»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_log/perl6 | UTF-8 is our friend! 🦋
Set by Zoffix on 25 July 2018.
00:01 tellable6 joined
SmokeMachine vrurg: who can vote on the extensions? 00:10
vrurg SmokeMachine: anyone, unless AlexDaniel decides otherwise. 00:11
SmokeMachine So I’m voting... if I can’t, please let me know that I remove my vote... 00:12
vrurg It's not a decissive vote, it would have only recomendational power.
SmokeMachine++ 00:13
SmokeMachine Yes, yes... 00:14
00:19 leont left
Xliff Extensions? 00:25
00:28 Kaiepi left
guifa Xliff: for file names (currently .p6, .pm6, .pod6, .t(6)) 00:32
00:32 Kaiepi joined 00:33 nekomune joined
AlexDaniel yeah, I think anyone can react with a thumbs up emoji :) 00:35
vrurg Xliff: github.com/perl6/problem-solving/issues/106
AlexDaniel though there's no guarantee that jnthn will go with the result that has most votes 00:36
I mean, with the option 00:37
00:47 p9s joined
AlexDaniel also: hello world 00:48
.tell also test test
tellable6 AlexDaniel, I haven't seen also around, did you mean nals?
00:52 lucasb left, p9s left 01:23 Kaiepi left, Kaiepi joined 01:24 Guest13777 left, success joined 01:25 success is now known as Guest49790, wildtrees left 01:34 Doc_Holliwood left 01:37 molaf left 01:43 Sgeo_ joined 01:46 Sgeo left 01:50 molaf joined 02:04 Sgeo_ left 02:06 Sgeo joined, Cabanossi left 02:15 Cabanossi joined 02:18 Seance[m] left, folex left, Matthew[m] left, matiaslina left, lance_w[m] left 02:19 roy[m] left, uzl[m] left, AlexDaniel` left, mack[m]1 left, rba[m] left, xliff[m] left, TravisRt2botio[m left, unclechu left, EuAndreh[m] left, sergiotarxz[m] left, Guest53157 left, aearnus[m] left, Demos[m] left, BlackChaosNL[m] left 02:25 AlexDaniel` joined 02:36 p9s joined 02:38 BlackChaosNL[m] joined, EuAndreh[m] joined, Matthew[m] joined, Demos[m] joined, TravisRt2botio[m joined, sergiotarxz[m] joined, lance_w[m] joined, Guest19607 joined, rba[m] joined, unclechu joined, roy[m] joined, aearnus[m] joined, folex joined, xliff[m] joined, Seance[m] joined, matiaslina joined, mack[m]1 joined, uzl[m] joined 02:41 Guest49790 left, p9s left, success joined, success is now known as Guest33122 02:47 p9s joined 02:50 rba[m] left, Demos[m] left, Guest19607 left, domm left, spacedbat left, Hotbees left, ChoppedBacon left, tbrowder left, mtj left, Bucciarati left 02:51 Demos[m] joined, rba[m] joined, Guest19607 joined, p9s left 02:53 khisanth_ left 02:56 domm joined, spacedbat joined, Hotbees joined, ChoppedBacon joined, tbrowder joined, mtj joined, Bucciarati joined 03:05 Actualeyes joined 03:06 squashable6 left 03:07 p9s joined 03:08 squashable6 joined 03:12 khisanth_ joined 03:18 domm left, spacedbat left, Hotbees left, ChoppedBacon left, tbrowder left, mtj left, Bucciarati left 03:21 p9s left 03:24 domm joined, spacedbat joined, Hotbees joined, ChoppedBacon joined, tbrowder joined, mtj joined, Bucciarati joined, Xliff left 03:31 mtdms left 03:33 cwilson left 03:34 cwilson joined
holyghost I've finished the statistical chaos theory book, if I comprehended noise in a stochastical dynamic system has Fermi-Dirac statistics 04:05
04:06 sno left
holyghost I can write a program for it e.g. to calculate noise in a ANN 04:06
it is however just a few functors for calculating noise with properties 04:08
So I am not going to do that
Next, I will work some more on my cell systems, maybe noise included 04:09
The noise function can also be used to trace in orbitals (of molecules) 04:12
I can use that in the cell system's base system with elementary particles 04:13
Hence the Fermi-Dirac statistics
05:04 Cabanossi left 05:18 Cabanossi joined 05:28 Sgeo_ joined, reach_satori left 05:30 Sgeo left 05:57 domidumont joined 05:59 sauvin joined 06:06 molaf left 06:17 mtdms joined 06:20 mtdms left 06:31 reach_satori joined 06:33 wamba joined 07:21 sena_kun joined 07:26 leont joined 07:27 Cabanossi left 07:32 Cabanossi joined 07:36 scimon joined 07:37 dakkar joined 07:38 leont left
xinming_ SmokeMachine: Why do you use the my $*RED-DB to indicate the different database? If an app wants to have 2 connections to the different database with same schema. Do we need to execute 'my $*RED-DB = xxx' before each database connection? 07:46
I just don't get it the design purpose.
