🦋 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.
00:15 swaggboi left 00:17 swaggboi joined 00:22 lichtkind left 01:08 dg joined 01:27 kylese left, kylese joined 02:15 kylese left, kylese joined
Voldenet librasteve: maybe exclude line numbers in the end of the article, because it's a bit tough to copy the code from it :P 03:27
04:02 guifa left 04:30 kylese left 04:34 kylese joined 05:03 abraxxa-home joined 05:30 kylese left 05:32 kylese joined 05:42 stanrifkin joined 06:04 merp left 06:20 abraxxa-home left 06:21 abraxxa-home joined 06:27 abraxxa-home left 06:28 abraxxa-home joined, abraxxa-home left 06:32 Sgeo left
librasteve voldemort: good advice .. tx! … should be ok now 07:17
lizmat that is a weird autocomplete librasteve :-) 07:21
07:54 melezhik joined 07:55 melezhik left
Voldenet it makes me feel more powerful, so I appreciate the typo :> 08:10
08:11 sena_kun joined 08:30 Aedil joined 08:32 lichtkind joined 08:41 Aedil left 08:44 Aedil joined 09:05 bdju left, jpn joined 09:07 bdju joined, sena_kun left
timo unfortunately you don't have a nose any more :( 09:54
librasteve i blame it on my small checker
10:44 mirHalel joined 10:47 mirHalel left 10:55 guifa joined
Voldenet oh no : ( 11:05
now I get why the code smell stopped bothering me
11:39 jpn left
Voldenet are those two pieces of code equivalent? I suspect they aren't: 11:47
m: my $x = ^10; my @y := $x; .say for @y;
camelia 0
1
2
3
4
5
6
7
8
9
Voldenet m: my $x = ^10; say @$x;
camelia (0 1 2 3 4 5 6 7 8 9)
Voldenet erm
m: my $x = ^10; .say for @$x;
camelia 0
1
2
3
4
5
6
7
8
9
11:49 apac left
Voldenet but I can't find docs for syntax on `@$x` 11:49
11:52 jpn joined
timo @$foo is like @($foo) is like $foo.list 11:55
Voldenet Makes sense, I found docs for it docs.raku.org/type/Any#index-entry...extualizer
11:56 apac joined 11:59 guifa left 12:18 apac left 12:22 apac joined 12:23 bbkr joined
bbkr m: "foo/../foo/file01.txt".IO.succ.say; 12:24
camelia "foo/../foo/file01.txt".IO
bbkr m: "foo/../foo/file01.txt".IO.basename.succ.say; 12:25
camelia file02.txt
bbkr ^ bug? happens only when path backtracks, basename is not bumped +1 despite being properly recognized. 12:26
Voldenet m: "x01".succ.say 12:29
camelia x02
Voldenet m: ".x01".succ.say
camelia .x01
Voldenet this looks deliberate
bbkr Why? Looks inconsistent with: 12:30
m: "foo/file01.txt".IO.succ.say;
camelia "foo/file02.txt".IO
Voldenet eh, I'm not sure in which context it'd make sense 12:31
but it smells like bugfix for some special case 12:32
bbkr succ() behavior on IO::Path is explicitly documented as "Returns a new IO::Path constructed from the invocant, with .basename changed by calling Str.succ on it.". So I'd expect foo/../foo/file01.txt successor to be foo/../foo/file02.txt. 12:33
There is no exception in documentation that path cannot contain backtracking. I just discovered this by locking production code in infinite loop because directory in which code was looking for next free file name was relocated to backtracked path and suddenly succ() was no longer returning successor. 12:36
12:38 merp joined
bbkr I've created github.com/rakudo/rakudo/issues/5842 12:39
timo sourceable6: "foo/../foo/file01.txt".IO.succ 12:52
sourceable6 timo, No build for revision “7f0474c”
timo sourceable6: 2025.03 "foo/../foo/file01.txt".IO.succ 12:53
sourceable6 timo, No build for revision “316ddd2”
timo linkable6: github.com/rakudo/rakudo/commit/30...a718ab1951
"Make IO::Path use its own .succ / .pred logic" "So it no longer depends on the Str.succ / .pred logic" 12:54
well, that might not actually matter
ah, right, so the .. somewhere else in the path confusing it is the bug i think 12:57
that really wants to have logic in it that properly resolves what part of the string is where the basename begins, and not look for any "." before that 13:01
I don't think I'm the best candidate to attempt to fixt his 13:06
