🦋 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.
00:00 sjaveed left 00:30 sjaveed joined 00:40 thaewrapt[m] joined 01:10 mensvaga left 01:17 molaf left 01:30 molaf joined, Tirifto left 01:54 aborazmeh left 02:04 Manifest0 left 02:05 stoned75 left 02:06 Manifest0 joined 02:12 stoned75 joined
[Coke] 47 ballots reporting (Average votes per ballot: 4.64) 02:26
Geth doc: 2ff7fb610c | Coke++ | doc/Type/Map.pod6
use preferred word
linkable6 Link: docs.raku.org/type/Map
02:34 stoned75 left 02:38 cpan-raku left 02:40 cpan-raku joined, cpan-raku left, cpan-raku joined 02:41 stoned75 joined 02:50 ex_nihilo joined 02:58 a3f left 02:59 a3f joined 03:02 zacts joined 03:29 sjaveed left 03:36 stoned75 left 03:42 stoned75 joined 03:43 zacts left 04:00 tejr left, tejr joined 04:56 sjaveed joined 05:00 sjaveed left 05:08 Kaiepi left 05:09 Kaiepi joined 05:11 bocaneri is now known as Sauvin 05:26 [Sno] left 05:34 sjaveed joined, [Sno] joined 05:40 sjaveed left 05:41 aborazmeh joined, aborazmeh left, aborazmeh joined 05:51 wamba joined
guifa rypervenche: check out Intl::X again. I have a localization editor mostly working — onlyt hing left to do is to enable save the edited stuff 06:13
06:16 aluaces joined 06:17 skids joined 06:39 stoned75 left 06:46 stoned75 joined 06:52 k-man left 06:55 aborazmeh left 07:02 skids left 07:09 jmerelo joined
jmerelo releasable6: status 07:09
releasable6 jmerelo, Next release will happen when it's ready. 3 blockers. Changelog for this release was not started yet
jmerelo, Details: gist.github.com/f32037527df8372727...fbc1ed1491 07:10
07:30 k-man joined 07:31 sena_kun joined 07:37 k-man left 07:48 stoned75 left 07:50 sjaveed joined 07:51 molaf left 07:54 sjaveed left 07:59 stoned75 joined 08:00 Kaiepi left 08:12 kylese joined 08:17 domidumont joined 08:19 wamba left, Black_Ribbon left 08:20 pecastro joined 08:24 domidumont left 08:25 wamba joined 08:31 aborazmeh joined, aborazmeh left, aborazmeh joined 08:41 molaf joined
jmerelo m : say v1.2.3 08:56
m: say v1.2.3 08:57
camelia v1.2.3
jmerelo m: say v1.2.3.parts
camelia v1.2.3.parts
jmerelo m: say (v1.2.3).parts
camelia (1 2 3)
09:07 telex left 09:08 telex joined
Geth doc/seq-op-variants: f8fd3cc35d | (Stoned Elipot)++ | doc/Language/operators.pod6
Revisit infix ...

  - mention Seq type
  - introduce all its variants, with examples
  - consider variants with an initial caret in wording
09:17 sjaveed joined 09:23 sjaveed left
lizmat rypervenche: that would be an option if there would be a unicode codepoint for 2 butterflies, but afaik there is none 09:28
there are codepoints for camels with one and with two humps 09:29
same for the beer: a unicode codepoint for 1 beer and one for 2 beers
09:35 wamba left 09:36 aborazmeh left 09:40 wamba joined 09:46 wamba left 09:58 Sgeo left
Geth doc/seq-op-variants: 1d85e896be | (Stoned Elipot)++ | doc/Language/operators.pod6
sequence operator with/out final ^ clarification
doc/seq-op-variants: ab328c9aec | (Stoned Elipot)++ | doc/Language/operators.pod6
Restore previous wording
10:05 k-man joined 10:09 rindolf joined
stoned75 sdsdf 10:10
10:12 jmerelo left 10:49 sjaveed joined
Geth doc/seq-new: 9d180cfcd2 | (Stoned Elipot)++ | doc/Type/Seq.pod6
mention the sequence operator

while here document the various way to build a Seq at the start of the document.
