[00:02] *** sena_kun left [00:05] *** xinming left [01:32] Anyone taking Timo's place on newie duty? [01:39] *** dominix joined [01:39] hi everyone [01:39] 2019-09-28T14:05:51Z #perl6 dominix https://mobile.twitter.com/zoffix/status/1159590283627155456 https://mobile.twitter.com/zoffix/status/1171193530624221184 [01:39] 2019-09-28T14:08:03Z #perl6 dominix It seems Zoffix is up to something entirely different. I guess the best we can do is wish Zoffix all the best in this new life endeavor. Although I hope he reconsiders. 😃� [02:02] *** ToddAndMargo12 left [02:07] I am observing a weird thing, and don't know if its a feature or a bug [02:07] if I write say (5,7,-3,2).minmax [02:07] I got -3..7 [02:07] but what is this result ? [02:08] p: say (5,7,-3,2).minmax.elems [02:08] m: say (5,7,-3,2).minmax.elems [02:08] rakudo-moar 7e7aa2626: OUTPUT: «11␤» [02:08] dominix: http://docs.perl6.wakelift.de/routine/minmax [02:08] so it is evaluated as -3 ... 7 ? right ? [02:09] I would have see it as a list of two elements, a min and a max. [02:09] dominix: the doc clearly states 'returns a range'. [02:10] m: say (5,7,-3,2).minmax.minmax [02:10] rakudo-moar 7e7aa2626: OUTPUT: «(-3 7)␤» [02:10] ^ this is how you get two elems. [02:10] ho man ! what a tricky [02:11] I would have made this two different functions, like minmaxrange and minmax (pair) [02:12] Range and Supply are special cases. minmax on Range is different. [02:12] m: say (5,7,-3,2).minmax.minmax.elems [02:12] rakudo-moar 7e7aa2626: OUTPUT: «2␤» [02:12] ho god ! [02:13] m: say (5,7,-3,2).minmax.bounds [02:13] rakudo-moar 7e7aa2626: OUTPUT: «(-3 7)␤» [02:13] ^ That'd be more correct use. [02:16] thanks, I got it now, but is was a WTF sequence for me to get to it. [02:17] greppable6: minmax [02:17] vrurg, 140 lines, 11 modules: https://gist.github.com/0d58f0013202a19a5dcf84cbdd7a3369 [02:52] uzl[m]: well, another thing we can do is make sure we don't cause such pain to devs in the future [02:53] uzl[m]: by pain I mean, like, in the bum [02:54] because Zoffix was doing a lot of great work but then they'd bump into issues that were a bit out of their control [02:54] e.g. the naming thing [02:54] at the time, we didn't have any framework on working on such issues [02:55] so Zoffix did their absolute best to improve things, which got a lot of backlash because some individuals thought that Zoffix is trying to impose their solution/view on others without consensus [02:56] I mean, “consensus” because there was nothing to define what it is [03:00] ¦ problem-solving: 7af1dfc3f9 | (Fernando Correa de Oliveira)++ (committed by Aleks-Daniel Jakimenko-Aleksejev) | README.md [03:00] ¦ problem-solving: Add my name to the list [03:00] ¦ problem-solving: review: https://github.com/perl6/problem-solving/commit/7af1dfc3f9 [03:00] ¦ problem-solving: 5e9d288a3b | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | README.md [03:00] ¦ problem-solving: Merge pull request #84 from perl6/merge-mishap [03:00] ¦ problem-solving: [03:00] ¦ problem-solving: Add FCO to the list [03:00] ¦ problem-solving: review: https://github.com/perl6/problem-solving/commit/5e9d288a3b [03:01] ¦ problem-solving/master: 4 commits pushed by (Vadim Belman)++, (Aleks-Daniel Jakimenko-Aleksejev)++ [03:01] ¦ problem-solving/master: 6a1a9d9086 | Add vrurg to the list [03:01] ¦ problem-solving/master: 3b70d4006b | Remove myself from label-assignment [03:01] ¦ problem-solving/master: 0b9098d3e6 | Merge branch 'master' into add_vrurg [03:01] ¦ problem-solving/master: 6d0c078ea0 | Merge pull request #87 from vrurg/add_vrurg [03:01] ¦ problem-solving/master: review: https://github.com/perl6/problem-solving/compare/5e9d288a3bdd...6d0c078ea04d [03:08] Congrats vrurg, SmokeMachine! 