🦋 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.
00:03 aborazmeh left 00:06 frost-lab joined 00:48 BenGoldberg left, Ben_Goldberg joined, Ben_Goldberg is now known as BenGoldberg 00:51 leont left 01:17 k-man_ joined 01:19 k-man left 01:55 molaf left 02:01 frost-lab left 02:04 Manifest0 left 02:06 Manifest0 joined 02:07 molaf joined 02:14 k-man__ joined 02:15 k-man_ left 02:36 squashable6 left 02:39 k-man__ is now known as k-man, squashable6 joined 02:40 BenGoldberg left, Ben_Goldberg joined, Ben_Goldberg is now known as BenGoldberg 02:55 sxmx joined 02:59 frost-lab joined 03:06 k-man_ joined 03:07 k-man left 03:28 sxmx left 04:28 bloatable6 left, bisectable6 left, statisfiable6 left, unicodable6 left, committable6 left, shareable6 left, releasable6 left, greppable6 left, squashable6 left, coverable6 left, benchable6 left, evalable6 left, tellable6 left, quotable6 left, notable6 left, nativecallable6 left, sourceable6 left, linkable6 left 04:29 sourceable6 joined, statisfiable6 joined, coverable6 joined, quotable6 joined, benchable6 joined, evalable6 joined 04:30 committable6 joined, squashable6 joined, releasable6 joined, nativecallable6 joined 04:31 greppable6 joined, bisectable6 joined, tellable6 joined, unicodable6 joined, BenGoldberg left, Ben_Goldberg joined, notable6 joined, Ben_Goldberg is now known as BenGoldberg, linkable6 joined, bloatable6 joined, shareable6 joined 04:34 BenGoldberg left, Ben_Goldberg joined, Ben_Goldberg is now known as BenGoldberg 05:08 aborazmeh joined, aborazmeh left, aborazmeh joined 05:11 BenGoldberg left, Ben_Goldberg joined, Ben_Goldberg is now known as BenGoldberg 05:31 BenGoldberg left 05:40 skids left 05:46 epony left 05:57 squashable6 left 06:00 squashable6 joined 06:22 Doc_Holliwould joined
Zero_Dogg Do I need to do anything in particular for changes to %*ENV to be propagated to the actual environment? 06:44
06:44 Altai-man joined 06:47 jmerelo joined 06:55 Sgeo left 07:00 abraxxa joined 07:03 abraxxa left, abraxxa joined 07:04 maettu joined 07:10 domidumont joined 07:15 abraxxa left 07:16 abraxxa joined 07:17 domidumont left 07:23 pecastro joined
Doc_Holliwould there is a leftover "perl6" in docs.raku.org/type/IO::Handle#Buff..._terminals 07:25
07:26 grep0r joined, grep0r left 07:29 epony joined
moritz Zero_Dogg: no 07:36
Zero_Dogg moritz: odd, if I write to %*ENV and then exec (from Native::Exec) the env in the exec-ed process doesn't have the var I wrote prior to exec 07:53
moritz Zero_Dogg: oh, I thought you were run()ing a command or so 07:55
Zero_Dogg moritz: mno, I'm actually exec()ing myself to replace the process with an updated version, so I wanted to just pass a few bits of info as an env var 07:59
timotimo yeah, you'll have to re-encode the environment yourself for that 08:08
08:13 Black_Ribbon left 08:22 BenGoldberg joined
Zero_Dogg timotimo: Okay, so, exec can take individual env vars, but then it overwrites the whole thing, and it expects key-value pairs and not a literal hash as its option. Can I somehow expand a hash into a set of key=>value arguments? 08:24
moritz .pairs ? 08:25
.kv ?
