02:30 nine left, nine joined 05:16 librasteve_ left 10:54 lizmat left, lizmat joined
[Coke] librasteve - thoughts on adding a link to an asciinema recording showing the REPL from raku.org? 15:37
15:51 librasteve_ joined
librasteve [Coke]: great idea! crag is my pastime and i have a bunch of asciinemas done , "$day-job" is (i) Air start, (ii) Air stubs (I have the design but need a block of time), (iii) get craking on sime DSL content 15:59
on the crag side, you are welcome to take a look here github.com/librasteve/raku-App-Cra...main/demos (asciinema play -q -s 2 demoN.cast is the cmd) --- my idea is to make Air::Plugin::Asciinema so you can easily play .cast files out of the website static files 16:01
oh and (iv) is vendoring the 3rd party css / js files 16:04
[Coke] nifty 16:08
iv?
patrickb 4 16:22
[Coke] ... missed the fact that it was part of a list. 16:23
timo what kind of infix operator is (iv)? ;) 16:43
[Coke] m: (iv) 16:46
camelia ===SORRY!=== Error while compiling <tmp>
Undeclared routine:
iv used at line 1. Did you mean 'kv'?
[Coke] m: (ā…£)
camelia WARNINGS for <tmp>:
Useless use of constant integer ā…£ in sink context (line 1)
timo it's the undeclared kind 16:51
nemokosch omg 16:52
timo (but also you put those in term position rather than infix, so ...) 17:01
ab5tract That isnā€™t pleasant to see 20:21
m: say (IV) + (III) 20:24
camelia ===SORRY!=== Error while compiling <tmp>
Undeclared names:
III used at line 1
IV used at line 1
ab5tract m: say (IV) 20:25
camelia ===SORRY!=== Error while compiling <tmp>
Undeclared name:
IV used at line 1
ab5tract Soā€¦ what?
20:29 shareable6 left, releasable6 left, committable6 left, evalable6 left, bisectable6 left, nativecallable6 left, notable6 left, tellable6 left 20:32 nativecallable6 joined, shareable6 joined, tellable6 joined, committable6 joined, notable6 joined, bisectable6 joined, evalable6 joined, releasable6 joined 20:44 nativecallable6 left, shareable6 left, tellable6 left, committable6 left, notable6 left, bisectable6 left, evalable6 left, releasable6 left, releasable6 joined, evalable6 joined, bisectable6 joined, notable6 joined, committable6 joined, tellable6 joined, shareable6 joined, nativecallable6 joined
timo how do you mean? 20:55
m: say ā…¢ + ā…£ 20:56
camelia 7
librasteve huh? 21:00
m: say IX 21:01
Raku eval Exit code: 1 ===SORRY!=== Error while compiling /home/glot/main.raku Undeclared name: IX used at line 1
librasteve m: say I + IX 21:02
Raku eval Exit code: 1 ===SORRY!=== Error while compiling /home/glot/main.raku Undeclared names: I used at line 1 IX used at line 1
nemokosch they are separate characters
ā…£ vs IV 21:03
librasteve m: say ā…¢.uniname
Raku eval <control-0003>
librasteve m: say ā…¢.uniprops
Raku eval (Nd) 21:04
librasteve lol
timo that gives you the properties of the number 3 literally 21:05
you need to give it a string with the character in it 21:06
m: say "ā…¢".uniname
camelia ROMAN NUMERAL THREE
nemokosch this can't be helped, although it's not a nice "feature"
timo would you argue that only arabic numerals should work in code? 21:08
or rather
only ascii digits 0 through 9
since there are arabic numerals outside of these ten as well
nemokosch that argument would have a long way 21:11
unicode-related features shouldn't be retrievable on numbers for sure, if numbers have multiple string representations
and some numbers will, either way
timo oh, you were talking only about 1234.uniname giving the name of unicode codepoint number 1234 21:12
rather than giving the names of arabic numerals 1, 2, 3, and 4 21:13
we do have allomorphs that allow us to keep the source string for a number around 21:14
nemokosch oh allomorphs are even worse 21:15
timo m: my $value = <0x55>; say $value; say +$value;
camelia 0x55
85
nemokosch a nice illustration of why the diamond problem is called a problem
timo i'm still not sure i understand what your position is 21:16
nemokosch that there should be way less magic about treating numbers and strings interchangably
no unicode-related methods on numbers and surely no allomorphs, god forbid 21:17
timo would you say 5.uniname should go but uniname(5) can be kept?
nemokosch no, surely not, no need for more EVAL kind of inconsistencies 21:18
timo i'm not okay with there being no direct way to get the name, or properties, of a unicode codepoint by its number 21:19
having to uniname(chr(6)) seems like BS
nemokosch why? 21:20
there is no "its number"
there is no good reason to write "what are the unicode properties of this non-string"
timo you've never iterated over a range of codepoints, or gone through a list of codepoints? 21:21
nemokosch how do codepoints come into play when we're talking about numbers? 21:22
timo codepoints are literally numbers
nemokosch and numbers aren't codepoints 21:23
timo of course not every 84 is a T
nemokosch then why would you want it to turn into that implicitly? 21:24
timo but when you're already asking "what's the unicode codepoint name for 84" it's quite obvious that that's what you're refering to
nemokosch no it's not, honestly
timo i guess we're just disagreeing then *shrug* 21:27
nemokosch for those who expect this, implicitly turning a list of integers into a string might also make sense 21:28
but why optimize for them by complicating the language and obfuscating the logic? we can agree that an explicit coercion or different sub name would be clear always 21:29
japhb As a data point: I *frequently* want to use a small uint as a codepoint, and I need such operations to be fast as heck. Please do not make this slow. 23:57
nemokosch it's not about making it slow, it's about making it explicit. These magic conversions have certainly nothing to do with speed. 23:59

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!