🦋 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.
discord6 <theangryepicbanana> yay my LLVM library is on github now github.com/ALANVF/Raku-LLVM 00:05
<theangryepicbanana> I still need to add the stuff needed to put it on CPAN though 00:06
00:06 Xliff joined 00:07 aborazmeh joined, aborazmeh left, aborazmeh joined 00:25 pecastro left, aborazmeh left 00:41 aborazmeh joined, aborazmeh left, aborazmeh joined 00:45 sena_kun left 00:46 benjif joined 00:49 gabiruh joined 00:54 lucasb left 00:59 sena_kun joined 01:03 Xliff left 01:04 do_you_thing joined
do_you_thing hi 01:09
Do you think Raku will kill Perl ?
discord6 <theangryepicbanana> Once it takes off maybe 01:15
do_you_thing Do you think Raku will kill Perl ?
discord6 <theangryepicbanana> I already responded
do_you_thing who is theangryepicbanana ?? 01:16
discord6 <theangryepicbanana> *and you already asked
<theangryepicbanana> uh, me?
<theangryepicbanana> I'm on the Discord bridge
do_you_thing I'm not sure if you are chatting with me
of another guy
discord6 <theangryepicbanana> ?
do_you_thing which post a message i don't know about
discord6 <theangryepicbanana> I'm talking to you through a chatbot
do_you_thing who *
oh ok 01:17
discord6 <theangryepicbanana> I'm on the Raku discord server rather than the IRC
do_you_thing why ?
discord6 <theangryepicbanana> because I use discord more
do_you_thing OK. I just can't go on Discord, otherwise my poor notebook will freeze 01:18
discord6 <theangryepicbanana> yea that makes sense
do_you_thing I don't understand why there web app is soo bloated 01:19
01:22 aborazmeh left
tadzik the web is a hilariously inefficient platform, unfortunately 01:26
01:26 do_you_thing left
Grinnz 🙄 01:30
01:30 mowcat left
discord6 <theangryepicbanana> I use the web version tho 01:52
<theangryepicbanana> the app kills my laptop 01:53
01:55 wildtrees left 01:59 molaf joined 02:19 molaf left 02:32 molaf joined 02:39 satori__ left 02:40 reach_satori joined 02:44 sena_kun left 02:56 reach_satori left 02:59 sena_kun joined 03:21 Cabanossi left 03:33 Cabanossi joined 03:40 Jack joined 03:41 Jack is now known as Guest73966
Guest73966 so you male just created a new regex 03:41
03:46 Guest73966 left
AlexDaniel maybe? 04:07
that was weird
04:46 linkable6 left, evalable6 left, linkable6 joined, sena_kun left 04:48 evalable6 joined 04:50 AlexDaniel left 05:00 sena_kun joined
tyil rypervenche: neat, so I don't need to look into anything? 05:17
05:23 ensamvarg joined 05:47 molaf left 06:01 squashable6 left 06:03 squashable6 joined 06:06 jmerelo joined 06:16 Doc_Holliwood joined 06:17 Kaffe joined 06:36 sauvin joined
Geth doc/infix-bool-and-op: 7e3cef998e | (Stoned Elipot)++ | doc/Language/operators.pod6
Add boolean logical AND operator
doc: stoned++ created pull request #3191:
Add boolean logical AND operator
doc: 459431d3cb | (JJ Merelo)++ | doc/Type/Int.pod6
Adds Int.new, commenting the new features. Refs #2632
linkable6 Link: docs.raku.org/type/Int
linkable6 DOC#2632 [open]: github.com/Raku/doc/issues/2632 [Hacktoberfest][RFE][big][docs][good first issue][help wanted][new][⚠ Top Priority ⚠] Checklist for 6.d
06:45 sena_kun left 06:46 cpan-raku left, [particle] joined 06:48 cpan-raku joined, cpan-raku left, cpan-raku joined 06:49 camelCaser joined 07:00 sena_kun joined 07:08 Kaffe left 07:12 rindolf joined 07:14 ensamvarg left 07:15 kensanata joined 07:20 Maylay left
Geth doc/infix-bool-and-op: 242dc90dfb | (Stoned Elipot)++ | doc/Language/operators.pod6
rephrase ?& operator description
07:22 Maylay joined
Geth doc: 7e3cef998e | (Stoned Elipot)++ | doc/Language/operators.pod6
Add boolean logical AND operator
linkable6 Link: docs.raku.org/language/operators
doc: 242dc90dfb | (Stoned Elipot)++ | doc/Language/operators.pod6
rephrase ?& operator description
doc: 21b691ccf6 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | doc/Language/operators.pod6
Merge pull request #3191 from Raku/infix-bool-and-op

Add boolean logical AND operator
07:33 jmerelo left, Doc_Holliwood left 07:34 domidumont joined 07:38 epony left 07:46 Black_Ribbon left, Cabanossi left 07:57 xinming_ left 07:58 xinming_ joined 08:01 Manifest0 left 08:03 Manifest0 joined
Geth ecosystem: rbt++ created pull request #481:
Moved to CPAN.
