Geth nqp: d40d6dc3c2 | (Jonathan Worthington)++ | tools/build/MOAR_REVISION
Bump to get latest MoarVM invocation opts.
 09:30
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...5-gc0f7a3b
jnthn NeuralAnomaly: status 09:31
NeuralAnomaly jnthn, [?] Next release will be in 1 week and 2 days. Since last release, there are 45 new still-open tickets (11 unreviewed and 1 blockers) and 139 unreviewed commits. See perl6.fail/release/stats for details
Geth rakudo/nom: f590863e17 | (Jonathan Worthington)++ | tools/build/NQP_REVISION
Bring in latest MoarVM invocation opts.

Plus various other specializer fixes. So they get a bit more testing.		 09:32
rakudo/nom: version bump brought these changes: github.com/perl6/nqp/compare/2017....2-gd40d6dc
cd7dc4ce93 | (Stefan Seifert)++ | lib/NativeCall.pm6

Instead of having a generic code block handle all NativeCall subs, we compile a specialized one for every sub. This way we can actually utilize the information from the signature to identify the arguments that need deconting and save the looping over the arguments at runtime.
This also leaves the argument checks to the call dispatcher.
nine This ^^^ of course passes all tests but could probably still use someone looking at it as I'm really just stumbling along here :) 11:02
dogbert17_ ZOFFLOP: t/spec/S03-junctions/autothreading.t 11:39
Geth rakudo/nom: 5e8d461325 | (Elizabeth Mattijsen)++ | src/core/Junction.pm
Fix Junction ~ Str and Str ~ Junction

The continuing story of concatenating a Str to Junction		 11:49
dogbert17_ lizmat: does that fix have something to do with the test failure? 11:50
lizmat dogbert17_: yes 11:53
if its about the auto-threading test you're talking about
Zoffix nine: so what's the new speed? :) 12:07
dogbert17_ yes, didn't work tough ... 12:14
still ZOFFLOPPING 12:15
atm 5-6 ZOFFLOPS 12:19
S03-junctions/autothreading.t, S05-modifier/ignoremark.t, S05-modifier/samemark.t, S32-io/IO-Socket-INET.t and integration/eval-and-threads.t 12:21
nine Zoffix: csv-ip5xs from 14.838s to down to 13.042s (best of 5, 100000 iterations) 12:47
Zoffix neat
nine Zoffix: was m17.483s 12:53
before I started optimizing on Monday :)
Zoffix :o 12:55
nine++
nine Also the code we produce now is much easier to optimize on the lower levels which is where I'll have to go next to make progress. 12:56
Zoffix sweet 13:00
dogbert17_ lizmat: not ok 88 - an & junction right of a | junction will be autothreaded first This type cannot unbox to a native string: P6opaque, Junction 13:14
Zoffix FWIW, jnthn's comments on the topic: irclog.perlgeek.de/perl6-dev/2017-...i_14992203 13:15
dogbert17_ m: use Test; sub tp($a, $b, $c) { "$a $b $c" }; my Mu $res = tp("dog", 1|2, 10&20); is $res.Str eq q{all(any("dog 1 10", "dog 2 10"), any("dog 1 20", "dog 2 20"))}, "an & junction right of a | junction will be autothreaded first"; 13:22
camelia not ok 1 - ??# Failed test at <tmp> line 1?This type cannot unbox to a native string: P6opaque, Junction? in block at /home/camelia/rakudo-m-inst-2/share/perl6/sources/C712FE6969F786C9380D643DF17E85D06868219E (Test) line 177? in sub is at /h…
stmuk I really think the html generation code in github.com/perl6/doc should be moved into another repo 13:28
Zoffix stmuk: well, don't leave us hanging. Why you think things 13:40
stmuk well it means we can't ship basic docs and the command line client in R* without shipping dependencies which aren't particularly useful (used as part the HTML generation) 13:45
also if the raw docs were on their own it would encourage others to write alternative viewers 13:46
also when distro packagers just want to package docs it makes more sense 13:47
it even contains coffeescript 14:06
Zoffix mmmm coffee. Good idea 14:08
nine Did I hear coffee?
