tbrowder AlexDaniel: i made the change to nqp PR #397 as you requested. 00:02
Zoffix ZofBot: status 00:32
ZofBot Zoffix, She said 'Good evening,' too, tying the score
AlexDaniel psch: post-release I think 00:50
Zoffix I'm guessing "foo {"bar"}" wrapping that QAST::Block into QAST::Stmts is an error... gonna remove em 00:54
samcv i'm getting failures on t/spec/S32-io/IO-Socket-Async.t 00:56
Parse errors: Bad plan. You planned 26 tests but ran 17
seems like it's not completing all of the tests. i ran it again to check and same thing 00:57
Zoffix samcv: is it immediate failure or does it sit and wait?
samcv Zoffix: immediate
Zoffix ah no idea then
samcv Unhandled exception in code scheduled on thread 4 00:58
Unhandled exception in code scheduled on thread 6
MoarVM 2018.01 has been released 🎉 01:15
Zoffix AlexDaniel: ^ :) 01:18
AlexDaniel yey!
samcv++
[Coke]: fwiw you tend to appear twice as a contributor. “Will "Coke" Coleda” and “Will Coleda” 01:20
perhaps you want to fix that
I mean, fix your name in one of the configurations :)
samcv you can add to the rakudo changelog that cf, sc, STerm, Upper, space and White_Space properties were fixed
or i guess i'll add it to the draft :) github.com/rakudo/rakudo/wiki/ChangeLog-Draft 01:21
AlexDaniel hehe whoops :)
samcv oh. where does it live? see nothing there
AlexDaniel because I was about to commit it
samcv :) 01:22
AlexDaniel ok added 01:23
samcv pretty happy i have all properties resolving right now though :) i fixed the property values during my grant but now both are fixed@ 01:28
was much easier to do now do to the huge cleanup of ucd2c.pl i did this last month 01:29
Geth rakudo: c359312b13 | (Aleks-Daniel Jakimenko-Aleksejev)++ | 3 files
Log all changes (+ announcement)

Deliberately not logged:
   c51f1796 7df750a1 8275c03d bbf95dbc cf7b01ef e5c38ad6 1f99aa14
   8943f7c5 57d164b4 f10e8b13 b7f93bee ec0021bb 9606aea2 9cb2170b
   98be6ab5 1ac1a125 ff34ab35 d9ce5e9b f35d0203 8746b534 e5bb9bd2
... (16 more lines)
01:35
AlexDaniel hm wow 01:36
one of these does not show up as a commit
Geth rakudo: 016b0c561c | (Aleks-Daniel Jakimenko-Aleksejev)++ | docs/release_guide.pod
Fix release date
01:37
Zoffix 'cause it's too short, there's another commit with same sha 01:46
*same short sha
maybe :) only one shows up in git log grep 01:47
works here tho github.com/rakudo/rakudo/commit/1ed8f5de5 01:49
AlexDaniel some time ago I wanted to bump the sha length in the next changelog, but for some reason decided against it
Zoffix m: my $m := Int; say "42".$m 02:24
camelia 42
Zoffix TIL
ah, it just invokes it as a coercer. nm 02:25
Geth nqp: b0c3964659 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/MOAR_REVISION
[release] Bump MoarVM revision to 2018.01
02:29
nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...902eb58... No newline at end of file
b9878442ef | (Aleks-Daniel Jakimenko-Aleksejev)++ | VERSION
rakudo: c82a698bc2 | (Aleks-Daniel Jakimenko-Aleksejev)++ | tools/build/NQP_REVISION
[release] Bump NQP revision to 2018.01
rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2017.......2018.01
fa4db18c76 | (Aleks-Daniel Jakimenko-Aleksejev)++ | VERSION
AlexDaniel \o/ 02:30
Zoffix wooo. AlexDaniel++ 02:31
