00:03 guifa joined 02:49 hulk joined, kylese left 03:15 hulk left, kylese joined 04:51 Aedil joined 04:52 Aedil left 04:58 Aedil joined 07:51 lizmat joined 08:27 sena_kun joined 09:22 lizmat left 09:43 Aedil left 10:58 Sgeo left 12:41 msiism joined
antononcube I only see “Net::IP::Parse” and ”IP::Addr” in raku.land. 13:15
Did you mean the former? 13:17
guifa whoa I missed IP::Addr 13:43
13:51 perlbot left 13:52 simcop2387 left
timo github.com/MoarVM/MoarVM/pull/1866 can i get some input from users regarding this? it's about the behaviour of Supply.interval and anything that uses ThreadPoolScheduler's $*SCHEDULER.cue with the "repeat on interval" feature. currently we get relatively even timings between ticks but always a little bit more than you ask for. with this pull request you'll get less even distances between 13:59
consecutive ticks, but the distance from an idealised "start time + interval * i" doesn't increase nearly as much over time
guifa hates that Nil throws an error on stringification
.?Str is basically an impossibility 14:00
timo m: say Nil.gist
camelia Nil
timo not an option for your case?
guifa my $a = Nil; my $b = $a.?Str.?parse-base(16) 14:01
m: my $a = Nil; my $b = $a.?Str.?parse-base(16)
camelia Use of uninitialized value $a of type Any in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something meaningful.
in block <unit> at <tmp> line 1
guifa my $a := Nil; my $b = $a.?Str.?parse-base(16)
m: my $a := Nil; my $b = $a.?Str.?parse-base(16) 14:02
camelia Use of Nil in string context
in block <unit> at <tmp> line 1
guifa the whole point of .? is that I should just end up with Nil at the end of that
right?
m: my $a := Nil; my $b = $a.?Str 14:03
camelia Use of Nil in string context
in block <unit> at <tmp> line 1
guifa oh wait no I dumb 14:04
.? is for if the method exists, not if the object exists
which would normally be the same with Nil, except we do have a method on Nil for Str
timo of course you can "quietly" 14:05
do not print quietly into that good method call
m: sub do-it($a) { my $b = .Str.parse-base(16) with $a; say $b.raku; }; do-it(Nil); do-it("99"); do-it("dog") 14:07
camelia Any
Cannot convert string to number: malformed base-16 number in 'd<HERE>og' (indicated by <HERE>)
in sub do-it at <tmp> line 1
in block <unit> at <tmp> line 1

153
14:08 simcop2387 joined 14:09 perlbot joined 14:11 Aedil joined 14:46 guifa left, guifa joined
ab5tract After using Kotlin a bit, I appreciate having syntax that means “call method if object is defined, else return undefined” 14:59
Maybe we could add a ‘.??’ … but then again, core may well be fully saturated already :) 15:01
15:19 sena_kun left 15:21 sena_kun joined
librasteve ab5stract: good idea! 15:38
ab5tract I’ve also toyed with a ‘///‘ op, which would be “defined-or-do’, but that falls more directly into the pure sugar category 15:44
$maybe-nil.??method /// return
timo so it'd be a bit like "orelse"? 15:46
msiism Hi, what's the difference between binding (`:=`) and using the `constant` prefix in declarations? 15:59
guifa you can still rebind 16:04
timo "constant" happens at compile time
guifa m: constant $a = 5; $a = 6;
camelia Cannot assign to an immutable value
in block <unit> at <tmp> line 1
guifa m: my $a := 5; $a := 6
camelia ( no output )
msiism I see, thanks. So, bound variables are sort of like "soft constants" then? 16:05
guifa sort of. It's more that they don't have containers 16:06
16:07 abraxxa-home joined
msiism Okay, I'll keep a note. 16:07
timo well, they don't come with containers of their own, even though you can bind a container into a variable 16:09
m: constant WHOA = Scalar.new(); say WHOA.raku; WHOA = 5; say WHOA.raku;
camelia ===SORRY!=== Error while compiling <tmp>
An exception X::Cannot::New occurred while evaluating a constant:
Cannot make a Scalar object using .new
at <tmp>:1
Exception details:
Cannot make a Scalar object using .new
in block at <tmp>…
timo ok makes sense
m: constant WHOA = my $ = 99; say WHOA.raku; WHOA = 5; say WHOA.raku;
camelia 99
5
msiism I don't get "doesn't come with a container", though. 16:10
timo here we have a container that was bound to a variable that would otherwise not have a container, and that makes it assignable "again"