08:13 polm joined 08:14 polm left, Doc_Holliwood joined 08:19 leont joined 08:27 stoned75 left, xinming_ left 08:30 xinming_ joined 08:37 leont left 08:43 leont joined 08:45 sena_kun left 08:58 xinming_ left 09:00 xinming_ joined 09:01 sena_kun joined 09:06 zakharyas joined, dakkar joined
jjatria I tried running `Supply.interval(1).tap: { say $^a }; sleep 4` and as expected I got numbers from 0 to 4 on different lines. But when I changed the `sleep` for `loop { }` the program kept running but stopped printing numbers at 2 09:13
What's making it do that?
dakkar you're running a single thread, I think 09:14
in 6.d, I'm pretty sure `sleep` goes to the scheduler, which can then run the supply in its event loop
if you're doing other stuff, and not yielding to the scheduler, it never gets the chance
(don't take my word for it, though!) 09:15
jjatria Hm, yeah... I thought it could be something like that. I'll read a little more. Thanks! 09:16
09:20 Doc_Holliwood left 09:32 pecastro joined 09:39 wamba joined 09:40 Sgeo__ joined
jjatria Yeah, if I do `my $p = start { Supply .... }; loop { await $p }` it works. But is that because it is yielding, or because it's no longer on a single thread? 09:42
09:44 Sgeo_ left
jnthn Hmm, loop {} should make no difference, except utterly burning CPU 09:46
I can reproduce the behavior though. 09:47
Oh dear. MVM_JIT_DISABLE=1 in the environment also fixes it 09:48
As does `Supply.interval(1).tap: { say $^a }; loop { [] }` 09:49
My theory would be that the JIT somehow forgets to emit the "check if we need to GC" at backbranches, but looking at the impl, it's right there. (Also, why'd we not have spotted this before...) 09:51
09:51 chloekek joined
chloekek Is there a situation in which you must use $¢ instead of $/? 09:52
Is it so that you can call functions within regex blocks, and then still access the match afterwards, even if the callee modified $/? 09:53
E.g. sub f { $/ = 1 }; /.{ f(123); $¢ neq $/ }/
09:55 leont left
jnthn jjatria: github.com/MoarVM/MoarVM/issues/1230 09:58
jjatria jnthn++ 09:59
10:02 Doc_Holliwood joined
jjatria "which is probably a nice easy speedup" <- ooh, I like that 10:03
Geth ecosystem: 0c5446e734 | (Rod Taylor)++ | META.list
Moved to CPAN.
ecosystem: 723d2237e1 | Altai-man++ (committed using GitHub Web editor) | META.list
Merge pull request #481 from rbt/master

Moved to CPAN.
10:23 antoniogamiz joined, reach_satori joined
cpan-raku New module released to CPAN! Math::Libgsl::Function (0.0.1) by 03FRITH 10:42
10:43 wamba left 10:45 sena_kun left 10:47 Doc_Holliwood left 10:53 ensamvarg joined 10:58 zakharyas left 10:59 sena_kun joined 11:00 zakharyas joined 11:05 mowcat joined 11:08 Doc_Holliwood joined 11:14 epony joined 11:15 aborazmeh joined, aborazmeh left, aborazmeh joined 11:28 domidumont left, shred_alert joined 11:30 wamba joined
shred_alert re: Discord. Their API "documentation" is possibly the worst I have ever seen. 11:31
It's basically a list of JSON objects with comments.
