🦋 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.
00:08 kjp joined 00:09 dbonnafo joined 00:13 dbonnafo left 00:31 jpn joined 00:36 jpn left 00:55 jpn joined 00:59 jpn left 01:44 jpn joined 01:49 jpn left
Geth ¦ raku.org: coke assigned to codesections Issue Rakudo™ github.com/Raku/raku.org/issues/192 02:09
¦ raku.org: coke assigned to dontlaugh Issue update main branch to 'main' from 'master' github.com/Raku/raku.org/issues/193 02:13
¦ raku.org: coke assigned to dontlaugh Issue FSCK errors in repo github.com/Raku/raku.org/issues/50
¦ raku.org: coke assigned to dontlaugh Issue Add friendly 404 message github.com/Raku/raku.org/issues/207
02:15 kylese_ joined 02:16 kylese left 02:39 jgaz left 02:45 jpn joined 02:50 jpn left 02:54 edr left 03:20 jpn joined 03:24 jpn left 03:47 jpn joined 03:51 jpn left 04:10 dbonnafo joined 04:14 dbonnafo left 04:17 jpn joined 04:21 jpn left 04:22 kjp left 04:29 kjp joined 04:37 kjp left 04:42 kjp joined 04:47 jpn joined 04:49 MasterDuke left 04:52 jpn left 04:57 dbonnafo joined 05:02 dbonnafo left, dbonnafo joined 05:48 jpn joined 05:53 jpn left 06:09 CIAvash joined 06:20 tobs left, tobs joined 06:49 jpn joined 06:52 jtza8 joined, jtza8 left, jtza8 joined 06:55 jpn left 07:02 jpn joined 07:04 CIAvash left 07:07 jpn left 07:08 CIAvash joined 07:37 melezhik joined
melezhik .tell @rawleyfowler - Humming-Bird build fails on sparrowci - ci.sparrowhub.io/report/3768 07:38
tellable6 melezhik, I cannot recognize this command. See wiki for some examples: github.com/Raku/whateverable/wiki/Tellable
melezhik .tell rawleyfowler: - Humming-Bird build fails on sparrowci - ci.sparrowhub.io/report/3768
tellable6 melezhik, I haven't seen rawleyfowler around
07:41 melezhik left 07:50 jpn joined 07:55 jpn left 07:56 jpn joined 08:01 jpn left 08:04 dbonnafo left 08:09 dbonnafo joined 08:17 dbonnafo left 08:28 jpn joined
patrickb melezhik: I think Rawley Fowler is rf in IRC. 08:30
tellable6 patrickb, I'll pass your message to melezhik
08:33 jpn left
nemokosch Coke: sorry for my side, for me, the ongoing weeks are a bit busy, going "on-site" in Germany next week and personal matters 08:34
08:35 CIAvash left 08:37 CIAvash joined 08:53 CIAvash left
tonyo: do dist versions have to increase on each new upload or is there a way to e.g maintain multiple main versions in parallel? 08:58
09:17 dakkar joined 09:26 Sgeo left 09:49 jpn joined 09:51 lichtkind joined 10:18 sena_kun joined 10:23 dakkar left 10:24 dakkar joined
Geth ecosystem: 2colours++ created pull request #626:
Remove Test::Output		 10:39
ecosystem/main: 52a85f76f6 | (Márton Polgár)++ (committed using GitHub Web editor) | META.list
Remove Test::Output

It lives in the zef ecosystem now		 10:40
ecosystem/main: c9b505f6ad | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | META.list
Merge pull request #626 from 2colours/patch-3

Remove Test::Output
nemokosch lizmat: strange, this time, I did manage with mi6, not sure what I set wrong last time 11:07
lizmat glad to hear it worked out now :-) 11:10
nemokosch by the way 11:32
github.com/Raku/doc/pull/4397 do yall know about this, or agree to begin with?
I tend to be okay with it but it would be good to kind of operate as one in these things 11:33
11:46 teatime joined 11:47 teatwo left 11:49 jpn left 11:55 Woodi left, Woodi joined
Xliff What's the best way to create an anonymous sub that will take any number of arguments. 12:02
lizmat sub (|c) { }
Xliff "sub ( *@a ) { ... }" -- seems too large. Is there something shorter?
lizmat if you only want positionals, then the slurpy is the best way
nemokosch and actually you might be better off using **@a... 12:07
I think *@a should be "soft deprecated" at this point, disadvised at least. Usually that's not what you'd want 12:08
librasteve i use *@a all the time - in most cases it does exactly what I want 12:10
(although I just discovered the +@a version so maybe that is better) 12:11
lizmat librasteve would **@a also do what you want ?