07:50 Ven_de_Thiel left 07:51 kensanata joined
sena_kun xinming_, it is an old API which will be likely replaced with red-do blocks, see github.com/FCO/Red/issues/153 07:51
SmokeMachine xinming_: you can set different $*RED-DB for different blocks... but you can also use `red-do` github.com/FCO/Red/blob/master/t/22-red-do.t 07:53
xinming_: Yes! that what sent_kun said! 07:54
xinming_ What if we need 2 connections for different host? 07:57
We'll have to nest the red-do?
I'd really like that we can have the $schema object just like DBIC does, instead of using the global $*RED-DB thing in this case. 07:59
Maybe something like, %RED-DB<a b c d>; red-do<b c d> -> $b, $c, $d { .... } is also acceptable. 08:00
SmokeMachine xinming_: you can do something like: `red-do host1 => { ... }, host2 => { ... }, host1 => { ... };`
xinming_ SmokeMachine: What if we want host1 and host2 needs to interchange data? 08:01
But the red-do<b c d> -> ... { } still uses the $schema object like thing. 08:02
SmokeMachine xinming_: `red-do host1 => { my $data; red-do host2 => { $data = ... }; use $data }`
xinming_: but I agree we can find a better way for doing it... 08:03
xinming_: the problem with `red-do<b c d> -> $b, $c, $d { .... }` is that we do not use $b, $c, $d as variables in Red, as we usually don't use $*RED-DB directly... 08:04
xinming_ Yea, I don't like the global $*RED-DB var design. 08:05
SmokeMachine xinming_: you can also do: `my $a = database ...; my $b = database ...; my $*RED-DB = $a; my $data = do { my $RED-DB = $b; ...; $data }` 08:06
xinming_: $*RED-DB isn't global...
xinming_ SmokeMachine: The problem with this design is, whenever we use the database, We'll need to set $RED-DB 08:07
scimon I think you need to bear in mind the general use case. Most people using an ORM don't need multiple db connections. So you need to jump through some hoops. 08:08
xinming_ Actually, I don't know why this design is implemented in first place. :-)
SmokeMachine xinming_: if red-do return data (that would make sense), you could: `my $data = red-do host1 => { ... }; red-do host2 => { use $data }`
xinming_ scimon: It's a bit unnatural to me to use it like this. 08:09
SmokeMachine xinming_: how would you do that?
xinming_ I personally like $db1.^find(xxx); $db1.^model('Person').delete; so, this way, We can use something like, my $rw-db = get-db('writer'); my $ro-db = get-db('reader'); 08:11
I personally like to think the database as an object, instead we initialize the $RED-DB before doing things. 08:12
red-do thing makes me think of use of cro. In cro, that works. But in db access, this makes me feel unnatural, Maybe I don't get used to it yet. 08:14
SmokeMachine xinming_: if you want use different databases to read and write, just for it, it's planned to `database :rw(Pg => (:host<host1>)), :ro("SQLite")`
08:15 Itaipu left, Itaipu joined
xinming_ SmokeMachine: I don't only mean the different db read/write, what I mean is, the get-db() function can return a "thing" that we can easily use. 08:15
SmokeMachine xinming_: or something like that...
sena_kun what about adding the db object as an optional argument to model methods? E.g. Person.^create($db, ...) and Person.^create(...) are easy to distinguish. Though I haven't looked at the sources to check if that's a sane option. Then you can `my $a = database ...; Person.^all($a)` and it isn't tied to dynamic variable. 08:19
xinming_ Also acceptable. 08:20
SmokeMachine xinming_: maybe we could to something like: `Model.^load: 42, :$db`, `Model.^all.grep: { .col = 42 }, :$db`, etc...
xinming_ What I thought that maybe we create a wrapper something like that.
and everytime we call the method on that object, It'll auto set the $RED-DB before calling any method in that class. 08:21
sena_kun SmokeMachine, that's a bad idea, because e.g. `Model.^create: foo => 42, :$db`. <- is this `$db` a part of the model or not? if yes, how to specify db?