doc: stoned++ created pull request #3649:
mention the sequence operator
10:55 sjaveed left 10:57 sjaveed joined 11:01 gnufr33dom left, sjaveed left 11:02 Altai-man joined 11:05 sena_kun left 11:12 stoned75 left 11:19 stoned75 joined 11:51 stoned75 left 11:52 mowcat joined 11:58 stoned75 joined 12:00 mowcat left 12:02 thaewrapt[m] is now known as ThaEwar, ThaEwar is now known as ThaEwat 12:08 sjaveed joined 12:12 sjaveed left 12:22 Tirifto joined 13:02 mowcat joined 13:48 wamba joined 13:56 MilkmanDan left 13:59 MilkmanDan joined 14:05 chloekek joined 14:10 ex_nihilo left 14:18 sjaveed joined 14:24 sjaveed left 14:40 kylese left 14:43 mowcat left 14:54 stoned75 left 15:01 sjm_uk left, stoned75 joined 15:03 sena_kun joined 15:05 Altai-man left 15:14 HeAvengin7_ joined 15:16 HeAvengin7_ left 15:28 kent\n left 15:29 kent\n joined 15:37 sjaveed joined 15:43 sjaveed left
[Coke] 50 ballots reporting (Average votes per ballot: 4.64) 15:48
(Reminder, some of those may be marked invalid if we can't verify a commit bit)
(And I may have accidentally marked some as OK because they had a commit, not a commit bit. Will have to make sure I'm looking at the right lists.) 15:50
16:07 MasterDuke left
stoned75 the documentation for Mu.Str says: "Returns a string representation of the invocant, intended to be machine readable". I'm wondering about the "machine readable part", what does it really mean ? 16:09
guifa stoned75: it just means something intended to potentially be processed by some other entity 16:11
it really contrasts more with gist
So for an array, .gist will only show the first 50 (or 100?) elements. .Str will give every single element
stoned75 hum.. ok. I see. it's machine readable as opposed to .gist :-) 16:13
now... I don't quite get what .Stringy is, wrt .Str 16:14
I mean why does the ~ operator uses .Stringy and not .Str ? 16:15
guifa Stringy is a role (like Numeric)
Str is just one type of Stringy thing, but it’s the most common of course
stoned75 I meant the Stringy method. 16:16
guifa .Stringy is just required to return something that does Stringy. That could be a Str (it is, in most cases), but it could also be something else
So using numbers as an example
the + prefix returns Numeric 16:17
timotimo m: say Blob ~~ Stringy
camelia True
timotimo ^- i think this is what makes ~ work as buf concat
16:17 aluaces left
guifa so Str.Numeric can return anything that’s Numeric — Real, Num, Rat, etc 16:17
timotimo github.com/timo/moarperf/releases/...4.AppImage - i would appreciate some testers for this AppImage of a Cro webapp
16:18 MasterDuke joined
timotimo after starting it, it should output a few lines and start a webserver, feel free to ctrl-c it directly after that 16:18
guifa timotimo: I *think* infix:<~> (Buf, Buf) is special cased
timotimo sourceable6: Buf.new ~ Buf.new 16:19
sourceable6 timotimo, github.com/rakudo/rakudo/blob/8e6f....pm6#L1097
16:19 stanrifkin joined
guifa The other Stringy is Uni 16:21
But it’s not used very often. It works more like Strs in other languages
16:22 wamba left
stoned75 ok so we have the Stringy role, for string like classes. so the Stringy method is the "go to" coercer to get a Stringy, as using '$whatever.Str' to get a Stringy object would have been counterintuitive 16:22
is that it ? :)
guifa More or less yeah. 16:23
$whatever.Str guarantees you a Str
And no other class
$whatever.Stringy could get you a Str, Uni, or Blob
stoned75 ok. thanks! 16:24
16:25 sjaveed joined
guifa (the reason most of those return roles is it makes it easier for you to write in classes that function *like* Strs but aren’t, and still get support if people just ask for Stringy) 16:25
stoned75 sure. got it! 16:26
16:30 [Sno] left 16:31 sjaveed left
rjbs Hello! I'm not really here to read replies, although I will see them in due course. 16:40
