🦋 Welcome to the IRC channel of the core developers of the Raku Programming Language (raku.org #rakulang). This channel is logged for the purpose of history keeping about its development | evalbot usage: 'm: say 3;' or /msg camelia m: ... | Logs available at irclogs.raku.org/raku-dev/live.html | For MoarVM see #moarvm
Set by lizmat on 8 June 2022.
02:27 MasterDuke left 03:11 MasterDuke joined
ab5tract Anyone know the specifics of why we don’t have `longdouble` as a native type? 10:17
10:20 sena_kun joined
lizmat is that the same as a num64 ? 10:23
or an int128 ? 10:24
Geth rakudo/main: eb032f0364 | (Márton Polgár)++ (committed using GitHub Web editor) | tools/templates/L10N/HU
Hungarian localisation of core (#5437)
10:41 sena_kun left
Geth rakudo/main: 3827870b24 | (Elizabeth Mattijsen)++ | tools/templates/L10N/HU
RakuAST: tweaks in the HU localization

  - add missing entries
  - fix some disabled entries that had German in them
  - enable some translated entries that were disabled
rakudo/main: 7584bd1c3e | (Elizabeth Mattijsen)++ | 5 files
RakuAST: make HU localization ready and tested
rakudo/main: 1604909bf1 | (Elizabeth Mattijsen)++ | 3 files
RakuAST: parse "use variables :U" and friends correctly

  - does not actually activate that feature yet
  - also make sure in other places to look at the Setting type
rakudo/main: 4f17fb5205 | (Elizabeth Mattijsen)++ | 4 files
RakuAST: introduce helper method to obtain setting types

Instead of repeating the same boilerplate over and over again
tbrowder__ ref bad published modules (mea culpa): how about some kind of repo like fez/zef, where an author with an account can publish bad modules ids (name, auth, ver). then, normal zef operation would ignore any module listed there. raku.land would do the same. options for both could enable full search if need be. 12:08
lizmat this is basically one of the things "supersedes" was intended to be doing 12:10
and producing a list for that was shot down :-(
tbrowder__ we could call the repo something like "NOTAMs" (from the aviation world) "notice to all module users"
the problem with supersedes to me is it would take more processing than just a database sql like my thought 12:13
one table with all modules, one table with bad modules, exclude the bad 12:15
using postgres it should be a breeze for our db experts 12:17
ugexe if anyone can modify an index for an item/distribution they don't actually own then we are just going back to p6c type insecure namespacing 13:24
it is probably ok for something like raku.land to use a list like that, but module installers shouldn't 13:26
that being said, if i ran raku.land i still wouldn't use it 13:27
raku.land probably needs to figure out a way to have user accounts, and a way for those user accounts to prove they own a specific zef:foo username (a naive way would be uploading a do-nothing distribution with some token that raku.land could fetch and check for), and then have a control panel that allows user to e.g. hide their own distributions on raku.land only 13:29
note that such a process would not allow hiding the distribution from a module installer resolving that namespace though 13:31
El_Che hi 13:37
tbrowder__ i’m talking about some way the owner could be proven via fez, and he would be the only person who could add it to the do-not-use list 13:58
15:15 El_Che left
japhb ab5tract: long double isn't a portable size, and it might expose some assumptions we have as well -- it would be equivalent to either num80 or num128, for architectures that support it at all, but obviously num80 isn't power-of-two sized, some architectures don't support anything bigger than num64, and it can be rather surprisingly slow when you can't output actual native code for it. 15:43
None of that would *prevent* us from supporting the size, but it certainly reduced the push to do so. 15:51
15:55 El_Che joined
El_Che (rakudo-pkg 2023.11 was released: github.com/nxadm/rakudo-pkg/releas.../v2023.11) 15:55
18:56 sena_kun joined 19:10 AntonOks joined
AntonOks Hi. I've copied over the new 2023.11 Rakudo Star release files to "rakudo.org". 19:16
I also logged in there and ran "update-rakudo.org.sh".
tellable6 2023-11-16T17:17:19Z #raku <andinus> AntonOks: should be fixed now
AntonOks Still the new release and files are not shown.
Can someone look into "rakudo.org" and check why the website is not showing updated releases anymore?
``` 19:17
rakudo.org@lavm-perl6infra-1:~$ ls -la public_html/downloads/star/*2023*11*
-rw-r--r-- 1 rakudo.org rakudo.org 30218811 Nov 22 18:54 public_html/downloads/star/rakudo-star-2023.11-01.tar.gz
-rw-r--r-- 1 rakudo.org rakudo.org      228 Nov 22 18:54 public_html/downloads/star/rakudo-star-2023.11-01.tar.gz.asc
linkable6 (2008-02-13) github.com/haytham43/ltk/commit/30218811bc Allowed multiple types of "choice" parameters to exist in a single llrp parameter.
AntonOks -rw-r--r-- 1 rakudo.org rakudo.org      922 Nov 22 18:54 public_html/downloads/star/rakudo-star-2023.11-01.tar.gz.checksums.txt
-rw-r--r-- 1 rakudo.org rakudo.org 38425454 Nov 22 18:54 public_html/downloads/star/rakudo-star-2023.11-01-win-x86_64-msvc.msi 19:18
-rw-r--r-- 1 rakudo.org rakudo.org      228 Nov 22 18:54 public_html/downloads/star/rakudo-star-2023.11-01-win-x86_64-msvc.msi.asc
-rw-r--r-- 1 rakudo.org rakudo.org      897 Nov 22 18:54 public_html/downloads/star/rakudo-star-2023.11-01-win-x86_64-msvc.msi.checksums.txt
19:22 AntonOks left 19:34 notna joined
ab5tract japhb:thanks for the explanation 19:53
20:35 notna left 22:18 vrurg joined 22:20 vrurg_ left 23:00 sena_kun left