SmokeMachine sena_kun: yes! something like that... that methods could just set $*RED-DB...
08:22 wamba left
sena_kun it just takes a new call candidate for every model method that'll be setting the variable and call the other candidate. 08:23
08:24 Itaipu left
SmokeMachine sena_kun: I was thinking on something like (idk if it works): `method create(:$*RED-DB = $*RED-DB, *%pars) {...}` 08:24
08:24 Itaipu joined
SmokeMachine sena_kun: I agree with you that's a bad idea... but I don't like that much the positional parameter... 08:25
sena_kun hm.... you can argue it both ways, I think. I just think that having it as first possible argument immediately gives you knowledge where the transaction goes, while having it at the end of an arbitrary length call is kind of harder to grasp. I don't know if that's worth anything, of course. 08:26
like `$a($db, $a, $b, $c, $d)` and `$b($db, $f, $g)` are pretty structured, while `$a($a, $b, $c, $d, $db)` and `$a($f, $g, $db)` are not, especially if they are not a single letter variables. 08:27
because people might write boilerplate like username => $user.username, email => $user.email and so on. 08:28
SmokeMachine sena_kun: yes, makes sense... 08:35
sena_kun: but still for me, the database that command is using seems for me as an adverb... 08:37
08:47 vike left 08:51 kensanata left
SmokeMachine `sub using-db($*RED-DB, &block) { block $*RED-DB } 08:53
`sub using-db($*RED-DB, &block) { block $*RED-DB }`
this could be the solution as well...
sena_kun, xinming_ ^^ 08:54
sena_kun `sub using-db($db = $*RED-DB, &block) { block}`? 08:55
more like `sub using-db($db = $*RED-DB, &block) { $*RED-DB = $db; block() }` 08:56
SmokeMachine sena_kun: could be... but passing $*RED-DB as first argument would make it possible to: `using-db $db, { .execute: "create table ..." }` 08:58
08:58 vike joined
SmokeMachine why not using the dynvar on the signature? 08:58
sena_kun aren't users suggested to use model API, not calling stuff like .execute? I mean, it is more like edge case to me than a common one 08:59
SmokeMachine m: sub bla { say $*v }; sub ble($*v) { bla }; ble 42
camelia 42
08:59 wamba joined
sena_kun ah 08:59
well, either way makes sense
SmokeMachine sena_kun: I know at least 1 user that keep using `.execute`... I'm not sure if I should incentive it or not... 09:00
sena_kun well, the API is useful for edge cases, no problem here
I meant that having it as $_ will not help much for calls like Model.^create, but I see now it doesn't matter much 09:01
SmokeMachine sena_kun: yes, I think there is nothing outside the block that should be passed as argument that wouldn't be better to be used as a simple variable... 09:04
sena_kun: so we can use $_ to pass something internal that could be useful...
sena_kun, xinming_: but what do you guys prefer? the $db on every method or a simple using-db? 09:06
scimon: ^^
using-db $db, *.execute: "create table ..." 09:11
scimon I think using-db: 09:13
Like I say. Generally it's going to be the exception not the rule. 09:14
IMO
09:19 ravenousmoose joined 09:25 dwarring left 09:27 HarmtH joined, zakharyas joined 09:34 Doc_Holliwood joined
SmokeMachine scimon: I agree... 09:51
vrurg: what do you think about that? ^^
09:59 MilkmanDan left, MilkmanDan joined 10:03 Sgeo__ joined 10:06 Sgeo_ left 10:08 reach_satori left 10:24 aborazmeh joined, aborazmeh left, aborazmeh joined 10:25 sftp left 10:29 sftp joined, sena_kun left
SmokeMachine xinming_: would you mind to open a issue about the using-db, please? 10:32
10:33 pat_js joined 10:44 kensanata joined 10:59 Black_Ribbon left 11:22 kensanata left 11:47 kensanata joined 12:09 sena_kun joined 12:12 sena_kun left 12:19 sena_kun joined 12:24 sena_kun left 12:29 sena_kun joined 12:30 sena_kun left 12:39 molaf joined 12:42 pat_js left 12:48 squashable6 left 12:51 pmurias joined
pmurias what's the recommended way of feeding large constant strings into EVAL? 12:51
in this case when evalling chunks of JavaScript
12:53 squashable6 joined, ChanServ sets mode: +v squashable6
pmurias I'm using heredocs but that requires a use MONKEY-SEE-NO-EVAL 12:55
moritz so use MONKEY-SEE-NO-EVAL 13:05
AlexDaniel pmurias: there's also EVALFILE 13:17
although that probably doesn't help you at all :)
vrurg SmokeMachine: I'm out of context and can't really read through the discussion now. If you get an issue with the proposal I'd check it out. 13:19
13:19 lucasb joined
tellable6 2019-09-17T08:28:45Z #moarvm <lizmat> vrurg I have 13:19
vrurg SmokeMachine: for the moment I'm not sure what's the reason?
