🦋 Welcome to Raku! raku.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: colabti.org/irclogger/irclogger_log/raku
Set by ChanServ on 14 October 2019.
00:16 JRaspass left
[Coke] why not use %*ENV? 00:25
m: say %*ENV{HOSTNAME}
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/rDetAVQbhA
Undeclared name:
HOSTNAME used at line 1
[Coke] m: say %*ENV<HOSTNAME> #oops
evalable6 (Any)
[Coke] so, no hostname there, but that should work for you.
I may be misundestanding a sublety of your request, though 00:28
00:31 squashable6 left 00:35 squashable6 joined, pecastro left
[Coke] m: run('hostname', :out).out.slurp(:close) # getting hostname from command 00:42
evalable6
[Coke] m: say run('hostname', :out).out.slurp(:close) # getting hostname from command
evalable6 superserver
[Coke] (you have to capture the output, get the IO object, and slurp the data (waiting for it to close first)
00:47 aindilis left
elcaro FWIW, on my CentOS server, when I type `env` at the shell, I see HOSTNAME, and `raku -e 'say %*ENV<HOSTNAME>'` works 00:47
on my Ubuntu server, no HOSTNAME seen in `env` output, and hence no 'hostname' in %*ENV
so it "works on my machine"... maybe check where your HOSTNAME var is being exported? is it under /etc/profile.d/* or /etc/bashrc or ~/.profile? 00:49
if you can find out where it's being exported, maybe that will point towards why Raku isn't seeing it (or reveeal a bug) 00:50
m: say qx[env].match(/HOSTNAME\S*/) 00:51
evalable6 Nil
01:47 evalable6 left, linkable6 left, linkable6 joined 01:48 evalable6 joined 01:56 rindolf joined 02:16 kvw_5_ joined 02:18 rindolf left, rindolf joined 02:19 kvw_5 left 02:22 Celelibi left 02:36 gnufr33dom joined 03:36 releasable6 left, unicodable6 left, bloatable6 left, coverable6 left, notable6 left, tellable6 left, sourceable6 left, committable6 left, evalable6 left, bisectable6 left, quotable6 left, greppable6 left, shareable6 left, benchable6 left, nativecallable6 left, statisfiable6 left, squashable6 left, linkable6 left, shareable6 joined, benchable6 joined, bisectable6 joined 03:37 sourceable6 joined, notable6 joined, bloatable6 joined, nativecallable6 joined 03:38 quotable6 joined, tellable6 joined, evalable6 joined, committable6 joined, statisfiable6 joined, coverable6 joined, releasable6 joined, unicodable6 joined 03:39 linkable6 joined, greppable6 joined, squashable6 joined 03:58 Garbanzo joined 05:00 pilne left 05:05 maggotbrain left 05:06 maggotbrain joined, aindilis joined, brtastic joined, maggotbrain left 05:08 maggotbrain joined 05:25 rindolf left 05:46 tejr left, tejr joined 05:58 parabolize left 06:13 xinming joined 06:17 xinming_ left 06:28 dudz joined 06:29 rindolf joined 06:31 __jrjsmrtn__ left, _jrjsmrtn joined 06:50 jmerelo joined 06:55 ufobat_ joined 07:09 wamba joined 07:10 mightypork left, ufobat_ left, Garbanzo left, mightypork joined 07:16 eseyman joined 07:29 ufobat joined 07:43 Sgeo left
andinus m: say "perl\nraku" 07:46
evalable6 perl
raku
andinus m: say "perl\nraku".comb
evalable6 (p e r l
r a k u)
andinus how can i ask comb to not consider newline as a characters?
i just want <p e r l r a k u> 07:47
07:48 sjm_uk joined
tyil m: "perl\nraku".comb(/\w/).say 07:49
evalable6 (p e r l r a k u)
07:52 webstrand joined
andinus m: "perl\nraku\n1234 \n0".comb(/\w/).say 07:53
evalable6 (p e r l r a k u 1 2 3 4 0)
andinus i see, thanks tyil
08:04 aluaces left 08:08 stoned75 left 08:23 MasterDuke joined
andinus i have defined a sub, can i add tests within the script for it? 08:27
so that i'm able to call it with : ./bin --test my-sub
MasterDuke you can always manually do that by handling the --test flag in your script. you may also find www.codesections.com/blog/raku-uni...mpilation/ useful 08:36
gfldex Is yellow considered the colour of warnings in most cultures?
08:37 domidumont joined
andinus i think so, yellow bg with black text 08:39
MasterDuk_e: i see, thanks
08:43 Scimon joined 08:45 dakkar joined 08:49 aindilis left 08:56 MasterDuke92 joined 09:00 JRaspass joined, MasterDuke left 09:04 aluaces joined 09:07 PimDaniel joined 09:09 PavelB joined
PimDaniel .tell Hi lizmat! How can i install locally your code github.com/lizmat/List-AllUtils with dependencies? , i need to play with it. Thank's in advance. 09:10
tellable6 PimDaniel, I'll pass your message to hi
PimDaniel .tell lizmat! How can i install locally your code github.com/lizmat/List-AllUtils with dependencies? , i need to play with it. Thank's in advance. 09:11
tellable6 PimDaniel, I cannot recognize this command. See wiki for some examples: github.com/Raku/whateverable/wiki/Tellable
PimDaniel .tell lizmat Hi lizmat! How can i install locally your code github.com/lizmat/List-AllUtils with dependencies? , i need to play with it. Thank's in advance.
tellable6 PimDaniel, I'll pass your message to lizmat
Scimon PimDaniel : Have you tried zef install List::AllUtils ? 09:12
PimDaniel Scimon: YES : does not found anything. :(
Scimon I'm running that now and it seems to be working fine. 09:13
Just installed the dependencies.
