Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
timotimo yes 00:02
there's lots of functions from gen/bootstrap visible in the profiler and i can't make those links to github 00:03
not directly anyway
MasterDuke huh, i thought they did work in nqp 00:04
00:09 SqrtNegI_ left
MasterDuke looks like not 00:10
Geth 6.d-prep: e8465bd8ac | (Zoffix Znet)++ | ChangeLog.md
[CL] Log up to c14b344a75815134402a762edd71e8d3c43
 00:11
MasterDuke timotimo: well, the implementation is all in nqp
00:14 lizmat left
timotimo cool 00:17
Geth rakudo: d73272e4c0 | (Zoffix Znet)++ | src/core/Any-iterable-methods.pm6
Consistify :U.pairup with :D one by returning a Seq
 00:21
roast: 6f96547445 | (Zoffix Znet)++ | S32-basics/pairup.t
Spec .pairup always returns a Seq

Modified test is a 6.d propspec PoV: github.com/rakudo/rakudo/commit/d73272e4c0		 00:22
Zoffix So whose is @perl6ide on Twitter? 00:30
timotimo i don't know
Zoffix Perl 6 IDE.. but which IDE? :)
Geth nqp/master: 5 commits pushed by (Timo Paulssen)++ 00:46
AlexDaniel maybe it's a misspelling of “die” 01:03
Zoffix: there's an issue with 33 modules so far 01:04
almost ready
well, issue*s* I guess 01:05
Zoffix Good :) 01:09
Especially if "so far" means after testing most modules and not like... first 33 modules :P
01:12 leont left
AlexDaniel yeah 01:19
Zoffix: ok here's the quick result: gist.github.com/AlexDaniel/3296f8c...e-bisected 01:29
01:30 cognominal-p6 joined
AlexDaniel can't gist the output yet 01:30
01:30 p6bannerbot sets mode: +v cognominal-p6
AlexDaniel IO::Glob “Too late to switch language version. Must be used as the very first statement.” 01:31
Path::Router “Too late to switch language version. Must be used as the very first statement.”
File::Compare “Too late to switch language version. Must be used as the very first statement.” 01:32
:)
Zoffix Those are easy to fix at least 01:33
AlexDaniel the graph on the bottom is kinda funny
(right click on it, view image)
Zoffix Fancy. It bisects
01:34 cognominal-p6 left
Zoffix And makes graphs :O 01:34
01:34 cognominal-p6 joined
AlexDaniel yeah, so that you can see if a failure in some module propagated to another module that depends on it 01:35
01:35 p6bannerbot sets mode: +v cognominal-p6
AlexDaniel if I just ignore Task::Galaxy for graphs it'll be much cleaner… 01:35
anyway I'll be improving it as we go 01:38
01:51 lizmat joined, p6bannerbot sets mode: +v lizmat
timotimo maybe also ignore JSON::Fast, lol 01:53
01:53 travis-ci joined, p6bannerbot sets mode: +v travis-ci
travis-ci Rakudo build passed. Zoffix Znet 'Consistify :U.pairup with :D one by returning a Seq' 01:53
travis-ci.org/rakudo/rakudo/builds/447571765 github.com/rakudo/rakudo/compare/a...3272e4c0f3
01:53 travis-ci left, cognominal-p6 left
timotimo maybe also uniquify the arrows? looks like Perl6::Tidy depends on Perl6::Parser like 5 times? 01:54
and it somehow also depends on Test::META as often? 01:55
Zoffix AlexDaniel++ # Blin is useful :) 01:56
01:56 Zoffix left
AlexDaniel timotimo: I don't know why :( 01:57
but it's the first time I see that graph, didn't even have a chance to debug
01:57 nebuchadnezzar left
AlexDaniel previous regressions were with modules that had no dependencies 01:57
so graphs with just 1 or 2 nodes were not very exciting
01:58 nebuchadnezzar joined, p6bannerbot sets mode: +v nebuchadnezzar 02:00 Zoffix joined, p6bannerbot sets mode: +v Zoffix 02:01 Zoffix left
timotimo Perl6::Tidy appears multiple times as a node, too 02:01
gray, red, gray, gray, gray
AlexDaniel ooooh it's probably different versions of perl6::tidy 02:02
the nodes on the graph just don't have the version appended
timotimo ah 02:04
yeah
that'd do it
i'll go to bed now :)
Geth roast: 9062bb39fc | (Zoffix Znet)++ | 2 files
[v6.d REVIEW] Appendicize Dateish.IO