right, so if you "my $foo = 5" you get a scalar container that the 5 is put into, and if you "my $foo := 5" you don't get the container because binding is what you use to get around containers / assignment semantics
but if you "constant foo = 5" there is no container there 16:11
even though you didn't use the := operator
msiism Okay, I need to read up on containers in Raku, apparently. 16:12
timo there's a documentation page on it under language/containers 16:15
feel free to ask any further questions
msiism Yeah, I surely will. :) Just started slurping the Raku Guide plus bits from the official documentation yesterday. 16:18
timo cool. i tend to be very interested in what people with fresh eyes stumble upon. that's an excellent source for potential improvements. so feel free to be very verbose about your journey of exploration, if that's OK with you of course 16:23
msiism Okay, so, basically, what I'm doing right now is going through the Raku Guide (which is very helpful to get started) and occasinally the official docs and building my private Raku quick reference maunual out of that. 16:26
When I have more experience, I'll probably be tempted to make some suggestions for the Raku Guide. 16:28
timo nice. the raku guide is nice, though i haven't looked at it myself in a good while 16:29
librasteve yeah - i like the raku guide --- its the "right" length and reminds me of the Rust book 16:30
msiism Just reading section "Literal lists" at docs.raku.org/language/list. 16:47
It contains the following piece of text:
"Because the semicolon doubles as a statement terminator it will end a literal list when used at the top level, instead creating a statement list. If you want to create a statement list inside parent hesis, use a sigil before the parenthesis:"
I'm not entirely sure, but I think this should rather be written something like: "Because the semicolon doubles as a statement terminator, it will end a literal list when used at the top level, instead of creating a statement list. If you want to create a statement list inside parentheses, use a sigil before the opening parenthesis:" 16:48
librasteve msiism: I agree ... we had a period of very helpful volunteers working on the docs (but they were not all native english speakers) and we could always use good feedback like this to improve them. When you see potentiak improvements like this, hit the pencil icon on the page and it will take you to the GH source ... the ideal way to propose a fix is to fork the raku doc repo, make the edits in your fork and pR back 16:52
for review...
16:53 sena_kun left, sena_kun joined 16:56 sena_kun left 16:57 sena_kun joined
msiism Okay, I'll keep a note. I'm kind of trying to avoid GitHub as much as I can now, though. Is there any other acceptable way? 16:58
timo I can understand wanting to avoid github 16:59
feel free to just have a clone hosted anywhere that allows any raku dev to pull from you, that should be fine 17:00
msiism Okay, fine.
timo for discussion / review we do have a mailing list if you find that acceptable, otherwise IRC / discord are also acceptable at least from my perspective
i'd like it if we migrated away from github more, but i'm afraid the right kind of volunteer-hours for that kind of project are scarce 17:02
msiism Mailing lists and IRC are both fine with me.
timo we did have RT instead of github issues before, and it was a big improvement to go to github instead of RT, but i guess that's a low bar to clear :D 17:03
msiism I don't have any experience with RT. Seems I can count myself lucky. :) 17:05
timo well, it's certainly no Jira 17:08
17:11 abraxxa-home left 17:24 lizmat joined
msiism Looking into lists, I encountered a problem with trying to extract elements of a list assigned to a scalar variable, which is probably not what you're supposed to be doing. 17:30
Here's an example: paste.debian.net/plainh/34d85391
timo that's coming from the assignment syntax unpacking your list on the RHS into what could be a list but is only one variable on the LHS 17:31
oh, is it the precedence 17:32
m: my $alist = (1,2,3,4); say $alist[1];
camelia 2
timo right, the precedence was making "my $alist = 1" the first element of the list the , operator is creating there 17:33
msiism Oh, okay.
