26
www.theonion.com/d-c-police-lose-c...1846002738
El_Che lol 22:28
Xliff timotimo: Yes. I'm living that. Right now. 22:43
I'm also black. And I jaywalk.
So DCPD scurs me.
Also... is it known that raku will choke on symlinks? 22:44
lizmat it does ?
El_Che Xliff: you're Washiington DC based if I remember corrctly? 22:45
lizmat wishes Xliff strength in the coming night of curfew
perryprog Xliff, stay safe <3 22:51
tony-o JJAtria[m]: submitted a mr for review for zef&p6c on raku.land 22:53
gfldex Thinking about The Situation In General, I just realised how much Raku helped me to stay sane lately. 22:56
JJAtria[m] tony-o++ 22:58
Xliff lizmat: Yes. If the file being compiled is a symlink, it gives a remarkable bit of LTA. 23:01
Can you check to confirm that on your end?
Xliff Thanks all for the safe wishes. I am in mah house in lockdown with plenty of beer. 23:02
lizmat sure... if you tell me what to check
Xliff: do you have a gist ? 23:07
Xliff lizmat: No impossible to gist. 23:11
Xliff Create a raku file. Symlink it to another location. Try to run raku on the symlinked file. 23:11
timotimo Xliff: i likewise wish you strength, all of this is very fucked, and has been for a long, long time 23:14
Xliff timotimo++ : thanks
lizmat Xliff: should it be in a different dir ? 23:16
Xliff: cannot reproduce on MacOS Mojave 23:17
Xliff lizmat: OK, thanks. 23:17
If I can confirm it again with concise reproduction steps, I'll bug it. 23:18
sortiz Xliff: Also I can't reproduce on Linux (Fedora 33)
Xliff sortiz++
sortiz: Are you on v2020.12-32-g8150a2864? 23:19
sortiz Yep, moar-blead. 23:20
lizmat also 23:22
sortiz Your symlinked file 'use' something relative?
melezhik tony-o I found another issue with fez installation - 161.35.142.50/report/fez-test/24 , I filled it to fez-GH 23:25
tony-o melezhik: just responded to it 23:26
melezhik tony-o ++
tony-o try a `zef update` and give it another go 23:29
melezhik yeah 23:31
melezhik did not help, the same error ))- 23:32
Xliff moritz: Um. -Ilib was specified, where the compunits in lib/ were symlinked 23:35
melezhik tony-o interesting ... - gist.github.com/melezhik/2bfea47bd...7f0d227eeb 23:36
it happens immediately after successful install , really weird
Xliff gist.github.com/Xliff/23d83c2ddcfc...c572caa87d 23:37
melezhik tony-o I created a SCM triggering CI on fez - 161.35.142.50/project/fez-test if you're interested ... :-)) 23:54

Powered by Cro and the Raku® Programming Language.

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

Thank you!