As there's no ruling or agreement on how it's meant to work R#2105 github.com/rakudo/rakudo/issues/2105		 02:06
synopsebot R#2105 [open]: github.com/rakudo/rakudo/issues/2105 [6.d review][@LARRY] DateTime.IO produces forbidden path on Windows
02:18 ggoebel joined 02:19 p6bannerbot sets mode: +v ggoebel
Geth roast: 9f35dc7855 | (Zoffix Znet)++ | S02-types/nil.t
[v6.d REVIEW] Precisify Nil.ords test
 02:24
6.d-prep: 6413b3c7cf | (Zoffix Znet)++ | ChangeLog.md
[CL] Log up to 2a3cdb2e257e97d8c6dafef
 02:26
japhb .ask Zoffix You explained SAP earlier; what is PoV? 02:34
yoleaux japhb: I'll pass your message to Zoffix.
japhb .tell Zoffix Nevermind, found it. 02:35
yoleaux japhb: I'll pass your message to Zoffix.
02:43 nebuchadnezzar left
Geth rakudo: jmaslak++ created pull request #2443:
Add type constraints to port numbers in IO::Socket::Async		 02:49
03:17 MasterDuke left 03:33 ufobat_ joined 03:34 p6bannerbot sets mode: +v ufobat_ 03:36 ufobat left 03:43 fake_space_whale left 04:39 AlexDaniel left 04:40 AlexDaniel joined, p6bannerbot sets mode: +v AlexDaniel 04:50 cognominal-p6 joined, cognominal-p6 left, cognominal-p6 joined 04:51 p6bannerbot sets mode: +v cognominal-p6 04:55 cog joined 04:56 p6bannerbot sets mode: +v cog 04:57 cognominal-p6 left 05:13 cog left 06:32 stmuk joined, p6bannerbot sets mode: +v stmuk 06:34 stmuk_ left 08:09 cog joined 08:10 p6bannerbot sets mode: +v cog 08:23 cog left 08:24 cog joined 08:25 p6bannerbot sets mode: +v cog 08:28 nebuchadnezzar joined, p6bannerbot sets mode: +v nebuchadnezzar 08:33 cog left 08:43 cog joined 08:44 p6bannerbot sets mode: +v cog 08:45 cog left, cog joined 08:46 p6bannerbot sets mode: +v cog 09:02 Ven` joined 09:03 p6bannerbot sets mode: +v Ven`
lizmat Files=1257, Tests=76411, 335 wallclock secs (15.61 usr 5.19 sys + 2366.98 cusr 225.22 csys = 2613.00 CPU) 09:05
Geth rakudo: 3a412be2ee | (Zoffix Znet)++ | t/spectest.data
Remove fully skipped test file
 09:08
roast: 1a88ef7e03 | (Zoffix Znet)++ | S16-io/quoting-syntax.t
Delete `:p` quoter skipped tests

