🦋 Welcome to the MAIN() IRC channel of the Raku Programming Language (raku.org). Log available at irclogs.raku.org/raku/live.html . If you're a beginner, you can also check out the #raku-beginner channel!
Set by lizmat on 6 September 2022.
[Coke] On behalf of the docs team, thank! 00:09
s!
guifa_ can we define constants with RakuAST? 00:16
ugexe looks like the google links to docs all 404 now because they don't have .html at the end
docs.raku.org/type/Str vs docs.raku.org/type/Str.html 00:17
Nemokosch yes, the server is in beta stage, simply put 00:19
Geth ¦ doc-website: coke assigned to dontlaugh Issue URLS shouldn't require .html github.com/Raku/doc-website/issues/147 00:26
[Coke] ugexe: opened github.com/Raku/doc-website/issues/147
coleman thanks ugexe 00:27
Geth doc-website/rewrite-rules: 9e98814f4c | (Coleman McFarland)++ | 3 files
Add rewrite rule
 00:50
doc-website/rewrite-rules: 1d9656ba8a | (Coleman McFarland)++ | default.conf
Add root
Geth doc-website/rewrite-rules: 3c537f74a6 | (Coleman McFarland)++ | default.conf
default.conf is what we need to edit

podman run -p 8080:80 -v ./default.conf:/etc/nginx/conf.d/default.conf \
   quay.io/colemanx/raku-doc-website		 01:30
doc-website/rewrite-rules: a94b7ac7f3 | (Coleman McFarland)++ | 2 files
Add rewrites for .html; switch to caddy

This might seem like a drastic change, but it's really not a big deal, since we aren't using advanced configs anyways.
Caddy's configuration language is much saner for our use-case, and it even opens the door to api-based config.		 01:52
coleman Opened a PR for the rewrite thing 01:55
Geth doc-website/main: a12436b52a | (Coleman McFarland)++ (committed using GitHub Web editor) | 2 files
Add rewrites for .html; switch to caddy (#148)