Zero_Dogg I thought so, but if I do .pairs it throws "This type cannot unbox to a native string: P6opaque, Pair", while .kv seems to pass them as command-line arguments. The signature for exec is (Str:D $file, *@args, :$nopath, *%env) 08:28
08:28 perfi joined 08:32 cpan-raku left 08:33 cpan-raku joined, cpan-raku left, cpan-raku joined
timotimo this is a native call thing, right? 08:36
oh, if it's a slurpy hash argument, you'll want to just |%foobar 08:38
Zero_Dogg oh, that did the trick, I'll have to go read about | 08:39
08:56 pecastro left, BenGoldberg left 09:01 pecastro joined 09:06 maettu left 09:38 domidumont joined 09:42 sena_kun joined 09:43 Altai-man left 10:02 domidumont left 10:10 molaf left 10:18 pecastro left 10:46 MasterDuke left 10:59 jmerelo left 11:02 AlexDaniel` left, CIAvash left, unclechu left, kiti_nomad[m] left, JJAtria[m] left, matiaslina left, suzusime[m] left 11:07 CIAvash joined 11:14 ThaEwat joined, AlexDaniel` joined, kiti_nomad[m] joined, unclechu joined, JJAtria[m] joined, matiaslina joined, suzusime[m] joined 11:18 rindolf joined 11:28 pecastro joined 11:30 maettu joined 11:33 holli__ joined 11:34 AlexDaniel` left, Doc_Holliwould left, shareable6 left, linkable6 left, notable6 left, unicodable6 left, tellable6 left, nativecallable6 left, committable6 left, evalable6 left, quotable6 left, coverable6 left, statisfiable6 left, sourceable6 left, sftp left, codesections left, finsternis left, hoelzro left, Altreus left, timlegge left, sftp joined 11:36 pecastro left, pecastro joined 11:40 AlexDaniel` joined, shareable6 joined, linkable6 joined, notable6 joined, unicodable6 joined, tellable6 joined, nativecallable6 joined, committable6 joined, evalable6 joined, quotable6 joined, coverable6 joined, statisfiable6 joined, sourceable6 joined, codesections joined, finsternis joined, hoelzro joined, Altreus joined, timlegge joined, AlexDaniel` left, AlexDaniel` joined 11:43 pecastro left 11:47 BenGoldberg joined 11:58 leont joined 12:04 donaldh joined
tbrowder .ask lizmat what does a blog (or blog site) need in its html to make it autodiscoverable for your weekly raku searches? 12:05
tellable6 tbrowder, I'll pass your message to lizmat
12:09 jmerelo joined
timotimo tbrowder: i would suggest at the very least an rss feed to be submitted to pl6anet 12:19
hm. that may want a new name post-perl6-rename, too
also, the perl6-webring
12:23 BenGoldberg left 12:33 jmerelo left 12:36 maettu left 13:00 skids joined 13:25 domidumont joined 13:27 abraxxa left 13:41 Altai-man joined 13:43 sena_kun left 13:51 frost-lab left 13:55 pecastro joined 14:05 maettu joined
[Coke] m: "â±°".uninames.say 14:07
14:16 perfi left 14:18 kylese joined
kawaii Asked in #cro originally but maybe someone here knows, I have a request body that will contain various fields, named like `url0`, `url1` and `url2`, but the frontend application allows for adding/removing as many as the user wants, is there some way to dynamically populate a `@url` array based on the content in the sample request body below? 14:18
Though, perhaps a hash is a better method of storing these blobs, as the user might go back in the UI and remove `url1`, meaning `url2` would then become `@url[1]`, kinda confusing 14:20
tbrowder timotimo: thnx, will do 14:22
14:28 kylese left, squashable6 left
[Coke] speeds up App::Uni a bit. whee 14:30
14:31 squashable6 joined 14:36 maettu left 14:40 orinthe left 14:41 orinthe joined 14:44 pecastro left
[Coke] is there a way to get a list of all unicode properties we know for a character? 14:50
m: dd "a".uniprop # this lets us ask for a specific one or get "General Category" 14:51
camelia "Ll"
moritz m: say 'a'.uniprops 14:53
camelia (Ll)
moritz hhhm
[Coke] (trying to implement "verbose" output for "App::Uni". I suppose I could (as a start) dig through rakudo and see which props we check there, and have a hardcoded list 14:55
m: dd "↉ ".uniprop("Numeric_Value_Denominator"); 14:59
camelia "1"
[Coke] ... I mean, I guess if the numerator is 0 it doesn't matter.