I caved in and just used a discord.py to build a bot for my friend. Couldn't bear to decipher those docs. 11:32
11:35 reach_satori left
shred_alert Then again, can't expect everything to have docs as great as Raku. =D 11:36
tyil is there not a discord module in raku? 11:37
shred_alert Welp. I should have Googled more. 11:38
Thanks tyil
tyil in case of doubt you can always ask around here ^_^
shred_alert On the bright side: it gave me an excuse to make something with Python. 11:39
tyil if you want a new project, you can make a bridge between discord and irc in raku
our current one is written in NodeJS, but I would of course prefer Raku so I can troubleshoot issues if any arise
shred_alert That sounds like a neat project, actually. Once I finish learning some other stuff I've got on the agenda, I'll start thinking about that. 11:40
tyil awesome!
shred_alert Currently learning some operations stuff with Go.
tyil (for future reference, the IRC module is called IRC::Client)
11:41 antoniogamiz left
shred_alert I've been planning on writing some utilities for FreeBSD in Raku as well. 11:41
tyil operations as in, how to do things with go, or using go to perform system maintenance/devops work
shred_alert systems stuff
tyil cool 11:42
shred_alert I'm trying to get my head around stuff like RabbitMQ, Opentracing, etc. and Go is pretty much the go-to for that field. 11:43
11:53 mcsvenster joined, mcsvenster left 11:57 McSvenster joined
tyil our application at work uses rabbitmq in perl and python, but I'm not on the dev team so I don't really know much about it 11:59
I'd still like to do some email stuff with Raku, but I have other projects that take priority ;~;
shred_alert Hope you manage to find the time soon 12:00
12:00 xinming_ left
tyil I have 3 weeks off of work, and I intend to spend it cooking and coding away my backlog :p 12:00
shred_alert Excellent! 12:01
cpan-raku New module released to CPAN! Desktop::Notify::Progress (0.0.2) by 03FRITH
12:01 xinming_ joined
cpan-raku New module released to CPAN! Migration::RTtoGithub (0.0.1) by 03ELIZABETH 12:01
tyil Im at schiphol now to pick up a friend, with some luck I can persuade him to make some time for raku during fosdem 12:02
lizmat tyil++ 12:06
rypervenche tyil: I'm going to test it again and on CentOS as well. I'll let you know in a few hours if I run into any problems. 12:07
tyil rypervenche: thanks a lot!
lizmat .tell AlexDaniel` modules.raku.org/dist/Migration::R...:ELIZABETH # for the next Squashathon ? 12:08
tellable6 lizmat, I'll pass your message to AlexDaniel
AlexDaniel` What's that
tyil also, I'm planning to make an IRC::Client plugin to play hangman on irc, and would like to abstract away the code to give people points for winning (so I can make other irc games later using the same score "database")
does anyone have a good name for the module to hold that abstraction for scores? 12:09
lizmat Game::Score ?
tyil I don't want to put it in the DB:: namespace
lizmat I guess it will be backend agnostic ?
tyil Game::Score could work, tbh
AlexDaniel` Oh, I see
tyil lizmat: I'm going for a postgres backend beacuse I already have postgres, but the intent is to write it such that other backends can be added
in case people prefer mysql, sqlite or a csv for instance 12:10
AlexDaniel` lizmat I thought we'll just attach squashable6 to the github repp
tyil lizmat: I think I'll go with Game::Score, thanks! 12:11
lizmat AlexDaniel`: well, I thought a library with the info would be nice to have
12:11 molaf joined
lizmat AlexDaniel`: I've used it for my own ticket updating 12:12
AlexDaniel` Okay
12:12 kensanata left
dogbert11 heh, just got a segv when building rakudo 12:12
lizmat interesting
AlexDaniel` lizmat maybe improve the readme?