stmuk no it was tea :P 14:09
Zoffix stmuk: yeah, all good points. I have another to add: I notice a few times new docs contributors struggling with build system because they want to "check" how their addition renders, even ater being told it's not that important to pre-check. Having docs as just docs in a repo would make it easier to contribute 14:18
Zoffix browses through the 263 docs issues 14:24
.... yeah... media.giphy.com/media/4pMX5rJ4PYAEM/giphy.gif
stmuk not sure how to get from current github.com/perl6/doc to two repos 14:27
maybe moving doc/doc and some support files to a new repo and renaming the old might be easier 14:28
so just-docs & web-docs exist (the latter having a submodule checkout of just-docs) 14:29
stmuk decides to email a git guru about it 14:36
Zoffix So the next step for modules.perl6.org will be rsyncing Rakudo dists off CPAN and unpacking them, giving users of the site the option to browse the files in the distro. The rsyncage and unpackage will be done on www.p6c.org box, under modules.perl6.org user. 14:40
(just mentioning in case someone sees a problem with that)
lizmat fwiw, it looks like test-t has become a lot slower on my machine after f590863e17 14:43
awwaiid stmuk: maybe isolate the build tools more in the current doc repo rather than being totally separate; like make building different outputs have different dependencies or something 14:47
Zoffix nine: lizmat BTW, if you see Damian, would you ask if "Quantum::Computing" module actually exists and if yes, where's the sauce? That question pops up not infrequently ( www.reddit.com/r/perl6/comments/6p...computing/ ) and there's never a good answer. (Perhaps he should incorporate a clear answer into the "On Shoulder of Giants" talk)
awwaiid stmuk: I mean, as a first step; another way to say this is instead of separate repos make them separate directories -- keep doc/ and move a lot of the other stuff into some other subdir like website/ 14:48
stmuk: then you'd keep things like the bit that runs the code samples to verify them 14:49
stmuk awwaiid: yes that sounds a good idea 14:50
I did consider pointing an installer at a parallel META6.json (but maybe too confusing) 14:52
certainly moving files around would be good in itself and probably help if moved into different repos
awwaiid I wouldn't worry too much about git slicing/dicing; my recommendation is to start a new repo called like doc-site or similar and copy files in; part of it's build process would be to clone the existing doc/ repo with it's current structure (ignore overlap). Once you get that going and you get the site building off of it, you then delete the overlap files from the doc/ repo 14:55
(if you go the new repo route)
Zoffix Hm, actually it's not just dists off CPAN, it'll also be git clones of GitHub/GitLab repos 14:57
Oh, actually not as needed as CPAN dists, so maybe not yet. 15:01
stmuk hmm if the doc-site was still called "doc" no server processes would have to be changed 15:05
doc & new p6doc?
Geth_ tap-harness6: 39b2469c1a | (Leon Timmermans)++ | bin/prove6
Allow multiple -I arguments
 15:21
Zoffix Geth is ded? 15:23
Just re-set the hook to new geth and re-delivered it. The hook said OK but nothing on IRC
Geth: he;[p
Geth: help
Geth Zoffix, Source at github.com/perl6/geth To add repo, add an 'application/json' webhook on GitHub pointing it to geth.perl6.party/?chan=#perl6 and choose 'Send me everything' for events to send | use `ver URL to commit` to fetch version bump changes
Zoffix Oh, I redilibered the wrong hook 15:24
Geth tap-harness6: 39b2469c1a | (Leon Timmermans)++ | bin/prove6
Allow multiple -I arguments
lizmat Zoffix: there is a tarbal at bit.do/Perl6SOG 15:27
Zoffix lizmat++ thanks 15:35
So it *is* a real module. Just done with classical methods
huggable: Quantum::Computation :is: Shoulder of Giants talk code exmples in bit.do/Perl6SOG the Quantum::Computation module is in lib6/Quantum/ 15:37
huggable Zoffix, Added Quantum::Computation as Shoulder of Giants talk code exmples in bit.do/Perl6SOG the Quantum::Computation module is in lib6/Quantum/
Geth tap-harness6: aa5dbdbabe | (Leon Timmermans)++ | bin/prove6
Fix empty $incdir in prove6
 16:26
rakudo/nom: ec7bc25c4f | (Stefan Seifert)++ | 2 files
Make RAKUDO_MODULE_DEBUG output reproducible

Turns out it's more helpful for debugging if one is able to diff the output of different RAKUDO_MODULE_DEBUG runs than having exact PIDs or timing information.		 17:13
timotimo aye! 17:17
AlexDaniel Zoffix: hey. What's the idea behind making it possible to browse the code from modules.perl6.org ? Is there a ticket with discussion somewhere? I don't see a problem with that, but at the same time I don't understand what problem it can possibly solve (normally I want to go the actual repo, and I'm fine with github or wherever it takes me). 17:55
just being curious. I may have misunderstood something though 17:56
Zoffix The problem being solved is there's no front end to browse Rakudo's CPAN dists 17:57
AlexDaniel aaaah
Zoffix++ 18:00
Zoffix And code/docs from github might be made browseable on the site as a matter of unified interface, but that's why I said "not yet", exactly because users can just go to the repo 18:01
Like once we figure out how to safely render POD6, for example
ugexe m: class Foo { has $.bar = 1; multi method gist(Foo:D:) { $.bar }; }; say Foo.new 18:16
camelia This type cannot unbox to a native string: P6opaque, Int? in block <unit> at <tmp> line 1??