librasteve no
lizmat or +@a for that matter
dakkar oh, TIL, `*@x` has perl5-style "flatten everything" behaviour
lizmat ack
librasteve +@a is a maybe for me
nemokosch yes, perl5-style "flatten everything" behavior, exactly 12:12
librasteve I respect that raku is "keen" to maintain nested structure, but in my meats and potatoes style I usually want to sledgehammer everything flat 12:13
nemokosch well, do you mind if I say
there is no slim chance you are rather the exception than the rule?
librasteve well me and everyone who has read Think Raku (11.3 and A.7.1) 12:18
lizmat nemokosch *I* mind: how can you know what is the exception and what is the rule in this context 12:19
do you have statistics?
nemokosch if you mean to say that you disagree, say that
librasteve I have yet to read a proper documentation of the pros and cons of maintaining the nested structures ... I think that with a bunch of nice examples (maybe passing yaml or json around) would be a nice addition 12:20
lizmat no, I'm *not* saying that: I'm saying that you should prove your statements, especially if they are directed at a person
nemokosch I'm fine with asking people around 12:21
lizmat so far I'm just seeing an opinion, rather than a statement of fact
"asking people around" is not statistics
nemokosch this is usually a false dichotomy
it is an opinion with a basis
not all "opinions" are equal
12:21 jpn joined
lizmat I get the impression that your opinion is more equal than others ? 12:22
librasteve www.youtube.com/watch?v=ohDB5gbtaEQ
lizmat librasteve so true
nemokosch I get the impression that I tend to put more effort into my "opinion" 12:23
while some others just tend to say "okay, it's just an opinion" and call it a day
I don't find this constructive 12:24
librasteve nss
nemokosch if we can agree that something has a right answer, let's make some effort to get to know it and see who is right, if anybody 12:25
but let's not just throw it all into the "opinion bin"
librasteve I think we all agree that in the set of *@a, **@a and +@a (and |c) there is a way to do this for everyone (even if you like your way and I like mine) 12:26
lizmat agree
12:26 jpn left
nemokosch we all agree about that but I still think that *@a doesn't do what most people naively assume at the very least, and therefore it shouldn't be introduced first 12:28
librasteve "interestingly" the python for anon slurpy sub is: lambda *args: args 12:29
12:29 jpn joined
nemokosch more objectively: it doesn't do what Xliff asked for, in this particular case, assuming any other context than oldschool Perl 12:29
and does it flatten? no
librasteve sadly not, one of the many reason I prefer raku 12:30
nemokosch it doesn't auto-flatting the individual entries in any other language I know of, with the notorious counter-example of the language that was meant to be fixed
auto-flatten*
and I don't think it requires a bad mindset to think that the behavior of *@a is really just a remnant of the pre-GLR times 12:31
so sure, statistics I don't have, but please, let's not act as if I was just making something random up, or talking from gut feelings 12:33
we can go through individual programming languages and check their behavior, we can go through code bases and see the common use cases etc 12:34
librasteve or we can put our effort into something constructive 12:35
nemokosch do you not think it is constructive to suggest something that helps people learning the language? 12:36
librasteve yes
yes I think it is easy to learn that there are three (or 4) variants of slurpy - and yes you have to learn it (it maybe that this is a throwback to perl5 ... one of the popularest languages of all time back in the day, but it is surely nto a viable request to swap *@a and **@a 12:38
)
nto => not 12:39
nemokosch I didn't suggest to swap them because I felt that would be way too sensitive anyway
just to turn the representation of them around, so to speak
like here, in this particular case: What's the best way to create an anonymous sub that will take any number of arguments was the question 12:41
without making further assumptions, **@a is a better answer
ugexe you probably would have done better to use the word assumption or presumption earlier fwiw 13:10
13:12 kylese_ left, kylese joined, kylese left 13:14 kylese joined 13:26 kylese left, kylese joined
nemokosch About what? 13:32
13:56 dbonnafo joined 14:00 dbonnafo left
ugexe well, i have an entire buffer of text with an argument over a matter of factly statement that was later admittedly an opinion with an informed basis 14:00
it seems that entire argument would not exist if the original statement made it clear it was presuming 14:01
14:06 edr joined, jtza8 left 14:13 andinus left, andinus joined
nemokosch Well, i don't think it's particularly useful to make "no further assumptions" out to be an assumption 14:32
Or something that can be actually argued for "just an opinion", and I think that was all the "argument" 14:35
ugexe I mean you are welcome to ignore my advice, but I get the impression your message is often lost in hubris when I suspect it would often be received more openly if it was communicated in a way that assumed good faith of everyone involved 14:44
nemokosch In the meantime, I don't really get what could have been done better. I started off with "I think" and made a statement based on statistical observation and precision about the given question 14:45
then somebody says that they like it otherwise, and it ends up turned into just "two opinions"
it's not rhetoric, I really cannot see why this is good. One could just say "stay out of it" by the same chance 14:46
ugexe I guess if you think your communication style is proving fruitful then indeed there is nothing you could have done better 14:49
nemokosch Seeing the sacrifices this community has already taken, I don't think I am the only one here who should think about these questions 14:51
ugexe We aren't talking about everyone else right now
nemokosch We might have to because communication is a bi-directional thing
14:52 dbonnafo joined
ugexe It is, but it doesn't validate what-about-ism 14:53
nemokosch it's not whataboutism that I don't think the people involved here have zero responsibility in the situation 14:55
I am not willing to claim full responsibility, and that seems to be what you expect or imply here
14:57 dbonnafo left
ugexe "Seeing the sacrifices this community has already taken, I don't think I am the only one here who should think about these questions" 14:57
this is a whataboutism
nemokosch the very same people are doing very familiar things 15:00
ugexe I am being pretty explicit in that I'm talking about the communication style you use and how it can be improved to get more people to be receptive to what you say. I have no expectations that you will follow through on that.
