00:00 raku-bridge left 00:01 raku-bridge1 is now known as raku-bridge 00:58 pamplemousse joined 06:09 pamplemousse left 06:20 jjmerelo joined
jjmerelo Hey 06:20
I'm trying to install a plugin in Comma that allows me to input unicode characters easily, don't know if that's achieved via plugin or not, actually 06:21
I seem to remember that was kinda easy, but in the current version, 2020.05, the marketplace is empty 06:22
[Tux] Rakudo version 2020.05.1-62-g2b38666f9 - MoarVM version 2020.05-10-g5fe4a8114
csv-ip5xs0.838 - 0.874
csv-ip5xs-208.168 - 8.317
csv-parser26.332 - 26.750
csv-test-xs-200.387 - 0.395
test7.412 - 7.548
test-t1.960 - 1.992
test-t --race0.940 - 0.977
test-t-2030.756 - 31.886
test-t-20 --race9.768 - 9.797
jjmerelo .seen vrurg 06:56
tellable6 jjmerelo, I saw vrurg 2020-05-10T01:57:51Z in #raku: <vrurg> thanks!
07:10 jjmerelo left 07:39 jjmerelo joined 07:58 Altai-man_ joined 08:15 tony-o left 08:17 camelia joined, sena_kun joined 08:18 Altai-man_ left
lizmat Files=1306, Tests=111275, 214 wallclock secs (28.69 usr 8.38 sys + 3008.86 cusr 272.37 csys = 3318.30 CPU) 08:53
jjmerelo OK, asking again: an easy way to enter unicode characters in comma? Plugin or otherwise? 08:57
if it's plugin, I can't seem to be able to list available plugins in the marketplace in this new version 2020.05 08:58
sena_kun jjmerelo, no. :( 09:01
jjmerelo sena_kun but I seem to remember there was a plugin for that.
sena_kun only in this version the plugin marketplace is empty
sena_kun jjmerelo, well, there is a bug in 2020.05 at the very least. OTOH, you can download some plugin by hand and install it from the disk. 09:02
jjmerelo sena_kun OK, got unicode_browser, which is the one I remember to have used 09:06
Now I'll have to find out how to use it. It's kind of old, too 09:07
sena_kun it opens a panel in the left. It's not perfect, but it's serviceable. 09:17
09:23 jjmerelo left
lizmat If I'm not mistaken, this line: github.com/rakudo/rakudo/blob/mast....pm6#L1275 09:29
will run every time at startup
feels to me that that should be run at compile time
but doing so breaks the core module installation
jnthn nine any ideas as to why that should not be run at compile time? 09:30
09:49 jjmerelo joined
jjmerelo Quite a few days without questions in StackOveflow... 09:49
lizmat some questions on reddit, though :-) 09:52
hmmm... could it be that modules.raku.org is not getting updated? 09:54
modules.raku.org/dist/Cache::Async...n:ROBERTLE shows version 0.1.6 09:55
but there is a mention of a version 0.2 on Twitter: twitter.com/raku_cpan_new/status/1...9793006592
09:57 bazzaar joined
Geth rakudo: 9a583dcbe9 | (Elizabeth Mattijsen)++ | 2 files
Move de-parentificiation of Mu to compile time

Not sure why this was needed to be run at runtime. Now it runs at the very end of settings compilation, which appears to not cause any problems.
jnthn lizmat: It probably should be at compile time, though I wonder if it helped avoid a bootstrapping issue... 10:08
jjmerelo lizmat no idea, really. Shouldn't that part be updated automatically?
lizmat jnthn: well, just adding a BEGIN did not work, moving it to core_epilogue did 10:09
so I guess it was a bootstrap issues originally
jnthn lizmat: Ah, BEGIN in core_epilogue, I assume?
lizmat yes
tyil: like the concept of "rakoons" :-) 10:11
tyil :D 10:12
not sure if there's an "official" way to refer to raku users yet
lizmat reddit uses Rakuers, but I don't particularly like that
tyil I've seen that as well, it doesn't roll off the tongue as easily as rakoons does for me 10:13
and ease of pronunciation is useful when talking
lizmat jnthn ^^ :-) 10:14
10:16 Altai-man_ joined
lizmat there appears to be a band called "Rakoon" 10:18
jjmerelo Of course
With umlauts liberally put over letters, I assume.
10:19 sena_kun left
jjmerelo Rakuifers? 10:19
lizmat rakuists feels elitist to me
jnthn Raccoons are kinda cute :)
jjmerelo Rakulottes?
lizmat right, and smart: "sly coon"
jjmerelo Rakutes? 10:20
tyil Raketjes, pronounced the dutch way, like the ice cream? :>
jjmerelo Racuckoos? 10:21
lizmat the more I think about it, I like Rakoon, for its similarity to raccoons
especially if you'd compare it to say, velociraptors
jjmerelo (y) 10:22
lizmat and of course, there's en.wikipedia.org/wiki/Rocket_Raccoon 10:25
Altai-man_ finds Raccoon City to be more widely known 10:26
timotimo uh oh :) :) 10:57
Altai-man_ Absolutely not in a bad way. :] 11:20
jjmerelo Flunked again Season of Docs 12:12
We need to keep trying, I guess... 12:13
Maybe third's the charm...
They only accept 50 applications, so it's highly competitive...
Maybe we should have put a little bit of more effort with the project. 12:14
You always learn.
12:17 sena_kun joined 12:18 Altai-man_ left
[Coke] the S.T.A.R.S of Rakuun City. 12:36
12:57 pamplemousse joined
rypervenche I like Rakuuns, or some other spelling of it. 13:02
13:02 jjmerelo left
lizmat notable6: weekly 14:02
notable6 lizmat, 9 notes: gist.github.com/c99a37e1e50dfd9e3c...cc0eaf2cdd
lizmat notable6: weekly reset
notable6 lizmat, Moved existing notes to “weekly_2020-05-11T14:02:57Z”
14:16 Altai-man_ joined
tobs is a raku-doer 14:16
14:19 sena_kun left
lizmat yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2020/05/11/2020-...e-release/ 15:36
15:49 sour32 joined
sjn lizmat++ # for doing rakudo weekly 15:57
16:16 patrickb joined 16:17 sena_kun joined 16:18 Altai-man_ left
Geth rakudo: 4297cb051d | (Elizabeth Mattijsen)++ | 2 files
Prevent $*OUT lookup for each eigenstate with put / print