would be cool to have a debug mode that annotates each QAST::Node with file/location of where it was created. 02:32
Geth nqp: zoffixznet++ created pull request #399:
Post release
02:34
nqp/master: 9 commits pushed by (Zoffix Znet)++ 02:35
AlexDaniel bah thunderbird wrapped the lines excessively in my email again. I thought that I fixed that…
Zoffix++ # post-release branch
Zoffix hm, weird. pulled all repos, rebuild all the things, but still getting "This is Rakudo version 2018.01 built on MoarVM version 2017.12.1-34-g4a0a912" 02:54
oh, nm, confused zscript commands and was only doing moarvm recompile and not rebuild (with configure run)
ZOFVM: Files=1287, Tests=152969, 159 wallclock secs (21.69 usr 3.20 sys + 3399.60 cusr 178.38 csys = 3602.87 CPU) 03:00
ZOFVM: Files=1287, Tests=153127, 159 wallclock secs (21.40 usr 3.27 sys + 3418.56 cusr 179.32 csys = 3622.55 CPU) 03:08
AlexDaniel releasable6: status 04:42
releasable6 AlexDaniel, Next release in 23 days and ≈14 hours. Blockers: github.com/rakudo/rakudo/issues?q=...%9A%A0%22. Changelog for this release was not started yet
AlexDaniel, Details: gist.github.com/e84bc43f84c6b58246...5699ac49ad
El_Che releasable6: status 08:11
releasable6 El_Che, Next release in 23 days and ≈10 hours. Blockers: github.com/rakudo/rakudo/issues?q=...%9A%A0%22. Changelog for this release was not started yet
El_Che, Details: gist.github.com/391d6e04ad840e6e11...084c3532d9
El_Che ah, 2018.01 released? nice
[Tux] Rakudo version 2018.01-26-g7c8bbc8da - MoarVM version 2018.01-17-g39928851f
csv-ip5xs0.998 - 1.031
csv-ip5xs-2010.684 - 10.704
csv-parser11.542 - 11.620
csv-test-xs-200.429 - 0.441
test10.881 - 11.199
test-t2.740 - 2.818
test-t --race1.144 - 1.346
test-t-2047.903 - 48.695
test-t-20 --race16.778 - 19.003
08:19
lizmat csv-ip5xs < 1 whee! 08:58
travis-ci Rakudo build passed. Aleks-Daniel Jakimenko-Aleksejev '[NQP Bump] Brings 11 commits 09:00
travis-ci.org/rakudo/rakudo/builds/332640077 github.com/rakudo/rakudo/compare/7...dce163e818
nine lizmat: csv-ip5xs-20 went down 2 seconds :) 09:23
jnthn nine++ # nice opt work :) 10:17
stmuk is the moar release RSN? 10:26
Geth star/master: 6 commits pushed by (Steve Mynott)++ 10:28
psch hrm, looking over nqp/docs/jvminterop-goals.md i feel like i didn't really implement anything at all yet hah 11:19
j: use java::util::ArrayList:from<JavaRuntime>; my $a = ArrayList.new; $a.add("foo"); $a.add("bar"); say $_ for $a 11:20
camelia ( no output )
psch j: use java::util::ArrayList:from<JavaRuntime>; my $a = ArrayList.new; $a.add("foo"); $a.add("bar"); say $_ for $a
camelia [foo, bar]
El_Che I see there is a new rakudo release, but no matching moarvm release
is this intended
?
psch j: use java::util::ArrayList:from<JavaRuntime>; my $a = ArrayList.new; $a.add("foo"); $a.add("bar"); say $_ for $a.iterator()
camelia java.lang.NullPointerException
in block <unit> at <tmp> line 1
11:21
El_Che (there is also a new nqp release)
psch j: use java::util::ArrayList:from<JavaRuntime>; my $a = ArrayList.new; $a.add("foo"); $a.add("bar"); say $_ for $a.toArray 11:23
camelia List interop NYI after GLR
in block <unit> at <tmp> line 1
psch oh right, there was something there...