m: dd "⅜ ".uniprop("Numeric_Value_Denominator");
camelia "8"
[Coke] yah, ok. :) 15:00
Altreus my @promises = @url.map: Cro::HTTP::Client.get(*); # Unable to parse URI '*': Malformed scheme 15:02
Why isn't this parsed as a whatever block?
in fact why *is* it sent as a string?
surely a bare * should not be treated as '*' 15:03
Array @url = ["codimd.opusvl.com/BxNe2xLDTi293aXy...download", "codimd.opusvl.com/oqVI12O-SaG1NXSj...download"]
[Coke] I know it doesn't answer your question, but as a workaround: use a block and $_ 15:04
Altreus Does this mean it's an array with an array in it? I still don't really understand what 'say' says
when we do that it gets ''
Unable to parse URI '': Malformed scheme
[Coke] use 'dd' instead of say for more debug-helpful output.
Altreus [16:03]
[Coke] my @promises = @url.map: {Cro::HTTP::Client.get($_)} ?
Altreus oh that was a dd 15:05
sorry, I'm curious about third-party issues :)
15:11 orinthe left 15:12 orinthe joined 15:14 BenGoldberg joined 15:15 Sgeo joined
lizmat tbrowder: what timotimo said 15:18
tellable6 2020-10-09T12:05:44Z #raku <tbrowder> lizmat what does a blog (or blog site) need in its html to make it autodiscoverable for your weekly raku searches?
[Coke] question for module authors. I wrote this ages ago: github.com/coke/raku-uni/blob/mast...ni.rakumod - the version # is in the source. Is this the desired approach? Keep this number in sync with the published META6.json #? 15:19
(and I let it get out of sync since then, oops) 15:20
15:21 kylese joined, jmerelo joined 15:24 maettu joined 15:32 maettu left 15:33 aborazmeh left
tbrowder [Coke]: i have started moving to mi6 which takes care of keeping all versions for main module and meta6 in sync. only downside (for some) is it uses cpan (afik). 15:45
15:48 BenGoldberg left 16:00 Noisytoot left 16:01 Noisytoot joined 16:04 Noisytoot left 16:05 Noisytoot joined
tbrowder as an aside, i think we ought to be encouraging authors to move away from our clunky github ecosystem to cpan until we get a raku equivalent (but improved). 16:14
El_Che please don't 16:16
cpan is a good fit for an established language like Perl 16:17
a *terrible* for a small lamguage iwth a small ecosystem
cpan is just a disk for raku and it does not have all the advantages and services it has for Perl 16:18
kawaii what's clunky about ecosystem?
16:22 MilkmanDan left, MilkmanDan joined
tbrowder well, for one, all the angst and problems with proper version selection 16:37
.tell moritz just heard from apress (springer nature): the print book price does NOT include the ebook with the purchase. 16:39
tellable6 tbrowder, I'll pass your message to moritz
El_Che tbrowder: that's not github fault, but the clunky meta6 format that predates the githubs of today 16:43
jdv79 both the github and cpan "backends" have their pros and cons. neither is terrible or teriffic. 16:44
16:45 domidumont left
tbrowder ok, i f 16:48
*i didn't mean to start a flame war.
