lucs Here's an example: gist.github.com/lucs/c6d470d030bc8...4dfe0f6c96 02:24
Going through the docs looking for something else, I found that having `--human > F` will produce the same as `--output-file=F` (the same as seen in the terminal when neither of those options is used). 06:14
And looking at the details of what `--human` does, I think I understand the why and how of the differences I was seeing. 06:30
Yay! :-)
06:57 wayland joined 10:27 [Coke] left 10:29 [Coke] joined 13:10 ChanServ sets mode: +o lizmat
[Coke] would be nice if the help gave an example of a callable for --dir 15:22
I got an error message that --ignore-dir in rak is --dir, tried that directly, no love.
(read the help to find it's a Callable)
ah, the deny above is helpful. so rak --dir='* ne "ignored_dir"' ... 15:24
shouldn't rak --dir=1 "thing" search recursively? 15:29
(also tried --dir=True)
lizmat trying to recollect atm 15:34
[Coke] I'm not blocked or anything, now down to curiosity 15:35
lizmat Specifying --dir as a flag will include 15:37
*all* subdirectories, while specifying --/dir (or --not-dir) will
cause no recursion to happen whatsoever
--dir=1 is not as a flag 15:38
[Coke] why is --dir=1 not a callable like --dir='*ne"skip-this"' ? 15:39
lizmat good question, I would have to spelunk 15:40
please make an issue :-)
[Coke] also --/dir=skip-dir doesn't recurse.
it only does the current dir 15:41
sure, opening issue.
done. 15:42
17:29 lizmat sets mode: +b *!~Nemokosch@2a02:ab88:388f:5100:3f10:8b08:ff1c:706 17:55 lizmat sets mode: +b TheUnforgiven!*@*
lucs Does App::Rak have a convenient API I can use in a Raku program, or am I better off just shelling (or "proc'ing") out to `rak` and collect the output? 19:15
lizmat lucs: App::Rak is actually a frontend for raku.land/zef:lizmat/rak 19:16
[Coke] raku.land/zef:lizmat/rak
lizmat yes, it has quite a number of arguments :-) 19:17
lucs Um, yeah, I know that! :-)
What I mean is, where could I go from here: `raku -e 'use App::Rak; ⋯` to invoke rak and grab its output in the program? 19:21
[Coke] you'd start with raku -Mrak -e'...', I think
don't use the front end as the module, use the low level, I would think 19:22
lizmat use rak; my $rak = rak / foo /, :file(/ \.txt $/); for $rak.result { ....
lucs Oh, plain `rak` there? Hmm...
lizmat yeah, the "rak" module is the engine, really
lucs Ah, I see.
lizmat App::Rak does a bit of pre- and postprocessing
lucs lizmat: That example you showed will get me started. Thanks. 19:23
lizmat just a copy/paste from raku.land/zef:lizmat/rak :-)
lucs Hehe! 19:24
Yeah, sorry, I hadn't looked at what you and [Coke] linked to: I was fixated on App::Rak being the correct place. 19:25
lizmat it happens :-)
19:48 lizmat sets mode: +b Nemokosch!*@*
lucs (General Raku question, I suppose, follows, but asking here anyway.) 20:41
Looking at `app.rakumod` from the `rak.git` repo, why are all subroutines declared with `my sub ⋯`?
Er, meant `rak.rakumod`, sorry. 20:43
Aren't all those top-level subs already scoped to the containing file? 20:49
(limited scope, that is)
Can a file's top-level `sub ⋯` be somehow reached from outside the file? 20:50
[Coke] look at the last paragraph here: docs.raku.org/syntax/my 21:12
docs.raku.org/language/variables#T...declarator slightly better URL
21:37 wayland left
lucs Hmm... In the `class Foo { my class Implem... ` example, I understand how the inner class becomes inaccessible to users of class Foo. 21:41
But (just now seeing) a bit higher it mentions that "`my` is the default scope for subroutines, so `my sub x() {}` and `sub x() {}` do exactly the same thing.", maybe those `my sub ⋯` in the code aren't really doing anything? 21:42
[Coke] just being explicit, as far as I know. 21:50
lucs Okay, nothing wrong with that, eh. 21:52
Another thing I noticed is that the code uses variable bindings much more often than assignments. 22:02
It had never occured to me to think about which to use in my code, I'm just used to assignments.
But seeing that code, it makes me think that when binding is appropriate, it probably saves some runtime (minuscule and perhaps negligible most of the time), but it expresses a kind of "logical" economy that I like.
Binding, here I come! 22:03