»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.
00:02 Manifest0 left 00:07 Manifest0 joined
AlexDaniel weekly: Time to blog about perl 6! Perl 6 is now in the TIOBE list, which searches for “Perl 6 programming”. So make sure to include that somewhere in your posts! 00:10
notable6 AlexDaniel, Noted!
AlexDaniel weekly: (currently on position 93)
notable6 AlexDaniel, Noted! 00:11
00:16 Manifest0 left 00:20 Manifest0 joined 00:24 kaare_ left 00:30 Manifest0 left 00:35 Manifest0 joined 00:40 Manifest0 left 00:44 Manifest0 joined 01:18 Manifest0 left 01:24 Manifest0 joined 01:25 molaf left 01:37 molaf joined, Manifest0 left 01:43 Manifest0 joined 01:57 jaldhar left 02:02 Manifest0 left 02:03 Manifest0 joined 02:08 Manifest0 left 02:13 Manifest0 joined 02:16 mowcat left 02:17 Manifest0 left 02:24 Manifest0 joined 02:33 Manifest0 left 02:37 Manifest0 joined 02:39 Cabanossi left 02:45 Cabanossi joined 02:46 Sgeo_ joined 02:49 Sgeo left 03:03 Manifest0 left 03:08 Manifest0 joined 03:15 Manifest0 left
irced readies a b-slap hand for Manifest0. 03:15
03:20 Manifest0 joined
irced slaps Manifest0. 03:20
03:27 Manifest0 left 03:32 Manifest0 joined 03:39 Manifest0 left 03:41 irced left 03:43 irced joined 03:44 Manifest0 joined 03:58 irced left 03:59 molaf left 04:02 dominix_ joined 04:06 dominix__ left 04:11 Manifest0 left 04:13 Sgeo_ left, Sgeo_ joined 04:17 Manifest0 joined 04:21 Manifest0 left 04:22 rindolf joined 04:28 Manifest0 joined 04:33 Manifest0 left 04:39 Manifest0 joined 04:44 Manifest0 left 04:47 MilkmanDan left 04:48 MilkmanDan joined 04:49 Manifest0 joined 04:56 Manifest0 left 04:58 curan joined 04:59 curan left 05:01 Manifest0 joined 05:10 Manifest0 left 05:15 Manifest0 joined 05:20 llfourn left 05:28 robertle left 05:29 Xliff joined 05:31 abraxxa joined, abraxxa left 05:32 abraxxa joined 05:35 reach_satori_ left, reach_satori_ joined 05:38 abraxxa left 05:39 abraxxa joined 05:40 llfourn joined 05:46 sno left, Manifest0 left 05:48 sno joined 05:49 jmerelo joined 05:51 Manifest0 joined, Xliff left
jmerelo releasable6: status 05:51
releasable6 jmerelo, Next release will happen when it's ready. R6 is down. At least 1 blocker. 578 out of 677 commits logged
jmerelo, Details: gist.github.com/83117f9c994a634908...2c7b12950c
05:58 Manifest0 left 06:03 Manifest0 joined 06:10 Manifest0 left 06:15 veesh joined 06:16 Manifest0 joined 06:21 tankf33der left 06:24 rfold joined 06:30 Manifest0 left 06:35 Manifest0 joined 06:42 Manifest0 left 06:47 Manifest0 joined, veesh left 06:49 domidumont joined, Black_Ribbon left 06:54 rfold left 06:55 sno left 07:02 Manifest0 left 07:05 llfourn left 07:06 llfourn joined, Manifest0 joined 07:10 patrickb joined
Geth doc: 923c2df0d2 | (JJ Merelo)++ | doc/Language/regexes.pod6
Reflow and clarifications refs #2887
07:13
doc: 56ce35b637 | (JJ Merelo)++ | doc/Language/regexes.pod6
Adds publication of variables, closes #2887
synopsebot Link: doc.perl6.org/language/regexes
07:16 llfourn left, llfourn joined 07:21 llfourn left 07:22 noisegul joined, llfourn joined 07:30 Manifest0 left 07:31 llfourn left 07:32 llfourn joined 07:35 Manifest0 joined 07:37 llfourn left
masak good antenoon, #perl6 07:40
timotimo greetings masak
07:41 zakharyas joined 07:45 sena_kun joined
El_Che masak: ante meridiem for you too 07:45
jmerelo hey everyone 07:47
07:47 llfourn joined
sena_kun enabled backup for our new server 07:47
jmerelo, o/
07:48 rindolf left 07:49 rindolf joined
timotimo so, with the optimization i have, that hits many, many methods in the core setting, how do i best figure out how big the win is? 07:49
lizmat test-t ?