PimDaniel Not possible :)
:)
Scimon Yup it's worked maybe your zef install needs updated? 09:14
PimDaniel I try again and give you the error!
zip update?
Scimon Good luck. 09:16
PimDaniel Scimon: is there a way to install it into my user land?
MasterDuke92 PimDaniel: if you want to make modifications to github.com/lizmat/List-AllUtils and test them, you'll need to clone it, and then in the clone directory, run `zef install .`
PimDaniel Because i need to play with it.
jmerelo PimDaniel: maybe it's not yet in Modules.raku.org 09:17
MasterDuke92 or run whatever code you're using with -I path/to/your/clone's/lib
PimDaniel MasterDuke : you reply to my question befor i ask! Goog guy!
Scimon What MasterDuke92 said. Clone the repo (I'd fork it) then you can install it locally.
PimDaniel I partially cloned it into /tmp dir but i cannot find the List-Util package on Github? 09:19
09:19 sena_kun left
Scimon If you do `zef install --deps-only .` in the folder it should get the dependencies for you. 09:20
Unless you want to modify List::Util as well?
PimDaniel > if you do `zef install --deps-only .` in the folder : let us be clear into List-AllUtils cloned dir? f 09:22
Scimon Yup 09:23
09:23 sena_kun joined
Scimon This should install all the dependencies. 09:23
PimDaniel Scimon: Ok thank's , i give a try!
zef install --deps-only => usage: .... :( 09:24
MasterDuke92 did you put ` .` at the end? 09:25
PimDaniel Hooo nooooooo. :( Sorry!
It worked but Question: i see theses modules dependencies into .zef is it the place and the only place it installed it? 09:29
tyil not a zef dev, but iirc it stores the dist in ~/.zef, but installs into ~/.raku 09:30
MasterDuke92 i believe that's the default. there is an option, something like '--install-to' if you want to change it 09:31
tyil the only thing that saddens me about this is that neither raku nor zef respect XDG basedirs 09:32
PimDaniel tyil and MasterDuke92 thank's. Yes look inside dirs, that's just curiosity because i do not really want to work on them.
09:33 dakkar left
PimDaniel Thank's to all of you, i'm less crazy than yesterday, i come back in one minute because i'm sure i'll be stucked since. :) 09:33
09:34 dakkar joined 09:38 PimDaniel left 09:43 PimDaniel joined 09:45 pecastro joined
PimDaniel I have many modules installed into ~/.zef but i cannot use them : what is the operation to do for it? 09:45
for sure i could use lib but no :( RAKULIB nor because there are many PATHS! 09:46
MasterDuke92 if they've been installed then you shouldn't need to modify RAKULIB or your PATH. a simple `use foo;` should work 09:52
PimDaniel MasterDuke: which path exactly? : i have a tree of libraries into .zef and i'll put my module main module into another local lib to play with it. 09:56
if i had $HOME/.zef to rakulib is it sufficiant? 09:57
MasterDuke92 that shouldn't be necessary
PimDaniel of is it $HOME/.zef/store ? 09:58
s/of/or/
or is it $HOME/.zef/store/<module1>:$HOME/.zef/store/<module2>, etc...?
MasterDuke92 if i understand what you're trying to do, you want to modify a module, and then use that modified module in some script? but the module you're modifying has some dependencies? if zef has installed those dependencies, you don't need to do anything with RAKULIB or PATH to use them 09:59
dakkar also, installed modules go into ~/.raku/, not ~/.zef/ 10:00
in ~/.zef/ you have zef's own working space
PimDaniel So why doesn't it work on my system? 10:01
dakkar please read what I just wrote
~/.zef/ is *not* where installed modules go
don't look there
look it ~/.rake/
sorry, ~/.raku/
10:01 a3f joined
PimDaniel i tried use File::Find and it hangs! 10:01
dakkar or even better, `zef list --installed`
MasterDuke92 what isn't working? using the module you modified? or is it complaining that the dependencies aren't installed? 10:02
dakkar PimDaniel: we don't have enough information to really help you
MasterDuke92 maybe it's just taking a long time to precompile the module, if it's the first time you're using it
dakkar "first time after upgrading rakudo", because zef pre-compiles on installation 10:03
PimDaniel zef list --installed returns File::Find:ver<0.1.1>, how do i import it?
dakkar `use File::Find;` 10:04
PimDaniel Doesn't work!