🎉🎉🎉 [03:08] AlexDaniel: thanks! [04:01] *** wamba joined [04:07] *** dominix_ joined [04:11] *** dominix left [04:33] dominix_: I think the confusion is that - on a list of numbers, .minmax returns a Range... and on a Range, .minmax returns a list of (2) numbers [04:35] ideally .minmax should be consitent... but if that was the case, Range.minmax would just return itself... which isn't very useful. maybe Range.minmax should be renamed to indicate it is a different method [05:00] *** wamba left [05:09] *** jaldhar joined [05:36] elcaro: yeah [05:37] that's really weird, actually [05:37] I don't think it was supposed to be this way [06:10] *** wamba joined [06:10] *** jmerelo joined [06:28] *** sena_kun joined [06:35] *** dominix_ is now known as dominix [06:51] *** wamba left [07:03] *** xkr47 joined [07:04] *** cpan-raku left [07:05] *** cpan-raku joined [07:05] *** cpan-raku left [07:05] *** cpan-raku joined [07:18] *** sena_kun left [07:25] AlexDaniel: :) thanks! [07:26] *** jaldhar left [07:26] *** jaldhar joined [07:33] *** jmerelo left [07:37] *** stoned75 left [07:46] *** dominix left [07:57] *** ufobat joined [08:06] *** wamba joined [08:27] *** kensanata joined [08:40] *** sena_kun joined [09:08] *** sena_kun left [09:43] *** wamba left [09:52] *** wamba joined [09:53] *** daxim joined [10:02] *** wamba left [10:10] *** wamba joined [10:20] *** sena_kun joined [11:29] *** zag joined [11:43] *** Altai-man_ joined [11:44] *** wamba left [11:45] *** sena_kun left [11:55] *** ggoebel joined [12:10] *** wamba joined [12:13] *** joule joined [12:13] *** wamba left [12:31] *** xinming joined [12:34] *** Wahnberger joined [12:34] jnthn: Sorry https://github.com/jnthn/p6-io-socket-async-ssl/issues/47 :/ [12:35] *** wamba joined [12:35] can I do something to provide more usefull Info to you ( i am not the best coder, so I dunno how to dig into that isse deeply) [12:57] *** a6502 joined [13:12] *** mensvaga joined [13:16] *** Wahnberger left [13:41] I saw that the grant proposal time has come, and I'd like to know if some one think that would good to have a grant proposal to let me spend more hours developing Red... Or if doing it on my spare time is good enough (I still don’t know if I can do it… but I’d like to know what youall think about this possibility...) [13:43] * [Coke] wonders what it is about email from ToddAndMargo12 to perl6-users that causes his gmail phone client to get confused about what is read/unread in that thread so he has to restart. [13:44] *** sena_kun joined [13:46] *** Altai-man_ left [13:59] *** wamba left [14:02] *** wamba joined [14:33] *** Wahnberger joined [14:33] yo [14:34] *** lucasb joined [14:35] <[Coke]> yo [14:36] *** ggoebel left [14:37] *** zag left [14:38] *** sarna joined [14:41] yay [15:10] SmokeMachine: I think you should apply. Red is assuredly one of the gems in Raku ecosystem. [15:16] vrurg: thanks! :) [15:24] SmokeMachine, you certainly should! I think that the whole community will only win if Red will be more polished and "production-ready" - less bugs, more documentation and possibly speed would be awesome. Just make sure to write it out properly: particular acceptance criteria and so on. [15:24] sena_kun: thanks! :) [15:29] *** cfa joined [15:30] *** kensanata left [15:30] it is also a cool project to show off raku in general, but I think it has its own papercuts. :/ e.g. IIRC I wasn't able to create a model named User with sqlite when I tried it out. [15:31] `'user' is an invalid table name for driver Red::Driver::SQLite` is the error [15:36] sena_kun: the database do not accept it… you can do `model User is table {…}` [15:38] SmokeMachine, while arguably more obscure than necessary, I'd expect it to automatically do another name for me, maybe _user or something. Maybe that's just me, but I found `class User`(`model User`) natural. [15:39] and when it says to me "You can't do that" I know why I can't, but it feels like not the best experience for the end