You can mention your module there 12:13
lizmat which readme are you referring to now ?
AlexDaniel` Old tissues tracker one
dogbert11 tries a rebuild 12:14
AlexDaniel` .... Geez I'm on my phone with autocorrect
.oO( old tissues should be thrown away )
AlexDaniel`: will do
AlexDaniel` It's a bit boring here 12:15
Can't sleep
shred_alert Try reading a book AlexDaniel`:p 12:16
AlexDaniel` Send me a link to something short :) 12:18
Geth old-issue-tracker: 636faa4fb7 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | README.md
Add mention of Migration::RTtoGithub
12:20 Maylay left, Maylay joined
shred_alert Uhhhh. Don't know of anything short 12:21
12:24 zakharyas left 12:25 aborazmeh left 12:26 shred_alert left, wamba left
discord6 <theangryepicbanana> how about the docs on raku's macros? that's pretty short 12:35
12:38 Sgeo_ joined 12:41 Sgeo__ left 12:44 kensanata joined 12:45 sena_kun left
AlexDaniel` Heh :) 12:47
12:47 lucasb joined 12:48 Kaiepi left 12:49 Kaiepi joined
cpan-raku New module released to CPAN! Math::Libgsl::Permutation (0.0.1) by 03FRITH 12:49
New module released to CPAN! Math::Libgsl::Matrix (0.0.1) by 03FRITH
discord6 <theangryepicbanana> actually more docs on macros would awesome 12:50
<theangryepicbanana> they seem really powerful
AlexDaniel` And incomplete ;) 12:53
13:01 sena_kun joined
discord6 <theangryepicbanana> yea lol 13:02
<theangryepicbanana> they seem similar to crystal's macros in a way
13:14 Doc_Holliwood left 13:16 titsuki joined 13:17 colomon left 13:18 WAAE5MS joined 13:21 gabiruh left 13:24 Kaiepi left 13:28 _jrjsmrtn joined 13:30 domidumont joined, __jrjsmrtn__ left, domidumont left 13:31 domidumont joined 13:32 gabiruh joined 13:37 chloekek left 13:40 mensvaga joined, titsuki left 13:44 domidumont left 13:56 zakharyas joined 14:00 titsuki joined 14:01 Kaiepi joined 14:02 domidumont joined, gabiruh_ joined, gabiruh left 14:09 Doc_Holliwood joined 14:12 gabiruh joined 14:13 gabiruh_ left 14:20 benjif left, benjif joined, aborazmeh joined, aborazmeh left, aborazmeh joined 14:22 aborazmeh left 14:35 kensanata left 14:36 wamba joined 14:39 Sgeo__ joined 14:43 Sgeo_ left, melezhik joined 14:45 sena_kun left 14:51 Doc_Holliwood left 14:57 kensanata joined 14:59 sena_kun joined 15:19 titsuki left
daxim I have a crash bug 15:19
put into parameterised-set.pm6: sub foo(--> Set[Str]) {}
run: rakudo -I. -e'use parameterised-set'
can you repro?