Recently, TPF has chartered an infrastructure group that will be providing hosting services and system administration for important software infrastructure -- both that used by the TPF and by the Perl and Raku communities. This could be static site hosting, web services, mailing lists, and so on. 16:41
It's early days, so I don't have many specifics to offer, but I wanted to get the word out that this was happening, and that anybody who wants to talk about how we might be able to help could be in touch. You can email [email@hidden.address] or hit me up on TPF Slack or here. (Email is probably best.) 16:42
16:42 cpan-raku left 16:43 cpan-raku joined, cpan-raku left, cpan-raku joined
moritz rjbs: thanks for letting us know! (not muching going on this time of day...) 17:08
17:09 patrickb joined
moritz rba: ^^ 17:09
patrickb nine: Any clues on how to reproduce the "Missing or wrong version" error you got? 17:13
timotimo: How did you trigger the "NQP_HOME is invalid" error? 17:15
rba rjbs: Thanks for letting us know. I will contact you. 17:17
17:34 mowcat joined 17:36 sjaveed joined 17:38 Chi1thangoo joined 17:40 stoned75 left 17:42 sjaveed left
patrickb nine: The only way I found to reproduce the error was by building and installing a current nqp + rakudo, then switching to a different (older) commit in nqp, installing again, and then re-running `make install` in rakudo. 17:48
nine: But (I think) the failure is expected in that case. 17:49
rjbs rba, moritz: +1 17:51
18:04 chloekek left, Sgeo joined 18:10 mowcat left 18:22 stoned75 joined
nine patrickb: but I didn't downgrade? I only updated my rakudo 18:22
Maybe from a version where the revert of the revert was not yet reverted? 18:23
18:24 jmerelo joined, mowcat joined
patrickb nine: I think that shouldn't make a difference. 18:25
nine: :-/ I'm grateful for any clue on how to reproduce this...
nine patrickb: maybe my reflog may give a hint: gist.github.com/niner/fd9d5285f61c...e23f28793f 18:27
18:27 sno joined
nine btw. a make clean fixed the build for me 18:27
18:32 chloekek joined
chloekek こんいちは楽、how’s it going 18:33
Haven’t used Raku in months
sena_kun chloekek, hey. With ups and downs, I guess. 18:34
jmerelo Hi... 18:35
m: say pi.^name
camelia Num
jmerelo m: say *.^name
camelia WhateverCode.new
chloekek m: say (*).^name 18:36
camelia WhateverCode.new
18:36 mowcat left
tobs m: say .^name with * 18:37
camelia Whatever
18:41 patrickb left
jmerelo tobs: good one 18:53
m: say .^name with *+3
camelia WhateverCode
timotimo jmerelo: hey you 18:54
Geth doc: 443e6a2b46 | (JJ Merelo)++ | doc/Type/Mu.pod6
Fixes typo, closes #3650
doc: 912c326caa | (JJ Merelo)++ | doc/Language/terms.pod6
Clarifies how * works as a literal

And when it becomes either Whatever or WhateverCode. Closes #3648
linkable6 Link: docs.raku.org/type/Mu
DOC#3650 [closed]: github.com/Raku/doc/issues/3650 [docs] Mu.raku description typo 18:56
Link: docs.raku.org/language/terms
DOC#3648 [closed]: github.com/Raku/doc/issues/3648 [bug][docs] Is Whatever (*) a term?
18:56 aborazmeh joined, aborazmeh left, aborazmeh joined 19:00 AlexDaniel joined, AlexDaniel left, AlexDaniel joined
stoned75 jmerelo: thanks for the typo correction for #3650. but I admit I do not understand the clause "or calls C<.raku> on the name of the type object otherwise" :-} 19:00
AlexDaniel [Coke]: I mean… technically… midnight was 19 hours ago 19:01
[Coke]: that's why midnight should not be used when setting deadlines :) 19:02
19:02 Altai-man joined, jmerelo left 19:03 chloekek left 19:05 sena_kun left, mowcat joined 19:06 aluaces joined 19:14 gnufr33dom joined 19:15 stanrifkin left
rypervenche Why does "say $*USER" give me Nil on some of my machines and my actual username on another? 19:30
All running 2020.08.2.