SmokeMachine vrurg: the reason is to make it simple to use Red with more than one database at the same time... 13:22
vrurg SmokeMachine: still, would be nice to see the proposal. But a way to do so is needed, no doubt. 13:23
13:24 p9s joined 13:25 AlexDaniel is now known as messagestoi, messagestoi is now known as AlexDaniel, aborazmeh left
SmokeMachine vrurg: first option: add a new candidate to Avery method waiting a database driver object, and that would set the $*RED-DB and call the original one 13:26
second: add a :$&RED-DB to the signature of every method
third: `sub using-db($*RED-DB, &block) { block $*RED-DB }`
vrurg SmokeMachine: using-db is the best, no doubt. I think with-db sounds better, but that's minor. 13:29
SmokeMachine vrurg: yes, makes sense: with-db... I liked it! 13:30
scimon Yup
SmokeMachine if the obj is not defined, should I just ignore the block? as the `with`? 13:31
AlexDaniel Geth: help 13:33
Geth AlexDaniel, Source at github.com/perl6/geth To add repo, add an 'application/json' webhook on GitHub pointing it to hack.p6c.org:8888/?chan=#perl6 and choose 'Send me everything' for events to send | use `ver URL to commit` to fetch version bump changes
vrurg SmokeMachine: not sure what you mean (sorry, have to do main work now). But if you make with-db multi-sig then it makes sense replacing red-do with with-db: with-db $db, { ... }; with-db { ... } – default 13:34
SmokeMachine ```sub with-db($*RED-DB, &block) { .&block with $*RED-DB }``` ???
vrurg SmokeMachine: I can't really focus now. I'd better get back later. 13:35
SmokeMachine vrurg: ok, np... thanks!
13:39 cpan-p6 left, cpan-p6 joined, cpan-p6 left, cpan-p6 joined
SmokeMachine with-db looks alot red-do... maybe both should be the same... and we should decide the best name for it... github.com/FCO/Red/blob/master/lib...Do.pm6#L68 13:40
vrurg SmokeMachine: BTW, red-do :db($other-db), { ... } or red-do $other-db, { ... } 13:45
SmokeMachine vrurg: yes... I think it would be better mantain the `red-do` and add this candidate... 13:46
but we don't realy need that... we could just: `red-defaults db1 => database ..., db2 => database ...; my $a = red-do "db1", { ... }; red-do "db2", { do something with $a }` 13:50
sorry: 13:51
13:51 p9s left
SmokeMachine ```red-defaults db1 => database ..., db2 => database ...; my $a = red-do :use<db1>, { ... }; red-do :use<db2>, { do something with $a }``` 13:52
pmurias moritz: feel wrong to use that pragma just because of a rakudo bug 13:53
13:54 isBEKaml joined 13:56 zakharyas left 13:57 sena_kun joined
MasterDuke pmurias: any objection to me trying to rebase npq's truffle branch up to master's HEAD? 14:00
14:02 pmurias left 14:03 cpage_ joined 14:04 pmurias joined 14:06 cpage left, cpage_ is now known as cpage
SmokeMachine just kidding: raku.do, raku.dolib, raku.dodoc, raku.dotest / raku.do, raku.dol, raku.dod, raku.dot 14:07
timotimo raku.club premium 14:09
SmokeMachine I thought of .ku, but the meaning of that in portuguese would not be very good...
timotimo $349.99*
14:11 nepugia joined
SmokeMachine red-defaultsdb1 => database ..., db2 => database ...; my $a = red-do :use<db1>, { ... }; red-do :use<db2>, { do something with $a } 14:12
sorry... 14:13
xinming_: would that be ok for you? www.irccloud.com/pastebin/466lwR2g/ 14:16
pmurias MasterDuke: none, go for it 14:21
MasterDuke pmurias: cool. i'm not going to make any promises, but i'll give it a try 14:22
vrurg SmokeMachine: red-do :with<db2> { ... } 14:23
SmokeMachine: or red-do db2 => { ... } 14:24
SmokeMachine vrurg: but `red-do db2 => { ... }`would return an array...