Although one probably shouldn't print or put a Junction, having a dynamic variable lookup for each eigenstate can be pretty expensive. Also, convert to calling .print immediately, for better interface with custom classes acting like $*OUT.
16:27 jjmerelo joined
Geth rakudo: e71d3cb076 | (Elizabeth Mattijsen)++ | 2 files
Move &printf to io_operators

Not sure why it was living in Cool. No functional change here.
jjmerelo lizmat at some point in time, there was a multi sub printf(Cool:D $format, *@args) 16:42
lizmat that's the one I moved
jjmerelo How independent routines end up in a particular file (and in a particular documentation page) is a mystery. 16:43
But heuristically, it's very likely that routines that have as a first argument the class the file is devoted to end up there.
lizmat ah, that way... but io_operators is about subs that output, and printf is one of them 16:44
and needs similar logic to print / put / say / note
jjmerelo right, it makes all the sense in the world...
lizmat so the Coolness of the format, is really just an extra "attribute" if you will 16:45
17:04 patrickz joined 17:05 patrickb left
Geth rakudo: e46d459021 | (Elizabeth Mattijsen)++ | src/core.c/io_operators.pm6
Make printf up to 5% faster

  - make printf an only sub
  - directly call $*OUT.print
  - unfortunately, could not go to | for signature, as some tests depend
   on the $, | signature.
This efficiency improvement is almost completely drowned out by the fact that the major part of doing a printf is still parsing the format string over and over again.
lizmat m: "%d\n".printf: 42 # TIL 17:32
camelia 42
lizmat I find this weird
sorta expected that: 17:33
m: 42.printf("%d")
camelia Your printf-style directives specify 0 arguments, but 1 argument was supplied
in block <unit> at <tmp> line 1
lizmat would work
nine Well the format string is the odd man out in the argument list
lizmat as that would be consistent with: 17:34
m: 42.print
camelia 42
lizmat m: $*OUT.printf("%d",42)
camelia 42
nine 42.print reads "print that 42", "%d\n".printf: 42 reads "print this string and fill in some placeholders with the provided values"
17:40 sour32 left
lizmat bisectable6: old=2020.05 dd sprintf "%d\n", 1 | 2 | 3 17:56
bisectable6 lizmat, On both starting points (old=2020.05 new=e46d459) the exit code is 1 and the output is identical as well
lizmat, Output on both points: «Directive d not applicable for value of type Junction (any(1, 2, 3))␤ in block <unit> at /tmp/wCV5cjpSfT line 1␤␤»
timotimo m: say 42.fmt("%d") 18:05
camelia 42
18:12 jjmerelo left
Geth rakudo: 110324d36e | (Elizabeth Mattijsen)++ | src/core.c/io_operators.pm6
Add support for printf($format, Junction)
18:16 Altai-man_ joined 18:18 sena_kun left
Geth roast: 8b060617e6 | (Vadim Belman)++ | S26-documentation/block-leading-user-format.t
Make the test use is_run from roast Test::Util

We better don't create extra copies of existing routines.
18:31 cognomin_ joined 18:33 cognominal left 18:34 cognomin_ left 18:35 cognominal joined 18:37 robertle left 18:38 robertle joined 18:52 patrickz left
nine Why didn't I realize that I would be doing BEGIN time EVAL all over again, just much harder, because it's in a NativeCall callback, thus in a nested runloop? 19:23
19:42 lichtkind joined 20:17 sena_kun joined 20:18 Altai-man_ left
lizmat sourceable6: my %h is SetHash 22:05
sourceable6 lizmat, No idea, boss. Can you give me a Code object?
22:09 bazzaar left 22:16 Altai-man_ joined
lizmat in NQP: $type.HOW.parameterize($type,@params); 22:18
how can I flatten the @params array into separate parameters to parameterize ?
timotimo don't we support | in nqp?
perhaps need to call .FLATTENABLE_MUMBLEE on it
22:19 sena_kun left
lizmat tries 22:19
indeed... but I first have to de-HLLize an Array into its reified 22:32
I guess nqp::getattr is the only way, right timotimo 22:33
22:33 Kaiepi left 22:34 Kaiepi joined 22:35 Kaiepi left, Kaiepi joined 22:48 Altai-man_ left 23:11 MasterDuke left 23:32 lichtkind left