timotimo that could be a way, yeah
lizmat make spectest ?
timotimo that takes ages :) 07:51
lizmat maybe a bit less age ? :-) 07:52
07:52 llfourn left
timotimo barely, i'm sure 07:52
lizmat you wanted suggestions :-) 07:53
timotimo true
08:00 Sgeo__ joined
sena_kun timotimo, I am also sure you're aware of callgrind to minimize noise 08:01
timotimo aye
08:02 llfourn joined
timotimo is the normal test-t-20 run with the 1_000_000 lines that README.speed suggests? 08:03
08:04 Sgeo_ left
timotimo aborted it after 2.5 minutes of run time 08:04
08:06 llfourn left
lizmat my t-20 run used a 200K line file 08:08
timotimo ok, i've got a 200000 one
d'oh, now i'm not sure if my rakudo at the time had the opt or not 08:13
08:13 nobody joined 08:15 nobody left 08:16 WhoIsNick joined 08:21 WhoIsNick left 08:23 Manifest0 left 08:26 dakkar joined 08:29 Manifest0 joined 08:35 Manifest0 left 08:37 dakkar left
timotimo i ran it a whole bunch of times in a loop with a sleep in between for thermal effects to settle down again 08:39
08:39 llfourn joined, dakkar joined 08:43 Manifest0 joined 08:44 llfourn left 08:49 Manifest0 left
timotimo i'm making extra sure and wiping a little bit of stuff because i've got some surprising behavior, like the first execution after rebuilding rakudo not being slower than the others, even though it would have had to rebuild Text::CSV as well as Slang::Tuxic 08:55
08:55 Manifest0 joined 09:01 Black_Ribbon joined 09:02 Manifest0 left
kawaii morning o/ 09:02
timotimo it looks like it's barely any faster, but earlier i was able to get it sub-30, now it no longer goes there at all 09:04
maybe that's what i get for trying to benchmark on a desktop computer?
09:07 Manifest0 joined
timotimo oh, haha, i'm running a moarvm with --optimize=1 and --valgrind 09:09
what am i even doing
what have i ever been doing
09:18 Manifest0 left
timotimo hm 09:19
could it be my optimization doesn't even impact Text::CSV at all?
09:23 Manifest0 joined 09:29 Manifest0 left 09:33 Manifest0 joined 09:39 Manifest0 left 09:41 kaare_ joined 09:46 Manifest0 joined 09:48 lord| joined
SmokeMachine m: my atomicint $a = 0; for 0 .. 1000 { start { $a⚛++ } }; say $a 09:51
camelia 1001
timotimo SmokeMachine: that is not likely to give you a lot of parallelism 09:52
a lot of overhead for a tiny operation 09:53
SmokeMachine www.irccloud.com/pastebin/R3A0eeuW/timotimo%3A
jnthn It'll give you precisely none, in so far as the CPU can't execute two such instructions on the same memory address in parallel :)
timotimo well, that's extra true 09:54
jnthn But of course, you'd usually do it to count as part of a lager operation 09:56
*larger
.oO( Off by one pint )
10:01 domidumont left 10:02 Manifest0 left 10:06 Manifest0 joined 10:10 llfourn joined 10:11 Manifest0 left 10:16 Manifest0 joined, llfourn left 10:20 llfourn joined 10:23 sena_kun left, sena_kun joined 10:28 llfourn left 10:29 llfourn joined, Black_Ribbon left 10:39 llfourn left 10:48 __jrjsmrtn__ joined 10:49 _jrjsmrtn left 11:03 llfourn joined
masak .oO( I thought I had 5 pints, but it turns out I had committed a fencepost error ) 11:06
11:08 llfourn left
lizmat weekly: blogs.perl.org/users/laurent_r/2019...lenge.html 11:10
notable6 lizmat, Noted!
lizmat weekly: dev.to/antoniogamiz/gsoc-report-week-6-28i4
notable6 lizmat, Noted!
lizmat weekly: www.braincells.com/perl/2019/07/pe...ek_15.html 11:24
notable6 lizmat, Noted!