nemokosch to be more specific: rendering a content problem as a tone problem
ugexe it doesn't matter how right you are if no one will listen to you 15:02
nemokosch you are right about it but does that mean I should just do whatever it takes all against my convictions? 15:03
15:05 Sgeo joined
ugexe I'm not sure what your convictions are so I can't say for sure. Would your convictions be furthered if people were more receptive to your message / opinions? Or are your convictions furthered more by communicating in an absolutist way? 15:07
Xliff Why is **@a better than *@a ? 15:14
And is it possible we could make something less annoying than either?
nemokosch lol 15:15
Xliff In this situation, I would think "sub { ... }" to be a nice any-arg way to specify "a sub with any number of arguments that uses *@ARGS semantings" 15:16
If you want a sub with no arguments, then that is spelled: "sub () { ... }"
nemokosch Anyway, what's so absolutist about saying that "I think" something should be done differently because it's "usually" not what people really want...
Xliff However this is language design, and something tells me someone has already had this discussion
ugexe Do you really think that is how your message came off? 15:17
nemokosch it's literally what I said, damn it
you know, this is why I'm skeptical at this point. It doesn't even matter what I say 15:18
tonyo nemokosch: you can publish multiple running versions, so if you're maintaining v2.X and still developing v3.X it should let you do that
nemokosch I do make the compromises and it's not received at all
tonyo: okay gotcha, thank you
ugexe It does matter what you say. I'm quite literally telling you this 15:19
librasteve nemokosch: I do listen to what you say - even if I think it is wrong 15:21
(and usually i will tell you honestly what I think)
what you (or anyone else) have not done is a sensible worked example of why **@a is a better style than *@a and I genuinley need to read (or write) this to be convinced that the regular *@ARGS syntax I have used for years could be done in a better way 15:24
nemokosch I don't think it's a problem if we disagree about something
librasteve **@a is a better stylel than *@a
^^ sorry Discord is swallowing my stars 15:25
nemokosch but I also don't think every time two people disagree about something, the right solution is to just announce that "it's a matter of preference" and call it a day
librasteve when you say "X should be deprecated" you are saying "anyone using X is doing it wrong and we will forbid this option since it is not in the zen of raku" 15:26
and I think "I want a programming language that does not force me to do things in a certain way" 15:27
nemokosch I don't know if I can even agree with this but actually I only said "soft deprecate", leaving it up to further specification, and generally talking about educating people to learn **@a as the basic thing 15:28
frankly, none of us are making claims only in order to make points, right? 15:29
librasteve ok - you need to convince me with some worked examples ... even better if your examples can be then repurposed directly in the docs to show the benefits of the double assterisk 15:30
nemokosch well, the biggest advantage is that it really just does what Xliff also asked about here 15:33
it really just collects positional arguments into one variable/symbol
15:34 jgaz joined
It is much harder to assume that somebody purposefully feeds nested structures in and wants them to be flattened. There are two parts: the potential (ie. making the abstraction over argument lists that might flatten for whatever benefit) and the actuality (ie. the caller knows the structure for sure) 15:37
For the latter case: if the caller knows it for sure what the arguments look like, then it doesn't make much of a difference. They actually have more fine-grained control without auto-flattening because they can either do something like .flat or something like .Slip for starters 15:39
For the former case: well, that's where I would like to see an example. In what situation do you think: "whatever the arguments may be, they definitely need to be flattened out"? 15:40
ugexe I would wonder if users lean towards *@_ due to their first introduction being `MAIN(*@_)` examples, where structure is not relevant 15:41
nemokosch for example, I found such an example for deepmap at least, after a pretty long time: tree traversal 15:42
but to use the arguments of a function to represent a tree, that sounds more than a bit peculiar
librasteve ugexe: I suspect that's where I got my bad? habit 15:49
15:49 sjn left
I am happy with the current design of raku and my current habit ... what you are saying is that the choice to flatten should be with the caller, what I suspect is sometimes yes sometimes no (ie the callee is the boss) depending on the API you want ... anyway I am not trying to convince you to not use double star @a 15:52
15:55 sjn joined
I will definitely use the double star when I have something like a set of recursive tree traversal functions 15:56
ugexe speaking of trees... does anyone use .tree? :P 15:59
I know Larry did...