user. [15:44] *** Altai-man_ joined [15:44] sena_kun: it makes sense... [15:45] *** sena_kun left [15:46] sena_kun: But what would be worst? not being able to create the table, finding out that you could use `is table<>` for that and making it work, or creating the table an not finding it because the name of the table is different of what you would expect? [15:46] *** ggoebel joined [15:46] SmokeMachine, I'll pass your message to sena_kun [15:47] sena_kun: I think the error message should suggest using `is table<>` what do you think about it? [15:47] SmokeMachine, I'll pass your message to sena_kun [15:48] *** wamba left [15:49] *** Altai-man_ is now known as sena_kun [15:50] SmokeMachine, if you insist on not doing magic, an error message would improve the situation, I think [15:50] 2019-11-04T15:47:22Z #raku sena_kun: I think the error message should suggest using `is table<>` what do you think about it? [15:50] hey sena_kun, you have a message: https://gist.github.com/504cc5826e0700dd93527f54ef988657 [16:11] *** wamba joined [16:20] releasable6, status [16:20] Kaiepi, Next release will happen when it's ready. There are no known blockers. 248 out of 454 commits logged (⚠ 9 warnings) [16:20] Kaiepi, Details: https://gist.github.com/d0ce22eb8bb5b4c3c045e3058dc5bea0 [16:30] *** moon-child joined [16:32] *** moon-child left [16:32] *** moon-child joined [16:34] *** Kaiepi joined [16:35] *** Kaiepi left [16:35] *** Kaeipi joined [16:39] *** Merfont joined [16:40] *** Kaeipi left [16:50] *** jaldhar left [17:05] *** wildtrees joined [17:14] SmokeMachine: in my view we need (at least) 1 good and feature complete webapp/rest app framework (with oauth/openid-connect support) and 1 db interface. With these 2 a lot of simple projects could be written in raku. [17:16] :) [17:17] SmokeMachine: not talking about ORM, per se. But something likethat lets you easily do CRUD, even close to SQL. [17:18] El_Che: so you think that shouldn’t be Red? [17:18] SmokeMachine: at the moment is RED the one that probably can get there the fastest [17:19] SmokeMachine: not, I think that Red had the biggest chance to become the DBI/DBM alternative [17:19] :) [17:19] (my point is that easy-db-support-for-trivial-thing is more important that the high-end orm stuff [17:19] in my view, of course) [17:20] I thinking small webapps/rest services where the db part can be rather trivial crud [17:23] SmokeMachine: so, if you have the time, I think most people would be thrilled by your possible grant [17:23] El_Che: I think makes sense (about `easy-db-support-for-trivial-thing is more important that the high-end orm stuff`) [17:25] So, if you keep that in mind, combined with Cro, it's something very significant for the ecosystem [17:28] El_Che: confirmed [17:29] the very most users are developing small stuff..and this has to done very quickly i guess [17:31] I ma doing small rest apis in Go nowadays, but more often than not they don't need the speed, so a few projects could be done in Cro+Red. I wrote a small Cro app in the pre-Red Days, but the db part wasn't great [17:31] so I put it on hold [17:32] i am still using perl on *nix boxes but I am working on to move all that (old crappy) stuff to perl6...atm a working cro setup on windows would be great :) [17:32] *** cognominal joined [17:33] Wahnberger: use docker to develop? Less painfull than native windows [17:33] Wahnberger: there is docker (with a smallish hyperv vm transparantly hidden) and the WSL [17:33] El_Che: I’ve already started plaing with that: https://github.com/FCO/Cro-HTTP-Session-Red [17:34] Wahnberger: I wouldn't bother with native Windows except if you plan to deploy on that OS [17:34] SmokeMachine: that is great! Session management is a PITA and nowadays you always deploy on several containers [17:35] yes, it sould run on that windows box..its all small