11:29 Manifest0 left 11:34 Manifest0 joined 11:38 domidumont joined 11:39 zakharyas left 11:42 llfourn joined 11:44 kaare_ left 11:47 llfourn left 11:49 llfourn joined 11:57 Sgeo__ left, Manifest0 left, Sgeo__ joined 11:59 llfourn left 12:02 Manifest0 joined 12:09 rfold joined 12:27 Manifest0 left
tbrowder kawaii: good morning! 12:29
12:33 Manifest0 joined
jmerelo tbrowder: hey, Tom 12:35
how're you doing?
tbrowder pretty good for my age, thnx! 12:37
trying to get excited about pod again.., 12:38
12:38 llfourn joined, jmerelo left
tbrowder jj, i appreciate all the work you are doing, especially the gsoc work! 12:39
.tell jmerelo ^^^ 12:40
yoleaux tbrowder: I'll pass your message to jmerelo.
12:45 llfourn left 12:46 llfourn joined 12:56 llfourn left 12:58 Manifest0 left 13:03 Manifest0 joined, nubie joined 13:05 llfourn joined
Geth doc: chloekek++ created pull request #2893:
Improve consistency and completeness of Packages document
13:06
13:15 zakharyas joined 13:16 Manifest0 left 13:19 llfourn left 13:20 vike left, llfourn joined 13:21 Manifest0 joined 13:25 Manifest0 left 13:27 nubie left 13:31 Manifest0 joined
Geth doc: tmtvl++ created pull request #2894:
Add the `S///` section from operators to regexes
13:34
13:34 llfourn left 13:35 llfourn joined 13:37 pamplemousse joined
Geth doc: ce50e11d49 | (Chloé Kekoa)++ (committed by Chloé) | doc/Language/packages.pod6
Improve consistency and completeness of Packages document

  - Include the output of the say calls in the name interpolation example.
  - Make spacing and capitalization consistent.
13:39
synopsebot Link: doc.perl6.org/language/packages
doc: 8b5e218271 | Chloé++ | CREDITS
Add chloekek to CREDITS
doc: 2f096dd914 | (Chloé Kekoa)++ (committed using GitHub Web editor) | 2 files
Merge pull request #2893 from chloekek/patch-1

Improve consistency and completeness of Packages document
13:45 llfourn left 13:47 llfourn joined 13:53 llfourn left 13:58 llfourn joined 14:04 Manifest0 left 14:08 Manifest0 joined 14:11 molaf joined 14:36 Manifest0 left 14:41 Manifest0 joined
sena_kun .tell timotimo will moarperf land in rakudo one day? 14:41
yoleaux sena_kun: I'll pass your message to timotimo.
timotimo sena_kun: it's a bit big to put into rakudo, IMO. maybe it'll land in star, though
yoleaux 14:41Z <sena_kun> timotimo: will moarperf land in rakudo one day?
sena_kun timotimo, but current angular one is in core, right? 14:42
timotimo yeah, though it grabs angular and the other libraries from the net
sena_kun I see 14:43
ok, trying it out again...
