[00:06] *** Doc_Holliwood left [00:19] *** squashable6 left [00:22] *** squashable6 joined [00:45] *** ggoebel joined [00:53] *** lucasb left [01:22] *** Doc_Holliwood joined [01:30] New module released to CPAN! Net::IP (1.1.0) by 03TBROWDER [02:27] *** cinch joined [03:14] *** ggoebel left [04:26] *** warriors_ joined [04:35] *** warriors_ left [04:35] *** warriors_ joined [04:37] *** warriors_ left [04:38] *** warriors joined [04:39] *** cpan-p6 left [04:40] *** cpan-p6 joined [04:40] *** cpan-p6 left [04:40] *** cpan-p6 joined [04:55] *** Doc_Holliwood left [05:04] *** warriors left [05:08] *** jmerelo joined [05:42] Instead of renaming the old advent calendar and losing links, I've created a new site https://rakuadventcalendar.wordpress.com/ [05:42] Anyone wants to help with managing users, themes, anything at all, just DM or send me the email address used in WordPress.com [05:56] ¦ doc: 51a68536bf | (JJ Merelo)++ | doc/Type/Proc.pod6 [05:56] ¦ doc: Reflows while revising #2632 [05:56] ¦ doc: review: https://github.com/perl6/doc/commit/51a68536bf [06:24] *** ufobat_ is now known as ufobat [06:35] *** jmerelo left [06:50] *** camelia joined [06:50] *** nine joined [06:51] *** Doc_Holliwood joined [07:04] ¦ problem-solving: hankache assigned to AlexDaniel Issue Perl 6 to Raku timeframe https://github.com/perl6/problem-solving/issues/118 [07:14] *** discord6 joined [07:15] *** sena_kun joined [07:18] *** releasable6 left [07:18] *** releasable6 joined [07:26] *** pmurias left [07:33] *** cinch left [07:38] *** domm joined [07:45] *** scimon joined [07:47] *** bbkr joined [07:56] Hi. How and when should ecosystem modules reflect name change? Will module authors get any tutorial describing files extensions, META6.json name, language name in travis-CI, etc? [08:00] Also will perl6/ecosystem repo be simply renamed to raku/ecosystem or will there be separate repository for adapted modules only? [08:04] in an ideal world someone would write a script that automatically does the lot of renaming work and creates pull requests [08:06] Maybe that someone... [08:06] is you [08:08] *** pmurias joined [08:14] *** bbkr left [08:32] i already have two yaks two shave [08:33] *** cinch joined [08:42] *** sena_kun left [08:43] *** b2gills joined [09:04] *** chsanch joined [09:17] *** pmurias left [09:22] For those asking what module authors should do, the draft advice is here: https://github.com/perl6/problem-solving/blob/master/solutions/language/Path-to-Raku.md#draft-guide-for-module-developers [09:38] jnthn: thanks :> [09:44] *** pmurias joined [09:49] *** cinch left [10:04] *** Doc_Holliwood left [10:07] *** Doc_Holliwood joined [10:14] *** cxreg left [10:40] *** pmurias left [10:43] *** cxreg joined [10:48] *** cinch joined [10:55] *** pmurias joined [11:01] *** Doc_Holliwood left [11:34] *** kawaii joined [11:38] *** diegok joined [11:41] *** shareable6 left [11:42] *** shareable6 joined [12:02] https://www.irccloud.com/pastebin/fRs8nF4z/ [12:03] I keep getting a segfault when trying to compile Rakudo on my Raspberry Pi :thinki [12:03] kawaii: maybe out of memory? [12:03] kawaii: does dmesg give a hint? [12:04] nine: sadly not, nothing in dmesg since boot [12:04] memory looks fine too [12:04] 100 mb used out of 1 gb [12:07] rakudo compile uses well over 1 GB of memory [12:09] ah that might be the issue then [12:09] let me get a reading while it compiles [12:09] kawaii: any reason you can't use an existing package? [12:10] `rakudo/stable 2018.12-5 armhf` is the latest in my repos [12:10] I suppose that will work [12:10] yup [12:10] ♥ debian :) [12:10] I've never used the packages before :D [12:10] kawaii: zef is also packaged, btw [12:11] say, why do we get a segfault for oom anyway, rather than a MVM_malloc calling MVM_panic somewhere? [12:11] ah it's packaged as `perl6-zef` [12:20] *** ggoebel joined [12:45] *** pmurias left [12:49] *** parkpirk joined [12:50] *** cinch left [12:55] https://www.irccloud.com/pastebin/PFQ4opcN/ [12:55] 🤔 [12:56] kawaii: looks like you need some additional libs [12:57] libffi, libtommath, libatomicops and libuv [12:57] *** pmurias joined [12:59] *** parkpirk left [13:00] tyil: I have all of those :( [13:00] depending on the distro you may need development packages of them [13:00] ah likely [13:01] libtoffee [13:01] libtomato? These seem food-oriented [13:01] maybe I'm hungry [13:21] *** cinch joined [13:41] New module released to CPAN! Gnome::Gtk3::Glade (0.8.8) by 03MARTIMM [13:41] New module released to CPAN! Gnome::N (0.13.8) by 03MARTIMM [13:41] New module released to CPAN! Gnome::Gtk3 (0.18.5) by 03MARTIMM [13:41] New module released to CPAN! Gnome::GObject (0.14.7) by 03MARTIMM [13:41] New module released to CPAN! Gnome::Glib (0.14.4) by 03MARTIMM [13:41] New module released to CPAN! Gnome::Gdk3 (0.14.11) by 03MARTIMM [13:41] STOP! [13:42] TY [13:42] VM [13:44] *** Doc_Holliwood joined [13:45] ZzZombo: ? [13:46] if you don't want to see updated modules on the ecosystem, you can filter/ignore *@perl6/bot/cpan-p6 [13:47] Well, not that. The bot has flooded my screen with notifications, and there was no visible end to the stream of messages. [13:47] is such as style of rakudo.js tutorial: https://gist.github.com/pmurias/7f5b9e2fffa4544f7a3be2fad8ec5608 okish? [13:49] ZzZombo: what would you propose when someone pushes a lot of modules at once? [13:49] if I add a (longer) delay, it'll just take longer, and if I start dropping messages, well, you lose updates :p [13:50] *** lucasb joined [13:59] pmurias: that would work for me (assuming this is just a start, otherwise I'd find it a bit short) [14:02] *** scimon left [14:07] *** cpan-p6 left [14:08] *** cpan-raku_ joined [14:08] *** cpan-raku_ left [14:08] *** cpan-raku_ joined [14:31] *** warriors joined [14:36] tyil: thanks [14:36] tyil: I think I'll add stuff how to add a -I when parceling and info on how to use rakudo.js under node.js [14:37] timotimo: reporting errors without allocating memory can be hard [14:37] tyil: but I don't mean it as a general introduction to using Perl 6 [14:37] ah [14:37] well Raku [14:37] it's just a "look at this" kind of thing, more than a tutorial on "this is how you can make something worthwhile" [14:38] kawaii: fun fact: since MoarVM byte code is architecture independent, you can just copy CORE.*.setting.moarvm from your desktop to the pi. No need to compile on a seriously underpowered machine. [14:38] so perl6 --version on windows is giving me This is Rakudo Star version 2019.03.1 built on MoarVM version 2019.03 implementing Perl 6.d., is this the latest release? [14:38] And compiling the CORE setting is the only part that really needs all that much memory [14:38] no release since march seems odd [14:39] yes, we were having A Time with releases this year [14:39] warriors: you're looking at rakudo star, not rakudo [14:39] rakudo star releases at different schedules than rakudo (without the star) [14:39] ah, yes, rakudo star releases; the rakudo compiler releases have not been as irregular [14:39] rakudo has a monthly release [14:39] i am very sure i installed the latest binary (msi) from perl6.org [14:40] warriors: that's rakudo star, yes [14:40] hmm [14:40] rakudo star is a distribution containing rakudo, moarvm, nqp, and a number of common modules [14:40] it's intended to be more useful for end-users, so they don't ahve to gather all parts themselves [14:40] (Y) [14:41] 👍 [14:41] if you want the montly releases, you'll have to gather all those parts yourself, and compile them yourself as well [14:42] (would not suggest as a "regular" user) [14:42] I'm actually not sure how hard it'd be on Windows, to be honest [14:47] warriors: rakudo 2019.03.1 is pretty good [14:47] warriors: surely the upcoming release this month will be even better, I think a star release will follow shortly (maybe next month?) [14:47] thanks , yes i think for me on windows, it should be good enough [14:47] tyil: intended it more to be a "look how you run you Raku code in rakudo.js" thing [14:48] but generally every other release from now on is just slightly more stable and slightly faster, in most use cases not a huge difference [14:48] I haven't tried in a while- is the entrypoint still a shell script? [14:49] the big difference will come once 6.e is released, which is going to happen in 2020 it seems [14:49] mspo: not anymore [14:49] AlexDaniel: that's good news [14:49] mspo: but that changed recently, yes, after 2019.03 [14:50] oh, that's pretty neat [14:50] AlexDaniel: is there a way to make a "packed app" like a jar? :) [14:50] mspo: why is it good news, btw? I know some benefits but wondering why it matters for you :) [14:50] I hadn't heard of that change yet, I'm also on R*'s latest :p [14:50] AlexDaniel: bsd didn't like it [14:50] ahh [14:50] the shell script [14:50] .seen patrickb [14:50] AlexDaniel, I saw patrickb 2019-10-15T22:59:32Z in #raku-dev: I'm heading off to bed. 'night everyone! [14:51] mspo: I remember people talking about packed binaries during perlcon, there was definitely some progress but I don't know the details [14:51] mspo: patrickb should know more, I think [14:51] vrurg: or maybe you? :) [14:52] AlexDaniel: I tried getting my head around the cached compile file names and stuff once [14:52] once was enough :) [14:52] still need to look into automated r* builds with gitlab-ci [14:52] so many projects ;~; [14:53] AlexDaniel: I'm only about makefile infrastructure. Builds are on Patrick. [14:53] what will happen in 6.e is this published somewhere, is there a roadmap? [14:53] m: use v6.e.PREVIEW; [14:53] rakudo-moar f3dda96a8: ( no output ) [14:53] warriors: you can do that to get the latest experimental changes today [14:53] maybe not in 2019.03 though… [14:54] warriors: but there are not a lot of changes right now, and you shouldn't depend on them anyway [14:54] m: use v6.e.PREVIEW; [14:54] rakudo-moar f3dda96a8: ( no output ) [14:54] one of the biggest changes I think is what Grammar.parse returns if it failed to parse [14:59] AlexDaniel: I would add the changes in roles support when merged. [14:59] vrurg: btw do we have a changelog for v6.e? [15:01] Not really. There is a template in the announce document. But I need to pick some time and fill it with updates. [15:01] * vrurg is afk [15:01] *** sena_kun joined [15:07] *** pmurias left [15:07] vrurg: just that alone is a lot of work :( [15:07] vrurg: I hope we can all help somehow [15:20] *** cpan-raku_ left [15:21] *** cpan-raku joined [15:21] *** cpan-raku left [15:21] *** cpan-raku joined [15:30] *** dogbert17 joined [15:33] *** timotimo[m] joined [15:33] hola [15:33] wow, that was a delay [15:40] timotimo: I don't know what's the current situation [15:41] I think it used to be that everything on matrix.org was slow, not just the bridge [15:41] *** jmerelo joined [15:41] too many users hugging a single server [15:51] ah, that makes sense [15:56] *** hankache joined [15:58] I think you meant "hogging", but it's a delightful mental image the way you wrote it :) [15:59] Good ol' hug of death [15:59] *** ufobat left [15:59] it got slashhugged [16:02] re: StackOverflow tag shift. We can simply make raku an alias for Perl6 (not the other way round), but in order to do that we need to have 25% more raku than perl6 [16:02] wouldn't that mean all of the posts still say "perl6"? [16:03] That means basically tagging everything that's been tagged with perl6 (and cro, and some other minoritary tags) with raku ASAP, and make people stop using perl6 tag from now on. Wait a bit (or a lot) until we get those 25% [16:03] Grinnz: I'm not sure you really can delete a tag, or move it. You can create synonims, though. [16:04] yes, but it's my understanding that if perl6 was a synonym for raku, all the posts would then say raku [16:04] At the current rate, we'll get to that 25% about Christmas, I'd say. [16:04] there's three concerns to think about really: what posts are visibly tagged