This might seem like a drastic change, but it's really not a big deal, since we aren't using advanced configs anyways.
Caddy's configuration language is much saner for our use-case, and it even opens the door to api-based config.		 05:38
[Coke] raku.org is pointing at colabti 06:11
coleman: before I press the button, it's "new build", yes? any env vars I need to set 06:14
?
ww
grondilu Hello. Anyone wants to help me figure out what I am doing wrong in this grammar: gist.github.com/grondilu/50e1143c3...20e344f659 ? 09:58
It's supposed to parse Wolfram language, the most simple form of it : Head[arg1, arg2, ...] 09:59
Where 'Head, arg1, arg2' are atoms (identifiers or numbers) or other expressions. 10:00
It shouldn't be hard, yet I can't for the sake of me write a working grammar. 10:01
guifa_ The thing that jumps out to me is 10:43
rule expression { ……… | <expression> …… }
that has the potential to hang because it'll generate an infinite loop 10:44
Anton Antonov Also, with rule you are saying — I think — that there should be a space between Head and square brackets. 10:45
grondilu yeah Nemokosch is currently trying to explain that to me on discord 10:47
grondilu left recursion and stuff. 10:47
guifa_ How about this? 10:48
rule expression { <atom> <argument>* }
token argument { '[' ~ ']' <expression>* % [',' \s*] }
that passes all of your tests
It also passes ok Wolfram.parse('Foo[bar][arg1, arg2, Foo[bar]]'); 10:49
Which I assume is also valid
grondilu it does, but as I was discussing with Nemokosh, it makes it difficult for me to extract the Heads.
in the correct order, that is. 10:53
like say I have 'foo[one][two]' 10:54
it's an expression whose head is 'foo[one]'
I'm not sure how I can get it from the match object.
Anton Antonov You need to change your rule definition to explicitly use rule for the “head’ expression. 10:55
grondilu that hangs too. 10:58
grondilu (see second revision of gist above) 10:58
Anton Antonov See here : github.com/antononcube/Raku-Mathematica-Grammar
grondilu ah, nice. I knew someone else had to have worked on this before. 10:59
Anton Antonov The section “Grammar derivation” discusses “your issue” in point 4. 11:00
grondilu I'll check this out, thanks. 11:01
Manifest0 Hi! The search function in the docs.raku.org isn't working anymore :-( 12:46
There are some errors in the browser console: i.imgur.com/lw0Wro8.png 12:48
oh... I needed to disable the localcdn extension. 12:50
It works now
lizmat Manifest0 perhaps make an issue for that so that it can be put in a FAQ ? 12:53
Manifest0 I have to say that i like this new website and how the documention is now organized.
lizmat: sure
Voldenet Wow, those new docs look amazing 12:54
Manifest0 done: github.com/Raku/doc-website/issues/150 13:00
Nemokosch it's safe to say that the search is a WIP 13:05
it was becoming kind of urgent to actually deploy so the compromise to make was that some things won't be perfectly sorted out at the beginning 13:07
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2023/02/27/2023-...u-renewed/ 13:12
Voldenet btw, regarding the styles on the page: developer.mozilla.org/en-US/docs/W...yle_sheets 13:15
lizmat Voldenet: you'd probably want to mention that on the #raku-doc channel, or make a doc-website issue ? 13:17
Voldenet lizmat: I will use that channel, thanks
Nemokosch what's with the styles on the page? 13:18
[Coke] guesses from the URL of the weekly that the doc site note was included. ;) 13:29
lizmat [Coke]: it was :-) 13:30
rf morning folks 13:55
frost lizmat++
Geth doc/main: c7f4f08475 | (Brian Duggan)++ | CREDITS
update credits
 14:13
Voldenet Nemokosch: basically, firefox has this theme switching feature (also other browsers) that didn't even require js 14:23
tellable6 Voldenet, I'll pass your message to Nemokosch
Voldenet so dark/light skins could be switched from the menu 14:24
moritz it still exists 14:25
View -> Page Style
<link rel="stylesheet" type="text/css" href="..." title="name to be shown in the menu" /> 14:26
Voldenet I know, that's why I mentioned it :P
but it doesn't exist in chrome, which became current version of "the internet"
moritz ah, why did I read "had" instead of "has"? :-)
haven't tried it, but I'd bet Opera had it at some point as well 14:27
Voldenet your instincts told you that this feature might've been gone… :)
[Coke] yah, the themes in new docs site are manual, if we can make it automatic in firefox, that'd be great.
grondilu /j raku 14:31
[Coke] can someone open a ticket on doc-website regarding the themes? 14:33
Nemokosch it would be good to have a solution that does not make the whole theme switching useless for those who don't have (or just simply don't use) this particular browser-specific feature 14:36
prefers-color-scheme is respected iirc, courtesy of Zephyr 14:39
shmup can i see a screenshot of what y'all are experiencing 14:41
[Coke] nemokosch - yes, we won't take away existing theme functionality on chrome. 14:44
Nemokosch 🍬 14:48
Voldenet [Coke]: added the issue with that (and my findings on the topic) github.com/Raku/doc-website/issues/151 15:00
el C:\Users\jack9\Documents\raku\Super>fez upload =<< ERROR: Failed to write 67 bytes to filehandle: Bad file descriptor 15:06
help
i upgraded the windows version 15:06
[Coke] is sad that App::Uni doesn't seem to work well on windows. /me wonders if it needs some kind of font backfill for emojis. 15:35
Geth doc/main: 9975aef048 | raiph++ (committed using GitHub Web editor) | CREDITS
Fix my email address

I don't want credit, but also don't want an email address that I consider lost to hackers (many years ago) to appear to belong to me.		 15:36
doc/main: e622fdff0f | raiph++ (committed using GitHub Web editor) | CREDITS
Fix my email address