14:44 AlexDaniel left, AlexDaniel joined
timotimo moarperf? 14:45
sena_kun: just pushed a few things i've been sitting on locally; nothing major at all, tho 14:47
14:49 molaf left
sena_kun >found 431 vulnerabilities (214 low, 2 moderate, 215 high) 14:54
timotimo shruuuuuuuuuuug :P 14:55
sena_kun indeed
timotimo i'm sure it's a matter of "npm update"
remote: GitHub found 4 vulnerabilities on timo/moarperf's default branch (1 high, 3 moderate). To find out more, visit:
14:55 Manifest0 left
sena_kun I'll risk my life to look at it! pls no hacks 14:56
kawaii I noticed on GitHub that Node and Ruby packages now have this 'used by X other projects' metric, I think it's based off how many 'package.json' files the repo is referenced in, I wonder if we could push for that using what we have with META6.json? 14:57
github.com/meew0/discordrb
Here's an example ^
usercontent.irccloud-cdn.com/file/...-57-34.png
Note the button only displays if you are logged in 14:58
It would be nice to get a dependency graph of ecosystem modules
15:00 Manifest0 joined
timotimo finanalyst.github.io/ModuleCitation/ - perhaps interesting 15:01
kawaii timotimo: yeah, that's great for you and me, but not Perl 6 outsiders who just use GitHub and want to see how popular a module is :) 15:05
timotimo oh, well
kawaii also congratz on having the most popular module? :)
timotimo in that case, we ought to build a tiny little service that just generates images from URLs and let everybody put that in their READMEs
like these badge services
kawaii I think going for official GitHub dependency integration might be something for us to aim for 15:06
timotimo ha, it was an easy task to achieve. also, lizmat recently basically completely rewrote one part of it, so it's not necessarily "mine" any more :)
kawaii You chose the low hanging fruit ;)
patrickb Alternatively push for metacpan integration. They have pretty dependency analysis too. 15:07
timotimo the fruit was extremely ripe for picking 15:08
json was already everywhere, and JSON::Tiny was, while very elegant in its implementation, sadly rather slow
kawaii I'm hoping that when Altreus and I 'finish' API::Discord, it'll attract some of the younger folk
timotimo you know, i used to be younger 15:11
currently i'm still with "it", but they're constantly tweaking what "it" is and soon "it" will be something that i don't understand and it will scare me
oh well, such is the tragedy of humanity 15:12
rfold Can I safely use Blob.allocate to create a blob, then pass a pointer to the blob to a C function that will mutate its contents? 15:13
In other words: is the data mutable? 15:14
timotimo you'll want CArray for that purpose, i believe
sena_kun wonders if something like Meshuggah will be considered a thing for old geezers in another 20-30 years 15:15
timotimo Blob is allegedly immutable, which is good because it won't allow the array to be resized, i.e. the pointer to change
15:15 Sgeo_ joined
rfold p6: use NativeCall; say(CArray[uint8] ~~ Blob); 15:16
camelia False
rfold The C function I'm calling doesn't change the size of the buffer, only the contents.
15:18 Sgeo__ left 15:23 jmerelo joined
ugexe patrickb: metacpan dependency analysis would require our dependency declaration format to be parsable by metacpan. so ultimately will require some perl 5 module for understanding how things work in perl 6 15:24
as a quick example consider "emulates"
kawaii note: I only raised this topic due to GitHub integration looking like a good way of getting more people interested in the language. I don't think creating MORE things within our own ecosystem helps exposure at all. 15:25
notable6 kawaii, Noted!
kawaii oh wait no
ugexe its not a monumental task though -- just means that there is more to it then us creating a META6 -> CPAN::Meta format converter
timotimo of course, to do that you'd have to pass a Pointer[CArray[int]] or make the CArray[int] argument "is rw" 15:26
jmerelo kawaii: re: your "used by", there are a couple of modules, my "river" and finanalyst's. They compute a score that indicates how much they are used and how deep
yoleaux 12:40Z <tbrowder> jmerelo: ^^^
jmerelo .tell tbrowder thanks!Ç 15:27
yoleaux jmerelo: I'll pass your message to tbrowder.
15:28 ctilmes joined
kawaii jmerelo: yep, but once again, these are tools within our own ecosystem - they don't help someone new to P6 gauge how popular something is at a glance, GitHub's integration provides an entire dependency graph right there :) 15:28
ctilmes rfold, see stackoverflow.com/questions/552093...-collector 15:29
rfold p6: say(Blob.allocate(20) ~~ Buf);
camelia False
jmerelo kawaii: github integration is nice, but never easy to do. We tried, and failed, to have Pod6 interpreted by GitHub