During IO::Grant it was decided we aint making `:p` quoter because it just adds yet another variant to the smörgåsbord of quoters without much benefit, since `.IO` is equally short to type already.		 09:10
09:38 pmurias joined, p6bannerbot sets mode: +v pmurias
pmurias jnthn: we have a designish problem 09:39
when we have a y(my int $z = 42) a nqp::assign_i is emitted which produces an int and *not* a lexicalref 09:40
Geth roast: 0488159e5e | (Zoffix Znet)++ | docs/release-guide.md
Include notes on upgrade guide
 09:48
10:04 cog left 10:07 robertle joined 10:08 p6bannerbot sets mode: +v robertle 10:11 pmurias left 10:17 pmurias joined, p6bannerbot sets mode: +v pmurias 10:25 Tux__ is now known as |Tux| 10:45 Zoffix joined, p6bannerbot sets mode: +v Zoffix
Zoffix .ask lizmat don't know how to changelog set ops changes in 6.d. Is simply saying "Major clarifications and redesign of all set operators" too broad? Are you able to commit a more detailed description? github.com/perl6/6.d-prep/blob/mas...angeLog.md 10:47
yoleaux Zoffix: I'll pass your message to lizmat.
02:34Z <japhb> Zoffix: You explained SAP earlier; what is PoV?
02:35Z <japhb> Zoffix: Nevermind, found it.
Zoffix All the terms are in the glossary docs.perl6.org/language/glossary 10:48
Geth 6.d-prep: 5e0bf8c6ba | (Zoffix Znet)++ | ChangeLog.md
[CL] Log up to 21fc586d7172eb36d645e58c610d44dafa952a2d
lizmat .
yoleaux 10:47Z <Zoffix> lizmat: don't know how to changelog set ops changes in 6.d. Is simply saying "Major clarifications and redesign of all set operators" too broad? Are you able to commit a more detailed description? github.com/perl6/6.d-prep/blob/mas...angeLog.md
lizmat Zoffix: will think about that after having done the pretty ginormous P6W 10:49
Zoffix .tell samcv would you be able to changelog log 6.d Unicode-related changes and collation stuff? It's over my head and I don't know how to document it all properly: github.com/perl6/6.d-prep/blob/mas...angeLog.md
yoleaux Zoffix: I'll pass your message to samcv.
Zoffix A ginormous P6W \o/
lizmat++
10:59 pmurias left 11:04 pmurias joined, p6bannerbot sets mode: +v pmurias 11:16 Geth left, undersightable6 left, undersightable6 joined, Geth joined, ChanServ sets mode: +v undersightable6, ChanServ sets mode: +v Geth 11:17 SourceBaby left, hoelzro left, Zoffix left, p6bannerbot sets mode: +v undersightable6, p6bannerbot sets mode: +v Geth, geekosaur left 11:18 hoelzro joined 11:19 p6bannerbot sets mode: +v hoelzro 11:22 pmurias left 11:26 pmurias joined, p6bannerbot sets mode: +v pmurias 11:27 pmurias left 11:29 AlexDaniel left, AlexDaniel joined, p6bannerbot sets mode: +v AlexDaniel, geekosaur joined 11:30 p6bannerbot sets mode: +v geekosaur 11:35 Ven` left 12:15 leont joined 12:16 p6bannerbot sets mode: +v leont 12:17 pmurias joined, p6bannerbot sets mode: +v pmurias 12:18 Ven` joined, p6bannerbot sets mode: +v Ven` 12:25 pmurias left 12:29 leont left 12:35 pmurias joined, p6bannerbot sets mode: +v pmurias 13:00 cog joined 13:01 p6bannerbot sets mode: +v cog
|Tux| Still slow :( 13:07
Rakudo version 2018.10-21-g3a412be2e - MoarVM version 2018.10-25-gfb57d40b8
csv-ip5xs1.027 - 1.061
csv-ip5xs-207.459 - 7.735
csv-parser20.304 - 20.898
csv-test-xs-200.426 - 0.434
test8.200 - 8.329
test-t1.837 - 1.840
test-t --race0.906 - 0.955
test-t-2029.417 - 30.079
test-t-20 --race10.052 - 10.181
lizmat wonders if El_Che is already working on 2018.10 packages 13:09
13:11 Zoffix joined, p6bannerbot sets mode: +v Zoffix
Zoffix |Tux|: what was the "fast" time? 13:11
|Tux| 1.667 13:12
Zoffix m: dd 1.837-1.667
camelia 0.17
Zoffix Well, we did de-opt $*PERL stuff and we now always load 6.d core, so that's probably where the lowness comes from 13:13
|Tux| 20181011 1.680❙20181001 1.674❙20180928 1.662❙20181015 1.659❙20181011 1.647
Zoffix R#2433 for $*PERL 13:14
synopsebot R#2433 [open]: github.com/rakudo/rakudo/issues/2433 $*PERL.version is wrong
Zoffix c: 46ef0ea08c,46ef0ea08c^1 say now - INIT now
committable6 Zoffix, ¦46ef0ea: «0.001359␤» ¦46ef0ea08c^1: «0.0013939␤»
Zoffix c: 46ef0ea08c,46ef0ea08c^1 say now - BEGIN now 13:15
committable6 Zoffix, ¦46ef0ea: «0.00735␤» ¦46ef0ea08c^1: «0.0078306␤»
Zoffix There goes that theory, I guess
Still slower, but not .17s slower 13:16
|Tux| 1.647 was the fastest run on 2018-10-11 13:20
Zoffix Don't really see a difinitive jump from faster to slower: gist.github.com/zoffixznet/9b77253...f2377990e3 13:23
Just on the 12th, 2 days after faster time we have a 1.818 result 13:24
13:24 cog left 13:27 pmurias left 13:48 SqrtNegInf joined 13:49 p6bannerbot sets mode: +v SqrtNegInf
dogbert2_ if the difference > 10% benchable should be able to find the problem 13:56
Geth specs: f424643a5b | (Elizabeth Mattijsen)++ | 9 files
Update email address
 14:14
14:35 |Tux| left, [TuxCM] joined 14:36 p6bannerbot sets mode: +v [TuxCM]
Zoffix Raku logo: temp.perl6.party/raku.jpg 14:39
14:41 |Tux| joined 14:42 p6bannerbot sets mode: +v |Tux|