stuff, but very important for my (small) company [17:35] and only inhouse [17:37] (new Cro is out!) [17:38] Wahnberger: if I look at our environment, Linux deployment is very automated (repos, CI, orchastration, monitoring, load balancing, etc etc). The Windows part is pretty... artesanal [17:38] cro is out? [17:43] *** Altai-man_ joined [17:43] Wahnberger: https://twitter.com/croservices/status/1191408589409333249?s=20 [17:44] ah..1. Nov on Github i see [17:45] mhm..try it tomorrot to see whether the ssl issue is gone :( [17:45] *** sena_kun left [17:45] has [17:46] Wahnberger: ssl issue? [17:46] uhm..no..thats another lib [17:46] yep.. its not Cro itself: https://github.com/jnthn/p6-io-socket-async-ssl/issues/47 [17:46] *** cpan-raku left [17:47] *** cpan-raku joined [17:47] *** cpan-raku left [17:47] *** cpan-raku joined [17:48] Wahnberger: and you windows has the CA used for www.perl6.org? [17:48] Wahnberger: I have seen several places where windows admin strip CA roots out of ignorance [17:49] i have tried it with several hosts - other libs, like WWW are working well [17:49] Windows is a tricky OS for a small footprint (now they have "core" it should be easier) [17:50] Cloudflaree it seems [17:50] *** wildtrees left [17:51] have a look in mmc, certificates snap-in, computer [17:51] (if it's a regular desktop, the ca will be there) [17:51] 2 das ago the lib failed to install while testing..this has already been fixed by johnathan..but at runtime the code fails as shown in the post [17:51] (I am referring to servers, where admins something remove CAs except the one from the company) [17:52] *** wildtrees joined [17:52] Wahnberger: ah, jnthn is aware. Cool. [17:52] El_Che: I have tested all that stuff on an windows7 box too [17:53] yes he is..i have to wait now :) I would do some tests if its helpfu [17:55] i could do the job with other libs or even curl on cli but I want to use Cro cus its damn nice to work with [17:55] <[Coke]> regarding Red - is there a list of what DBs it works with? [17:56] sqlite and postgres [18:12] *** Tirifto joined [18:16] *** Merfont is now known as Kaiepi [18:37] jnthn, IO::Socket::Async::SSL's taking a very long time to run tests for someone on linux mint https://hastebin.com/teyeyodoxu.makefile [18:38] *** chloekek joined [18:40] *** wamba left [18:42] m: say (1, 2) == (2, 1); [18:42] tmtvl, rakudo-moar 7e7aa2626: OUTPUT: «True␤» [18:42] Hm... [18:44] m: say (1, 2) eqv (2, 1) [18:44] rakudo-moar 7e7aa2626: OUTPUT: «False␤» [18:44] m: say (1, 2) eqv (1, 2) [18:44] rakudo-moar 7e7aa2626: OUTPUT: «True␤» [18:45] tmtvl: eqv is for more tight structural equivalence here [18:46] Yeah, switching between languages on Monday evening is a recipe for disaster. [18:46] oh wait i thought the version they were using was 2019.03, not 2018.03 lmao [18:46] nvm then [18:46] [Coke]: what DB are you looking for? [18:57] <[Coke]> I have an interest in SQL Server and mongo. (I get that mongo would be a weird fit) [19:09] *** jmerelo joined [19:10] *** joule left [19:31] hmm [19:31] 2019-11-04T19:24:15Z #raku-dev cfa it's done by hand for the time being. Automatic deployment is yet to come... [19:31] it looks like LibraryCheck's tests are failing against rakudo HEAD [19:31] namely, the bogus library checks [19:31] anyone able to reproduce that? i've encountered on openbsd and macos [19:32] jmerelo: thanks, noted [19:32] jmerelo: i wasn't aware of process changes (i thought the existing pipeline had been transferred to the new server) [19:33] Where can I read an example of a class that has another class as an attribute? [19:33] Kaiepi: i'd recommend they try El_Che++'s repos to get a newer rakudo before any further debugging [19:33] class Equipment { has PhysicalConnections Array $physical_connections} [19:33] or something [19:34] Ah. How do I specify that a class has as an