codesections jdv79: there's a lot of truth to that. But having a split also isn't great: it makes it harder for us to teach people how to publish packages, makes discoverability harder, and leads to some modules being listed twice 16:49
El_Che "we're not perl, we're raku now. Oh, and you'll find ou libraries on the most famous Perl infra" 16:50
not a lot of genius in that statement :)
jdv79 its quite a nuanced subject 16:51
tbrowder well, afterall, raku is the sister of perl
El_Che ir is, but it's muddling the water of the new language story
tbrowder: that narrative went the way of the dodo :) 16:52
tbrowder well i am a dodo now :-D
El_Che tbrowder: we are all dodos here :) 16:53
part of the charm
jdv79 i wasn't aware of the raku communities wishes to hard sever from all things Perl
El_Che I don't think there is a consensus on that 16:54
jdv79 yes, its a new name, perhaps a "new" lang, but history isn't being rewritten
16:55 approaching236 joined
El_Che spinning off is not rewriting history 16:55
while maybe trying to keep Perl and Raku in one box is
I have the feeling the communities already split
(with many people being in 2 separate communities at the same time, not to different as being part of the Go and JS community at the same time) 16:56
anyway food :)
jdv79 sure. all i'm saying is i wouldn't not use cpan just because of this "spinning off" should we not interact with YAS/TPF?...
El_Che I --hence no once-- would prefer to not have a Perl Fa;ily layer indedd 16:57
don't get me wrong, I love Perl
*Family 16:58
I just don't see an overlap anymore besides some syntax familiarities
I don't even see a big overlap in people anymore
16:59 aluaces left
El_Che (because one community is very small and the other bigger but getting smaller) 16:59
I sincerily hope Perl 7 will turn things around for Perl
jdv79 its def shaking things up over there 17:00
El_Che certainly
it's not clear if it will revitilize the community (modernizing stuff) or speed up the downfall (nothing sustantial compared to other langs) 17:01
jdv79 yup
El_Che but I think that a close association with Perl could damage Raku in both scenarios 17:02
if sucessful the return of the "Perl 6 smokeware" story
if failed, the guilty by association problem
so hence I feel that a more separate identity is warranted 17:03
ok, now food :)
codesections re YAS/TPF, I think the plan is to have a The Raku Foundation as one of the groups under YAS, which seems like a good balance to me 17:04
tbrowder any more chatter about the proposed raku version of cpan (tony-o?)
codesections we can all share infrastructure/resources without needing to share identity or confusing outsiders
17:06 MasterDuke joined
tbrowder on subject of modules, i filed an issue a few years ago about sorting modules.raku.org by name as default. i got pushback from znet because it should be by popularity. at the time, lizmat mentioned that cpan raku modules wouldn't show true popularity but with current tooling that should no longer be true, no? 17:13
sorry, it was zoffixnet who said that 17:14
jdv79 codesections: good luck with that last statement. as you just witnessed, its not as cut and dry as that. 17:15
lizmat codesections
tbrowder see modules.raku.org issue 101
lizmat codesections: talk about another shim of YAS has now been going on for almost a year :-)
tbrowder: which issue tracker ? 17:16
jdv79 tbrowder: just curious - what's stopping someone from doing a PR for that as an option or something?
tbrowder repo modules.raku.org on github/raku 17:17
jdv79 you're talking about a "sort by" select list for search results?
tbrowder github.com/Raku.org/modules.raku.org 17:18
for the record, i have never like cpans search, either 17:19
jdv79 metacpan even? not just the old search.cpan.org? 17:20
tbrowder well, i am usually too impatient (or lazy) for that...
i go to cpan.org and fire away in the search window 17:21
Grinnz that is... just metacpan's search 17:22
jdv79 unless you do it it will likely not get done:) i personally don't feel a big need for this feature.
codesections lizmat: yeah, but the past year (really, past few months) have seen the launch of the Raku Development Fund and the Raku Steering Council 17:26
lizmat true, but the latter didn't have anything to do with TPF?
codesections the RST isn't really related to YAS, I know, but it would make co-ordinating easier 17:27
lizmat I hope so!