(if add a `do` at the beginning) github.com/FCO/Red/blob/master/lib...Do.pm6#L78 14:25
vrurg m: sub foo ( *%named ) { say %named }; foo db => { say "Ok" } 14:28
camelia {db => -> ;; $_? is raw { #`(Block|60037944) ... }}
vrurg SmokeMachine: ^
SmokeMachine vrurg: that way you loose the order...
vrurg SmokeMachine: it makes possible paralellizing of bodies on different databases: red-do db1 => { ... }, db2 => { ... }
SmokeMachine: sure, but one not always require it. Plus, see my previous comment. :) 14:29
SmokeMachine it is possible to run that way now, but it respects the order you added...
vrurg Besides, you could have variants. red-do with a positional block parameter would require :with or :use, all named – the above semantics. 14:30
SmokeMachine vrurg: github.com/FCO/Red/blob/master/t/2...d-do.t#L38 14:31
vrurg m: sub foo ( *%n ) { .say for %n.key }; foo( db1 => 1, db2 => 2, db3 => 3, db4 => 4)
camelia No such method 'key' for invocant of type 'Hash'. Did you mean 'keys'?
in sub foo at <tmp> line 1
in block <unit> at <tmp> line 1
vrurg m: sub foo ( *%n ) { .say for %n.keys }; foo( db1 => 1, db2 => 2, db3 => 3, db4 => 4)
camelia db2
db1
db4
db3
SmokeMachine m: sub foo ( *@n ) { .say for @n }; foo( "db1" => 1, "db2" => 2, "db3" => 3, "db4" => 4) 14:32
camelia db1 => 1
db2 => 2
db3 => 3
db4 => 4
SmokeMachine `red-db :parallel` and `red-db :serial` ? 14:34
vrurg Ah, you mean this. Ok. :) Implementing red-do :with(db) would require breaking backward compat now, isn't it? 14:35
SmokeMachine `red-db ({...}, db1 => {...}, {...}).race` ?
vrurg SmokeMachine: :async is shorter and :serial is to be the default. 14:36
SmokeMachine vrurg: yes... but, at least it was still experimental...
vrurg SmokeMachine: I have to go now. It's interesting subject, worth an issue to consider all details. 14:37
SmokeMachine vrurg: yes, I agree!
14:45 xq joined 14:57 AlexDaniel left, AlexDaniel joined 14:58 AlexDaniel left, pmurias left, AlexDaniel joined, AlexDaniel left, AlexDaniel joined 15:00 ft4l joined
lucasb tyil: hello. are you the one responsible for Geth? I added a new channel to its config.json. Is a 'git pull' enough to make it join the new channel? 15:01
tyil I am currently hosting it, I'll need to git pull it and build a new docker image 15:02
guifa Ugh, Comma hung on me, and after I forced quit, it won’t launch anymore.
jnthn: sadly console output isn’t giving me much of anything I could send you, have you seen this problem before, though? (launch and immediately exists of its own volition, it seems) 15:03
15:03 AlexDaniel left
jnthn guifa: No, never seen (or heard reported) that one 15:04
guifa: Though it does maybe have an "am I already running" check, so check it's not somewhere in the process list (and send it a kill if so)
15:05 AlexDaniel joined, AlexDaniel left, AlexDaniel joined
guifa Err okay I’m going to go put my dunce hat on. Last night I tried deleting the app support folder and rebooting, but forgot totally about prefs folder 15:07
Lemme see if I can recreate by replacing the old prefs folder. 15:08
15:11 AlexDani` joined
guifa Bingo. Can consistently cause it by putting in this particular preferences folder, when I take it away, and it creates preferences from scratch, launches no problem. I don’t have any sensitive information in it, so I can zip it up and send it to you if you’d like (I’ve tried playing around to see which file is toxic to the launch process but haven’t figured it out just yet) 15:11
15:12 Geth left, Geth joined 15:13 AlexDaniel left 15:18 isBEKaml left 15:20 APic joined 15:21 domidumont left 15:23 Geth left, Geth joined 15:25 APic left
[Coke] .seen finanalyst 15:36
tellable6 [Coke], I saw finanalyst 2019-08-26T10:02:26Z in #perl6: <finanalyst> jjmerelo: hi.