dakkar please describe exactly what you do, and what happens
"doesn't work" is rarely useful to people trying to help you
tyil PimDaniel: take a step back, show an example of some code you're trying to run, and show us the error you're seeing 10:05
you can do this on StackOverflow too, if you want more space to properly outline your issue
PimDaniel here is what i do : pastebin.com/A13QUBHM 10:08
MasterDuke92 is your ~/.perl6 a symlink to ~/.raku? or vice versa 10:10
dakkar ok, that's a bug in zef
it should list ~/.perl6 if rakudo doesn't look there 10:11
I think you can `mv ~/.perl6 ~/.raku`
PimDaniel NO my ~/.perl6 is a dir
with thinks into it 10:12
dakkar PimDaniel: have you noticed that rakudo shows you where it's looking for modules, and it's very clearly *not* looking in ~/.perl6 ?
try renaming that directory to .raku
PimDaniel but i yet have a .raku dir too 10:14
dakkar oh joy 10:15
only suggestion I have: delete ~/.perl6, re-install the modules you need
PimDaniel so i'll rename .perl6 temporarily to nimportequoi and make a link to try...
dakkar no 10:16
*DELETE* it
it's not used by anything, it confuses zef, and it confuses you
get rid of it
PimDaniel which one i delete? .perl6?
dakkar yes
PimDaniel ok for now i move it into /tmp/
dakkar do whatever you want 10:17
PimDaniel I'v got the same problem with simlink applied and .perl6 dir desapeared! :( 10:18
dakkar I don't understand what that sentence means 10:19
what symlink?
nobody told you to create a symlink
PimDaniel i made a simlink : ln -s .perl6 .raku like MasterDuke92 proposed to me. 10:20
MasterDuke92 PimDaniel: how did you install raku and zef?
dakkar no, that symlink must not be there
there must not be anything named ~/.perl6
PimDaniel MasterDuke: dont't remenber how.
dakkar MasterDuke92 *asked* if it was there, didn't tell you to create it
PimDaniel Ok i remove it. 10:21
Same problem :(
dakkar have you re-installed all the modules you need?
as I told you to?
PimDaniel dakkar: YES.
dakkar ok
I can't see what you're doing, so you have to tell me 10:22
can you re-run `zef list --installed` and show us the output?
PimDaniel The result is here : pastebin.com/A13QUBHM posted half an hour ago. 10:23
dakkar ys
but that was before you deleted ~/.perl6 and re-installed modules
the output must be different now
PimDaniel YES : only 2 lines. :( 10:24
dakkar can you please show that output?
PimDaniel pastebin.com/z0ETd6wY 10:25
dakkar now install File::Find
`zef install File::Find`
(that's what I meant by "reinstall the modules you need") 10:26
PimDaniel OK I don't need any perl6 seemlink? 10:27
dakkar no
there must be nothing named ~/.perl6 10:28
PimDaniel I installed File::Find zef install created a .perl6 dir. 10:32
dakkar ugh 10:33
PimDaniel I think i found the problem : my version of zef is Debian package one i thing not custom install raku.
Scimon Ahhhh....
dakkar yeah, I was going to ask what version of zef you were running
PimDaniel i check that.
Scimon It may not include cpan support.
dakkar probably ancient, if it's the standard Debian stable package
PimDaniel OK that's it! 10:34
dakkar mixing system packages and self-compiled bits is always a bad idea
MasterDuke92 yeah, you should uninstall that zef package 10:36
PimDaniel dakkar: yes but i discovering things, now how do i get zef for my custom installation? 10:37
dakkar github.com/ugexe/zef follow these instructions
*after* you uninstall the system zef
(ignore the `rakubrew` bit on that page if you are not already using rakubrew to build your rakudo) 10:38
10:38 brtastic left
PimDaniel ok thank's! 10:38
10:40 vike left 10:47 PimDaniel left 10:53 vike joined 10:55 PimDaniel joined
PimDaniel \o 10:55
zef works!
Sorry for the noise. 10:58
Scimon Glad it's working :) 10:59
PimDaniel but i'v lost my installed modules list :(
10:59 MasterDuke joined 11:01 MasterDuke92 left
PimDaniel Confused 11:05
Now back to the beginning where are installed so i can modify one of them just to play with it? 11:06
* where are modules installed?
11:10 MasterDuke53 joined 11:12 MasterDuke left, PimDaniel left
MasterDuke53 you can't modify modules that have been installed via `zef install Foo::Bar`. the source is copied to your system so you can see it, but raku actually loads the immutable bytecode 11:13
so if you want to test modifications you have to clone the repo (probably fork it first), make your modifications, then do `zef install .` (or just override the zef installed one with `-I path/to/the/modified/module/lib`) 11:14
11:24 wamba left 11:26 MasterDuke53 left 11:27 gnufr33dom left, MasterDuke joined 11:29 PimDaniel joined
PimDaniel \o 11:29
11:30 El_Che left, vike left
PimDaniel MasterDuke53: Like told me somebody here (you may be) i gitcloned the main package, went in the directory, then zef installed only its dependencies. Now i can play with the main package. 11:31
tellable6 PimDaniel, I'll pass your message to MasterDuke53
11:31 El_Che joined
PimDaniel Now i can play with the module because i just want to understand its code and adapt some part of code for myself. 11:33
I had to add the RAKULIB PATH on the main package of course 11:34
11:34 PimDaniel left 11:44 vike joined 12:00 tejr left 12:05 tejr joined 12:07 JRaspass left 12:15 Black_Ribbon left, Black_Ribbon joined 12:29 PavelB left, PimDaniel joined
PimDaniel \o , Hi! 12:29
my new question is here : pastebin.com/haixwFJv 12:30
I feel there's a kind of cache to manage the Module imports in Raku.
