AlexDaniel Failed to find dependencies: DateTime::Parse 00:07
(╯°□°)╯︵ ┻━┻
raw.githubusercontent.com/sergot/d.../META.info 404 00:09
Unsuccessful HTTP response: 404 404 Not Found at server/updatelist.pl line 43, <$_[...]> line 877.
that's from ecosystem-api.p6c.org/update.log
right, because it's META6 00:10
ok, should be good now 00:13
Geth nqp: 8192977b92 | (Tom Browder)++ (committed using GitHub Web editor) | docs/ops.markdown
document removal of shell and spawn
01:03
nqp: cbfe319ce2 | (Tom Browder)++ (committed using GitHub Web editor) | docs/ops.markdown
remove entry for external commands
01:09
travis-ci Rakudo build passed. Zoffix Znet 'Merge pull request #1441 from rakudo/post-release 02:15
travis-ci.org/rakudo/rakudo/builds/333092767 github.com/rakudo/rakudo/compare/d...8bbc8dad6a
Rakudo build passed. Elizabeth Mattijsen 'Give a bare 'fc' the same treatment as bare 'lc' or 'uc'' 04:02
travis-ci.org/rakudo/rakudo/builds/333262169 github.com/rakudo/rakudo/compare/7...fbc43b411c
Rakudo build passed. Pepe Schwarz 'Fix basic list marshalling' 05:17
travis-ci.org/rakudo/rakudo/builds/333271941 github.com/rakudo/rakudo/compare/8...499c90fc06
stmuk subjectively without benchmarking 2018.01 pod generation seems faster to me 08:31
p6doc is more useable than before
AlexDaniel stmuk: that's very interesting. If that's true, then I think it's because of general performance improvements in rakudo. All pod changes were about its correctness and not performance, IIRC 08:47
committable6: HEAD say 42 09:00
committable6 AlexDaniel, ¦HEAD(a2499c9): «42␤»
AlexDaniel committable6: HEAD say ‘hello world’ for ^20
committable6 AlexDaniel, gist.github.com/93babf61ea049e30d8...cce2fbbec0
lizmat Files=1233, Tests=76092, 339 wallclock secs (15.07 usr 5.47 sys + 2303.60 cusr 237.49 csys = 2561.63 CPU) 09:55
stmuk this "say" truncating output via .gist is going to catch some p5 people out 10:40
Geth nqp: e4f79cbae1 | (Zoffix Znet)++ | tools/build/MOAR_REVISION
[MoarVM Bump] Brings 21 commits

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...1-gd3ce626 d3ce626 Add JIT template for atkey_o ac9f241 Update release_guide.md w/ info on generating releases.html 65273a4 Add script that generates moarvm.org/releases.html f1197e4 Edit the ChangeLog to include the original release 3992885 Fix MVM_spesh_log_decont not logging anything 7e7ee8d Run basic facts discovery on inlined code 00d19a6 Fix segfault in JIT bytecode dumper on code without static frame 279d7d7 Fix a typo in a comment 63f92e0 Only BAIL on DEOPT_ONE annotations on actual guard ops in expr JIT d3f38a1 Expr JIT template for create 72626cc Expr JIT template for capturelex 4d71947 Expr JIT template for takeclosure ... (9 more lines)
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...1-gd3ce626
rakudo: a5c2398cc7 | (Zoffix Znet)++ | tools/build/NQP_REVISION
[NQP Bump] Brings 4 commits

NQP bump brought: github.com/perl6/nqp/compare/2018....4-ge4f79cb e4f79cb [MoarVM Bump] Brings 21 commits cbfe319 remove entry for external commands 8192977 document removal of shell and spawn b930e35 No need to load ModuleLoader multiple times
MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...1-gd3ce626 d3ce626 Add JIT template for atkey_o ac9f241 Update release_guide.md w/ info on generating releases.html 65273a4 Add script that generates moarvm.org/releases.html f1197e4 Edit the ChangeLog to include the original release 3992885 Fix MVM_spesh_log_decont not logging anything 7e7ee8d Run basic facts discovery on inlined code ... (15 more lines)
rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....4-ge4f79cb
1ee89b5407 | (Zoffix Znet)++ | src/Perl6/Actions.nqp