attribute an array of other classes? [19:35] has OtherClass @otherclasses [19:35] thanks [19:35] [Coke]: i can make xoos work with SQL server fairly quickly [19:36] hardest part would be making a sql server to test against [19:38] ( https://github.com/jonathanstowe/LibraryCheck/issues/4 ) [19:38] doesn't MS provide images for testing on linux? [19:38] <[Coke]> Could probably fire one up in azure... [19:38] <[Coke]> (images for testing) ah, even better. [19:39] I see that I can say: has Int $.age; has Str $.name [19:39] Is there a way for me to specify that it has to be a number? And I don't care if it's an int, or a float? [19:39] has Numeric $.num [19:39] *** wamba joined [19:39] m: say Int.^mro; say Num.^mro [19:39] rakudo-moar 7e7aa2626: OUTPUT: «((Int) (Cool) (Any) (Mu))␤((Num) (Cool) (Any) (Mu))␤» [19:39] Numeric. Who would have guessed? ;) [19:39] m: say Int.^roles; say Num.^roles [19:39] rakudo-moar 7e7aa2626: OUTPUT: «((Real) (Numeric))␤((Real) (Numeric))␤» [19:40] ah, yea that'd be even better [19:40] Nice. Learning some internals. [19:40] i'll take a look tonight [Coke] [19:42] *** Wahnberger left [19:44] MasterDuke, everything seems to work on a more recent rakudo version [19:45] *** Altai-man_ left [19:47] *** wamba left [19:48] *** jmerelo left [19:49] *** wamba joined [19:50] that's good [19:54] *** stoned75 joined [20:11] <[Coke]> tony-o: no pressure. I'm just thinking about what it would take for me to be able to use raku at work. [20:11] <[Coke]> (but it wouldn't happen quickly anyway) [20:13] *** ufobat_ joined [20:14] if i do it now it'll give you time to play around with it so i can make it as production ready as possible [20:14] (and others, at that) [20:17] *** ufobat left [20:19] and another Rakudo Weekly hits the Net: https://rakudoweekly.blog/2019/11/04/2019-44-comma-cro/ [20:20] <[Coke]> lizmat++ Thanks! [20:23] Nice, thanks! lizmat++ [20:33] *** Doc_Holliwood joined [20:33] * Doc_Holliwood says `hello`from hungary [20:34] any of you guys located in the balaton area? [20:39] Kaiepi: Probably I'd just retry it. [21:06] *** dkoi joined [21:07] *** dkoi left [21:23] lizmat++ [21:31] *** LtStaffel joined [21:36] Anyone have any ideas why the discord module https://github.com/shuppet/p6-api-discord does not work with the latest cro (off github)? [21:37] Yeah it's broken currently [21:37] We know that much [21:38] Uh something something breaking change in JSON::Fast happened originally and then some other stuff broke too [21:38] I haven't had the time or energy to look into it [21:38] Altreus: ping ^ [21:39] I'm nearly a complete noob with Raku but if there's a way I could try to help I'd like to [21:40] [Coke]: it's not that difficult to write a Red Driver... (I mean for SQL... NoSQL I don't know yet...) [21:44] *** lucasb left [21:48] *** LtStaffel left [21:52] kawaii: hiii! :) [21:52] Yes hello Alex I've been away for a while :) [21:53] It's nice to be missed [21:53] You can have a Blin tomorrow [21:53] kawaii: no need! We figured it out, I think! [21:53] kawaii: El_Che++ finished the dockerizing, maettu++ has a machine set up to run it [21:53] * kawaii vanishes forever [21:54] nooo [21:54] well, we still need to somehow run it automatically and share the results, or something like that [21:55] kawaii: as far as the release goes, technically it's ready, just waiting for moarvm release and then there's some paperwork left to do (like changelogging and stuff) [21:56] but all blockers are resolved and results from Blin seem to be clean [21:56] kawaii: so, I'm merging this? https://github.com/perl6/problem-solving/pull/83 [22:02] *** wildtrees left [22:06] *** Tirifto left [22:08] *** chloekek left [22:27] *** wildtrees joined [22:30] *** sarna left [22:38] *** sarna joined [23:00] *** AlexDaniel left [23:23] *** marcusr joined [23:26] *** cfa left [23:36] *** wamba left