lizmat El_Che: afaik there's a new MoarVM release as well 11:24
except that moarvm.org/releases.html is not updated yet, and moarvm.org hasn't been for a few months :-( 11:25
psch 2018.01 is tagged on github at least
stmuk moarvm.org/releases/ is empty
well not actually empty at all 11:26
El_Che lizmat: my build scripts retrieves that tars from moarvm.org/releases/ 11:28
lizmat: is has worked sinds I started building packages
lizmat El_Che: ok, noted, perhaps ping samcv / jnthn on #moarvm
El_Che thx
left a note there 11:29
psch hrm, either my approach of mimicking List.new in RakudoJavaInterop to fix list marshalling is wrong or i need IterationBuffer during bootstrap to add it to GlobalExt 12:13
somewhat related: is ListIter still around? 12:15
m: say ListIter
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared name:
ListIter used at line 1
lizmat psch: no, ListIter is pre GLR afaik 12:18
psch we're still carrying it around in GlobalExt is all, guess i can toss it there 12:20
lizmat sounds like it, but I'm a JVM backend noob 12:23
psch yeah, it's an empty field during runtime in the GlobalExt instance we have in the GlobalContext 12:24
the eventual port to java 1.9 is gonna be lotsa fun too :/ 12:26
cause sun.misc.Unsafe gets replaced :x
huh. i have a perl6 compiler during the interop wrapper generation for ArrayList, but i don't have the perl6 hll and thus can't get IterationBuffer via gethllsym :/ 12:37
ah, no, term just cut off output, never mind 12:38
so i have the hll perl6 but IterationBuffer isn't registered as a symbol
lizmat psch: in the setting IterationBuffer is first stubbed with a REPR, and then later actually defined
perhaps that is related ? 12:39
psch lizmat: i'm not sure. as in, i don't know how far through the settings we are any way 12:40
lizmat: fwiw, the hll syms i have are the following:
main[2] print tc.gc.hllSyms.get("perl6").keySet()
tc.gc.hllSyms.get("perl6").keySet() = "[ModuleLoader, $!ARGITER, MD_Pos, MD_PBF, PROCESS, GLOBAL, $COMPILER_CONFIG, JavaModuleLoader, &THE_END, P6EX, @END_PHASERS]"
lizmat: which, to me, looks like either we're not done with the setting yet, or hllsyms wouldn't have what i need in the first place 12:41
lizmat: ideally i'd have IterationBuffer in GlobalExt, but we register types there during BOOTSTRAP, and IterationBuffer isn't in BOOTSTRAP
i think we did have ListIter in BOOTSTRAP, but i don't wanna assume that moving IterationBuffer there is fine and dandy 12:42
jnthn: does gist.github.com/anonymous/31f1d15b...9a644eaa28 -- with moving IterationBuffer to BOOTSTRAP -- seem sensible to you? 12:43
err, and replacing the getcurhllsym call with a GlobalExt field lookup too... :) 12:44
ooh, android oreo finally got indy. i totally missed that 13:07
well, something equivalent at least
jnthn psch: I'd prefer not to move it to BOOTSTRAP if we don't have to, but we can still register it with GlobalExt without doing that perhaps? 13:12
psch jnthn: i think we'd need a new op to register it later than BOOTSTRAP? like, either similar to p6setitertype but for IterationBuffer, or something more generic which would have to rely on reflection to map type to gcx field 13:16
maybe turning GlobalExt into a Map makes sense to not need reflection, though that gives lots slower lookups, probably 13:17
jnthn psch: Ah, may need an op for it, yes 13:25
psch jnthn: right, i'll do that then, thanks :) 13:26
Geth rakudo: 8ffbc43b41 | (Elizabeth Mattijsen)++ | src/Perl6/Grammar.nqp
Give a bare 'fc' the same treatment as bare 'lc' or 'uc'
[Coke] AlexDaniel: what is the list of configurations where I have to verify the listing? 13:30
(I thought our canonical listing was the rakudo contributor's list. 13:31
)
AlexDaniel [Coke]: it's from a commit to rakudo/nqp/moarvm/doc/roast repo I think, so I guess one of the PCs (or OS) you use has something different in git config 13:32
AlexDaniel looks
[Coke] ah. so this isn't part of what you're doing to get a list of names for a release email 13:34
May be the web commits vs. actual clones/pushes 13:35
I don't care about that level of consistency. email address is fine.
AlexDaniel Str $author = "Will \"Coke\" Coleda"
Str $author = "Will Coleda"
from the doc repo
[Coke] Sure, that's fine. don't care unless it's making your life difficult. 13:36
AlexDaniel [Coke]: well, I have to touch the file manually to delete the duplicate…
I did not notice once IIRC
[Coke] which filie?
AlexDaniel the announcement
[Coke] why are you pulling names directly from there and not using the contributors file with the preferred names?
(pretty sure that's how that worked back in the day)
AlexDaniel I use tools/create-release-announcement.pl6, it lists you twice 13:37
[Coke] (if it's docs, it's almost certainly web commits vs. not)
AlexDaniel i.e. it creates an announcement that has you listed twice
[Coke] ok. it's a bug in contributors.pl6 then. I'll see if I can get you a patch. 13:38
sorry, CREDITS is the file I'm thinking of. 13:40
logic I'm thinking is: you get the email from the commit message, look it up in that list; if you find it, use that Name; else use the name from github.