Last "fix" wasn't correct. Sorry.		 15:37
Nemokosch is there no verbose mode? 15:40
el no
[Coke] wonder if it's possibly a MAX_PATH issue. 15:42
ugexe el gatito: when you said you upgraded windows version what exactly did you do 15:44
like a patch level upgrade, 10 -> 11 upgrade, etc
el updated from windows 1507 to 1803
[Coke] what does this say: systeminfo | findstr /B /C:"OS Name" /C:"OS Version" 15:45
(my version of sw_vers for windows) 15:46
ugexe I don't see where uploading in fez would open a file handle
[Coke] (I ask because my number is an order of magnitude bigger than 1803)
ugexe: doesn't it upload a tarball?
ugexe it looks like its just passing a path as an argument to curl (which should interact with the handle, not raku itself)
[Coke] ah 15:47
el [Coke
ugexe unless "Failed to write 67 bytes to filehandle: Bad file descriptor" is a curl error
el [Coke
[Coke]: OS Name: Microsoft Windows 10 Pro OS Version: 10.0.17134 N/A Build 17134 OS Manufacturer: Microsoft Corporation OS Configuration: Standalone Workstation OS Build Type: Multiprocessor Free 15:48
[Coke] Thanks. 15:49
I personally have not used zef on windows.
ugexe github.com/tony-o/raku-fez/blob/4d...akumod#L30 removing the :out, :err from that line should output whatever curl would to stdout 15:50
[Coke] You can clone fez locally, make that mod, and either run it from the clone or install the modified version.
ugexe you could also clone it locally, make that change, and then run `RAKULIB=/path/to/cloned/repo fez upload ...` I think 15:51
it might also help if it was ran with --ll-exception, but alas there is no env flag for that so I think you'd have to do this little trick of loading the CLI as a module: `raku --ll-exception -MFez::CLI -e '' upload ...` 15:53
[Coke] oh, I've been running irssi on a remote, old mac mini, and ssh'ing in and running tmux from various clients. using windows terminal (available from MS store) is a huge quality of life improvement when sshing from windows. unicode support, and links in irc are clickable! 15:54
ugexe winget is pretty cool too 15:55
i use it to install various packages (similar to chocolately)
liek visual studio
now if rakubrew could be installed via winget... 15:56
[Coke] 👍 15:57
Nemokosch well, what needs to be done for that? 16:04
Geth doc-website/clean-up-caddyfile: c73c8ee592 | (Coleman McFarland)++ (committed using GitHub Web editor) | Caddyfile
Clean up Caddyfile

Remove unused "named" matcher rule. Fix spacing.		 16:06
el no same error message PS C:\Users\jack9\documents\raku\Super> fez upload =<< ERROR: Failed to write 67 bytes to filehandle: Bad file descriptor 16:08
Geth docker/2023.02: 2f4a2aa423 | (Daniel Mita)++ | 5 files
Bump to 2023.02
 16:10
docker: m-dango++ created pull request #53:
Bump to 2023.02		 16:11
[Coke] docs.raku.org/type/Int now working in prod. 16:19
Nemokosch I haven't seen this footer before, looks cool 16:21
Geth ¦ doc-website: dontlaugh self-assigned How to test changes to Tooling without affecting production? github.com/Raku/doc-website/issues/152 16:51
tonyo el gatito: do you have a non-typical tmp dir setup? 16:55
Geth ¦ doc-website: coke assigned to dontlaugh Issue Proc::Async is generated as Proc/Async github.com/Raku/doc-website/issues/149 16:59
Geth doc-website/reflog_defaults_to_doc_without_value: f833cb9f21 | finanalyst++ | Website/plugins/git-reflog/add-ref-hash.raku
make reflog fail softly if no reflog obtained
 17:20
doc-website/caddycolons: 50c6cfe515 | (Will Coleda)++ | Caddyfile
Support old-style Proc::Async URLs

go live only supported /Proc/Async, this restores the old functionality - many links to the docs depend on this style.
Part of #149		 17:24
doc-website/main: f05bc2f743 | (Will Coleda)++ (committed using GitHub Web editor) | Caddyfile
Support old-style Proc::Async URLs (#158)

go live only supported /Proc/Async, this restores the old functionality - many links to the docs depend on this style.
Part of #149
[Coke] Oof, so many lines. 17:27
Geth doc-website/clean-up-caddyfile: f05bc2f743 | (Will Coleda)++ (committed using GitHub Web editor) | Caddyfile
Support old-style Proc::Async URLs (#158)

go live only supported /Proc/Async, this restores the old functionality - many links to the docs depend on this style.
Part of #149		 17:29
doc-website/clean-up-caddyfile: 7380f0ff27 | (Coleman McFarland)++ (committed using GitHub Web editor) | Caddyfile
Merge branch 'main' into clean-up-caddyfile
doc-website/main: c73c8ee592 | (Coleman McFarland)++ (committed using GitHub Web editor) | Caddyfile
Clean up Caddyfile

Remove unused "named" matcher rule. Fix spacing.		 17:32
doc-website/main: 7380f0ff27 | (Coleman McFarland)++ (committed using GitHub Web editor) | Caddyfile
Merge branch 'main' into clean-up-caddyfile
doc-website/main: 0f73ed75aa | (Coleman McFarland)++ (committed using GitHub Web editor) | Caddyfile
Merge pull request #154 from Raku/clean-up-caddyfile