15:37 Ven_de_Thiel joined
[Coke] .tell finanalyst Sent you several emails, looks like you missed the first one, check your spam filter? 15:37
tellable6 [Coke], I'll pass your message to finanalyst
15:44 MilkmanDan left 15:46 MilkmanDan joined 15:49 pat_js joined 15:51 mowcat left 16:02 ft4l left 16:04 kensanata left
SmokeMachine vrurg xinming_ sena_kun scimon: github.com/FCO/Red/issues/400 16:13
scimon Neat :) 16:15
16:26 scimon left 16:30 AlexDani` is now known as AlexDaniel, AlexDaniel left, AlexDaniel joined, dakkar left 16:51 p9s joined 16:55 leont joined 16:56 p9s left 17:07 molaf left 17:09 pat_js left 17:10 natrys joined, robertle joined 17:17 wildtrees joined 17:26 APic joined 17:30 Poohman joined
Poohman hello all, do we have something in perl6 like type providers in F# 17:31
17:31 Doc_Holliwood left 17:32 Doc_Holliwood joined
moritz what's that? 17:34
Poohman using these type providers it is possible to create types in F# using excel tables (headers/labels), Json files, SQL etc 17:36
tadzik oh, you can write those no problem :) 17:37
Poohman using one statement it eliminates all boilerplate and directly creates the type which is then available in Visual Studio using Intellisense etc
but this is then somehow used in compile time 17:39
jnthn Sounds like what ASN::META does
17:42 Poohmaan joined
Poohmaan tadzik: cool, pull something from excel tables generated from SAP tables?? 17:43
17:43 Poohman left
Poohmaan tadzik: will that be trivial or will need to invest sometime? 17:44
I was wondering if RED did something like that - but I think it is only for databases 17:46
17:52 ravenous_ joined 18:00 pmurias joined 18:01 sno joined 18:08 MasterDuke left 18:12 wamba left 18:19 MasterDuke joined 18:20 ravenou__ joined 18:23 ravenous_ left 18:26 sauvin left 18:28 cwilson left
sena_kun sounds like what ASN::META does. :) I think you need to invest the time, though, to support exact descriptions you need to support 18:33
18:34 aborazmeh joined, aborazmeh left, aborazmeh joined
sena_kun >More formally, type providers are a form of compile time meta-programming. 18:39
hmmmmm...
18:44 AlexDaniel left 18:45 AlexDaniel joined, AlexDaniel left, AlexDaniel joined 18:50 Guest33122 left 18:55 aborazmeh left 18:58 mowcat joined 19:03 success joined, success is now known as Guest27953 19:04 Ven_de_Thiel left 19:12 Itaipu left 19:13 pmurias left 19:14 squashable6 left 19:15 squashable6 joined, ChanServ sets mode: +v squashable6 19:17 wamba joined 19:18 Guest27953 left 19:21 pmurias joined
[Coke] .seen jmerelo 19:22
tellable6 [Coke], I saw jmerelo 2019-09-12T16:09:23Z in #perl6: <jmerelo> antoniogamiz: :-)
tadzik Poohmaan: well, you need to do the excel part yourself I suppose 19:27
but creating the types themselves in code is mostly trivial
pmurias MasterDuke: I haven't touched nqp-truffle code in a while but if you need help with anything I'll try to help with that
19:28 wamba left, wamba joined
MasterDuke pmurias: i've got the rebase up to the significant changes in the build system 19:28
and thought i'd addressed the first conflict in Configure.pl 19:29
and it started building, but now the compilation itself is errorring out
pmurias: what's happening right now gist.github.com/MasterDuke17/bbdc0...cd69a10532 19:30
pmurias that a java.lang.ProcessHandle not being present error 19:32
MasterDuke missing import? 19:33
pmurias GraalVM seems to be JDK 8 based
and ProcessHandle is a JDK 9 thing
19:34 ravenou__ left
MasterDuke yeah, looks like graalvm probably won't support jdk 9 until graalvm 20. might be simpler to put off the rebase until then 19:35
19:36 failure joined, failure is now known as Guest45192
pmurias MasterDuke: is that something they have commited to doing? 19:36
MasterDuke pmurias: github.com/oracle/graal/issues/651 19:37
so yes? but not quickly? 19:38
19:39 Guest45192 left, success- joined
Doc_Holliwood is it possible to chain multiple greps, using the whatever form? `list.grep: * (elem) foo;` i would like to filter add an additional grep 19:41