with, what tags people put on posts, and how people can find the tag [16:04] Grinnz: be that as it may, it's still the same requirement. [16:05] Grinnz: with respect to the second concern, from today I'm deleting perl6 from all new posts and tagging them with raku. [16:05] If anyone else's got the privs, it's probably a good idea to do it too. [16:06] There are now, however, 177 questions tagged "raku" but 1201 tagged "perl6" [16:07] simply updating all existing posts and watching for use of the wrong tag in the future is an option that solves the first two things, for the third i guess you'd have to rely on the tag description [16:09] Additional problem is that you need to be a "Stack App" to be able to write to StackExchange. So adding tags is not really so simply. I wonder if scraping could do the job... [16:38] *** wildtrees joined [16:39] *** wildtrees left [16:40] *** wildtrees joined [16:59] *** robertle joined [17:07] AlexDaniel: the best help would be to remember to mark 6.e specific changes in release changelogs, commit messages. [17:22] *** mensvaga joined [17:29] timotimo: btw you might want to accept the invite to perl6/raku community on matrix :) [17:30] i don't know where to find that [17:31] nheko is my matrix client [17:33] it says “Basic communities support” [17:33] *** broquaint joined [17:35] Is this clickable? +perl6:matrix.org [17:38] AlexDaniel: BTW, have you seen comments under opennet.ru rename announce? [17:39] it's clickable, but it opens a browser [17:39] vrurg: yeah [17:40] "Refreshing" experience. ;) [17:42] vrurg: I better liked the positive spin by Andrew with an oyster and pearls [17:43] or whatever that was [17:44] this one: https://www.facebook.com/groups/raku.perl6/permalink/2453536504912694/ [17:45] *** hankache left [17:45] *** hankache joined [17:48] AlexDaniel: his one too. I could add few more connotations. In my childhood many used to pervert anything one says, so have good experience in this. But what the hell.... [17:55] ¦ doc: 4356fbfabc | ab5tract++ | doc/Type/Callable.pod6 [17:55] ¦ doc: Clarify Callable requirement [17:55] ¦ doc: review: https://github.com/perl6/doc/commit/4356fbfabc [17:59] *** asie joined [18:00] vrurg, AlexDaniel quality of some Russian-speaking communities is as low as it goes, with literaly everything being mocked / hatred there, so while it is painful to see for sure, no real need to be offended, I guess. [18:01] sena_kun: I'm not offended by the "community". I rather laugh of them as in circus. [18:02] What Andrew does worries me because he's a recognized person in the perl community. [18:04] vrurg, ah, then good for you. I am mostly sad to see the blatant ignorance without any curiosity paired with "I am the smartest mind on this planet" syndrome (not related to raku, in general). [18:06] sena_kun: so do I. Down with them, we have stuff to do. :) [18:07] vrurg, true, better to strive for writing code that can impress, than spend time arguing with people who won't change their minds no matter what you tell them. [18:08] (even though /me did not write a lot code recently) [18:09] *** hankache left [18:13] *** MasterDuke joined [18:26] *** Xliff joined [18:42] *** cognominal joined [18:43] just reinstalling zef. I can't find what I have to do for rakudo to access libs installed (where?) by zef. [18:44] Do I miss something obvious in the README.pod [18:44] *** reportable6 left [18:49] *** reportable6 joined [18:50] *** guifa joined [18:58] cognominal: do you mean `zef list --installed`? Or what? [19:00] vrurg: it's interesting how people say that they want perl to just die on a website I haven't even heard about, which also has an interface that looks like it needed a redesign for about 10 years [19:03] *** mla joined [19:04] vrurg: habr.com seems generally more civilized, at least based on previous perl 6 mentions there [19:04] vrurg: but I'm not following Russian websites at all, last time I checked habr.com was in… uh… English? [19:04] seems