15:24 molaf left
[Coke] Anyone here do .net development? (Esp. on a mac/) 15:28
dakkar daxim: ah yes, the "Missing serialize REPR function for REPR MVMContext" 15:29
*some* parametrized types can't be serialised, which breaks precompilation
15:31 aluaces left 15:32 mahmudov joined 15:35 Doc_Holliwood joined 16:01 kensanata left 16:12 McSvenster left
cpan-raku New module released to CPAN! Math::Libgsl::Combination (0.0.1) by 03FRITH 16:17
16:31 wamba left 16:44 domidumont left
TreyHarris How could I find the last grapheme Raku can encode in three bytes? 16:45
16:45 sena_kun left
TreyHarris I've been staring at the pack and unpack docs but without a code example I'm getting answers I know aren't right, like "ÿ\0\0" 16:47
From simple arithmetic and knowledge of UCS it should be somewhere around Linear-B 16:48
Definitely not Extended Latin :-)
moritz graphemes? How do you even enumerate them to create a "last" one? 16:50
TreyHarris codepoint, rather, then I can work backwards to see if it's a grapheme by itself 16:51
if it isn't, I can check the pred codepoint, recursively 16:53
16:53 ensamvarg left
TreyHarris I'm just trying to figure out what it's even reasonably possible to encode in a three-byte fixed UTF field 16:53
(It came up in p6-u today as an alleged part of a Microsoft registry API, and it seems batshit to me) 16:54
Grinnz three bytes are enough for up to U+FFFF 16:55
rypervenche So 0xffff.chr then? 16:56
TreyHarris ah, then the missing-grapheme replacement character at U+FFFD, then
U+FFFE and U+FFFF are not valid 16:57
16:58 wildtrees joined
TreyHarris Until OS X 10.4 (I think), OS X displayed a version of this that had the actual codepoint(s) it was standing in for in teensy tiny type printed into the character you could only see by zooming in to 128-pt (on a 72 dpi display; this was pre-Retina I'm pretty sure) 16:59
17:00 cpan-raku left
TreyHarris I think that was actually early PDF behavior which OS X got from its NeXTSTEP Display PostScript roots 17:00
17:01 sena_kun joined
TreyHarris They got rid of it when Retina displays came along because the text didn't disappear at small sizes anymore, it just looked like weird display-artifact fuzziness around the perimeters of the � 17:03
17:03 cpan-raku joined, cpan-raku left, cpan-raku joined
Geth problem-solving/front-matter-for-routines: 4982b8254f | (JJ Merelo)++ | 2 files
Adds proposal for solving #155
El_Che lo
problem-solving: JJ++ created pull request #156:
Adds proposal for solving #155
17:11 pyrimidine joined 17:12 Doc_Holliwould joined, wamba joined 17:13 jmerelo joined 17:14 pyrimidi_ left, Doc_Holliwood left, patrickb joined 17:16 wamba left 17:28 shred_alert joined
Geth problem-solving/front-matter-for-routines: a0ed4da650 | (JJ Merelo)++ | solutions/documentation/front-matter-for-routines.md
might → will
shred_alert I was wondering why unicode wasn't working on my terminal emulators today. Then I realised I forgot to setup locales on my new FreeBSD install......... 17:31
17:33 dakkar left
TreyHarris shred_alert: oops. Default isn't C.UTF-8? Most Linuxes that had default C locale have switched to that quite a few years ago... 17:36
17:36 xinming_ left
TreyHarris And I think the ones that do it in setting time zone all set *.UTF-8 17:36
17:37 aluaces joined, xinming_ joined
shred_alert TreyHarris: default is C >.< 17:37
Recommended procedure is to use login classes 17:41
I need to automate and version control all this. Becoming a pain in the rear.
TreyHarris Huh, I would've thought embedded Linuxes would be the holdouts, but no, AFAICT, they default to C.UTF-8 too 17:44
17:45 Kaiepi left
shred_alert Interesting 17:45
17:45 Kaiepi joined
TreyHarris I could've sworn it was necessitated by zoneinfo changing to include non-ASCII for place names. But nope, it's still ASCII AFAICT 17:51
Oh, maybe it was the non-ASCII tld's
That would make it more universal such that the embedded devices would make sense
shred_alert Funny thing is that Unicode was working just fine inside Emacs. 17:53
'Graphical' Emacs, that is.