Geth doc: 3573f9b36c | Altai-man++ (committed using GitHub Web editor) | doc/Type/IO/Socket/Async.pod6
Fix link
linkable6 Link: docs.raku.org/type/IO::Socket::Async
Altai-man cannot reproduce on gnu/linux 19:37
19:39 epony left 19:40 epony joined
[Coke] AlexDaniel: I didn't pick it. 19:43
also, your subject on the email was incorrect, but I'll allow it.
AlexDaniel I mean, you shouldn't allow it because it was too late xD
[Coke] also I'm not sure if your send was in reply to something. 19:44
did you deliberately vote after what you thought the deadline was just to cause controversy?
AlexDaniel -_-
[Coke] Not that it matters to the voting.
AlexDaniel no
nine Huh...I've not voted yet. And I think I'm actually not going to. 19:45
stoned75 rypervenche: what does the id command return on the "faulty" machines ?
nine Well if the deadline is already past, I'm definitely not going to :D
I wouldn't be able to decide anyway. All candidates have merits and I can imagine all of them contributing greatly 19:46
AlexDaniel nine: technically it's already past, but I guess the common understanding of “midnight” is different from what it actually is, so it's fine
[Coke] I never did say voting was closed or open, though, so not sure where AD led in from. :) 19:50
but yes, makes sense to let it go until end of day due to common confusion about midnight. 19:51
AlexDaniel [Coke]: just to clarify, it says in bold “Voting is open between midnight on September 6th UTC & midnight on September 20th UTC.” 19:52
so that's where I got it from :)
[Coke] I don't have admin rights on most repos, so I cannot check voter privs through the GUI, and I'm not sure the API is giving me what I'm thinking it is.
AlexDaniel: yes. I know what it says.
Just feel like you're trying to give me a hard time here, which I'm sure you're not, intentionally. 19:53
I agree: it should have said 11:59:59PM on one of the days to avoid confusion. (That's what I used to do for the TPF GC submitals)
AlexDaniel no, just wanted to let you know in case you missed this detail. You can decide yourself what to do with that info.
either way is fine to me 19:55
19:55 squashable6 left
[Coke] so, looking at github.com/Raku/nqp/settings/access - who has commit bits? 19:56
19:56 squashable6 joined
[Coke] ah. it's basically AlexDaniel, samcv (admins), and "everyone in Raku". 19:57
ok. I can verify this against jdv79's script...
AlexDaniel [Coke]: this should be the list: github.com/orgs/Raku/people?type=source
aaah, you're actually more correct, yes: github.com/orgs/Raku/teams/raku/members
although… why are these lists so different… 19:58
[Coke]: OK, hold on, this is not entirely right
one second let me try something… 19:59
19:59 squashable6 left
codesections rypervenche: sorry if this it too obvious a question, but is the USER environmental variable set to your username outside of Raku on the machines where `say $*USER` outputs «Nil»? 20:00
20:01 squashable6 joined
AlexDaniel [Coke]: I'll try to explain. So at some point in the past everyone in the org had write access. That was changed at some point because there are (or were) some limitations to this approach (can't have locked repos that only privileged users can write to) 20:01
[Coke]: I believe the default privilege was changed from `write` to `read`
[Coke]: and the Raku team was created. Then I wrote a script which adds all people to the selected team 20:02
[Coke] I only wanted to check who had privs on that team: the list I get back from jdv79's script has 177 people with admin or write access to nqp, so it seems to match up. the details of how they're assigned probably doesn't matter.
AlexDaniel [Coke]: however, people forget to add people to the Raku team when they intend to give write access 20:03
so I reran the script now, and the Raku team has 203 members now
this is how it is supposed to be, I believe
[Coke] 203 or 213?
AlexDaniel there's still a difference (213 vs 203)
let me check why these differ real quick… 20:04
[Coke] ok. i'll re-re-run jdv79's script...
still 242 people.