to be back to normal now [19:05] vrurg: I juste wanted to know where they are installed [19:05] *** jmerelo left [19:05] AlexDaniel: I just keep RSS subscribtions from like pre-historical times. They're half-dead and I always forget to unsubsribe. [19:06] is there a way to print that info from zef itself ? [19:06] cognominal: You don't need to know this unless you're going to manage the installations somehow. If you need to look into sources then better reach their repos via modules.perl6.org (modules.raku.org soon) [19:07] cognominal: the point is that modules are installed in a special format which is neither way human-readable. [19:11] vrurg, So do I need to go to #raku-dev to get a straight answer to a straight question and not be patronized ? :) [19:13] cognomial: What are you trying to do with the modules? [19:14] The problem is that module locations are non-trivial to find. At least as, far as I know. [19:14] All modules are stored on disk in a hash. [19:14] You can not find them via a nice name like Method::Also. [19:14] or Inline::Perl5 [19:16] So what is the purpose of PERL6LIB ? or soon RAKULIB ? [19:17] That's where the hashes can be found. [19:18] The compiler converts the module name to its associated hash and then looks for that hash on disk. [19:18] It will look in the directories specified by that environment variable. [19:19] cognominal: PERL6LIB does pretty exactly the same as PERL5LIB. But Raku supports different module repository implementations. In addition to the good old "bunch of module files in a directory" structure you know from Perl 5 or Python (CompUnit::Repository::FileSystem), there's also CompUnit::Repository::Installation which is a bit like an indexed data base. [19:19] Please note that I am just passing on this information to you. This is what I've learned about the module installer. I do not know the whys as to this design. [19:19] nine++ [19:20] Xliff: http://niner.name/talks/A%20look%20behind%20the%20curtains%20-%20module%20loading%20in%20Perl%206/Module%20loading%20in%20Perl%206.pdf [19:20] AKA the "please don't hate me for all those SHAs" talk [19:20] ROFL! [19:20] nine: we try :) [19:21] What are the most essential Unicode chars needed for P6? I’ve got 「 」 « » ≤ ≥ ≠ ∩ ⊍ ∪ ⊎ ∖ ⊖ ∈ ∉ ∋ ∌ ⊂ ⊄ ⊆ ⊈ ⊃ ⊅ ⊇ ⊉ Any others? [19:21] I won't blame you. I just talked of the benefit of indirection from some unique identifier on #raku-dev [19:21] cognominal: Sorry, I didn't mean to patronize. Just replayed own experience with same question. :) [19:22] guifa: that silly atomic one? [19:22] oh right! [19:22] * guifa is still super bitter about Unicode making that a by-default emoji and requiring a tag to unemoji it [19:22] ⚛ [19:23] what, you don't want emojis in your source code? ⚛️ [19:23] Grinnz: only if I put them there [19:24] die “☠️ You shouldn’t do that” [19:24] m: say ‘#’.uniprop(‘Emoji’) [19:24] rakudo-moar f3dda96a8: OUTPUT: «True␤» [19:24] ☠️ is for try/catch :) [19:24] warn “There be 🐉 here” [19:24] guifa, rakudo-moar f3dda96a8: OUTPUT: «There be 🐉 here␤ in block at /tmp/4GzzZtMOoA line 1␤» [19:24] I didn't find a way to get non-emojy version of ⚛. But this one works and I'm ok with it. [19:25] *** sena_kun left [19:26] say “⚛”; say “⚛\xfe0e” [19:26] guifa, rakudo-moar f3dda96a8: OUTPUT: «⚛␤⚛︎␤» [19:27] I wonder if the VS impacts the code [19:27] woah what's that small version [19:27] ⚛︎ [19:27] my $a = 1; say $a ⚛︎+ 2; [19:27] it's not as wide? [19:27] Oh, I missed the keywords: it's on macOS and it's about keyboard layout. [19:27] p6: my $a = 1; say $a ⚛︎+ 2; [19:27] rakudo-moar f3dda96a8: OUTPUT: «5===SORRY!5=== Error while compiling ␤Confused␤at :1␤------> 3my $a = 1; say $a7⏏5 ⚛︎+ 2;␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ statement modif…» [19:28] p6: my $a = 1; say $a ⚛+ 2; [19:28] rakudo-moar f3dda96a8: OUTPUT: «5===SORRY!5=== Error while compiling ␤Confused␤at :1␤------> 3my $a = 1; say $a7⏏5 ⚛+ 2;␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ statement modifier…» [19:28] oh oops [19:28] m: my atomicint $a ⚛=1; [19:28] rakudo-moar f3dda96a8: ( no output ) [19:29] my $a = 1; say $a⚛︎++; [19:29] p6: my $a = 1; say $a⚛︎++; [19:29] rakudo-moar f3dda96a8: OUTPUT: «5===SORRY!5=== Error while compiling ␤Bogus postfix␤at :1␤------> 3my $a = 1; say $a7⏏5⚛︎++;␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ statement mo…» [19:29] Ok, after all, it seemingly inserts non-emoji but it always displays it as emoji except for the IRC client. [19:29] * vrurg goes insane about this... [19:29] vrurg: the problem is the way that Unicode handled the stupid emoji non sense [19:30] They retroactively added emoji status to some characters [19:30] You can choose emoji display or text display by inserting VS15 or VS16 after the character [19:31] guifa: which we don't support yet, unfortunately [19:31] * guifa is making a keybaord and thought about including the VS15 (text selection) by default but it doesn’t like [19:31] *** cognomin_ joined [19:32] Handling VSes are weird. On the one hand, they’re not supposed to be semantic at all so theoretically they can be ignored. OTOH, by virtue of that, people use them to distinguish and they become significant [19:32] guifa: I had a custom layout with a bunch of unicode stuff, I didn't use most of the keys [19:33] guifa: somehow using compose key is way easier [19:33] *** cognominal left [19:33] It would be great if I had access to the hold-key-down menu that Mac OS will pop up [19:34] But it actually is possible for me to have someone type option-S for Set operators, and then type out whatever it is like “element” and then the ∈ shows up [19:35] guifa: I made a special Perl 6 keyboard a while ago with most important (as I thought) symbols accessible via Opt. Hold down would be better but don't know yet to make it. [19:37] What annoys me is that there is no way to use to use touch bar. Emojis? Yeah, plenty, use them! Serious stuff? What for??? [19:37] Agreed. The inability for writing quick plugins for the touchbar is like a giant WTF Apple [19:38] OTOH it seems like it wouldn’t be too hard to implement something for Comma. But that would definitely be bottom of the priority list [19:41] Aha. So far, I'm ok with opt-key mappings and perhaps will find some time to have hold-down work too. [19:42] I think the hold down requires programming a IME [19:42] The basic keyboard layouts are just XML files which are easy to edit [19:43] (Especially using Ukelele) [19:44] macos is the worst I have seen in my life for keyboard setup [19:45] they just took qwerty shortcuts and expect it to be ok on other layouts [19:45] ó == alt + shift + 1 + o [19:45] that's crazy stuff [19:46] I had to dive in the karabiner xml stuff to add usefull shortcuts for azerty [19:46] (luckily it's not my main os) [19:47] What? I’ve never seen that for ó. On US keyboard it’s opt-E, o, on the Spanish, it’s ´, o [19:47] exactly [19:47] qwerty is OK [19:47] other layouts: use qwerty shortcuts but don't take the new positions of the chars in accounts [19:47] -s [19:48] you know alt+tab alternative alt+` ? [19:48] the ` is on the right side of the keyboard [19:48] so you need 2 hands [19:48] wtf [19:49] utter completely garbarge [19:49] I am one of those people that can fix that (and uploaded the stuff to github :) ), but regular mac users can't. The settings are crap. [19:51] intenational qwerty with dead keys works well in Linux and Windows [19:51] so moving from azerty to qwerty there was ok [19:57] *** ufobat joined [20:02] *** Doc_Holliwood left [20:04] *** Doc_Holliwood joined [20:05] [20:21] *** chloekek joined [20:22] [20:30] *** mensvaga left [20:37] *** cognomin_ left [20:40] *** chloekek left [20:54] *** ggoebel left [20:56] *** moon_child left [21:01] *** moon_child joined [21:23] *** ggoebel joined [21:32] *** ggoebel left [21:38] *** patrickb joined [22:34] *** patrickb left [22:56] *** Xliff left [23:04] *** wildtrees left [23:06] *** MasterDuke left