🦋 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.
00:42 tadzik left
antononcube @calma 1) hi; 2) ok; 3) no; 4) depends on your interests . 00:52
01:22 tadzik joined 01:57 calma joined, hulk joined, kylese left
scullucs I'm so unhappy (refraining from swearing here) with all the "smart" formatting helpers out there. 02:05
Be it Markdown, Pod, whatever, they always get in the way of what I'm trying to do. 02:06
02:15 hulk left, kylese joined 02:26 calma left 02:30 kylese left 02:35 kylese joined
Maybe I'm just ignorant. 02:41
Here's what I want to do: 02:42
Paragraph reads:
There are also variations of the methods where their name can be changed by either, or both:
(doing Pod here)
I want that to be followed by three indented paragraphs describing those variations. 02:43
But if I indent the paragraphs in Pod, they get formatted as 'code' and all the C<baz> stuff appears literally. 02:44
Can this be done? 02:46
Indenting a paragraph?
Just indenting a paragraph?
Haha! Okay, ignorant I am. In desperation, I prefixed each paragraph with - and they got indented (at least with the shell script that converts the Pod to Markdown then to HTML -- not sure how robust that all is). 03:20
04:43 skyesoss joined 05:08 wayland left, wayland joined 05:26 skyesoss left 06:08 Aedil left 07:01 sena_kun joined 07:43 calma joined 08:10 calma left 08:19 Sgeo left 08:23 Aedil joined 08:32 Aedil left 09:33 Aedil joined 11:13 xinming left 11:15 xinming joined
tbrowder scullocs: the rendering from pod to Markdown is a basic translator and i don't think a random '-' is significant in pod so it doesn't get any special handling. 13:00
you should be able to indent =item blocks which is kind of like indenting paras 13:02
try: 13:06
following your intro para, skip a line and start your first subpara with "item blah blah", skip a line and start subpara "=item blah..." etc. 13:10
antononcube @lizmat It did not occur to me to do a poll. And after you pointed that out, now I see the poll button... 🙂 13:18
lizmat TYL :-) 13:19
antononcube Thank You, Lord! 13:22
lizmat Today You Learned ?
antononcube Ten Years Later -- it did't take me that long... 13:23
@lizmat 🙂 I figured -- so, TYL on "TYL", also.
@lizmat and @librasteve I deleted my previous post on Reddit -- thank you for your comments on it! 13:24
My new Reddit post has a proper poll. (As prompted by lizmat.) 13:25
lizmat :-)
13:27 foxieze joined, foxieze left 14:15 hulk joined, kylese left
lucs tbrowder: Turns out that ‹raku --doc=Markdown ⟨rakumod file⟩› converts "=item Subpara..." to " * Subpara...", but leaves "- Subpara..." unchanged. 14:57
But coincidentally, both forms get converted to "<li><p>Subpara…" in HTML produced by ‹markdown-calibre ⟨Markdown file⟩›.
Not that I want "list item" indented paragraphs, but that will do.
(I'll use "=item ...", it's more meaningful than "- ..." -- thanks for the tip.) 14:59
tbrowder lucs: can you show a gist of your pod file? something doesn't sound right. 16:59
[Coke] ok, I like rak, but it definitely takes longer to install than ack. :) 17:00
tbrowder and the second transformation is a different ballgame 17:02
with the =item you can create something like a titled subpara if you use =begin item followed immediately by a line with a subpara title, then a blank line, then paras separated by blank lines, finally a =end item closes it 17:06
lizmat [Coke]: there's quite a lot tests, have you tried with --/test 17:24
[Coke] I hesitate to do that, but yah, makes sense. 17:27
lizmat BTW, the next release of App::Rak will allow you to specify: "s:foo &!bar" as a search query 17:28
the s: for split on spaces for multiple queries
the & to indicate the next needle should be and-ed with the previous, instead of or-ed 17:29
and the ! means: should not
so in other works: give me all the lines that contain "foo" but not "bar"
17:30 xinming left
lizmat *words 17:30
antononcube @lizmat Maybe a guide why using "App::rak" is more efficient than ack might convince Coke to rak it.
17:32 xinming joined 17:47 Sgeo joined, eseyman left 17:52 manu_ joined, manu_ is now known as eseyman 17:57 xinming left 17:59 xinming joined 18:47 tadzik left, tadzik joined
[Coke] a guide is not necessary 18:56
change is annoying. I'm still using ack v1 muscle memory
lizmat [Coke]: with the --save feature you can adapt rak to match your ack muscle mempry 18:59
*memory
and if you do, please publish your list of aliases, I may well put that into rak as a feature 19:00
e.g. --ack : accept all following arguments with ack semantics 19:01
hmmm...
19:06 rypervenche left, rypervenche joined
lucs tbrowder: It's this file (er, I'd said it was a rakumod file, but it's actually a rakudoc): github.com/lucs/Debugging-Tool/blo...ol.rakudoc 19:11
At line 101 is the first of three '=item'.
That file rendered to Markdown is github.com/lucs/Debugging-Tool/blo.../README.md 19:12
antononcube @Coke "a guide is not necessary", "change is annoying." -- This means a guide to ease the change is necessary. 19:46
[Coke] ... no 19:56
it's not annoying -because I don't have a guide-
having a guide -will not make it less annoying-. I already know how to use rak. :) 19:57
but I appreciate you trying to help me out. thank you
lizmat github.com/lizmat/App-Rak/issues/53 [Coke] 19:58
antononcube @Coke Honestly, I am looking for an excuses to revisit and extend this post/work: rakuforprediction.wordpress.com/20...e-manuals/ 19:59
lizmat afk&
[Coke] lizmat: You don't have to do that for me, I'm not sure I'd use that. 20:07
But again I appreciate you.