SmokeMachine Zoffix: I liked it! it has fun with it! :) 14:43
s/fun/ofun/
lizmat timotimo: this is the repo for the profiler UI work, right: github.com/timo/moarperf 14:45
timotimo that is correct 14:46
i should really change the readme, lol
gotta go AFK for a while first, though 14:47
maybe i can do it on the go, though
how long do i have, approximately, before the post lands? 14:48
Zoffix: do the red - and horn symbolize something particular? 14:49
AlexDaniel Zoffix: I like it 14:51
lizmat timotimo: I was planning on publishing around 7pm tonight
Zoffix timotimo: it's a flame/thumbs up, not horn. Red symbolize the hotness
And fastness 14:52
:)
tyil I also saw a little horn in it :p
but it looks p good
8/8
lizmat Zoffix: fwiw, a little horn was also my first interpretation
Zoffix lol 14:53
lizmat if you want to go for flame, I guess it needs to be less uniform and have more yellow in it
and potentially larger / less smooth 14:54
also: what is the significance of the horizontal bar in the R ?
AlexDaniel ah yeah I didn't think of it as flame also, but it didn't jump at me
14:55 patrickb joined, p6bannerbot sets mode: +v patrickb
Zoffix lizmat: it's a database icon and R makes a vice: for all the data crunching. And there are two blank stripes for this being the 2nd name for the language 14:56
AlexDaniel notable6: weekly 15:03
notable6 AlexDaniel, 23 notes: gist.github.com/dc93f1ba2ad9632306...596c0e7fbe
Zoffix lizmat: but don't put the logo into weekly :) I was just being bored :) 15:04
15:04 pmurias joined, p6bannerbot sets mode: +v pmurias
lizmat Zoffix: I most definitely won't :-) 15:07
15:08 fake_space_whale joined
AlexDaniel weekly: last two weeks separately: gist.github.com/738c6f6056072ac702...b2c1713c6b gist.github.com/be13c1b83da2182204...36a8b545c1 15:08
notable6 AlexDaniel, Noted!
AlexDaniel do people even like these reports? I never got any feedback :D
15:08 p6bannerbot sets mode: +v fake_space_whale
lizmat AlexDaniel: if you do thing because of feedback you expect, then nothing will happen 15:09
Zoffix I glance at them once in a while, yeah
lizmat *things
it's sad, but it's true
15:14 Zoffix left 15:15 evalable6 left, evalable6 joined, ChanServ sets mode: +v evalable6
dogbert2_ AlexDaniel: did you see the discussion on [Tux] test above? 15:15
15:16 p6bannerbot sets mode: +v evalable6, releasable6 left, releasable6 joined, ChanServ sets mode: +v releasable6, Ven` left 15:17 p6bannerbot sets mode: +v releasable6 15:18 cog joined 15:19 p6bannerbot sets mode: +v cog 15:20 leont joined 15:21 p6bannerbot sets mode: +v leont
japhb Zoffix: I liked the treatment of the butterfly very much, both directly visually and as a reference to Rorshach blots, because I definitely feel like Perl 6 is a language where everyone can see their own thing in it. The flame did look like a horn to me, and so I associated that with the red in Optimized for Fun, so figured there was a "go wild" aspect there. Definitely like having Optimized for Fun 15:23
right there in the logo. The database crunching vise thing wasn't obvious to me, though I've seen that metaphor before elsewhere (especially for compression programs). Overall a really good starting point I think.
15:24 cog left, cog joined 15:25 p6bannerbot sets mode: +v cog
pyrimidine I like it! Maybe make the bottom part of the 'R' the handle of the vice? 15:28
AlexDaniel lizmat: you misunderstood me a bit, I think. It's open-source, I get it. I don't mind doing it, but question is if I should be doing it at all 15:29
lizmat AlexDaniel: I think it's a nice addition... and it doesn't cost much effort, right ? 15:30
AlexDaniel yes
lizmat then let's keep it for now :-)
AlexDaniel there was a person working on automating it, by the way… and then they disappeared
15:31 pmurias left
lizmat yeah, it *is* open source, after all 15:31
AlexDaniel last words were something like “I'm not committing yet, wanna add something. You will love it!”
haven't seen that person around since then :D
dogbert2_: I did see the discussion, but generally IMO for that benchmark we need proper data on every single commit 15:32
dogbert2_: here's how you can run something on a random rakudo version without compiling it: gist.github.com/AlexDaniel/fb18003...926f706ba0 15:33
as long as you're on linux and stuff, yeah, but still
and I definitely can't come up with a reason why we can't run a test that takes less than 2 seconds on every single commit 15:34
multiple times even, to denoise it
of course if you give me a golf I can bisect it 15:36
15:37 pmurias joined, p6bannerbot sets mode: +v pmurias
AlexDaniel but likely with bisectable6, not benchable 15:37
timotimo lizmat, cool, ill be able to write a better redme until then 15:39
16:08 patrickb left 16:10 pmurias left 16:16 pmurias joined, p6bannerbot sets mode: +v pmurias 16:25 fake_space_whale left
Geth rakudo: 92aa4a52c3 | (Zoffix Znet)++ | t/spectest.data
Remove S06-signature/caller-param.t