(yah, the 2015.12 version used the CREDITS file...) 13:42
(but probably didn't check enough repos)
AlexDaniel another option is to add “U: Will Coleda” I think
[Coke] ah, looking for the right filename, there is a CREDITS in there! Sure, if adding an alias works, please do. 13:45
AlexDaniel [Coke]: but I like your proposal
[Coke] U is "username" though, which is coke ?
AlexDaniel yes, but you can have more than one it seems 13:46
[Coke] ok. I opened a ticket for that logic, will poke at it tonight.
"will coleda" isn't my username though
even with the web edits. (that's name which is something else, I think)
but yah, let me try to see if I can patch contributors to be more concise. 13:47
psch ugh, i apparently forgot my ssh key passphrase :l 13:49
AlexDaniel [Coke]: fwiw, a warning if two different names for the same email are found should be good enough too 13:50
Geth nqp: b88de49aad | (Pepe Schwarz)++ | 3 files
[jvm] Add some cleanup when intercepting exits.

The EvalServer adds a few minor headaches with memory usage, which we have to take care off manually. There's probably other ways still out there to reduce memory usage, but this set at least finishes the Perl 6 spectest without hangs or notable OutOfMemory exceptions.
psch keepassx++ >_>
jnthn
.oO( Version 10 of the best tool for keeping your donkey! )
13:51
Geth rakudo: 1676359bcd | (Pepe Schwarz)++ | tools/build/NQP_REVISION
Bump NQP_REVISION for nqp-j memory improvements
13:52
rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....gb88de49aa
a2499c90fc | (Pepe Schwarz)++ | 4 files
jnthn psch++ # tending the JVM backend
lizmat psch++ indeed
psch :) 13:53
DrForr jnthn: Aha, found you here :) 13:56
jnthn ? 13:58
yoleaux 13:53Z <DrForr> jnthn: 'static $filename;' ignores encoding, at least for UTF-8 HTML files. I'm using "content 'text/html', slurp $filename;" as a workaround.
13:54Z <DrForr> jnthn: (at least for Cro)
jnthn Hm, I think you can also tell static how to map extensions into content-type, and override it that way 13:59
If you think it should be doing something smarter, please file an issue 14:00
DrForr The content-type is getting mapped correctly, but it appears that it's being sent over the wire as Latin-1 when the file is UTF-8. 14:01
I'm doing some React work (didn't initially see your React+Cro tutorial, hastily abandoned most of what I was doing and rewrote with that, many thankks) 14:03
jnthn Are you sure it's being sent over the write that way rather than just no charset being set in the Content-type header? 14:07
DrForr No, I'm not. Let me recreate the error, oe mo.
*one
Geth nqp: tbrowder++ created pull request #400:
add new test file for nqp using perl
14:10
DrForr Claims that no character encoding was set, but .. Ah. I missed it when I moved to your static/index.html file. I'd set <meta content-type...> out of habit, not so much in the sample app. 14:11
My bug, although if ou want I'll file a bug against the sample React app... 14:13
jnthn Well, mebbe static should be doing something more sensible too
DrForr I think HTML should default to UTF-8 unless overridden by the content-type heder, but I'd want to reread the spe to confirm. 14:14
*spec
psch j: use nqp; my $a = { a => 1 }.keyof; say $a; say nqp::eqaddr($a, {a => 1}.keyof) # wow RT#132694 is weird :l 14:15
camelia (Str(Any))
0
14:16
synopsebot RT#132694 [new]: rt.perl.org/Ticket/Display.html?id=132694 [JVM] Coercion type Str(Any) returned from .keyof is not the same object as Str(Any)
geekosaur I would not want to bet on that. Browsers will do what they will do, and assuming sanity is not a safe option
DrForr Oh, of course, browsers can and will do what they want with text, but the spec is at least a starting point.