lizmat PimDaniel: BEGIN blocks are run at *compile* time 12:31
the first time you call it, the module gets (pre-)compiled, and runs the BEGIN blocks 12:32
PimDaniel Ok i see!
lizmat the second time, it is already compiled, so the BEGIN blocks are *not* run
s/BEGIN/INIT/
PimDaniel and INIT behavec like BEGIN except it will rune each time? 12:33
sjn PimDaniel: docs.raku.org/language/phasers <- here's the full overview of BEGIN and other phasers :)
lizmat INIT is run every time execution is started
PimDaniel Yes but it must be run first!
Ok thank's!
lizmat Ah, you mean to set up the stuff to EXPORT ? 12:34
that'd be too late for that
sjn .o(Should EXPORT be mentioned in the phasers docs?) 12:35
PimDaniel lizmat: since yesterday morning i'm looking for a simple way to export_ok like in Perl5 into a :TAG some constants! 12:36
what you mean too late?
12:37 Black_Ribbon left
lizmat the module's EXPORT is run at compile time of the "use"ing code 12:37
which would be before any INIT of the module
12:37 aluaces left 12:38 brtastic joined
PimDaniel So you mean INIT would not do the stuff! 12:38
lizmat the code in INIT will run *after* compilation of the code that does the -use-
PimDaniel ok 12:39
lizmat and that will have run the EXPORT sub of the module already
PimDaniel Ok now i understand too late!
lizmat in any case: so "is export(:TAG)" is not something you want to do 12:40
PimDaniel I try to adapt your code so that i can into a loop export_ok some of my CONSTANTS into a some :TAGS, is it the good way to proceed?
lizmat I would say, yes 12:41
please note that the EXPORT sub can be a multi
so you can use multiple dispatch on a tag
sub EXPORT(:$TAG!) { }
multi sub EXPORT(:$TAG!) { }
12:43 Black_Ribbon joined
PimDaniel But are there not yet some exporters written in Raku? That can to that? 12:43
12:43 Black_Ribbon left
PimDaniel On the other hand, it not lost time because it helps me to understand things. 12:43
What is the ! after the positional $TAG? The 1st time i see that! 12:45
moritz it marks the argument as required 12:47
named args are optional by default
12:47 wamba joined
PimDaniel moritz: thanks! 12:47
I'm learning: sorry!
lizmat hmmm... looks like :$TAG! *will* look at the standard mechanism 12:48
12:48 aluaces joined
lizmat I have this in a module: 12:48
constant A = 42; constant B = 666; multi sub EXPORT('TAG') { (A,B).Set.list.Map }
in a file lib/Foo.rakumod 12:49
and I can call it with:
raku -Ilib -e 'use Foo <TAG>; dd A,B
PimDaniel Hooo! So simple!
lizmat PimDaniel: would that do what you want ?
12:50 grayrider left
lizmat before long, you should be able to the ".list" in there 12:50
PimDaniel lizmat: I think it is what i need but i MUST test your solution. 12:51
lizmat sure :-) you should never believe what you're being told just by someone 12:52
or hesitate to tell them that they're wrong
just because of perceived seniority
such behaviour was responsible for the greatest airline disaster of all time :-( 12:53
PimDaniel No problem : I never believe others and i never test my own code because i'm too proud for this!
12:54 dataangel left
lizmat en.wikipedia.org/wiki/Tenerife_airport_disaster 12:54
"the KLM flight engineer expressed his concern about the Pan Am not being clear of the runway by asking the pilots in his own cockpit, "Is he not clear, that Pan American?" Veldhuyzen van Zanten emphatically replied "Oh, yes" and continued with the takeoff" 12:55
but enough of history :-)
12:55 JRaspass joined
PimDaniel I know what you tell about, i saw all plane disasters, it's very fun, and i remember it. 12:57
well very fun. Hum :/ 12:59
lizmat sometimes learning things can be fun, even if they are not funny in of themselces 13:02
*selves
in this particular case, the junior officer did not dare to go against the most senior officer, and it got them killed 13:03
13:04 JRaspass left 13:10 PimDaniel left, sjm_uk left
lizmat clickbaits rakudoweekly.blog/2021/02/01/2021-...proposing/ 13:12
13:14 PimDaniel joined
PimDaniel lizmat: Your solution does not seam to work. :( 13:16
hooo wait: The problem is My constants are into a module. 13:19
13:24 PimDaniel left 13:27 PimDaniel joined
PimDaniel I works only when my constant are out of my package but i do not want that. :( 13:27
I works only when my constants are out of my package but i do not want that. :(
13:29 PimDaniel left 13:36 PimDaniel joined 13:38 xelxebar left, xelxebar joined
PimDaniel lizmat: 1/ echo 'constant A = 42; constant B = 666;' > lib/Foo.rakumod 13:39
PimDaniel lizmat: raku -Ilib -e 'use Foo <TAG>; dd A,B , works without export sub,NO? But nothing is tagged and protected thought. :( 13:41
lizmat constants are by default "our", and thus part of default exporting
should have made them "my" :-) 13:43
argh,.. of course, it's taking the values, not the names: must not have been awake yet 13:46
PimDaniel Notice that to complicate things : i have them under an "unit module <name>;" , but i can change for a module <name> { } if necessary.