timotimo that's not "the whatever form", that's the "colon form" :)
you can use parenthesis and put the whatever code inside that
pmurias MasterDuke: it seems we need jdk 9 (only?) for the getppid op 19:42
MasterDuke pmurias: over in #perl6-dev vrurg says he has plans for further build system changes. so i think i won't bother with the rebase right now, and will wait until after his changes and java 9+ support in graalvm
19:43 domidumont joined
MasterDuke but i do hope to do more with the truffle branch (just implementing more stuff, not the build system revamping) 19:43
so just leave the branch as is and add more commits (that shouldn't effect any future rebase) 19:44
vrurg MasterDuke: you're pushing me to work on NQP sooner than I'm ready for it. Seeing faster jvm backend is so exciting! 19:45
19:46 Ven`` joined
MasterDuke you should chat with pmurias, he actually understands it, i just did some simple stuff 19:46
but truffle/graalvm is way cool
you can create native executables that start up almost instantly 19:47
Doc_Holliwood when I do `constant \alphabet = ( 'a' .. 'z', 'A' .. 'Z' ).Set;` within a sub, is the work done once or on each call?
MasterDuke and i had some simple number crunching examples that where faster on that branch than on moarvm (though that was a while ago, moarvm has had a lot of optimizations since then) 19:48
cpan-p6 New module released to CPAN! XML::Actions (0.3.3) by 03MARTIMM
timotimo constant runs at BEGIN time, i.e. during compilation
i think you need a "flat" in there, though
19:48 pmurias left
Doc_Holliwood Sweet 19:48
19:54 pmurias joined 19:55 success- left 19:58 success joined
SmokeMachine github.com/FCO/Red/issues/400#issu...-532377649 19:58
19:59 success is now known as Guest70324 20:00 domidumont left
xq there is a weird issue on docs.perl6.org; example: I open docs.perl6.org/type/Range press ctrl-f to open my browser's search bar, enter some string, press F3 several times to find a needed match, then press escape to close browser search bar, my browsers viewport moves to the top of the page, and focusing the pages own search bar 20:00
effectively discarding all my search efforts 20:01
I think this is a serious usability problem
rba xq: Just tried the same on Chrome on my Mac. Can not confirm the behaviour. 20:04
xq chromium 76.0.3809.100 (Developer Build) built on Debian bullseye/sid, running on Debian bullseye/sid (64-bit)
20:06 Cabanossi left
rba Google Chrome Version 76.0.3809.132 (Official Build) (64-bit) on macOS High Sierra 10.13.6 20:06
xq my exact keypresses after opening the page are: ctrl-f -> main -> f3 -> f3 -> f3 -> esc 20:07
20:08 Black_Ribbon joined, Cabanossi joined, wamba left, wamba joined 20:10 leont left
rba Can confirm your finding on Windows 10, Google Chrome Version 76.0.3809.132 (Official Build) (64-bit) 20:11
xq: May I ask you to open an issue under github.com/perl6/doc/issues ? 20:12
xq ok 20:13
rba Would guess, that ctrl-f is in javascript to lead to the search box. On my Mac it's cmd-f, yet ctrl-f doesn't work either.
xq it's probably not only ctrl-f but f3 as well, if I have some "stored" search inside browser already 20:15
i.e. pressing f3 instead of the initial ctrl-f produces the same end result
Elronnd xq: I have the same issue as you. I made a userscript to disable the behaviour 20:21
20:21 pmurias left 20:22 zakharyas joined, sena_kun left
Elronnd 0x0.st/zy5J.txt here 20:22
20:22 Guest70324 left 20:24 success- joined
xq submitted github.com/perl6/doc/issues/3025 20:27
20:29 molaf joined, krychu joined 20:33 pmurias joined 20:36 zakharyas left
Elronnd the javascript is actually listening for esc 20:36
You can try pressing esc anywhere on the page, and it'll take you to the website's search bar
Why it hears that esc even when it's used to close the *browser*'s search box, I don't know 20:37
20:37 melezhik joined
xq ouch 20:37
it does indeed 20:38
melezhik I need to define signature for method that accept both array and single parameter
foo($value); foo(@value)
how do I do that?