17:33 aluaces joined
[Coke] codesections: the RDF is a rebrand/relaunch of ye olde P6DF. 17:36
17:42 sena_kun joined 17:43 Altai-man left 17:45 rindolf left 17:52 jmerelo left 17:53 rindolf joined 18:01 wamba joined 18:32 approaching236 left 18:36 approaching236 joined 18:38 telex left 18:40 BenGoldberg joined, telex joined 19:13 BenGoldberg left 19:27 donaldh left 19:34 wamba left 19:47 agentzh left, agentzh joined, agentzh left, agentzh joined
codesections weekly: www.codesections.com/blog/raku-manifesto-3/ 19:57
notable6 codesections, Noted! (weekly)
codesections [Coke]: thanks -- somehow I'd misunderstood/missremembered that (I was thinking that there was just a PDF before, not separate P5 and P6 DFs) 19:59
20:07 MasterDuke left 20:12 kylese left 20:17 Black_Ribbon joined 20:23 wamba joined 20:32 patrickb joined
patrickb releasable: status 20:48
releasable6 patrickb, Next release in ≈14 days and ≈22 hours. There are no known blockers. 0 out of 25 commits logged
tellable6 2020-10-08T07:41:07Z #raku-dev <moritz> patrickb I've given rakudo and nqp the labels; docs and roast already had them
releasable6 patrickb, Details: gist.github.com/c7c710dfb81a28c260...1dfd1b2770
20:58 niceperl joined
JJAtria[m] I've been playing around with porting HTTP::Tiny to Raku, and it's starting to take some shape. I'd love to get some feedback, particularly on whether the API needs any more changes to make it feel natural in Raku: gitlab.com/jjatria/http-tiny 20:58
I still haven't ported all the features in the original library (I haven't managed to get HTTPS to work yet, for example) but I feel it's promising 20:59
Grinnz not sure how it applies to raku, but it's important to note that `content` in the result is bytes 21:00
LWP has decoded_content which will guess based on the content-type and decode to chars if possible, HTTP::Tiny does not have that functionality 21:02
JJAtria[m] Yeah, I've gone back and forth on that... 21:04
Part of me thinks it should be simple to just make `content` always a Blob, and let the user decode it if they want a string 21:05
Grinnz it's a bit of a messy prospect, but to start with content needs to be bytes and the docs should reflect that
if you want to add something like LWP's guessing then that should be explicitly requested
21:05 rindolf left
JJAtria[m] Yeah, I've tried to stay away from making too sophisticated guesses about encodings 21:06
I feel like the heuristics in the `content` field of the request are already a bit overloaded, but I wanted to have at least one way to intuitively support stuff like multipart file uploads, etc 21:08
Grinnz Mojo::UserAgent also provides decoding based on content-type, but as a separate method as well - there always needs to be the ability to get the raw bytes
yeah perl's HTTP::Tiny doesn't do any of that itself, but there is metacpan.org/pod/HTTP::Tiny::Multipart 21:09
21:10 sena_kun left
JJAtria[m] Currently, I do it based on whether the user passed a IO::Path value: gitlab.com/jjatria/http-tiny/-/blo...ile-upload 21:11
21:12 patrickb left
JJAtria[m] But you can also set the content type to `multipart/form-data` manually and any value that has a `slurp` method will get similarly slurped 21:12
21:17 cydf joined, holli__ left 21:21 aluaces left 21:27 MasterDuke joined 21:48 wamba left
JJAtria[m] Grinnz: I've updated the code and documentation to clarify that the response will always be in bytes. Thanks for the nudge! 21:56
21:56 mowcat joined
Grinnz 👍 21:59
22:05 BenGoldberg joined 22:10 pecastro joined
JJAtria[m] codesections: reading part 3 of your manifesto, and really liking it so far. I did find a small mistake at the beginning: "why I've convinced" should probably be "why I'm convinced" 22:12
22:13 cydf left
codesections JJAtria[m]: Thanks, fixed 22:34
JJAtria[m] 🎉 22:35
22:36 niceperl left 22:39 BenGoldberg left
lucs codesections: Very interesting manifesto. 23:07
23:15 BenGoldberg joined
codesections thanks :D 23:16
23:28 grep0r joined, grep0r left 23:35 pecastro left 23:47 approaching236 left