Adding the <meta/> tag cleans things up. 14:19
I'd suggest adding it to the demo, just as a preventative measure though. 14:20
And it *does* let me get a tour.golang.org quasi-lookalike working with React+MUI-CSS and a working sidebar. 14:21
And jnthn++ # I really wasn't looking forward to sorting out React toolchain stuff on my ow. 14:23
*own
jnthn I think "ow" is also right :P 14:26
DrForr Oh, agreed. I started sorting through it, and I think a few times my reaction was "*really*? So much for *that* idea..." 14:29
Geth nqp: tbrowder++ created pull request #401:
improve pod panic msg
14:33
psch humm, that coercer bug seems to be about serialization contexts? 14:42
jnthn DrForr: I thought I was in for an easy task as a year or so back I'd been on a react-using project at $dayjob. But no, the right way to do things had changed in that time. 14:45
DrForr Yeah. I'm using MUI for this project because I want a responsive UI that I can run on a tablet, but squaring what I have in the HTML and the React code and what jnthn is doing makes that obvious. jnthn++'s stuff works fine, and I'm happy to use it as a demo for what I'm doing, but it's definitely a different lyout than I'd anticipated. 14:48
tbrowder Zoffix: where should nqp tests be placed when they need perl ? 14:56
Zoffix tbrowder: TBH, I don't see why they *need* perl 15:00
tbrowder ok, how then do i exec nqp with nqp to test the failure msg! spawn 15:01
nqp::spawn is gone, nqp::spawnasyncproc is too much overhead (and it’s not documented). 15:02
Zoffix tbrowder: did you see my comment on your PR? It links to a run_command routine that captures STDOUT. Just copy it and modify it to capture STDERR 15:03
tbrowder looking...
Zoffix tbrowder: github.com/perl6/nqp/blob/b88de49a...nqp#L4-L69
tbrowder: "stdout_bytes" would be "stderr_bytes": github.com/perl6/nqp/blob/b88de49a...nd.nqp#L27 15:04
psch aha, the CoercionHOW root doesn't have the setting Str(Any) for some reason..? 15:05
tbrowder i’ll see what i can do... 15:07
travis-ci Rakudo build canceled. Aleks-Daniel Jakimenko-Aleksejev 'Fix release date' 15:08
travis-ci.org/rakudo/rakudo/builds/333077599 github.com/rakudo/rakudo/compare/c...6b0c561c8b
buggable [travis build above] ☠ Did not recognize some failures. Check results manually.
Zoffix tbrowder: basically, I see this as an argument of all maintainers need to learn a whole new language to maintain the tests vs. a test writer needs to learn how to write STDERR capture routine in NQP. The latter is a much smaller maintenance burden. Also, if we later decide to swap Configure.pl to some other system we won't need Perl 5 for anything, but with tests requiring multiple languages we would. 15:09
jnthn If this is something needed in multiple tests, I don't have a problem with this going into NQP's setting, fwiw 15:11
psch hm, so when we run into e.g. the Str(Any) coercer in the mainline on r-j we have a CoercionHOW on the root type that has an empty VMHashInstance as its WHO..? 15:36
i feel like that shouldn't be empty, but contain e.g. the setting Str(Any) coercer 15:37
oh, nevermind, it has at least the root 15:38
Geth nqp: b930e35e2c | (Stefan Seifert)++ | src/NQP/World.nqp
No need to load ModuleLoader multiple times

When loading a setting we can just check if the ModuleLoader is already loaded. When loading a module we have already loaded a setting and thus the ModuleLoader must be available already.
15:46
nine timotimo: ^^^
No measurable difference though
tbrowder Zoffix: what about the DRY principal? why can’t we add the spawn command back to nqp? 15:52
and shell... 15:53
*principle 15:54
timotimo 0.54 before, 0.26 after 16:09
japhb timotimo: What do those numbers represent? 16:13
timotimo miliseconds spent doing stat and lstat calls 16:15
for perl6 -e ''
japhb Oh, that's a nice savings! 16:21
Is this with a kernel updated post-spectre/meltdown?
nine timotimo: sooo..it does save a quarter of a millisecond? At least you could measure it :D 16:24
timotimo nine: *shrug*, plus/minus overhead from tracing with perf trace?
japhb I'll take 100 of those quarter-millisecond improvements, please. That would be *great*. 16:29
Also, there's just something that chafes in my mind knowing that we're doing a lot of completely wasted effort somewhere. 16:30
Even if the total time spent on it "isn't that big".
nine japhb: do you need it gift wrapped?