Phixes github.com/rakudo/rakudo/issues/1212 Fixes RT#130575: rt.perl.org/Ticket/Display.html?id=130575 Fixes RT#132337: rt.perl.org/Ticket/Display.html?id=132337 Fixes RT#131548: rt.perl.org/Ticket/Display.html?id=131548 Fixes RT#132211: rt.perl.org/Ticket/Display.html?id=132211 Fixes RT#126569: rt.perl.org/Ticket/Display.html?id=126569 Fixes RT#128054: rt.perl.org/Ticket/Display.html?id=128054 Fixes RT#126413: rt.perl.org/Ticket/Display.html?id=126413 Fixes RT#126984: rt.perl.org/Ticket/Display.html?id=126984 Fixes RT#132172: rt.perl.org/Ticket/Display.html?id=132172
All of the bugs in the tickets are caused by mis-scoped blocks resulting in p6capturelex failing to set outers correctly, and thus ... (39 more lines)
Zoffix stmuk: considering they stole `say` from Perl 6, I'd say it's their problem :P
yoleaux 25 Jan 2018 21:23Z <Xliff_> Zoffix: HAH! I almost guessed Russian.
stmuk maybe it should be renamed "mumble" 10:42
Zoffix -1 10:44
Based on what people ask for help with, I wouldn't say it causes a lot of issues. 10:45
ZofBot: hold my eBeer and watch this 10:48
ZofBot Zoffix, There are some that are todo'd for os x, IIRC [Coke]: why cant the fudge step happen at runtime? i thought it just inserts todos /skips it's rewriting the source code
synopsebot RT#130575 [new]: rt.perl.org/Ticket/Display.html?id=130575 [BUG] Variables inside a try {} block used with `andthen` don't always get updated to correct values
synopsebot RT#132337 [new]: rt.perl.org/Ticket/Display.html?id=132337 [BUG] Code block inside `andthen`/`orelse` doesn't close over lexical variables correctly
synopsebot RT#131548 [new]: rt.perl.org/Ticket/Display.html?id=131548 [BUG] andthen weirdness with lexicals
synopsebot RT#132211 [new]: rt.perl.org/Ticket/Display.html?id=132211 [BUG] combination of `andthen` and `orelse` causes closed over variable to never update
synopsebot RT#126569 [new]: rt.perl.org/Ticket/Display.html?id=126569 [BUG] xx-repeated expression enclosed in parens can't access the topic $_ of a statement modifier
synopsebot RT#128054 [open]: rt.perl.org/Ticket/Display.html?id=128054 [PARSER] In a string enclosed in parens, a {}-interpolation can't access the topic $_ of a statement modifier
RT#126413 [new]: rt.perl.org/Ticket/Display.html?id=126413 [BUG] WhateverCode loses topic inside given modifier
RT#126984 [open]: rt.perl.org/Ticket/Display.html?id=126984 [BUG] WhateverCode in given doesn't get refreshed when entering surrounding block in Rakudo
RT#132172 [new]: rt.perl.org/Ticket/Display.html?id=132172 [BUG] `orelse` + block in a string gets the wrong $_
Zoffix M-M-M-MONSTER KILL! 😎
Geth roast: 2f2998733a | (Zoffix Znet)++ | MISC/bug-coverage.t
Test lexical scoping in thunks