20:21 donaldh joined 20:22 donaldh left
librasteve @antononcube I voted 20:23
antononcube Yes, please “vote often, and vote early.” 20:24
20:31 Aedil left
tbrowder lucs: are you happy with the results of the =items? they look okay to me. buit i do notice the later chunk of examples in the large code block need # in front of the internal comments. 20:36
lucs tbrowder: Yes, I'm okay with the '=item' renderings. 20:39
The missing '#' are not accidental. The reader will have to think for half a second to realize that those lines are not code that should be commented out, but just comments. 20:44
tbrowder ah, i see. did i help or hinder? 20:45
lucs Suggesting '=item' helped, it makes my Pod better than the '- ' I had found to render similarly. 20:46
tbrowder how do you actially generate the Markdown README? 20:47
*actually 20:48
lucs I did 'raku -doc=Markdown .../Tool.rakudoc'
And I realized this afternoon that doing 'rakudoc Debugging::Tool' won't show the contents of the Tool.rakudoc file :-( 20:50
I'm guessing there's a good chance that can be arranged, but I have no idea how. 20:51
tbrowder reason i ask is your code looks like using App::Mi6 would help you do that automatically. i started using it when it was introduced some years ago and love it. @lizmat does too!
i'll be happy to help if you'll accept a PR? 20:52
lucs Um, I do it kind of automatically, I'm Perl-lazy, so I wrote shell scripts to do it for me :)
tbrowder arg!!
lucs No worries, it works fine. 20:53
But like I said I would like to know how to get that Tool.rakudoc file rendered when invoking 'rakudoc Debugging::Tool'. 20:54
tbrowder ok, but you're missing out. among other things it make testing and publishing real easy. for instance, you want to rebuild all your docs: mi6 build 20:55
are your shell scripts in the repo? 20:56
lucs I used to use mi6. I don't anymore, but I used to. (sorry Mitch Hedberg)
No, they're not. 20:57
I understand where you're leading, and you're probably correct.
tbrowder do you mind saying why you quit mi6? it's improved a lot since skaji first released it 20:58
lucs I can't remember. I should look at it again, it has been a while.
tbrowder for instance, a one-liner in yr dist.ini file can turn the pod in one file to its own Markdown. add more one-liners for other inputs 21:01
more complex stuff can also be done 21:02
lucs Yeah, it would make things better self-contained. 21:04
21:20 sena_kun left
[Coke] having trouble golfing this, but have a script that is erroring with: 21:57
No such method 'mutate' for string '`*%args`,'
(the code is: $word = $word.mutate(:g, /<-alpha>+/, " ").words[0]; 21:59
antononcube @Coke Is subst doing what you want? 22:09
[Coke] ... wait, mutate is just wrong, I had subst-mutate at one point and incorrectly updated it. 22:10
... why is only dying after so many values of $word?! weird. 22:11
antononcube I just read in the docs that subst-mutate is obsolete...
[Coke] Yup, used it in a oneliner, tried to update it in the script, and ... what a strange failure mode. 22:13
(ah. maybe it was the first file being tested that met any of the conditions (but not literally the first file in the list)) 22:14
antononcube 🤷‍♂️ 22:23
[Coke] that was it. sorry for the spam 22:25
weird. using subtest and in some case it's nesting the subtest instead of completing it 22:40
ab5tract Maybe I'm holding things wrong, but this looks like a deeply insidious bug to me... 22:42
m: say from-json(to-json(%( array => [ %( i => "d", a => %( b => "c" ), d => %( e => "f" ))] ) ))<array>[0]
camelia {d => {e => f}}
ab5tract that should be `%( i => "d", a => %( b => "c" ), d => %( e => "f" ))`
this also happens with JSON::Fast, I haven't checked other libraries yet 22:43
doesn't matter if you change things to `{ }` brackets, either 22:44
m: my %h; %h<array>.push: %( i => "d", a => %( b => "c" ), d => %( e => "f" )); say from-json(to-json(%h))<array>[0]; 22:46
camelia {a => {b => c}, d => {e => f}, i => d}
tbrowder ref my roundtrip problems the other day, they seem to be solved. using :bin on all pertinent files tested now seems to work. part of my badness was awkward use of slurp and spurt. i implememted subs slurp-file and spurt-file to have stricter control over the inputs. 22:48
ab5tract sigh, I guess I'm holding it wrong, but it's late and the subtlety absolutely escpaes me...
m: my %h; %h<array>.push: %( i => "d", a => %( b => "c" ), d => %( e => "f" )); say from-json(to-json(%h))<array>[0]; say %h ~~ %( array => [ %( i => "d", a => %( b => "c" ), d => %( e => "f" ))] )
camelia {a => {b => c}, d => {e => f}, i => d}
False
[Coke] <-alpha> is "not alpha". how to combine with <-digit> in the same class? 22:52
ugexe <-alpha -digit> ? 22:56
(i am indeed guessing)
ab5tract into a single class, not sure (but also curious why that would be necessary?) 22:59
m: "9e" ~~ /<-alpha>&<-digit>/ ==> say()
camelia Nil
ab5tract surprised this didn't work: 23:02
m: "9e" ~~ /<!:L+:!N>/ ==> say()
camelia 「e」
ab5tract but I'm frustrated enough at the moment so I'll avoid diving deeper
I hope you find a suitable solution [Coke]
ugexe m: for " ", 1, "a" { say so $_ ~~ m/<-alpha -digit>/ } 23:03
camelia True
False
False
ab5tract ugexe++
23:38 modula joined, defaultxr left