I also whatched code with loops around "MY::" but don't worry too mutch for me... i'll find a workarround for now if needed like use constant Foo is export(':tag') = 'TRUC'; 13:49
on each of them. 13:50
lizmat my constant A = 42; my constant B = 666; multi sub EXPORT('TAG') { <A B>.map({ $_ => ::($_) }).Map }
this will create the mapping you need for the given lexical constants
13:51 wamba left
PimDaniel does ::($_) looks into the inner module? 13:51
lizmat it does a lexical lookup 13:52
PimDaniel Ok but no, i did not well express my need:
hoo wait...
ok you declare each constants my but where? 13:53
heres's the stucture of my package : package Foo { my constant A= 42 ; my constant B= 666} thez are inner , or unit module Foo; my constant ... etc... 13:55
13:56 wamba joined
lizmat PimDaniel: if that is in a module to be loaded, then you should put the constants *outside* of the package Foo() for visibility 14:00
so: my constant A = 42; package Foo { .... }; multi sub EXPORT('TAG') { ... } 14:01
PimDaniel why cannot we export them by some trick like if we would do that : unit module MyModule; constant A is export(":tag") = 42; in that case , use Module :DEFAULT,':tag' works. 14:03
and then A.say.raku; on the script side. 14:04
what i wonder is what trick make raku on the constants to make them visible. 14:05
raku is powefull but that part seams really obscure. :( 14:06
i must leave a few time : thanks lizmat, see you later... 14:07
lizmat later!
14:08 PimDaniel left
lizmat afk& 14:23
14:32 k-man joined 14:51 MasterDuke left 14:55 parabolize joined 14:57 MasterDuke joined 15:12 finsternis joined 15:14 Sgeo joined 15:17 mowcat joined 15:22 JRaspass joined, sno left 15:32 Scimon left 15:42 sno joined 15:49 guifa joined
guifa o/ 15:50
our %foo is export := Map.new(…); is supposed to work no problem, right? Or is there a known bug with precompilation of it? 15:55
I’ve been getting this weird bug where when I pull it into a module file, %foo is empty, but if I just C&P back into the main script, it’s got what I expect. The values are Callables, though, and I know there are some issues with precompilation and callables 15:56
s/there are issues/there were (are?) issues/
vrurg guifa: doesn't look like it's something about callables. You either get %foo exported or not. But it being empty is something different. 15:59
guifa vrurg: yeah, I’m going to do some golfing today to try to track it down further 16:08
lizmat I think the main issue is here that %foo is getting exported before it is getting initialized 16:16
our %foo is export := BEGIN Map.new(...)
guifa confirmed, actually, with and without BEGIN. Just did a quick module with my %foo is export := [BEGIN] Map.new: ‘a’, 1, ‘b’, 2, ‘c’, 3; and %foo.keys returns an empty list 16:18
vrurg guifa: try replacing := with assignment. If it fixes the problem then this is the export/initialization race.
guifa vrurg: it does 16:19
vrurg lizmat: BEGIN before Map would not help because the actual binding takes place after is export picks the container.
guifa: then see my note for lizmat. Not sure if `BEGIN our $foo is export := ...` will work, but you could try. 16:20
lizmat our %foo is export *is* already compile tim 16:21
e
guifa vrurg: BEGIN our %foo … gets the same result.
I can use plain old assignment here no problem
vrurg lizmat: that's why I doubted `BEGIN ...` will work. :) 16:22
Ok, I have to go now...
guifa I just instinctively use binding for array-to-array and hash-to-hash assigning so it never really occured to me to use plain assignment
16:23 sjm_uk joined
lizmat well, if you use assignment, %foo won't be a Map 16:23
m: my %foo is Map = a => 42; dd %foo 16:25
evalable6 Map.new((:a(42)))
lizmat m: our %foo is Map = a => 42; dd %foo
evalable6 (exit code 1) ===SORRY!===
First child of a 'bind' op must be a QAST::Var, got QAST::Op
lizmat that is definitely LTA and a bug
need to go afk again& 16:26
guifa found a decent enough work around (this is going in a submodule file, so I don’t mind a bit of boilerplate for cleaner code elsehwere).
my \foo = Map.new: … ; sub EXPORT { Map.new: ‘%foo’ => foo };
16:45 neshpion joined
JRaspass Writing a graphical program in Raku and I'd like to change the name that appears in the gnome top bar, atm it's just "rakudo". I presumed I would write to $*PROGRAM-NAME like you'd write to $0 in Perl, but that's readonly now :-( Is there a supported way? 16:49
tadzik what are you using for that graphical program? Gtk? 16:50
JRaspass sdl
i've changed the window name
tadzik is the gnome top bar showing something else than a window name?