nemokosch I didn't know it existed 16:04
only Data::Dump::Tree
the examples didn't help me understand how it works 😅 16:06
anyway, my biggest concern isn't that somebody uses *@a in an informed way; more that people who just want to learn how to collect all arguments into a single list will pick this one up and later find that "something weird happened" 16:11
16:25 bigdata joined 17:16 bdju left 17:20 soundtoxin joined 17:26 jpn left 17:27 soundtoxin left 17:30 bigdata left 17:32 codesections left 17:33 codesections joined 17:35 bdju joined 17:36 dakkar left 17:41 jpn joined 18:33 dbonnafo joined 19:40 melezhik joined
melezhik . 19:40
tellable6 2023-11-09T08:30:54Z #raku <patrickb> melezhik: I think Rawley Fowler is rf in IRC.
melezhik .tell rf: last Humming Birds fails on sparrow  - ci.sparrowhub.io/report/3768 19:41
tellable6 melezhik, I'll pass your message to rf
nemokosch 😄 19:45
melezhik hi nemokosch - thanks for mentioning  my  works on the last raku conf 19:54
20:03 jpn left 20:06 melezhik left
antononcube Well deserved. 20:36
20:46 Xliff left
tbrowder__ lizmat: mentioning Raku Advent in weekly i think would be timely 21:02
21:15 dbonnafo_ joined
Geth ¦ problem-solving: librasteve assigned to codesections Issue Data::Dump::Tree in Rakudo core github.com/Raku/problem-solving/issues/399 21:17
21:18 dbonnafo left
Geth ¦ problem-solving: librasteve assigned to codesections Issue Mixed Fraction Support for Rats (Feature Request) github.com/Raku/problem-solving/issues/400 21:27
nemokosch yes, it was well deserved
and Sparky works fine for me right now anyway
tbrowder made a good point though
21:29 jpn joined
antononcube How easy it is to use the Selenium webdriver (of Chrome) in Raku? I see "Selenium::WebDriver" at raku.land -- has anyone used it? 21:31
librasteve how can we leverage the Advent Calendar to attract some converts? 21:32
[since most languages do their own advent these days having copied the idea from perl/raku] 21:33
antononcube Easy -- make LLM-based applications and AI-image-generation workflows. 21:34
librasteve okaay - I had a couple of ideas for LLM "apps" coming off your preso ... so I could pick up and learn that stuff and write my Raku Advent submission as an LLM oriented piece 21:38
certainly its the hot topic ...
question for the wider community: can we / should we each focus our submissions on LLM (we need ~ 25 in total and a willingness to suborn our personal priorities to the LLM theme) 21:40
? 21:42
nemokosch that's a good question... 21:54
librasteve [btw I am happy to take on the advent github "marshall" role if that is needed] 21:57
Nemokosch: ... well what do you think? 21:58
nemokosch I think that I'm both fairly clueless with LLM and largely uninterested 22:05
so it's kinda hard to be enthusiastic
22:05 jpn left
but I'd be a hypocrite to not offer a helping hand if I see something gets going with it 22:05
22:11 nicole left, nicole joined
librasteve is anyone interested / enthusiastic in Antons idea? 22:22
(off to bed now will look for replies in the morning my time)
antononcube I will consider making a related blog post. 23:15
23:23 sena_kun left 23:46 teatwo joined 23:49 teatime left 23:55 Xliff joined
tbrowder__ nemokosch: i enjoyed yr presentation at raku conf. i like the idea of defining some workgroups for ppl to help in areas you pointed out. lizmat, [coke], fianalist, and the RSC can't do it all. 23:57
(and many others carrying a heavy load)
roping in new ppl is cool 23:58

Powered by Cro and the Raku® Programming Language.

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

Thank you!