ugexe m: class Foo { has $.bar = 1; multi method gist(Foo:D:) { ~$.bar }; }; say Foo.new
camelia 1?
jdv79 uh. so i updated to the latest and tried to install JSON::Tiny and all my gnome-terminal instances disappear whilst trying 18:17
that's impressive
Zoffix There is (was?) a ticket for .gist and my response to it was that .gist is on the same level as .Str; there's an expected return type for it and if you don't obey it, you get to keep the pieces. 18:18
Zoffix doesn't get what the problem is in rt.perl.org/Ticket/Display.html?id...et-history 18:21
I think that ticket is closed now; it was by smsl 18:23
geekosaur Zoffix, they want the default gist behavior for classes to defer to .perl on a per-attribute basis 18:25
Zoffix ok 18:26
Zoffix tries 18:28
geekosaur (I actually suspect what they *really* want is that .gist maps to .perl on non-collective types and to recursive .gist on collectives) 18:35
Zoffix damn lots of failures. 18:49
travis-ci Rakudo build failed. Stefan Seifert 'Make RAKUDO_MODULE_DEBUG output reproducible 18:59
travis-ci.org/rakudo/rakudo/builds/263181009 github.com/rakudo/rakudo/compare/5...7bc25c4f4d
buggable [travis build above] ? All failures are due to timeout (0), missing build log (0), GitHub connectivity (1), or failed make test (0).
Zoffix fucking 'ell. I have more failures than I can count by just glancing and that's without my patch. 19:06
gist.github.com/zoffixznet/091f2ba...88aa99dd1d
What happened to master-always-passes :/ 19:07
Kinda pointless in having tests if I can't tell whether the change I'm introducing makes stuff fail :/
"You planned 106 tests but ran 88." can't even test 19:13
And google killed my VM while I was trying to figure out which failures were mine... 19:15
(?°?°??? ???
Commented my patch on the ticket, if anyone wants to tinker with it: rt.perl.org/Ticket/Display.html?id...xn-1480860 19:17
ugexe is it a moar/nqp/rakudo version mismatch thing? 19:27
Zoffix The test failures? No idea. I just pulled in latest changes and ran `make stresstest` 19:29
ugexe whats your moarvm version? curious if your command ultimately did `moar HEAD` or `moar whatever-the-last-bump-was` 19:39
Zoffix Don't know. I didn't restart the VM. The command is perl Configure.pl --gen-moar --gen-nqp --backends=moar; make; make test; make install 19:40
Are you saying your stresstest passes clean on rakudo HEAD? 19:41
ugexe install/bin/perl6 -v 19:42
im not able to stress test right this moment 19:43
Zoffix <Zoffix> Don't know. I didn't restart the VM.
ugexe vm is getting too ambiguous :/ 19:44
Zoffix heh 19:45
Zoffix relocates
dogbert17 Zoffix: I have the same failures, nested.t is a flapper though
they have been present a couple of days 19:47
I don't think they are in any way related to the recent spesh changes though 19:48
ugexe bisect: bisect: use lib "inst#/home/camelia/star-2017.07/share/perl6/site"; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; } 20:01
bisectable6 ugexe, Bisecting by output (old=2015.12 new=ec7bc25) because on both starting points the exit code is 1
ugexe bisect: use lib "inst#/home/camelia/star-2017.07/share/perl6/site"; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; }
bisectable6 ugexe, bisect log: gist.github.com/e6c720be2d5c078841...9e38a873d3
ugexe, (2016-07-16) github.com/rakudo/rakudo/commit/b4...55e66fb646
ugexe, Bisecting by output (old=2015.12 new=ec7bc25) because on both starting points the exit code is 1
ugexe, bisect log: gist.github.com/5b2edae66e9adb6816...6784f3c508
ugexe, (2016-07-16) github.com/rakudo/rakudo/commit/b4...55e66fb646
ugexe bisect: old=2017.07 use lib "inst#/home/camelia/star-2017.07/share/perl6/site"; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; } 20:03
bisectable6 ugexe, Bisecting by output (old=2017.07 new=ec7bc25) because on both starting points the exit code is 1
ugexe, bisect log: gist.github.com/006657caab9b57a136...764c2c840b
ugexe, (2017-07-15) github.com/rakudo/rakudo/commit/d3...0de30596dd
ugexe star: use lib "inst#/home/camelia/star-2017.07/share/perl6/site"; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; } 20:04
camelia 1..4?ok 1 - ?ok 2 - ?ok 3 - ?ok 4 - ?