So I think Emacs must be doing its own thing
TreyHarris It does. Editors mostly have to, especially if they're cross-platform and can deal with disparate input and output encodings (and where applicable, different ones for different files simultaneously open) like Emacs. Ones that aren't tend to use an HTML or Gtk or Wx, etc., widget that does it itself. 17:56
terminal emulators are probably the only thing that will make an effort to _not_ do something different than locale and the shell prescribes 17:57
No, the presumption goes the other way for congress people; if you have a clearance you're under different rules. Whistleblower provisions or nothing, I'm afraid. 18:07
sorry for the pastebomb
18:19 wamba joined 18:21 shred_alert left
cpan-raku New module released to CPAN! Math::Libgsl::Multiset (0.0.1) by 03FRITH 18:23
18:37 chloekek joined 18:44 kybr left 18:46 sena_kun left
Geth ¦ problem-solving: Kaiepi assigned to jnthn Issue IO::Socket::INET's methods for socket creation may need reconsideration github.com/Raku/problem-solving/issues/157 18:48
18:54 zakharyas left 18:55 vrurg left 18:56 vrurg joined 18:58 Bname joined 19:00 sena_kun joined 19:01 kybr joined 19:07 eseyman joined 19:23 wamba left 19:25 sauvin left
rypervenche .tell tyil So installing rakudo star using moar worked fine on 3 machines. I tried jvm on one and it failed the first time. I'm trying again. It's possible that I don't have the correct environment to be successful though. 19:31
tellable6 rypervenche, I'll pass your message to tyil
19:31 dansolo42 left
tyil rypervenche: awesome 19:32
I never tried the jvm variant
rypervenche This is the output I get when I try jvm, if anyone cares: gist.github.com/rypervenche/7ae2fe...df7037dd86 19:35
19:38 wamba joined 19:40 jmerelo left 19:53 aborazmeh joined, aborazmeh left, aborazmeh joined 20:01 MasterDuke joined
rypervenche Question: I see in this code that something is being assigned to $_ and then some search and replace regex lines are below that. What would one of those lines look like with an explicit $_ ? Would it be something like $_ ~~ s:g/'-'//; ? I'm trying to figure out what it's short for. 20:16
It might help if I paste the link: github.com/Raku/whateverable/blob/...ble.p6#L47
lizmat m: $_ = "foo"; s/foo/bar/; .say 20:20
camelia bar
lizmat m: say "foo" ~~ s/foo/bar/ 20:21
camelia Cannot modify an immutable Str (foo)
in block <unit> at <tmp> line 1
lizmat m: $_ = "foo"; $_ ~~ s/foo/bar/; .say
camelia bar
[Coke] rypervenche: can you open a ticket for the JVM build failure? 20:28
lizmat I bet bartolin is aware of it 20:30
rypervenche [Coke]: At github.com/rakudo/star ?
[Coke] or rakudo/rakudo 20:31
(looked like a compiler error, no?)
rypervenche It was testing tyil's rakudo start RC.
rakudo star*
lizmat I'm not sure JVM even builds atm
[Coke] lizmat: thought it was just fixed.
(like, in past 2 weeks)
lizmat I hope it is...
[Coke] rypervenche: I'm sure either is fine. 20:32
bartolin rypervenche: the error you're seeing with the JVM backend has been fixed yesterday: github.com/rakudo/rakudo/commit/a08f043953 20:33
(on HEAD) 20:34
I'm running a spectest right now (but that takes quite some time)
rypervenche Looks like tyil might need to update the RC then. 20:35
20:36 libertas_ left, libertas left
bartolin rypervenche: i'm not really sure about the state of things with the 2020.01 release. I *thought* there was a release branch and the latest commits (including the commit that broke the JVM build) didn't make it into the release. But now I wonder why tyil's RC for rakudo star has the error. I expected it to be based on the 2020.01 release. 20:44
20:45 sena_kun left 20:46 Kaeipi joined, Kaiepi left
rypervenche Actually, it looks like that file has the parenthesis in it in the rakudo-star's rakudo source. 20:49
20:49 Bname left
rypervenche Oh! It's removing it. Derp. 20:49
21:00 sena_kun joined 21:01 xinming_ left 21:02 xinming_ joined 21:03 aborazmeh left 21:11 rindolf left
tyil bartolin: it is based on 2020.01 for nqp and rakudo, and 2020.01.1 for moarvm 21:13
rypervenche: I have no clue what to fix, but there's bound to be a core dev that has some insight about it 21:14
rypervenche tyil: It looks like it was already fixed in github.com/rakudo/rakudo/commit/a08f043953 . I'm testing the install now with the change.
tyil awesome work 21:15
rypervenche Assuming I have enough RAM to build it...
bartolin tyil: I see. (And that seems very sane). So the problem is in 2020.01, obviously.