moritz m: sub f($x) { say $x.perl }; my @a = <b c d>; f(@a) 20:39
camelia $["b", "c", "d"]
moritz melezhik: just use the fact that a scalar can also hold an array
melezhik Type check failed in binding to parameter '@cmd'; expected Positional but got Str ("C:\\Users\\melezhik/.spa...) 20:40
method !run-bash-command-async (@cmd) {}
on this metod 20:41
method
I get this error when I call it as run-bash-command-async("foo")
timotimo maybe you want a slurpy. perhaps the + slurpy even 20:42
melezhik how do I change run-bash-command-async then? 20:43
timotimo try +@cmd
or *@cmd
melezhik github.com/melezhik/Sparrow6/blob/...on.pm6#L94 20:44
timotimo are you talking about signature?
timotimo yes 20:45
melezhik ok
20:46 Geth_ joined, Geth left, robertle left 20:47 netrino joined 20:48 Geth joined, Geth_ left 21:21 stux|RC-only joined, wamba left, pmurias left 21:22 wamba joined 21:28 pmurias joined 21:30 melezhik left 21:31 Geth_ joined, Geth left 21:35 natrys left 21:36 pmurias left 21:38 pmurias joined 21:44 agentzh left 21:46 agentzh joined, agentzh left, agentzh joined 22:09 reach_satori joined 22:13 netrino left, pmurias left 22:15 success- left
tony-o @ AlexDaniel` is there a way on gh issues to allow people to vote and while closing the issue? (looking at 106 specifically) 22:16
AlexDaniel tony-o: you can react regardless of the status, so you can still 👍 comments on closed issues 22:17
tony-o: but what do you have in mind? I'm not sure I understand :) 22:18
22:18 success joined, success is now known as Guest56967
tony-o i'd like to force that vote to happen so we can move forward, 50+ comments on file extensions and it's mostly (at this point) noise 22:28
AlexDaniel tony-o: that vote is not strictly necessary, just talk to jnthn and lizmat to decide which extensions should go into the PR 22:29
xq are you voting on the file extension, meaning that the new name has been decided already?
AlexDaniel xq: no, there's a PR to rename the language, and we want to include new extensions in that PR 22:30
tony-o ah, i'm going to mute that thread then
xq thanks for clarifying AlexDaniel
(I personally like camelia, btw)
AlexDaniel xq: so we're trying to decide how exactly to modify the PR, and there's still some room for new ideas too :)
xq: as for camelia, that's unlikely to happen 22:31
xq :(
AlexDaniel the current state of the PR is pretty much final, except extensions
see github.com/perl6/problem-solving/pull/89/files
xq I see. 22:32
gdonald Are there any other debuggers besides p6-app-moarvm-debug?
xq I guess I'll just hope it doesn't get approved
timotimo perl6-debug-m exists, but won't debug into libraries or the core setting; comma can also debug perl6 code on moarvm 22:33
AlexDaniel xq: that's also somewhat unlikely from what I see right now, any reason for you to think that? :) 22:34
xq personal preference only
AlexDaniel xq: yeah, well, there's something about “camelia” that made me happy, I'm not sure what. But I'm fine with Raku 22:35
xq it's name of the original perl 6 mascot, it has connection to camel, and it sounds good 22:36
raku is a strange, alien, meaningless word 22:37
(at least to me)
naming things is hard
22:37 mowcat left
AlexDaniel xq: I mean, maybe sometimes it's better when the word does not immediately associate with something specific. Like, there's this: www.camelia.com/ 22:39
I totally wouldn't mind if our language associated with female products, I even think that it's pretty cool. But having something neutral is probably better 22:40
22:53 Guest56967 left 22:54 success joined, success is now known as Guest5539 22:56 krychu left 23:13 Itaipu joined 23:18 Itaipu left 23:23 Itaipu joined 23:24 Ven`` left 23:53 wamba left
rindolf TimToady: here? 23:57
23:59 wildtrees 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!