ugexe m: use lib "inst#/home/camelia/star-2017.07/share/perl6/site"; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; }
camelia 1..4?ok 1 - ?ok 2 - ?ok 3 - ?Input (2 characters) is not a valid JSON string? in sub from-json at /home/camelia/star-2017.07/share/perl6/site/sources/9B467EEF9267A777BB53BAA2F19BE2C9D756BEED (JSON::Tiny) line 59? in block <unit> at <tmp> li…
ugexe i thought I could get bisectable to figure it out :/ 20:05
MasterDuke ugexe: bisectable6 doesn't have a "inst#/home/camelia/star-2017.07/share/perl6/site", but you could create an issue github.com/perl6/whateverable/issues about making modules available 20:59
yoleaux 17:51Z <AlexDaniel> MasterDuke: I've created a ticket. Please leave some comments there: RT #131879
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131879
AlexDaniel there was this idea but unfortunately nothing came out of it yet: github.com/perl6/whateverable/issues/55 21:00
MasterDuke: buuuuuuuuuuuut why can't we do it already? There's perl6-all-modules thingy almost in the same directory 21:01
c: HEAD use lib ‘data/all-modules/moritz/json/lib/’; use JSON::Tiny 21:02
committable6 AlexDaniel, ¦HEAD(ec7bc25): «»
AlexDaniel there you go
however, note that precomp files may be left hanging
MasterDuke yeah, shouldn't be too bad
AlexDaniel bisect: HEAD use lib ‘data/all-modules/moritz/json/lib/’; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; } 21:03
bisectable6 AlexDaniel, Bisecting by output (old=2015.12 new=ec7bc25) because on both starting points the exit code is 1
AlexDaniel, bisect log: gist.github.com/26e25d33e728b74379...bcc1c34fb3
AlexDaniel, (2015-12-25) github.com/rakudo/rakudo/commit/07...dc61f84053
AlexDaniel, The result looks a bit unrealistic, doesn't it? Most probably the output is different on every commit (e.g. ?bisect: say rand?)
AlexDaniel my bad…
bisect: use lib ‘data/all-modules/moritz/json/lib/’; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; }
bisectable6 AlexDaniel, Bisecting by exit code (old=2015.12 new=ec7bc25). Old exit code: 1
ugexe old=2017.07
bisectable6 AlexDaniel, bisect log: gist.github.com/6c72470cbdefdd5e30...22e1a35c25
AlexDaniel, (2017-08-10) github.com/rakudo/rakudo/commit/f5...6e3529060e
AlexDaniel ugexe: what do you think about this? 21:04
we can even check what's going on there
c: f590863e17^,f590863e17 use lib ‘data/all-modules/moritz/json/lib/’; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; }
committable6 AlexDaniel, gist.github.com/43c5f3fadc0a2f6dce...b732b8af68
ugexe well it just figured out what broke JSON::Tiny 21:05
jdv79: ^
AlexDaniel is it correct that one test is failing both before and after f590863e17?
“Input (2 characters) is not a valid JSON string” is the new part though 21:06
ugexe m: use lib "inst#/home/camelia/star-2017.07/share/perl6/site"; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; }
camelia 1..4?ok 1 - ?ok 2 - ?ok 3 - ?Input (2 characters) is not a valid JSON string? in sub from-json at /home/camelia/star-2017.07/share/perl6/site/sources/9B467EEF9267A777BB53BAA2F19BE2C9D756BEED (JSON::Tiny) line 59? in block <unit> at <tmp> li…
ugexe oops
star: use lib "inst#/home/camelia/star-2017.07/share/perl6/site"; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; }
camelia 1..4?ok 1 - ?ok 2 - ?ok 3 - ?ok 4 - ?
ugexe they all pass here
AlexDaniel m: say 'bå'.uninames 21:08
camelia (LATIN SMALL LETTER B LATIN SMALL LETTER A WITH RING ABOVE)?
