01:00 sibl joined 01:43 ShimmerFairy left, lue joined 02:01 human-blip left 02:03 human-blip joined 02:30 lue is now known as ShimmerFairy 02:46 Sgeo left 02:50 kylese left 02:52 kylese joined 02:57 Sgeo joined 03:15 kylese left, kylese joined 03:30 human-blip left 03:32 human-blip joined 03:39 human-blip left 03:40 human-blip joined 03:42 Aedil joined 03:46 human-blip left 03:48 human-blip joined 04:44 sibl left 04:49 sibl joined 04:53 human-blip left 04:55 human-blip joined
disbot6 <shalokshalom> Raku is on the last place 🙁 github.com/niklas-heer/speed-comparison 07:33
08:02 Sgeo left 08:30 sibl left, sibl joined
Voldenet it's kinda expected, if I had computational problems I would use nativecall 08:36
and C probably 08:37
however `Are the compile times included in the measurements?` `No` 08:40
and also, the code isn't measuring jit
disbot6 <antononcube> Worse, it does not have Mathematica / Wolfram Language in the comparison. 08:42
Voldenet also, benchmark is cheating by classifying numpy code as python 08:44
…inline python could use numpy as well
SmokeMachine www.irccloud.com/pastebin/6GgqJJqC 08:45
I know that’s not related to Raku, but I’d really love to hear your thoughts on my new project: github.com/FCO/git-issue?tab=readme-ov-file 08:46
It’s a way to have issues inside a git repo without depending on the host’s issue system (like GitHub). The issues live inside the git repository itself, so you can move your repo anywhere and the issues come along with it.
It also has a React page (which I asked an AI to build) that reads the refs from the host (for now it only works with the GitHub API, but the idea is for it to support any host) and lists the issues. Ideally it should also be able to create new issues, but I’m not sure that’s possible without a backend.
This creates “a new git command” (git issue) with a few subcommands (I’ll add more soon) to manage issues.
What do you think?
Voldenet it sounds like a good way to make huge repo 08:49
and in the worst case painfully slow
08:59 dakkar joined
SmokeMachine Why? I have worked with repos with many refs and never got slow… and the issues refs are not pulled by default, only when you ask for specifically that. 09:02
Voldenet define "many", 1000? 10000? 100000? 09:07
it could be enough for projects with under 100 issues total 09:08
SmokeMachine I can let it to be worried about that when any repo using that gets more than 1000 issues… 09:09
(That would be a problem for the js) 09:10
Voldenet js gets them one by one so perhaps with just 30 it'd get very slow 09:11
SmokeMachine Yes, the js will be slower as more issues you have… and may not work passing the GitHub api limit… 09:12
Voldenet well, with more than 30 you'd get banned from github api quickly ;)
so – consider keeping issues in one file 09:13
basic one-item-per-line markup with title, some tags that would be an index
SmokeMachine I plan to make it prepare the file when you work on issues locally, that way you would have most of the issues already on the file… 09:14
Voldenet it'd still get large fast
in a lot of projects I work with, I see people write more text about the code than the actual code 09:15
SmokeMachine But it’s already added to the html file when the issue is created, it would load it fast, right? 09:16
09:17 tgt joined
Voldenet at some point you'd see huge lag in browser because of many items 09:18
09:20 abraxxa-home joined
SmokeMachine I can have arrays of issues, one array for each page, and the js would load only 1 each time 09:20
Voldenet (actually loading 200MB of json is fast btw)
09:21 abraxxa-home left 09:22 abraxxa-home joined 09:25 sibl left 09:29 melezhik joined
melezhik librasteve - pictures on Raku advent calendar posts are just beautiful 😍 09:31
09:36 tgt left
SmokeMachine librasteve: ir seems the link for HARC is wrong… 09:52
librasteve: I have a question: do you think something like Crolite could help on harc examples? I mean, it would make it easier to run the examples? 09:53
10:00 mahlay left 10:01 mahlay joined
disbot6 <librasteve> SmokeMachine: thanks! link should be OK now 10:04
<librasteve> on Crolite - not really - I think that we could move a bunch of Cro tests from t/ to xt/ so that we do not have to advice --/test on install 10:05
<librasteve> melezhik: kudos to @antononcube for that! 10:06
ab5tract That would be an amazing change for the db library that wants pg installed and running too 10:38
SmokeMachine I’m still plan to do that on Red… but never got time to have it done… if some one would like to PR that change, I would like very much… 11:20
disbot6 <librasteve> sorry to pile in, but to help people use your great code, it would really help to fix this ... right now, I have to advise zef install Red --exclude="pq:ver<5>" 11:22
<librasteve> (and it took me a while to work out that this is fairly safe option ... here are the notes I keep in my local drive gist.github.com/librasteve/e5dd78c...f39856c08) 11:24
<librasteve> so I am worried that as soon as someone wants to try the Red in HARC (hint - what colour is Rudoplhs nose?) and goes slightly off piste from my install advice, they will suffer 11:26
11:39 melezhik left 11:58 tgt joined
tbrowder g'day all. regarding putting an article on the advent as draft. would this be a workable solution: (1) create a Markdown article. (2) publish it on Github. (3) copy the rendered result, and (3) paste it to an empty draft article on the Worspress site. 12:02
disbot6 <librasteve> tbrowder: if you can do (1) -- just create a markdown article, I will happily do the rest ... 12:04
12:04 tgt left
tbrowder that's a deal, how shall i get it to you? 12:21
disbot6 <librasteve> can you put it in the GH repo (eg via PR) here, please: github.com/Raku/advent/tree/main/r...5/articles 12:22
tbrowder wilco 12:23
12:28 melezhik joined
tbrowder i'll call mine "tbrowder1.md" 12:31
SmokeMachine Yes… I need to get some time for that, or a PR… :) 12:53
Voldenet > cd Cro; mv t/* xt/*; git add -A; git commit -m "Made test fast" 12:54
what could go wrong 12:55
truth be told, people never run these tests anyway, so maybe simple sanity test "start http server, do http request to it" would be enough
and probably more useful
tbrowder ok, an empty place holder is in that spot...will post a look-see draft this pm cdt 13:45
or soon, but it's not ready for prime time yet, pm critiques welcome... 13:47
disbot6 <antononcube> The so called "full list of articles" of the Raku advent blog has only three article links. 13:48
13:53 melezhik_ joined
tbrowder my rough draft is there now 14:12
14:32 melezhik_ left, melezhik_ joined 14:38 melezhik_ left, melezhik_ joined
disbot6 <antononcube> @librasteve I scheduled my Dec 9th Raku advent post. I use "Hilite::Simple" in it, but I have not figured out how its output to be properly shown. (I.e. the output now is black-and-white.) 14:41
14:44 melezhik_ left, melezhik_ joined
disbot6 <librasteve> Voldenet: actually the start server, check page is opne I would put in xt - because eg localhost may not be configured, or firewall posrt may be closed, etc 14:46
<librasteve> @antononcube thanks - i'll take a look in an hour or so
14:50 melezhik_ left, melezhik_ joined
[Coke] antononcube - ML::AssociationRuleLearning is failing tests due to missing "Data::ExampleDatasets" 14:53
Seen in brownie testing, working on extracting a reproducable test for you. 14:54
disbot6 <antononcube> @Coke I will take a look.
14:56 melezhik_ left, melezhik_ joined
[Coke] github.com/antononcube/Raku-ML-Ass...g/issues/2 14:58
Voldenet librasteve: It's possible to deny local connections and so on, but In practice 127.1:random would cover most useful test case I think (+ you can still skip tests!) 15:05
disbot6 <librasteve> Voldenet: i am not convinced - i think the failure mode of "my firewall is set to deny all inbound ports" is quite frequent in particular for new, inexperienced users, I would recommend that zef succeeds (ie the libraries and apps install in any case), but that the cro run or equivalent fails (perhspas with some fairly prominent "have you checked this" kind of help in the Getting Started) 15:09
Voldenet Yes – deny all inbound ports, but not on loopback connections 15:11
15:14 melezhik_ left, melezhik_ joined
Voldenet a lot of things would fail without 127.1 properly working though it's possible to run tests on entirely synthetic http request I suppose 15:16
so that Cro gets tested and not tcp
15:36 melezhik_ left 15:37 melezhik_ joined 15:39 melezhik left 15:42 melezhik_ left 15:46 melezhik_ joined 15:57 avuserow left 15:59 avuserow joined 16:04 RandalSchwartz joined 16:05 RandalSchwartz left 16:16 melezhik joined 16:17 hvxgr left 16:18 hvxgr joined, tgt joined 16:23 tgt left, hvxgr left 16:25 melezhik_ left 16:35 leah2 left 16:37 leah2 joined 16:41 hvxgr joined 16:56 abraxxa-home left 16:58 abraxxa-home joined 17:13 itaipu left
Geth advent/main: 77100ed93d | librasteve++ (committed using GitHub Web editor) | raku-advent-2025/authors.md
MOve xmas day back to 25th