kawaii: but as long as there's some repo we can do a pull request to, telling it how to compute dependencies (basically dependency declaration, I guess), we could do it. 15:30
lizmat and yet another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2019/07/08/...ythifying/
jmerelo lizmat++
kawaii jmerelo: I'm just trying to look at what other popular languages are doing and hope that we can replicate that somehow, we're all smart and capable people here so I don't doubt it's possible 15:31
jmerelo kawaii: :-) We are in the top 100 already. We can only go uphill from there :-) 15:32
kawaii: there's this thing: github.com/sourcegraph/about/blob/...-badges.md I'm not sure it's the official one 15:35
kawaii jmerelo: it's not, I have no idea how the one I linked on that Ruby gem repository actually works 15:36
I assume GitHub themselves wrote some kind of parser for popular languages
but as an OSS community, we should be pushing them to help us get this stuff working for us too :)
I'm sure they have some kind of open source liaison we can contact, show what we have with the `depends` section of META6.json, and have them implement something similar for Perl 6 modules :) 15:38
ugexe they would need to resign themselves to running perl6 code to do so, which is the reason they rejected pod6
timotimo and then have them curse us for allowing the "depends" to ask for specific auths, like ones from cpan :)
kawaii after all, there are over 1300 modules now, that's not an insignificant number
ugexe github.com/niner/Inline-Python/blo...6.json#L16 15:40
they cannot parse dependencies unless they can do $*DISTRO.name in that case
there is also $*KERNEL.*, etc
timotimo for a dependency graph, i'd assume they'll want to display all of the possible ones? 15:41
they're not trying to resolve one list of candidates that should be installed
ugexe that is true
timotimo just measuring how many modules depend on what modules, that should in theory be a lot easier
but i haven't thought it through much yet
ugexe at the very least they need to be able to parse perl6 versions 15:42
and compare them the same way
timotimo what would truly impress me would be if they manage to chase down native dependencies through linux distro package repos back to other github repos
ugexe yeah, that would mean there is a mapping of names (instead of every package manager having a slightly different name for a given package) with money behind it 15:43
kawaii FWIW: The node.js integration of their 'used by' feature seems to only scan the package.json file in each repo and check for the dependencies listed by the author 15:44
that would satisfy my requirement, and then we can prompt module maintainers to be good people and fill that information out :) 15:45
ugexe just encode all dependency source code as base64 and store it in the META6.json
timotimo ha
15:46 Cabanossi left
kawaii github.com/meew0/discordrb/network/dependents 15:48
^ here's a good way of looking at it, each of those repos have a Ruby 'Gemfile', which then states that the repo is a package dependency of their own
ugexe I wrote a module for doing that for the `provides` awhile back -- github.com/ugexe/zef2/blob/master/...m6#L54-L58
15:48 Cabanossi joined
ugexe github.com/ugexe/zef2/blob/5a9c581...#L110-L130 15:52
timotimo oh, the contents literally go into the meta6 as base64? 15:53
ugexe yeah. its purpose was to make writing tests for Distribution easier (not having to create all those files)
timotimo NGL that's pretty neat 15:55
if you use that, will it Just Work™ if a following repo has working precomp storage? 15:56
or, wait, you'd usually install that distribution, wouldn't you? 15:57
ugexe that isn't for loading.
right
a CompUnit::Repository could be made to load it however though (like the one for tar files)
timotimo this is now the p6i file format; Perl 6 Installer :P
15:59 zakharyas left 16:03 dominix__ joined 16:05 robertle joined 16:06 dominix_ left, patrickb left 16:10 mowcat joined 16:18 Manifest0 left 16:23 Manifest0 joined 16:28 Manifest0 left
zostay is there a way to use experimental features in the REPL aside from prepending every line with `use experimental :pack`? I want to play with pack/unpack, but can't figure out how to get the exports to last longer than a line. 16:30
16:30 dakkar left 16:33 Manifest0 joined
timotimo perhaps -Mexperimental=pack, no clue :( 16:36
zostay Yeah, tried that. ===SORRY!=== 16:37
Could not find experimental=pack at line 1 in:
ugexe that syntax wouldnt work for perl6 anyway
m: module ::("Foo=bar") { }; ::("Foo=bar").WHAT.say
camelia (Foo=bar)
timotimo mhh 16:38
zostay You can do the same in Perl 5.