Clean up Caddyfile
tbrowder .seen ufobat 17:37
tellable6 tbrowder, I saw ufobat 2020-10-08T09:28:19Z in #raku: <ufobat> can one write libs in raku, compile them into a jar and use it in java/kotlin projects?
[Coke] moving doc-website & doc geth updates to just raku-doc...
[Coke] ... per request, leaving doc here. -website is moved 17:39
tonyo looks good 17:40
tbrowder anyone know of @ufobat's status? (Time::Crontab) 17:43
lizmat they seem to have become a lot less active on Github since 2020 17:51
cfa morning 18:11
lizmat cfa o/ 18:27
El_Che hello 18:28
cfa hey lizmat, El_Che
[Coke] docs.raku.org/type/Proc::Async and realted URLS should be working again. 18:33
*related
cfa looks like there's a regression in 2023.02, found via a failing doc examples test 18:36
m: multi sub infix:<||=>($a, $b) is equiv(&infix:<+=>) { $a || $b }; my $foo = 0; $foo ||= 1;
camelia ===SORRY!=== Error while compiling <tmp>
No such method 'op_props' for invocant of type 'Block'
at <tmp>:1
cfa bisectable6: multi sub infix:<||=>($a, $b) is equiv(&infix:<+=>) { $a || $b }; my $foo = 0; $foo ||= 1;
bisectable6 cfa, Will bisect the whole range automagically because no endpoints were provided, hang tight
cfa, Output on all releases: gist.github.com/2fee0c5bf19d849b59...462c98261e 18:37
cfa, Bisecting by exit code (old=2022.12 new=2023.02). Old exit code: 0
cfa, bisect log: gist.github.com/0f5e79e87f0273fbd1...ea5f7e19d1 18:39
cfa, Bisecting by exit code (old=2016.12 new=2017.01). Old exit code: 1
cfa, bisect log: gist.github.com/c3c1aac555c8f401c3...5bf8073dcb
cfa, (2017-01-09) github.com/rakudo/rakudo/commit/a7...cbbf8694a6
cfa, Output on all releases and bisected commits: gist.github.com/56a6497914763c6a0e...d696228cfd
shmup cfa: i have things to fix. sorry if you see weird output from the irssi script. aka an extra > throws off the grep. gotta fix that 18:44
aka i.imgur.com/iF2GM2A.png 18:45
cfa shmup: sure thing 18:49
shmup should be better now cfa. if ur perl is strong, maybe you'll see a prob idk. github.com/shmup/irssi-scripts/blo...mer.pl#L33 19:11
just lazied it
cfa bisectable6: multi sub infix:<||=>($a, $b) is equiv(&infix:<+=>) { $a || $b }; my $foo = 0; $foo ||= 1; say $foo 20:11
bisectable6 cfa, Will bisect the whole range automagically because no endpoints were provided, hang tight
cfa, Output on all releases: gist.github.com/76206dee2e26324b4b...ccad04701e 20:12
cfa, Bisecting by exit code (old=2022.12 new=2023.02). Old exit code: 0
cfa, bisect log: gist.github.com/eb01bf61dcb88f5a7a...9425b60cc8 20:13
cfa, Bisecting by exit code (old=2016.12 new=2017.01). Old exit code: 1
cfa, bisect log: gist.github.com/0574ca270d5d9b027e...f7d3fae0d8 20:14
cfa, (2017-01-09) github.com/rakudo/rakudo/commit/a7...cbbf8694a6
cfa, Output on all releases and bisected commits: gist.github.com/9516f200dc2befcf57...c3743e9935
cfa hopefully captured in github.com/rakudo/rakudo/issues/5221 20:17
siberia-man hello, everyone 21:07
Nemokosch 👋 21:08
siberia-man Would you be so kind as to let me know the meaning of the "@@" in raku, please? And where can I read more about it and possibly other magics? 21:10
Nemokosch can you point at the context where it appeared? 21:12
Geth raku.org: zag++ created pull request #180:
Update index.html		 21:16
rf How can I append a string to a buf 21:35
Buf.append('str'.ords); 21:36
I feel like that kind of isn't great
tonyo m: my Buf[uint8] $b .=new; $b.append('str'.encode); $b.decode.say; 21:46
camelia str
ugexe $buf ~= "str".encode
tonyo siberia-man: in regexes? 21:47
tellable6 tonyo, I'll pass your message to siberia-man

Powered by Cro and the Raku® Programming Language.

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

Thank you!