00:00 jmcgnh left 00:03 Guest3992 left
tbrowder antononcube: i have worked witha 00:09
ray tracing. would that help with the maze solution? 00:11
intersections of rays in a plane with known planar figures (curved or straight, connected or disconnected) 00:15
00:31 LainIwakura joined 00:35 LainIwakura left 00:54 jrjsmrtn joined 00:57 hvxgr joined 01:04 jmcgnh joined 01:07 ilogger2 joined 02:30 kylese left, kylese joined 03:12 jgaz_ left 03:15 kylese left, kylese joined
disbot7 <aruniecrisps> @jubilatious1_98524 here's what i wrote, note that it only works for one character glot.io/snippets/hea42gkiew 03:48
Voldenet sadly it's limited to what unicode knows to be decomposable 04:01
m: "ł".NFD.list[0].chr.say 04:02
camelia ł
Voldenet it's clearly l with a slash
disbot7 <aruniecrisps> @voldenet yeah I'm not sure why that is 04:06
<aruniecrisps> For my use case it works fine because that's just how the Indic languages work, they're all decomposable
Voldenet unicode in general loves theoretical linguistics, which is correct
but for practical ones you might want to use things like metacpan.org/pod/Text::Unidecode 04:07
though in general you can't translate unicode to ascii sadly, modules like above usually handle just a subset and do really wild stuff, like translating 十 to + 04:16
transliterating, even 04:17
04:33 lichtkind_ joined 04:36 lichtkind left
guifa the L with stroke doesn't really decompose because there's no equivalent diacritic 05:09
the stroke combining diacritics aren't actually the same thing so it's not a decomposition candidate
05:50 Aedil left
grondilu m: role R[::T] { has %.h{T} }; class A does R[Str] {} 05:59
camelia MoarVM panic: Memory allocation failed; could not allocate 131072 bytes
grondilu m: my %h{Str} = foo => "bar"; 06:01
camelia ( no output )
grondilu m: my %h{Str} = foo => "bar"; print %h
camelia foo bar
06:04 Aedil joined 06:35 dg left