PS lizmat putting your round up on 24th		 17:22
17:41 dakkar left 17:50 itaipu joined 18:05 itaipu left 18:23 itaipu joined 18:33 user47384 joined 18:53 user47384 left 18:58 librasteve_ joined
librasteve_ notable6: weekly 18:58
notable6 librasteve_, 2 notes: 2025-12-03T10:10:27Z <librasteve_>: glot.io/snippets/hdjxplylcc ; 2025-12-03T14:44:59Z <librasteve_>: www.youtube.com/watch?v=LvFTyvYM-ds
librasteve_ notable6: weekly reset 19:00
notable6 librasteve_, Moved existing notes to “weekly_2025-12-08T19:00:22Z”
tbrowder i am continuing to update my Christmas/Holiday card generator system which i intend to publish part of as a free mailing list label generator to compete with Microsoft’s mail merge and its parts. if anyone wants to get a Christmas card from me send me your name and postal mail address with a subject line of “add me”. in the contents please show exactly how the address should look (including desired salutation and name, in the
format for your country. email to: <tom.browder@gmail.com>.
librasteve_ rakudoweekly.blog/2025/12/08/2025-...s-rolling/ 19:09
19:17 mehbark left, mehbark joined 19:18 mehbark left, mehbark joined 19:31 itaipu left 19:33 itaipu joined 19:35 abraxxa-home left 19:36 abraxxa-home joined 19:51 itaipu left 19:56 itaipu joined 20:07 Aedil left 20:23 abraxxa-home left, abraxxa-home joined
Geth advent/main: 0fea6545ad | librasteve++ (committed using GitHub Web editor) | raku-advent-2025/authors.md
update scheduled status
 20:37
21:49 melezhik left 21:52 abraxxa-home left 22:01 itaipu left 22:16 itaipu joined 22:20 avuserow left
Geth advent: tbrowder++ created pull request #137:
Update my article		 22:49
advent/main: 7 commits pushed by (Tom Browder)++ 22:50
22:50 phogg joined 23:16 justache left 23:22 Sgeo joined 23:29 justache joined 23:34 itaipu left

Powered by Cro and the RakuÂŽ Programming Language.

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

Thank you!