(those extra few may have already been in one of the other repos?) 20:05
er, 240. some had trailing commas missing. 20:06
(so they showed up twice)
20:06 patrickb joined
AlexDaniel [Coke]: github is showing 213 in some places and 203 in others, something to do with child teams and stuff… weird 20:07
either way more is probably right-er 20:08
patrickb nine: I've been thinking about your "wrong version" error a bit more. If I interpret it correctly, it means that 'blib/Perl6/Pod.moarvm' (compiled from 'gen/moar/Pod.nqp') was built against a different version of NQPs 'gen/moar/stage2/NQPHLL.nqp'. If my understanding of linking in Raku is correct, then that's an expected error. When making changes to nqp and *not* doing a `make clean` in rakudo afterwards, the files in 'blib' will still expect 20:14
the older nqp files and fail with the error you observed. make simply doesn't know that it needs to rebuild the files in `blib` when there are changes in NQP.
nine: As the expert of Rakudos module handing, is the above correct? 20:15
.ask timotimo Do you have a specific question about relocatability or do you want a general explanation? 20:16
tellable6 patrickb, I'll pass your message to timotimo
patrickb is off to bed now.
20:16 patrickb left 20:17 rindolf left
[Coke] AlexDaniel: if you can find any of these people, I'll let them know their ballot can count after all: gist.github.com/coke/07f167179135f...d526d5437e 20:28
but with them removed, we're at:
46 ballots reporting (Average votes per ballot: 4.70)
(discounting their ballots didn't change the top 7, but it did change the order slightly) 20:29
nine patrickb your understanding of where the error message is coming from is correct. However while make itself isn't smart enough to catch this, our build system used to be. That's the whole point of the check-nqp-version.pl script - to trigger a full rebuild if NQP got updated
[Coke] 3.5 hours left if we go with the charitable deadline of the upcoming midnight.
AlexDaniel [Coke]: can I invite some of them? For example, alabamenhu wrote an advent post, did some PRs, wrote a module 20:31
20:31 aborazmeh left
AlexDaniel [Coke]: it's just that nobody bothered to invite them yet 20:31
[Coke] I am just tallying votes here.
20:32 aborazmeh joined, aborazmeh left, aborazmeh joined
[Coke] if you need signoff on that, please ask someone else. 20:32
AlexDaniel well, I'm inviting alabamenhu anyway… 20:33
[Coke] If you're considering one, I ask you consider them all on their merits (no clue which orgs they may have submitted things to)
AlexDaniel [Coke]: alabamenhu is the only nickname I recognized immediately, trying to find the others though 20:34
[Coke] OK. I feel like getting the invite meets the spirit of the request, even if they haven't ACK'd it yet 20:35
AlexDaniel: you invited them to the Raku org?
AlexDaniel [Coke]: alabamenhu, yes. The invite does not add them immediately to the Raku team but that's just because I want them to have a chance to turn autosubscribe feature off 20:36
github.com/Raku/please-turn-off-th...ch-feature :) 20:37
[Coke] ok, let them know their ballot is counting.
*I* let them... 20:38
Channel - please speak up if you know the four remaining voters and can point at contributions. you can go to their user page and look at contributions... 20:39
AlexDaniel camstuart submitted a PR to Raku/examples, created some tickets for Cro, submitted a PR for raku-community-modules/raku-mime-types and already has write access to that particular repo 20:41
maybe more. Seems like a new user acting in good faith, I'll invite them
that's it. I don't know the others and can't find their contributions 20:42
guifa [Coke]: sorry about that!
and AlexDaniel ty for the invite 20:44
AlexDaniel guifa: so that I don't forget later, can I invite you to the Raku team now? (is autosubscribe feature turned off?) 20:47
guifa AlexDaniel: you just did :-) (I’m alabamenhu)
someone squated guifa on GH haha 20:48
AlexDaniel guifa: no, I only added you to the “Everyone” team
guifa ah
AlexDaniel which has access to just docs and github.com/Raku/please-turn-off-th...ch-feature :)
guifa Aaaah haha 20:49
AlexDaniel reason for that is that people were very unhappy in the past :)
guifa AlexDaniel: I’ve made the mistake before of not turning off notifications. CLDR lists will be quiet for weeks and then next thing you know you’ve got 200 messages in less than an hour 20:53
El_Che I just disabled mails 20:57
rypervenche codesections: Ahhh, I've got all of my machines in Chinese (locale zh-TW) except for the one that works, which is in English. The $USER variable works correctly on all machines. But the output of "id" gives "使用者id=1000(rypervenche)" instead of "uid=1000(rypervenche)" 21:10
And I don't set the USER variable anywhere. 21:11
stoned75 $*USER is set with a regex on id's output... 21:14
codesections rypervenche: Interesting! I wonder why `id` is involved? I would have guessed that $*USER was just reading USER directly
stoned75: any idea why?