timo m: my @alist = 1,2,3,4; say "the alist with at sign is ", @alist.raku; 17:34
camelia the alist with at sign is [1, 2, 3, 4]
timo when assigning to a @-sigiled container it uses a different precedence level
$ is meant for single items generally, that's probably the reason for this difference 17:35
m: my $alist = 1..4; say $alist.raku
camelia 1..4
msiism But it works when you use binding. 17:36
timo m: my $alist = (1..4).list; say $alist.raku
camelia $(1, 2, 3, 4)
msiism So, when I want to deal with list-like data, is it generally better just to go with an array? Seems so.
timo m: my $alist := 1,2,3,4; say $alist.raku
camelia (1, 2, 3, 4)
timo often, yeah
there's also the difference between List and Array to consider, which has implications for the mutability 17:37
m: my $alist = (1,2,3,4); $alist[2] = 99
camelia Cannot modify an immutable List ((1 2 3 4))
in block <unit> at <tmp> line 1
timo m: my @alist = (1,2,3,4); @alist[2] = 99; say @alist.raku
camelia [1, 2, 99, 4]
timo a @-sigiled variable will be an Array by default, which makes nice writable containers for you in every slot 17:38
msiism It seems to me like lists in Raku are primarily useful in contrcuts like for <list> { … } and probably a dozen others, but they're not lists like in Tcl or Scheme.
timo m: my $alist = (1..*).list; say $alist[99] 17:39
camelia 100
timo lists can also be lazy lists
lucs I just fixed an issue reported in one of my github repos. 17:40
I'm quite certain that the fix is correct and I commented on the issue as such and closed it.
Is that proper etiquette or should I have waited for the person who opened the issue to close it?
timo since there's also the feature where putting a line like "Fixes: #1234" into your commit message will automatically close issues, I think it's fine to close the issue and if necessary the user can re-open it 17:41
lucs Oh, I didn't know about that feature. 17:42
Next time, I'll do that (pretty sure there'll be a next time :-) 17:43
timo I tend to put a "thank you" message in the issue anyway
lucs That, I did.
timo that's a good place to put a bit to the effect of "feel free to re-open if the fix is not enough" or something like that
[Coke] (fork doc repo) also happy to add privs for repeat contribs. 18:22
18:31 lizmat left 18:45 lizmat joined 19:38 Aedil left
librasteve msiism: the raku.guide presents a coherent subset of raku as an introduction - per timo comment and the guide, just use an @ sigil where you need many things and it provides a typical Array with similar features to other languages 19:40
once you are conversant with the guide stuff, then you are ready for the language references - docs.raku.org/reference
in this case docs.raku.org/language/list Lists, Sequences and Arrays is a good next level deeper imo 19:42
19:43 Sgeo joined
you may also like docs.raku.org/language/containers based on the above discussion 19:43
I prefer to use the @ sigil and Arrays over Lists since the raku designers provided lots of features to make that do what I mean (but I respect that the other animals in the List zoo have a role to play depending on the task) 19:46
20:07 Sgeo left
antononcube It seems numerical properties of 2025 provoke(d) fair amount of interest. 20:09
Is there are a more elegant way to write the this in Raku: (e ** π + π ** e).floor ** 2 ? (Gives 2025.)
I tried to use different superscript entering methods without successs. 20:10
20:10 Sgeo joined
I know this works: floor(e ** π + π ** e)² 20:10
msiism librasteve: How about course.raku.org/? 20:34
librasteve Andrew Shitov is a great raku contributor, sorry to say I have not read this course so not able to comment 20:40
on the content
21:35 sena_kun left 21:38 Sgeo left 21:41 Sgeo joined
leont Is there something nowadays that prevents me from having to manually type my fez password? 22:06
tellable6 2024-12-30T15:12:41Z #raku-dev <finanalyst> leont I solved my module inside prove6 problem. Turns out it was a path spelling error on my part. Everything fine now
msiism What is the "telemetry snapper" in Rakudo? 22:09
patrickb leont: AFAIK it won't ask you again once logged in (and until the token expires, unsure what the lifetime is though). 22:17
leont patrickb: yeah, it expires 22:18
Just give me a configuration file until the thing is smart enough to understand key rings.
23:36 sivoais_ left 23:38 sivoais joined 23:40 sivoais left, sivoais joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!