sounds weird
JRaspass yeah, just "rakudo" 16:51
16:57 PimDaniel joined
PimDaniel \o, Look at that piece of code : pastebin.com/CsFXUW3b, the question is the last one : TODO, thank's! 16:58
JRaspass tbf it's probably a gnome thing, i suspect a .desktop file would fix it
(am slightly surprised we can't rename our processes anymore though) 17:00
the perl way used to affect what htop/ps etc showed
PimDaniel The question is the last *LINE , #TODO: 17:01
Juste for information : this is not my code, I grabed this code somewhere, and adapted it but i did not change many things. 17:03
17:08 Scimon joined 17:09 domidumont left
PimDaniel I'm disappointed to find that my problem is of no interest. :( 17:09
Ohhh sorry: Tea time probably. :/
17:11 andrzejku joined
andrzejku hello 17:11
PimDaniel \o
andrzejku what's current raku status? 17:12
PimDaniel andrezjku: are you talking to me? 17:13
for myself : status is 0 and the posted code works, but there's a question at the end of it. 17:15
andrzejku PimDaniel, yes you can answer 17:18
I meant more globally
17:19 PimDaniel left 17:22 PimDaniel joined
guifa andrzejku: in what sense do you mean? 17:23
(there are a lot of ways to answer the question ha) 17:24
andrzejku guifa, I mean if it is production ready 17:25
guifa I would say it is — there are a lot of us that are using it now in production. 17:28
I think the only issues people have right now with it are speed (which is constantly being improve step by step) and ecosystem (which has made a lot of strides over the past year, in large part because 6.d is more production ready) 17:30
17:30 Garbanzo joined
PimDaniel Sorry andrzejku : I think it has been released as a stable one but i'm not an expert of the subject. 17:30
guifa: where are speed problems? particularly? 17:33
guifa PimDaniel: there’s not any one single area, but for one example, run time type checking is done a lot where it might not strictly be necessary 17:34
PimDaniel Ok i see... well for now and for myself and what i need to do , learn Raku at most, it is not yet a problem. 17:36
guifa Because Raku is written in Raku, optimizations can thankfully have a compounding effect :-)
PimDaniel ok ok : je comprends! :)
but the time you loose at running your programs will be earn in programming because it is a really most powerfull langage, no? ;) 17:39
17:39 dakkar left
guifa Indeed — that’s generally the trade off you have with most languages 17:40
C can be made to be very fast — but it requires a lot more work to develop in
PimDaniel True: We often loose compilation time and test time. 17:41
so i can say all languages i use are very fast because i never make tests. 17:42
we ofter *forget ...
*often
El_Che PimDaniel: that's a silly statement, the never test one 17:43
guifa The amount of code that is *really* speed critical in the world is probably fairly limited.
Heck, Python is the most popular language for ML, and you’d really think for training massive datasets and stuff you’d want to use insanely optimized C for that. 17:44
PimDaniel Sorry : what is ML? 17:46
JRaspass machine learning
PimDaniel Ok the new fashion. 17:47
tadzik I think all python modules for ML are actually using insanely optimized C(++) under the hood, merely exposing a python interfaec
17:47 gnufr33dom joined
PimDaniel but When we use NaviveCall with raku, is it as fast as C? I mean if you do not do many calls into a loop? 17:49
Even in Perl5, for some Modules, the underhood was compiled libraries, no? 17:51
guifa tadzik: are they? I’ve seen a number of them that list the Python API as the fullest, with the C (and other) APIs as incomplete, so I just assumed that they were written in python and were being ported slowly
NativeCall has a little bit of overhead when you call it, but I don’t use it enough to tell yo how bad it is 17:52
PimDaniel Or how good is it, no?
you seam to have an advice. 17:53
Well these thinks are interesting but not the most important for me now. I like the language anyway! 17:54
El_Che If your main concert is speed, I would pick another language. But most use case, there are external factors that make speed relative like IO or network 17:57
PimDaniel Damn : my pastebin disapeared :(!
hooo no, here it is my i post it again? pastebin.com/CsFXUW3b 17:58
17:59 Black_Ribbon joined
guifa does a dance 17:59
Skeleton patterns are now (mostly) supported in Intl::Format::DateTime
format-datetime DateTime.now, :language<es> :skeleton<HHmmB> —> 13:00 de la tarde 18:00
now I just need to figure out some decent named options to make the skeleton less… cryptic (I’ll see myself out) 18:01
PimDaniel me gusta la mentequilla , me gusta!