tyil I can make a branch of 2020.01, cherry pick that commit and use that, but if possible I'd like an official 2020.01.1 of rakudo 21:16
of course, coredevs willing, I'm not sure on their policies on it
(I haven't been doing rakudo star for that long yet)
bartolin actually, I doubt a point release is worth the trouble. the JVM backend isn't in a good shape, anyway. 21:17
21:21 Kaeipi left 21:22 wamba left
lizmat rypervenche: there was one more commit and one potential issue that bartolin mentioned tonight 21:24
21:26 Kaiepi joined, veesh_ joined
rypervenche I'm guessing I should stop worrying about JVM at this point. I was just trying to be thorough. :P 21:26
21:28 veesh left
[Coke] I would treat JVM issues as needing to be fixed but not blockers. 21:28
so, keep worrying but only a little. 21:29
21:29 veesh_ is now known as veesh
rypervenche lizmat: Oh, and thanks for the response on the $_ search and replace thing. So it's like doing $_ = before it. Got it. 21:30
21:30 vike joined
lizmat :-) 21:30
21:30 stoned75 joined 21:45 xinming_ left, xinming_ joined 22:19 stoned75 left 22:46 sena_kun left 22:56 NODE left 22:57 NODE joined 22:58 Kaeipi joined 22:59 Kaiepi left 23:02 sena_kun joined
sena_kun re point release for JVM: if we can put a monkey patch into star, that'd be nice. OTOH in some time I'm here I didn't see a person using this backend and all the answers are "Well, it's there BUT", so point release seems a bit redundant. The good news is that 1)We are back to montly releases anyway, so one doesn't need to wait for a couple of months to. 20just ping me this weekend with instructions on how to test if the thing builds at all 23:02
(I am not sure releasable knows about it?) and we'll do some magic to ensure jvm backend builds now and forever.
23:08 titsuki joined 23:09 MasterDuke left 23:17 chloekek left
SmokeMachine vrurg: hi! Just to be sure, so that was not a Red bug? 23:27
vrurg SmokeMachine: luckily, not 23:28
tellable6 2020-01-29T21:11:48Z #raku-dev <lizmat> vrurg but that piece of code has been there for years :-)
23:29 wildtrees left
vrurg Though it looked suspicious at the first glance, but that's because I didn't know how exactly 'build' attribute is been used. 23:30
SmokeMachine vrurg: great!! I’ve been without time... I still have to fix the tests running on Pg... :(
23:31 wildtrees joined 23:33 wildtrees left
vrurg SmokeMachine: BTW, do you have a suggestion how to speed up storing thoughands of records in a sqlite DB? 23:33
23:33 wildtrees joined
SmokeMachine vrurg: where are you getting the records from? 23:34
vrurg I generate them. :) The scheme is: an app produces tracing records (debug prints). A dedicated thread accepts them via a channel and sends to a DB. 23:36
SmokeMachine: the problem arises when app finishes and the writer thread takes ages to dump out all remaining records.
23:39 oodani joined
SmokeMachine vrurg: have you tried `PRAGMA synchronous = OFF`? 23:40
vrurg SmokeMachine: No. Didn't know it exists. Is it sqlite feature? 23:41
SmokeMachine: BTW, Red itself is not thread-safe, as far as I understand? 23:42
SmokeMachine Yes... it will make SQLite not flush the fh
vrurg: to tell the truth, I don’t know... I’ve never tried (seriously) make it parallel... (besides some tests with `red-do :parallel` (if I remember it well) 23:43
vrurg Didn't work for me, so – likely it isn't :) 23:44
I'll try the pragma. Thanks! 23:45
SmokeMachine vrurg: this may be faster: cs.stanford.edu/people/widom/cs145...eLoad.html
vrurg SmokeMachine: I'd like to keep the tracing in production to provide debugging capabilities to a user. So, better be producing the DB myself. 23:47
23:51 patrickb left
guifa sena_kun: there was actually someone in here not long ago asking about the JVM backend 23:53
also, Perl/Rakucon 2020 in Houston will have at least ONE Raku talk. Mine got accepted today yay 23:54