m: say "/foo/bar/hi.01.bla/test01.txt".IO.succ # probably also not what you want or expect 13:13
camelia "/foo/bar/hj.01.bla/test01.txt".IO
timo m: say "/foo/bar/hi01.bla/test01.txt".IO.succ # it just uses the first dot it finds in the whole path, which is not what the docs say it should do
camelia "/foo/bar/hi02.bla/test01.txt".IO
timo i also put it in the ticket, apparently at the exact same time as oyu 13:16
bbkr I've added this discovery of "first dot logic" to GH issue.
:)
13:16 JimmyZhuo joined
timo in any case, "next free file name" is not a great thing to be doing for a few reasons 13:17
bbkr agree. however in my use case with guaranteed single instance and controlled directory sizes this is justified - I get benefit of having sequential files without any race confitions or slowdown risks. 13:25
timo any benefit over just counting the number of files in the dir to get the next number? there could be holes in the sequence? 13:27
i guess it's nicely compact to write like ""file01.txt".IO ... !*.IO.d"
bbkr in this tool user can refer to files using negative notation (--file-index=-1) because in this particular case this is super convenient. 13:28
I've tripped on counting files once because stupid macOS filesystem added .DS_Store junk. So I'm leaning towards succ() method. 13:29
13:30 lichtkind left
timo right, you'd want dir() with a test that regexes only the filenames you're interested in 13:31
m: "test99.txt".IO.succ.say
camelia "tesu00.txt".IO
timo ^- another thing that's probably not what you want 13:32
m: "test_99.txt".IO.succ.say
camelia "test_100.txt".IO
bbkr True. i'm aware of it and it is already under control. However I think this quirk with "a99" vs "a_99" should be better documented n IO::Path. It's explainable but not DWIMy. 13:36
To be fair it is documented in Str.succ, but from my experience users don't like to drill documentation if they think they already understood something from examples. 13:39
github.com/Raku/doc/issues/4566 ^ this 13:48
13:50 JimmyZhuo left
timo good 14:01
14:10 bbkr left 14:30 ACfromTX left 14:43 ACfromTX joined
lizmat PSA: this week's Rakudo Weekly News will most likely be published tomorrow 14:49
14:50 jpn left
[Coke] lizmat++ 14:51
14:55 jpn joined 15:22 apac left 15:26 apac joined 15:34 jpn left
[Coke] Any idea how to satisfy this requirement on mac? 15:35
"pq:from<native>:ver<5>"
apparently not with `brew install postgresql@15`
(libpq?) 15:36
nope.
15:37 apa_c joined 15:38 apac left, lichtkind joined
holmdunc libpq is apparently "keg-only" docs.brew.sh/FAQ#what-does-keg-only-mean 15:41
[Coke] I tried running the zef command after that with an updated LDFLAGS, adding the path to .zshrc, etc. 15:42
(tried all the env vars in the brew info notes) 15:43
... I'll do the testing on ubuntu, I guess.
16:20 apa_c left 16:26 jpn joined
jdv linux is the way 16:27
well, the easier way:)
librasteve [Coke]: I found a way to get it to work and documented here github.com/raku-community-modules/...issues/242 18:35
18:36 MoC joined
[Coke] thanks 18:37
librasteve yw
I didnt test adding the specific location to one of PATH, LD_LIBRARY_PATH, DYLD_LIBRARY_PATH or DYLD_FALLBACK_LIBRARY_PATH - which maybe easier, and clearer to explain... 18:39
19:05 Guest53 joined 19:07 Guest53 left 19:31 jpn left 19:51 abraxxa-home joined 20:17 jpn joined 20:21 jpn left 20:47 MoC left 20:51 Guest28 joined, Guest28 left 20:52 jpn joined 20:57 guifa joined 21:44 abraxxa-home left 21:47 jpn left 21:58 stanrifkin left 22:13 sena_kun joined 22:14 Sgeo joined 22:18 jpn joined 22:22 jpn left 22:23 stanrifkin joined 22:52 sena_kun left 23:01 Guest92 joined 23:03 Guest92 left 23:43 jpn joined 23:48 jpn left 23:56 lichtkind left 23:57 dgl joined, dgl left

Powered by Cro and the Raku® Programming Language.

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

Thank you!