|
This channel is intended for people just starting with the Raku Programming Language (raku.org). Logs are available at irclogs.raku.org/raku-beginner/live.html Set by lizmat on 8 June 2022. |
|||
|
01:51
MasterDuke joined
02:17
teatwo joined
02:19
teatime left
03:31
Heptite left
08:05
p6steve left
09:21
habere-et-disper joined
|
|||
| habere-et-disper | I was expecting zero: | 09:23 | |
| m: say ().min | |||
| camelia | Inf | ||
| Nemokosch | Imo zero is a much worse expectation than Inf but even Inf is troublesome | 09:25 | |
| github.com/Raku/problem-solving/issues/354 | 09:26 | ||
| habere-et-disper | (y) | 09:27 | |
| Nemokosch | I think these corner cases are usually defined (in math as well) in a way that you can apply the essential operation even between the "zeroeth"/default element and the first element added | 09:29 | |
| so the product of 0 numbers is 1 because 1*x will be the "product" of x | |||
| the sum of 0 numbers is 0 because 0+x will be the "sum of x" | 09:30 | ||
| and similarly the minimum of 0 numbers is positive infinity because the smallest of positive infinity and x will be the "smallest of x" | |||
| lizmat | and vice-versa for .max | 09:31 | |
| m: say ().max | |||
| camelia | -Inf | ||
| Nemokosch | the problem I see here is rather that 0 somethings cannot be assumed to be 0 numbers in Raku | ||
| strings can be ordered but the smallest of zero things is most definitely not positive infinity; arguably something like that doesn't even exist, the abstraction cannot be extended | 09:32 | ||
| tbrowder__ | hi, if i have a sub like this: "sub foo(Str $s?, :$kern) {...}" and inside the curlies i call another sub, say "bar", with the same signature, how can i write that bar call inside foo? i read the docs, and i have done it successfully before with other subs, but am not certain how to handle the optional $s | 10:47 | |
| Nemokosch | Does something like foo(|args = :(Str $s?, :$kern)) not work? Not sure about the exact syntax but you get the idea... | 10:54 | |
| In particular maybe it should be without the = and the colon | 10:55 | ||
| tbrowder__ | m: sub b($s?, sub f($s?, :$k) { bar | 11:04 | |
| camelia | ===SORRY!=== Error while compiling <tmp> Invalid typename 'sub' in parameter declaration. Did you mean 'Sub'? at <tmp>:1 ------> sub b($s?, sub⏏ f($s?, :$k) { bar |
||
| tbrowder__ | arg, i'll practice at home some more. thnx | ||
| i got it working | 11:32 | ||
| i'm confusing working with multi methods, using a single signature for different subs works fine | 11:41 | ||
|
11:41
habere-et-disper left
|
|||
| Nemokosch | hm, I'm not sure what the goal was | 11:47 | |
| but in case anybody is interested: sub funky(|all (Str $optional?, :$named)) is valid, after all, and then you can refer to the whole capture as all, or pass it to another function as f(|all) | 11:48 | ||
|
12:26
jgaz left,
jgaz joined
|
|||
| tbrowder__ | then is |c the same thing (i.el., the bare text after the | is not unique)? | 13:46 | |
| iow, it could be |a or |foobar | 13:47 | ||
| Nemokosch | what does iow mean? | 13:50 | |
| lizmat | in other words | 13:51 | |
| tbrowder__ : yes, the "c" is a sigilless identifier | |||
| could be "a" or "foobar" | 13:52 | ||
| tbrowder__ | great, thanks to you both | 13:53 | |
| sorry, "iow" = | |||
| = in other wods | 13:54 | ||
| *words | |||
| Nemokosch | 👍 | 14:00 | |
|
14:16
tea3po joined,
tea3po left
14:17
tea3po joined
14:19
tea3po left,
teatwo left,
tea3po joined
|
|||
| tbrowder__ | back, eaample Nemokosch is using is not what i originally showed. the bar sub is independent and defined outside foo. | 14:46 | |
| but their signatures are the same (not their definitions). os, is there a clever way to pass the args other than just repeating them in the call | 14:47 | ||
| lizmat | no, there currently isn't really | 14:48 | |
| tbrowder__ | e.g., sub foo($s?, :$kern) { bar($s, :$kern); } | 14:49 | |
| lizmat | well, what Nemokosch showed | 14:51 | |
| " but in case anybody is interested: sub funky(|all (Str $optional?, :$named)) is valid, after all, and then you can refer to the whole capture as all, or pass it to another function as f(|all)" | 14:52 | ||
| tbrowder__ | ok, that's what my tests showed. however, i have had probs unpacking foo's args to pass to a multi | ||
| say i have "multi foo($s?, :$kern){}" and "multi foo($s, $a?, :$kern)" | 14:55 | ||
| are those valid multi signatures? | |||
| if so, how do i call on multi from the other? | |||
| *one | 14:56 | ||
| ok, now i see that construct Nemokosch passed answers the second part of my example. i'm not sure that's in the docs that clearly | 15:00 | ||
| but how about the first part, is it a valid multi? | 15:01 | ||
| (of my second example) | |||
| given "multi foo($s?, :$kern) {...}" and "multi foo($s, $a?, :$kern) {...}", are they valid multi signatures? | 15:03 | ||
| Nemokosch | how are you planning to disambiguate calls with one positional argument? | 15:07 | |
| tbrowder__ | that answers the question then: it's not a valid multi example | 15:16 | |
| thank you Nemokosch and lizmat for helping me understand signatures much better! | 15:17 | ||
| Nemokosch | I don't know what will happen, to be honest; all I know is that it seems conceptually problematic | 15:19 | |
|
15:28
Heptite joined
15:58
habere-et-disper joined
16:49
habere-et-disper left
16:57
jumpnbrownweasel joined
17:05
tea3po left
20:16
deoac joined
20:41
jumpnbrownweasel left
21:20
teatime joined
|
|||
| tbrowder__ | yes, i expected that | 21:47 | |
|
22:44
Ebudae joined
22:47
Heptite left
|
|||