Closes github.com/rakudo/rakudo/issues/1212 RT#130575: rt.perl.org/Ticket/Display.html?id=130575 RT#132337: rt.perl.org/Ticket/Display.html?id=132337 RT#131548: rt.perl.org/Ticket/Display.html?id=131548 RT#132211: rt.perl.org/Ticket/Display.html?id=132211 RT#126569: rt.perl.org/Ticket/Display.html?id=126569 RT#128054: rt.perl.org/Ticket/Display.html?id=128054 RT#126413: rt.perl.org/Ticket/Display.html?id=126413 RT#126984: rt.perl.org/Ticket/Display.html?id=126984 RT#132172: rt.perl.org/Ticket/Display.html?id=132172
Rakudo fix: github.com/rakudo/rakudo/commit/1ee89b5407
10:50
jnthn Zoffix++ :D 10:51
Ven`` Zoffix++ skids++
Zoffix :) 10:52
Geth roast: e39e5d72ea | (Zoffix Znet)++ | 2 files
Unfudge now-passing tests
11:02
dogbert2 ZOFFLOP: t/spec/MISC/bug-coverage.t
Zoffix++
lizmat expects a small gain for test-t after these commits 11:45
tbrowder nqp err msg: “write_fhb requires a native array to write from” 11:52
i’m just trying to write to an open file handle... 11:53
AlexDaniel Zoffix++ this is awesome :D 11:58
tbrowder nqp 2017.11, moarvm 2017.11 12:01
jnthn tbrowder: What did you try to write? 12:02
tbrowder sorry, moar version 2018.01 12:03
AlexDaniel nqp 2017.11 and moar 2018.01?
is that supposed to work? 12:04
jnthn nqp: nqp::writefh(nqp::getstdout(), "no, can't do a string")
camelia write_fhb requires a native array to read from
at <tmp>:1 (<ephemeral file>:<mainline>)
from gen/moar/stage2/NQPHLL.nqp:1537 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/NQPHLL.moarvm:eval)
from gen/moar/stage2/NQPHLL.nqp:1774 (/home/camelia/rakudo…
tbrowder i tried to write a string to the open file
jnthn Then the error is accurate, write is for byte buffers
tbrowder that has changed from prev behavior
jnthn the nqp::writefh op has never worked with strings 12:05
Are you working with a handle you got from open, or nqp::open?
If the former, then you chould have an object wrapping the handle that has a print method 12:06
tbrowder nqp::open
the docs say writefh takes a str
jnthn Then they're wrong 12:07
Consider using open instead
tbrowder so that’s another builtin?
jnthn That gives you an NQP::FileHandle object, which has a print method
Sort of
nqp:: ops are the interface to the VM
NQP, for convenience, also has a setting 12:08
Which has various higher-level built-ins
At the nqp:: op level, all I/O is binary
That's a fine design for the VM inter-op layer, but not very convenient for writing NQP code
tbrowder :-D 12:09
jnthn Thus open returns an NQPFileHandle object which has methods like print and say, which do the encode and write for you.
tbrowder like this: $fh := open($fnam, “w”); print($fh, $str). ?? 12:11
jnthn $fh.print($str) 12:19
Possibly also :w instead of "w", but I forget
lunch, bbiab
tbrowder this works: $fh := open(“f.txt”, :w); $fh.print(“text\n”); 12:26
thnx 12:27
Geth nqp: d71bd7334c | (Zoffix Znet)++ (committed using GitHub Web editor) | docs/qast.markdown
Fix typo
14:18
|Tux| Rakudo version 2018.01-31-g1ee89b540 - MoarVM version 2018.01-21-gd3ce626b4
csv-ip5xs1.026 - 1.034
csv-ip5xs-2010.821 - 11.063
csv-parser11.626 - 11.936
csv-test-xs-200.449 - 0.521
test10.556 - 10.608
test-t2.608 - 2.687
test-t --race1.104 - 1.110
test-t-2046.828 - 47.088
test-t-20 --race16.274 - 18.746
14:19
2018-01-18 18:56:57 test-t 2.704
2018-01-19 13:30:08 test-t 2.691
2018-01-26 15:13:38 test-t 2.687
2018-01-19 13:33:23 test-t 2.683
2018-01-26 15:16:48 test-t 2.608
timotimo way cool. 14:23
lizmat test-t about 4% faster, I'd say 14:24
ilmari buggable: speed :4 test-t 14:25
buggable ilmari, ▅ █ ▄↑ ▇▅ ▄ ↑▆ █ dates: 2018-01-12–2018-01-26
ilmari, █▆█ ██▇▅▅▅██ ▃▆▂ ▂█ ▄▆ ▄▄▂ ▃ ██▅█▄ ▃ range: 2.608s–3.397s
ilmari, ███▇▅████████▅███▅███▄▄██▁▃ ███▇▆▇▂███████▅▆▄▄█ speed: 7% faster (widths: 9/3)
ilmari, ███████████████████████████▇▇███████████████████▇▁
dogbert2 Zoffix++ for Unreal fixes :-) 14:36
lizmat I think MasterDuke also had a bit in this one :-)
afk for a few hours& 14:37
dogbert2 was referring to the M-M-M-MONSTER KILL! fix
.oO(one more and it would have been an ULTRA KILL)
14:38
Geth rakudo: 5bd3366921 | (Stefan Seifert)++ | src/core/CompUnit/RepositoryRegistry.pm
Speed up repo chain setup by avoiding costly regexes