It's going out of 6.d spec into appendixes		 16:39
16:42 cog left
Geth roast: 7780ce3980 | (Zoffix Znet)++ | 4 files
Move $CALLER::_ to SAP

It's being deprecated in 6.d		 16:44
lizmat notable6: weekly 16:50
notable6 lizmat, 24 notes: gist.github.com/bfe4e801efa4a4967d...521610d40a
17:08 stmuk_ joined, p6bannerbot sets mode: +v stmuk_ 17:10 stmuk left 17:26 ExtraCrispy left 17:27 ExtraCrispy joined, p6bannerbot sets mode: +v ExtraCrispy 17:50 cog joined 17:51 p6bannerbot sets mode: +v cog 18:03 robertle left 18:18 cog left, cog joined
Geth rakudo: 8ea13823c4 | (Zoffix Znet)++ | 3 files
Add missing plans to `subtest`s
 18:19
18:19 p6bannerbot sets mode: +v cog 18:27 cog left
Geth star/2018.10: 2a2661f278 | (Steve Mynott)++ | 3 files
bump versions
 18:46
star/2018.10: ab47dde7e4 | (Steve Mynott)++ | 11 files
bump modules
star/2018.10: f27a5282c1 | (Steve Mynott)++ | 3 files
remove dead patches
18:47 ExtraCrispy left 18:51 patrickb joined 18:52 p6bannerbot sets mode: +v patrickb 19:12 AlexDaniel left, AlexDaniel joined, p6bannerbot sets mode: +v AlexDaniel
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/10/29/...proaching/ 19:22
notable6: weekly reset 19:27
notable6 lizmat, Moved existing notes to “weekly_2018-10-29T19:27:23Z”
20:09 cog joined, p6bannerbot sets mode: +v cog 20:34 cog left 20:56 cog joined, p6bannerbot sets mode: +v cog 21:03 stmuk joined 21:04 p6bannerbot sets mode: +v stmuk 21:05 stmuk_ left 21:07 cog left 21:08 cog joined, p6bannerbot sets mode: +v cog
Geth rakudo: 3995e42354 | (Elizabeth Mattijsen)++ | 4 files
Always conditionally add objects to the serialization context

So we don't run the risk of having another case of R#2400.		 21:12
synopsebot R#2400 [closed]: github.com/rakudo/rakudo/issues/2400 [fudged tests committed][regression][⚠ blocker ⚠] Module Crane is failing tests on HEAD
21:19 stmuk_ joined 21:20 p6bannerbot sets mode: +v stmuk_, pmurias left 21:21 stmuk left 21:26 pmurias joined, p6bannerbot sets mode: +v pmurias 22:00 lizmat left 22:13 [TuxCM] left 22:34 pmurias left 22:39 [Coke]_ is now known as [Coke]
[Coke] docs folk, testing a change to xt/examples-compilation that lets you mark a test as requiring its own file for testing; so the majority go through the EVAL path we have, but you can mark problematic ones like those containing "unit" or "is export" as requiring a separate file to run them in. (lets me get xtest passing again) 22:41
... EWRONGCHANNEL 22:43
23:10 lizmat joined, p6bannerbot sets mode: +v lizmat 23:16 patrickb left

Powered by Cro and the Raku® Programming Language.

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

Thank you!