japhb I mean, perl5 starts in like 3ms, so saving you saved almost 10% of a perl5 startup. ;-)
japhb: With a bow, thanks. ;-) 16:31
timotimo perf says MVM_validate_static_frame takes 1.85% of time in perl6 -e '', someone want to optimize the bytecode validator? :P 16:32
maybe come up with some vectorization opportunities or something
lizmat m: my @a; @a[4] = 42; dd @a; my @b = reverse @a; dd @b # this feels like a bug 17:04
camelia Array @a = [Any, Any, Any, Any, 42]
Array @b = [42, Mu, Mu, Mu, Mu]
lizmat my @a; @a[4] = 42; dd @a; my @b = reverse @a; @b[10] = 666; dd @b 17:05
m: my @a; @a[4] = 42; dd @a; my @b = reverse @a; @b[10] = 666; dd @b
camelia Array @a = [Any, Any, Any, Any, 42]
Array @b = [42, Mu, Mu, Mu, Mu, Any, Any, Any, Any, Any, 666]
lizmat GH #1444 17:10
PSA: I won't be able to do the Perl 6 Weekly on 5 February 17:29
timotimo won't be able to do it either because of personal situation (yeah, timing is bad) 17:30
is there someone here who would want to try to do the Perl 6 Weekly once?
If we can't find anybody, then we'll have to make it a bi-weekly once :) 17:31
Zoffix lizmat: Yeah, I can do it. I'd need access to post on the weekly blog tho. 17:56
wordpress username @zoffix 17:57
lizmat Zoffix: cool!
[Coke] Zoffix++ 17:58
El_Che Zoffix: you're change to rename the "_Perl 6_ weekly news"
I mean, alias
;)
timotimo Zoffix: check your emails or something :) 18:04
lizmat Zoffix: to be clear, the 29 January one I'll do still 18:05
Zoffix timotimo: got the invite. Thanks. 18:07
lizmat: understood.
There's no easy way to set a dynvar for just the first arg of an op, is there? I need to mark first arg of `andthen` as $*IN_STMT_MOD := 1; 18:16
lizmat draws a blank 18:21
perl -E 'say length "🦋"' # 4
*sigh*
ilmari $ perl -Mutf8 -E 'say length "🦋"' 18:22
1
Zoffix :) 18:23
lizmat aaaaahh
ilmari++
dinner& 18:26
stmuk pl6anet.org/drop/rakudo-star-2018.01-RC0.tar.gz 19:28
(I may have to add some more test deps for HTTP::UserAgent in RC1)
nine Why on earth do we assemble strings representing our standard repositories just to parse them again when we assemble the repository list? Could anything be more wasteful? And why have I stared at this code for 2 years without realizing that? 20:13
geekosaur because this was all slapped together rapidly post-Christmas >.> 20:14
AlexDaniel stmuk++ 20:27
[Coke] nine++ 20:34
dogbert17 AlexDaniel: hasn't GH #1430 been fixed? 21:05
AlexDaniel GH#1430 21:27
synopsebot GH#1430 [open]: github.com/rakudo/rakudo/issues/1430 [UNI][regression] “1” is whitespace according to uniprop
AlexDaniel yeees
samcv: ↑ 21:28
I marked it as testneeded because I don't know if there's a test
lizmat hmmm... do we have a link to an explanation of how precompilation works in Rakudo Perl 6 ? 22:01
Google only comes up with docs.perl6.org/programs/00-running...ompilation
and a bunch of (old) bug reports about precompilation 22:02
stmuk lizmat: I found www.youtube.com/watch?v=uu8auCKCUN4 useful 22:03
travis-ci Rakudo build passed. Zoffix Znet '[NQP Bump] Brings 9 commits 22:49
travis-ci.org/rakudo/rakudo/builds/333092649 github.com/rakudo/rakudo/compare/f...a48efc975a
lizmat calls it a day& 22:53
samcv AlexDaniel: there is a test 23:33
AlexDaniel samcv: yeah, Zoffix++ found it already
samcv cool :)
AlexDaniel thanks!
samcv i'm working on a script that will update the moarvm.org page after release. since it doesnt have any automated way to do it
AlexDaniel awesome! 23:34
samcv woo moarvm.org/releases.html 23:48
i think everything should be ok.
had to modify ChangeLog to make it render properly with a MD rederer and make it retain the same format 23:49