back later ... 18:02
18:02 PimDaniel left 18:09 aindilis joined, Scimon left 18:14 mowcat left 18:31 ufobat left 18:36 neshpion_ joined, neshpion left 18:40 PavelB joined 18:51 sjm_uk left 18:53 andrzejku left, MasterDuke left 18:56 jmerelo left 19:03 andrzejku joined 19:06 neshpion_ is now known as neshpion, xelxebar left 19:07 xelxebar joined 19:09 gnufr33dom left 19:12 patrickb joined 19:17 MasterDuke joined 19:18 kaiwulf left 19:55 Sgeo left 19:56 wamba left 19:58 Sgeo joined 20:14 Sgeo left 20:19 Sgeo joined 20:20 rindolf left 20:25 sjn left 20:33 sjn joined 20:53 thundergnat joined
thundergnat .tell JRaspass > "I'd like to change the name that appears in the gnome top bar" If you are using SDL2, take a look in github.com/thundergnat/Raku-SDL2-applets and see if anything there helps. (polyspiral.raku is my favorite) 20:56
tellable6 thundergnat, I'll pass your message to JRaspass
JRaspass ta, without running them it looks like they do the same, just name the window, but thanks 20:57
21:00 PavelB left 21:13 PimDaniel joined
PimDaniel \o 21:13
I need to valorize a part of package name with a loop variable : MAINPACKAGE::NAME::$loopvar:: , what is the solution to do this? when i write a valid value in place of $loopvar it works but not with the variable this way. 21:17
tyil ::("MainPackage::Name::$foo")
iirc 21:18
PimDaniel tyil: ok thank's i'll try this. 21:19
21:24 PimDaniel left 21:26 PimDaniel joined
PimDaniel tyil: no it does not work this way. :(. 21:30
tyil can you send me the code you're working on for context? 21:31
MasterDuke docs.raku.org/language/packages#In...into_names 21:32
tonyo m: class A { our $loop = 0; method inc-loop { $loop++; } }; say A::<$loop>; A.inc-loop; say A::<$loop>; 21:33
evalable6 0
1
PimDaniel :MasterDuke i was just reading this part of the reference. 21:36
Thank's tyil but not for now, not sure it is usefull. 21:37
tonyo PimDaniel: ^
21:43 PimDaniel left 21:44 andrzejku left 21:48 brtastic left
guifa Is there anyway we could provide hints to comma what exported symbols might be for modules that use EXPORT? 22:05
22:17 Garbanzo left 22:18 Garbanzo joined 22:19 Garbanzo left, Garbanzo joined 22:20 Garbanzo left 22:21 Garbanzo joined, Garbanzo left 22:22 Garbanzo joined, Garbanzo left
gfldex guifa: if you stick metadata into &EXPORT.WHY Comma should be able to retrieve it. If that works it's just a matter of defining a protocol for Comma to learn. 22:24
guifa That actually could be a pretty simple way. OBviously, for non-static exports, doesn’t help, but for all the rest it should be fairly simple enough. 22:26
22:26 kaiwulf joined
guifa #| foo bar $scalar @array; sub EXPORT { Map.new: foo => …, bar => …, ‘$scalar’ => …, ‘@array’ => … } 22:28
guifa . o O ( maybe even a regex in the WHY if they’re predictable ) 22:29
mst so, #raku needs to decide whether n*gg*r is an acceptable thing for the channel to upload to cpan 22:34
hermit1.scsys.co.uk/~matthewt/raku
gfldex guifa: Or mix a role into &EXPORT to provide something that is specific to Comma. After all .WHY is for human consumption. If I would set it to anything then to a URL for the docs of that module.
mst lizmat forced me to allow this becuase apparently "it's just a word" is an acceptable response to 'n*gg*r' 22:35
but here's a full explanation
I recommend you work out whether you actually give a shit about having a code of conduct or not
if you don't, I will happily step down as freenode group contact
but I' 22:36
but I'm done with having such behaviour defended and being attacked for criticing it
HTH, HAND
22:37 donaldh joined
mst nine: ^^ 22:37
22:38 HaraldJoerg joined 22:41 HarmtH joined, donaldh left 22:43 patrickb left
guifa mst: reading over the conversation, I have to agree with what El_Che says in it: “I think the story is to confusing to make much sensen to a third party reading it now” 22:45
mst guifa: I've had to keep it confusing
guifa: because lizmat attacked me in a way I can't defend myself without attacking somebody who doesn't even like me
guifa: just because she insists on doing thatr doesn't mean I won't avoid it 22:46
tyil how is creating drama on #raku making things better?
22:46 mowcat joined
mst tyil: you kept attacking volunteer conference organisers by yelling them about having a code of conduct 22:46
tyil kept?
and "attacking" seems rather strong 22:47
mst tyil: because you believe that a code of conduct is evil because you should be allowed to say n*gg*r because it's "just a word"
MasterDuke is this recent? the references to #perl6 make it seem not so
tyil that's not the reasoning, but eh...
mst tyil: nah, not strong at all based on your behaviour in person or on mailing lists repeatedly
tyil MasterDuke: unless he made up a new account of it, about half a year to a year ago
[Coke] Not this again. I thought we hashed this out years ago.
mst [Coke]: no, because liz continued to say she wouldn't attack -me-, instead she was going to attack perl people who were nothing to do with me 22:48
[Coke] mst: Where's the upload?
mst [Coke]: after she deleted all her cpan modules without warning and trashed lots of perl5 people's dependency chains
[Coke]: I am so done with people defending this
all I wanted was people to attack -me-
rather than treating other people as acceptable collateral damage 22:49
but apparently that was too hard
22:49 kline joined
MasterDuke if you really want to be attacked...your mother was a hamster and your father smelt of elderberries 22:50
mst MasterDuke: go away, or I shall taunt you a second time! 22:51
kline hey guys, is everything ok?
i got reports of racist language or something
22:51 spurdo joined
MasterDuke silly english knigget! 22:51
kline MasterDuke: im not sure what the context is but i think we should stay well clear of anything that could be misconstrued 22:52
mst kline: don't worry about it
kline: we're discussing internally a possible mistake that happened a while back 22:53
MasterDuke oh, ha, whoops. shouldn't have spelled that phonetically
mst kline: the reports are either people who're confused or people being knobheads
[Coke] steps away from IRC for a bit and will review.