timotimo well, we gotta have something :D
--Module( Foo=Bar :pack :foo )
zostay Given our preference for adverbials, I would go for -Mexperimental:pack:macros
it's just a short hand anyway
if you need to be fancy, just -e it 16:39
ugexe then what about -M "Foo:ver<1>"
zostay sure, again, it's just a short-hand, it doesn't have to do everything
ugexe it should be consistent
zostay agreed
timotimo oh, right, just that would probably be fine
zostay Anyway, I'll just write a script for now. It looks like I could do muck around with OUR variables to make it work, but I don't want to mess with it further for now. 16:40
16:41 vrurg left
timotimo may i suggest you try out the jupyter notebook kernel? 16:42
it's more reliable than the REPL, at least sometimes
there's even a service that hosts it for you, including the perl6 kernel 16:43
zostay hmm... i can't even get unpack to work in a script... seems hosed at the moment, i'll just use the Blob.read-*/Buf.write-* thing thing. Blah. 16:44
*thing then
ugexe should work. Net::HTTP uses it among others 16:46
it doesn't actually use it, so i take that back
timotimo App::MoarVM::HeapAnalyzer will use it for the new heap snapshot format
16:51 vrurg joined, Manifest0 left
timotimo the code is already there, but since moarvm doesn't spit out the new format except in a branch... :) 16:53
16:57 Manifest0 joined 17:06 veesh joined
Geth doc: 17a102abdb | (JJ Merelo)++ | doc/Language/regexes.pod6
Adds examples using backreferences in regexes

This closes #1545. Since this part is going to probably be moved dur to #2890, it will move with them; this seemed the most reasonable part to document these back references, since it's where they are actually used.
17:08
synopsebot Link: doc.perl6.org/language/regexes
17:10 lord| left 17:22 Manifest0 left, domidumont left 17:27 Manifest0 joined 17:30 Sgeo_ left 17:31 Sgeo_ joined, patrickb joined 17:38 Black_Ribbon joined 17:46 [Sno] joined 17:55 molaf joined 17:56 lord| joined 18:14 ctilmes left, Cabanossi left 18:22 [Coke] joined, [Coke] left 18:27 Cabanossi joined 18:28 sauvin left 18:34 epony left 18:42 molaf left
rfold Given that accessing POD from Perl 6 is relatively straightforward I wonder how hard it would be to extract code from SYNOPSIS sections and run it as tests. 18:49
18:51 empee0 joined, Manifest0 left
rfold Set $*OUT to a buffer, run the synopsis with EVAL, check if it gave the same output as the sequence of the ‘# OUTPUT: «...»’ comments in the synopsis. 18:51
18:55 Manifest0 joined 19:00 Manifest0 left
ugexe there is no way to extract the pod from a namespace 19:01
i dont think
unless the author explicitly creates a e.g. `our sub POD { $*POD }`
rfold I wouldn't mind the module having to do something like: use Test::Synopsis; 19:02
ugexe that puts a strict test dependency into your runtime dependencies which is not exactly ideal
lizmat weekly: blogs.perl.org/users/laurent_r/2019...s-pie.html 19:04
notable6 lizmat, Noted!
rfold That's true.
19:06 Manifest0 joined 19:08 netrino joined 19:11 dominix__ is now known as dominix 19:18 Manifest0 left 19:24 Manifest0 joined 19:37 Manifest0 left 19:38 pamplemousse left 19:41 Manifest0 joined 19:42 empee0 left 19:59 jmerelo left, molaf joined 20:08 Sgeo__ joined 20:11 Sgeo_ left 20:12 robertle left 20:20 Manifest0 left 20:25 Manifest0 joined 20:33 haukex joined 20:36 rfold left 20:38 kst` joined 20:40 kst left 20:41 haukex left
lizmat weekly: yakshavingcream.blogspot.com/2019/...etely.html 20:42
notable6 lizmat, Noted!
20:52 vike joined 21:07 afresh1 left 21:08 afresh1 joined 21:18 Manifest0 left 21:23 Manifest0 joined 21:31 Manifest0 left 21:39 Manifest0 joined 21:41 patrickb left 21:43 patrickb joined 21:45 kst` is now known as kst 21:46 Manifest0 left 21:51 Manifest0 joined 22:01 cpage_ joined 22:02 cpage_ is now known as cpage 22:16 kaare_ joined 22:19 Manifest0 left 22:20 ctilmes joined 22:24 Manifest0 joined 22:33 Manifest0 left 22:39 Manifest0 joined 22:44 sena_kun left 22:46 japhb joined 22:51 patrickb left 22:52 dominix left, AlexDaniel left 22:53 AlexDaniel joined 23:17 smash left, jnthn left 23:18 jnthn joined 23:19 smash joined 23:21 veesh left 23:27 mowcat left 23:31 veesh joined 23:33 Manifest0 left 23:38 Manifest0 joined, aborazmeh joined, aborazmeh left, aborazmeh joined 23:39 reach_satori_ left, reach_satori_ joined 23:40 veesh left 23:43 epony joined 23:44 Manifest0 left 23:50 Manifest0 joined