AlexDaniel e: say 'bå'.uninames
evalable6 (LATIN SMALL LETTER B LATIN SMALL LETTER A WITH RING ABOVE)
AlexDaniel MasterDuke: do you see anything?
ah sorry 21:10
MasterDuke looks the same to me?
AlexDaniel because that's exactly how I copy-pasted it
because I copy-pasted it guess from where
c: f590863e17^,f590863e17 use lib ‘data/all-modules/moritz/json/lib/’; use JSON::Tiny; use Test; my @t = '{ "a" : "b\u00E5" }' => { 'a' => 'bå' }, '[ "\u2685" ]' => [ '?' ], qq{"x\c[ZERO WIDTH JOINER]a"} => "x\c[ZERO WIDTH JOINER]a", qq{"\c[ZERO WIDTH JOINER]"} => "\c[ZERO WIDTH JOINER]"; plan (+@t); for @t -> $p { my $got = from-json($p.key); is-deeply $got, $p.value; } 21:11
committable6 AlexDaniel, gist.github.com/2a9088ddcff37fd6f7...7ba3339af2
MasterDuke emacs?
AlexDaniel MasterDuke: that's a very interesting idea, but no :) irclog.perlgeek.de/perl6-dev/2017-...i_14998134
(it has ? instead of ?) 21:12
jdv79: can you submit a ticket for this? 21:13
.tell jnthn FWIW some kind of breakage due to f590863e17 gist.github.com/Whateverable/2a908...7ba3339af2 21:15
yoleaux AlexDaniel: I'll pass your message to jnthn.
jdv79 should i? 21:43
thinks its pretty obvious but sure
[Coke]: shouldn't it be possible to log a bug directly on rt? 21:46
the web interface i mean
MasterDuke jdv79: that was disabled because of spammers 21:48
jdv79 AlexDaniel: RT#131881 21:56
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131881
japhb Wow, that's an impressive bug. 22:13
.oO( Expected result: JSON parses correctly. Actual result: Moon alters orbit, solar eclipse suddenly moved to today. )
 22:14
geekosaur actual result: that one stackoverflow >.> 22:17
MasterDuke
.oO(isn't there an emacs command for altering the moon's orbit?)
AlexDaniel MasterDuke: no, that's a vim thing when you attempt to exit it 22:19
github.com/search?q=%3Awq&type=Commits
MasterDuke ha! 22:20
AlexDaniel by the way, is there any good reason to use JSON::Tiny nowadays? 22:27
instead of JSON::Fast
timotimo its code isn't messy :) 22:36
jnthn AlexDaniel: To see how much harder we need to optimize grammars yet ;) 22:38
timotimo very true 22:39
ugexe Grammar::HTTP still doesn't work from after braids stuff was merged in :( 22:41
someone finally opened an issue about it the other day 22:42
nine Ok, with the latest Inline::Perl5 commits, I'm now at 9.881s for csv-ip5xs (best of 5 runs, 100000 iterations). Used to be 13.042s earlier today and 17.483s last week. 22:45
jnthn :) 22:46
jnthn wonders if his efforts today might have shaved something off test-t
Nice if *that* number goes down as well as the one relating to Inline::Perl5 :D 22:47
nine I hope we'll see tomorrow :) But I should have a comfortable 50+ % advantage still ;)
jnthn :P
nine I'm now at a point where the startup overhead accounts for 30 % of the runtime
jnthn Anything that makes native calls faster is very good for us
In that it makes a ton of modules faster 22:48
nine in the standard 10000 iterations tests that is
jnthn Including such stuff as IO::Socket::Async::SSL :)
nine jnthn: have you looked at my NativeCall commit from today? github.com/rakudo/rakudo/commit/cd7dc4ce93 22:49
Actually already yesterday ;)
jnthn nine: Not yet, trying to inline closures was fairly absorbing... 22:50
Looks very much along the right lines though :) 22:51
nine exhales a little
Ok, will now let my subconcious practice this breathing thing a little ;) Good night! 22:52
jnthn 'night o/ 22:53

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!