stoned75 see src/core.c/Process.pm6 21:15
no I do not known why it is implemented this way
github.com/rakudo/rakudo/blob/mast...ss.pm6#L81 21:16
I supposed a quick fix would be something like: 21:18
- once if !Rakudo::Internals.IS-WIN && try { qx/id/ } -> $id {
+ once if !Rakudo::Internals.IS-WIN && try { qx/env LANG=C id/ } -> $id {
timotimo we would possibly want to use getgrgid or something on linux, but putting nativecall directly into the setting .. dunno if great idea 21:19
tellable6 2020-09-20T20:16:26Z #raku <patrickb> timotimo Do you have a specific question about relocatability or do you want a general explanation?
timotimo moar could expose it, or a rakudo extop
codesections Hmm, either way, it seems like a bug. Either that quick fix, an update to the regex, or a fallback to the USER env var
stoned75 I suspected as much (that using nativecall is not a good idea) 21:20
codesections the *very* quick fix for this particular situation would be adding `.*?` to the start of the regex 21:21
But I have no idea if the output of `id` can vary in other ways in different locales
stoned75 hence fixing the locale :) 21:22
rypervenche I'll submit an issue at least. 21:24
stoned75 that's the thing to do :)
m: use NativeCall;sub getuid() returns int32 is native { * }; say getuid; 21:26
camelia 1017
stoned75 works nicely :)
ah well $*USER is the username... by bad :) 21:27
21:29 Altai-man left 21:30 aluaces left
MasterDuke fecundf has created a bunch of issues in rakudo/problem-solving/etc. don't know about commits 21:37
rypervenche Mkay, submitted github.com/rakudo/rakudo/issues/3927 21:38
stoned75 eh. mmm raku/doc. docs.raku.org/language/nativecall#...rd_library ;) 21:43
22:07 pecastro left 22:12 Tirifto left
[Coke] got another non-member vote an hour ago. 22:48
last entry on gist.github.com/coke/07f167179135f...d526d5437e
22:49 Black_Ribbon joined
[Coke] little under an hour to go. 22:49
little *over* an hour to go.
23:03 aborazmeh left
AlexDaniel [Coke]: I did invite camstuart (reasoning above) 23:21
as for everyone else in that list, I think they should submit some PRs first :)
codesections weekly: www.codesections.com/blog/raku-unicode/ 23:37
notable6 codesections, Noted! (weekly)
moon-child ah - does 'commit bit' just mean 'has a commit' (merged by someone else)? 23:44
I took it to mean 'has commit privileges' 23:45
[Coke] correct. for voting, you need a bit, not just a merged PR 23:47
sorry: no, "commit bit" does not mean "has a commit" 23:48
it means you have permission to commit. You can have a merged PR (and therefore a commit) without having the "bit" 23:49
that leaves 4 IDs that had ballots but no commit bits. Can we get someone other than AlexDaniel to look at those IDs, please (not that i don't appreciate AlexDaniel doing the work on those 2, but he's a candidate, so...) 23:51
48 ballots reporting (Average votes per ballot: 4.67) 23:53
AlexDaniel moon-child: it's a bit random. Sometime people submit one PR and get invited, other times people do a lot of noticeable contributions but don't get invited 23:54
I think it's a good opportunity to notice that some people were not invited, so I invite them :) 23:55
perl6 org had 323 members, it is very inclusive 23:56
[Coke] of the remaining ones, fecundf has the most activity, it seems. 23:57
AlexDaniel (fwiw all of them were invited to Raku/)
[Coke] voting closes in 2 minutes. 23:58
I'll keep the results under wraps for a day, giving us some time to deal with the remaining voters who don't have bits. If we decide they get bits, I'll enable their ballots that were submitted on time.
AlexDaniel 👍 23:59