Seems wasteful to assemble repo spec strings just to parse them again later on. Better to create the repository objects directly and assemble the chain manually.
15:44
rakudo: 6fa1e78f80 | (Stefan Seifert)++ | src/core/CompUnit/RepositoryRegistry.pm
Avoid costly regexes if include spec is a plain path

If there's something to parse, the spec contains a '#' which we can look for quite cheaply. In the very common case of -Isome/path/to/lib we don't have to parse anything.
rakudo: CurtTilmes++ created pull request #1446:
Fix function signature
16:49
rakudo: 44e7220fca | (Curt Tilmes)++ | t/04-nativecall/01-argless.c
Fix function signature

Current signature is int*, but it actually returns a char* which gives a warning. It still works because Perl declares the nativecall as a Str anyway. This just avoids the warning.
16:50
rakudo: 38ae3b9c69 | (Zoffix Znet)++ (committed using GitHub Web editor) | t/04-nativecall/01-argless.c
Merge pull request #1446 from CurtTilmes/master

Fix function signature
lizmat nine: How difficult would it be for repo's to handle :api<perl5> in use statements ? 20:15
nine lizmat: that should already work? 21:18
lizmat m: use Test:api<perl5> 21:20
camelia ( no output )
lizmat hmmm
m: class Foo:api<perl5>
camelia 5===SORRY!5=== Error while compiling <tmp>
Cannot use adverb api on a type name (only 'ver' and 'auth' are understood)
at <tmp>:1
------> 3class Foo:api<perl5>7⏏5<EOL>
expecting any of:
generic role
lizmat m: class Foo:api<perl5> { }
camelia 5===SORRY!5=== Error while compiling <tmp>
Cannot use adverb api on a type name (only 'ver' and 'auth' are understood)
at <tmp>:1
------> 3class Foo:api<perl5>7⏏5 { }
expecting any of:
generic role
lizmat ok, so it's just matter of getting this parsed ?
nine lizmat: note that the FileSystem repo ignores :ver, :auth and :api. That may be why it finds Test anyway 21:22
lizmat: api is part of a dist's META 21:23
lizmat well, would be nice if you could specify that also in the class / module statement, no? 21:24
would help mi6
nine certainly 21:27
lizmat ok, I looked at it earlier today... 21:29
nine: question: I'm trying to re-export imported subs 21:30
do we have a recipe for that anywhere that you know of ?
nine lizmat: github.com/LLFourn/p6-CompUnit-Util#re-export 21:32
lizmat cool
hmmm... isn't this just a matter of finding the code refs and putting them in a map in EXPORT ? 21:36
nine Could very well be. It's quite unmagical actually 21:39
lizmat guess I'm too tired now to figure this out, but this works: 21:47
use P5tie;
BEGIN %export{'&' ~ .name} := $_ for &tie, &tied, &untie;
and then sub EXPORT { %export }
Geth rakudo: c268f55c7c | (Stefan Seifert)++ | src/core/CompUnit/RepositoryRegistry.pm
Micro optimize parse-include-specS
21:55
rakudo: 5be174bbe4 | (Stefan Seifert)++ | 4 files
Speed up repository chain setup

Instead of specially encoded repository specification strings, we now parse the input into CompUnit::Repository::Spec objects as soon as possible and only deal with those. This saves quite some string mangling and parsing cutting a couple of milliseconds from repository chain setup overhead.
lizmat whee! :-) 21:56
nine My commits today amount to about 10ms savings when running perl6 -Ilib -MTest -e ''
I.e. about a second saved when running a test suite of 100 files ;) 21:57
What makes optimizing this part really hard is that it's code that runs only once and that very soon. So none of our spesh and JIT stuff helps and even multi dispatch shows its cost 21:58
moritz nine++
lizmat www.reddit.com/r/perl/comments/7t8..._and_cpan/ 22:07
before anybody gives it a different title 22:08
jnthn nine: I suspect we could at least do the candidate sort at compile time, even if we can't prime the cache. 23:04
japhb jnthn++ # A very thoughtful, calm, measured response. This is what being a project leader *should* look like. 23:06