kline mst: if thats the case im sure it can probably be discussed without trigger words?
since thats i think whats the issue
mst kline: uh, if 'n*gg*r' is a trigger word then I dunno how to avoid it when having such conversations 22:54
kline: cos nobody in this entire conversation has used that word without the asterisks
22:54 throwaway joined
spurdo ****** 22:55
kline mst: i think if people find it objectionable you might want to find some other way of talking about the expression
im sure everyone how "needs to know" and is relevant to the discussion is already aware of the word without it being repeated 22:56
Grinnz "the n-word" is sufficiently descriptive at least to americans
guifa notes that “the N-word” is the most common way to refer to the word itself
kline thats not to say its not an important issue, im just wondering if people need to be as explicit as has been the case so far, since its upsetting people
mst kline: if people are bitching about the asterisked version that's new to me
kline: is that really what's happened? 22:57
kline yes
22:57 spurdo is now known as Ny, Ny is now known as Nyx
mst kline: jeez, ok, I'll switch to saying that tyil nearly uploaded a module to cpan using the N-word as a directory name\ 22:57
sorry about that
tyil I'm incredibly sure that issue was supposed to be buried at least twice now 22:58
and you promised twice now to stop bringing it up
care to hold you to your own promise?
kline mst: is there anyone involved/relevant who is unaware of the language used?
mst tyil: I did, and then liz decided to bring it back up repeatedly
kline: tyil's argument continues to be "it's just a word"
kline mst: im just trying to get to the nub of this. if its not necessary to repeat it, i think its the case that you shouldnt 22:59
mst kline: and "I don't understand why I'm supposed to be sorry about this"
tyil no, my argument at this point is "why bring it up again when it has been delt with countless times now"
at the very least, be honest
mst tyil: this isn't your fault this time mate
tyil it was only the first time
mst it's because liz is insisting on attacking other people
tyil the other times it wasn't me bringing it up to stir up some channel drama
mst because she knows that attacking other people will silence me 23:00
tyil if you have a problem with liz, pm her
if you have a problem with me, pm me
mst tyil: look at the thing I linked
tyil: I never menti8oned your name at all
tyil starting a shitshow in a public channel is a surefire way to get people against you
mst tyil: you're the one who brought -you- up
tyil: maybe don't do that if you don't want mentioning when I was careful not to
tyil (especially since you yourself promised you wouldn't, and are breaking this promise a second time) 23:01
mst tyil: no, I'm not, you broke it yourself, son
tyil you got less careful when kline asked you to stop being racist
so while you were careful, you at least strongly imply you were talking about me
mst maybe you shouldn't've summoned kline then
tyil what are you accusing me of, exactly? 23:02
it isn't your first false accusation with no proof whatsoever
so don't hold back
mst whine whine whine
kline mst: since this is clearly between the two of you 1) it wasnt tyil who brought me here (but its not important to go into that further) and 2) perhaps it could be done in a query between the two of you
El_Che mst: the discussion of not getting clearer if the problem is not tyil or the past action itself
kline without having to disrupt a channel
El_Che s/of/is/ 23:03
mst kline: yes sorry
I didn't mention tyil until tyil started stropping about it though\
and yet somehow it's me that started it?
lol
kline im not sure i really care or that its relevant, but i dont think either of you need to do this in public when nominally there is raku related discussion that could happen 23:04
tyil I can literally see the evidence of that not being true without scrolling
but eh, as always, being decent isn't going to change the course of the story
kline tyil: cmon, you dont need to white knight yourself, can you just call each other out in private
tyil kline: nah, I'll put him back on ignore 23:05
mst has yet to prove he can uphold a single word of what he promises
kline that would be sensible, thanks
tyil no point in trying to argue anything with such a person
23:08 throwaway left
mst kline: I agreed to leave it be 23:09
kline thanks
mst then tyil decided to keep attacking community organisers and conference orgaminsers
he broke the deal and so I reminded people of his mistakes
if he'll stop it, I'll drop it
I've always been happy with that
kline mst: i just want to be clear that i dont really want to be involved or to judge project-internal matters 23:10
mst tyil doesn't seem to be capable of it
kline: course you can't
kline: you can hold no opinion about this at all
kline i just want people to not use any language that can cause friction unless its absolutely necessary
and as far as i can tell thats not been the case tonight
mst of course, and I accused nobody of using such language tonight
kline ace, then i think this topic is probably done 23:11
mst really, I never accussed anybody of using such language during this convewrsation 23:12
kline please
stop
El_Che mst: once you and lizmat both calm down, we may have a calm talk if you both want it. It's too soon now. The discussion about the attack by proxy is difficult to grasp because you need to stay vague to respect someone elses privacy. The previous discussion with the racist name seems closed some time ago, and conflating the whole raku community with it seems counter-productive.
mst El_Che: I've requested this repeatedly. 23:13
El_Che mst: I honestly don't think you nor lizmat are there yet
mst kline: fair enough
23:17 aluaces left, squashable6 left 23:19 squashable6 joined 23:33 MasterDuke left, MasterDuke joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!