Geth rakudo: 53130587f7 | (Nick Logan)++ (committed using GitHub Web editor) | t/02-rakudo/04-diag.t
Allow test to find libs before a 'make install'
02:10
rakudo: 1aa83a789c | (Nick Logan)++ | 6 files
Remove tests for no-longer-core module
03:28
rakudo: b2725c1228 | (Nick Logan)++ (committed using GitHub Web editor) | t/03-jvm/01-interop.t
skip jvm test unless running on jvm
03:54
rakudo: 4849c73355 | usev6++ | src/core/Kernel.pm
Change lookup of OS name on JVM backend

nqp::jvmgetproperties is where we have that information on JVM. We use the same differentiation in src/core/VM.pm.
Unbusts the JVM build.
06:59
[Tux] Rakudo version 2017.10-173-g4849c7335 - MoarVM version 2017.10-58-gad8618468
csv-ip5xs1.100 - 1.105
csv-ip5xs-2012.719 - 13.048
csv-parser12.287 - 12.385
csv-test-xs-200.438 - 0.438
test12.200 - 12.448
test-t3.236 - 3.276
test-t --race1.306 - 1.388
test-t-2059.942 - 60.079
test-t-20 --race21.049 - 23.185
08:05
lizmat Files=1229, Tests=75812, 319 wallclock secs (14.95 usr 5.37 sys + 2195.32 cusr 215.97 csys = 2431.61 CPU) 10:43
Geth rakudo: 3590beaa99 | (Elizabeth Mattijsen)++ | src/core/Kernel.pm
Remove Kernel.memory

Apparently the ru_maxrss element in nqp::getrusage is 0 on Windows, so this is nowhere near a general solution to providing memory size info. So remove it for now until we can come up with something better that also works on Windows.
11:00
rakudo: 5f4b61b120 | (Elizabeth Mattijsen)++ | lib/Telemetry.pm6
JVM-proof Telemetry

We have already abstracted getting the osname in VM, duh!
11:01
Zoffix lizmat: would you mind reminding people to sign up for Advent Posts in the Weekly. We still have lots of space open. Maybe prod the beginners who are too shy too by saying something like all skill-levels welcome; write about something cool you learned in Perl 6 recently. The schedule is at github.com/perl6/mu/blob/master/mi...7/schedule and original blog post is at 11:41
rakudo.party/post/Rakudo-Perl-6-Ad...or-Authors
lizmat Zoffix: will do 11:44
Zoffix lizmat++ 11:45
Geth rakudo: 68b2891d96 | (Elizabeth Mattijsen)++ | tools/build/moar_core_sources
Move atomicops before Thread

So we can use atomic ops inside Thread for providing telemetry data.
12:16
rakudo: 77048b6acb | (Elizabeth Mattijsen)++ | src/core/Thread.pm
Add some telemetry to Thread

  - started, aborted, completed, joined, yields
  - class method "usage" returns a struct similar to nqp::getrusage
12:42
rakudo: cb4d8b6608 | (Elizabeth Mattijsen)++ | src/core/Thread.pm
Add highest OS thread ID to telemetry
13:47
rakudo: e5912337e7 | (Elizabeth Mattijsen)++ | lib/Telemetry.pm6
Add Telemetry::Instrument::Thread

To allow for reporting Thread.usage data within the Telemetry framework.
13:57
Zoffix "Introspection into objects and meta-objects (turtles all the way down)" 14:04
turtles? :)
My google-fu gives www.amazon.com/Turtles-All-Down-Jo...0525555366 "It’s the most powerful and open book about mental illness that I have read, and it’s required reading for everyone" 14:05
Found it in the brochure github.com/tpf/marketing-materials...60page.pdf 14:06
(page 8) 14:07
lizmat Zoffix: I think this is the source of that meme? en.wikipedia.org/wiki/Discworld_(world)
Zoffix Ah: en.wikipedia.org/wiki/Turtles_all_the_way_down 14:08
Geth rakudo: de961b8146 | (Elizabeth Mattijsen)++ | lib/Telemetry.pm6
Oops, T:I:Usage.cpus didn't work in reports

  - because it was missing from T:I:Usage.formats
  - also slightly speed up calculation
14:24
rakudo: 246f410134 | (Elizabeth Mattijsen)++ | t/06-telemetry/02-usage.t
Add basic Telemetry::Instrument::Usage testing
14:51
rakudo: 06cbe9c29c | (Elizabeth Mattijsen)++ | t/06-telemetry/01-basic.t
Add editing footer
lizmat afk& 15:26
Zoffix .ask can you run the first draft of the brochure remake by wolfy? It doesn't have to become the new brochure, so if you hate it, just say no. I'll also DM it to Mark Keating on Twitter. Link: github.com/perl6/marketing/raw/mas...ochure.pdf 16:57
yoleaux Zoffix: I'll pass your message to can.
Zoffix is & for ~6h 17:04
[Coke] Zoffix: '.ask can' ? 17:08
Zoffix .ask lizmat can you run the first draft of the brochure remake by wolfy? It doesn't have to become the new brochure, so if you hate it, just say no. I'll also DM it to Mark Keating on Twitter. Link: github.com/perl6/marketing/raw/mas...ochure.pdf 17:17
yoleaux Zoffix: I'll pass your message to lizmat.
Zoffix Mark said it looks OK and he's fine with splitting off the book excerpts into a separate booklet (listing another benefit of doing so: ease of translation) 18:21
lizmat . 18:25
yoleaux 17:17Z <Zoffix> lizmat: can you run the first draft of the brochure remake by wolfy? It doesn't have to become the new brochure, so if you hate it, just say no. I'll also DM it to Mark Keating on Twitter. Link: github.com/perl6/marketing/raw/mas...ochure.pdf
Zoffix I haven't actually read any of the text from the original brochure... I notice some things still refer to content/structure of old brochure :P 18:37
change intro para to read this instead (removed all the layout/books mentions) i.imgur.com/RPGtiNr.png 18:41
*changed
lizmat Zoffix: there seems room in "Built-in types and operators for operations with Sets, Bags, and Mixes" to mention examples, such as ∈ (is element of) and ∩ (intersection) 18:47
Zoffix: s/elminating/eliminating/ 18:49
[Coke] laminating 18:51
lizmat Zoffix++ # way cool :-) 18:55
Zoffix lizmat: added, though I used ASCII ops: i.imgur.com/iNXP7in.png a bit worried of using Unicode ops in there and distracting with "how the hell do I type that" thoughts 18:56
lizmat Fine by me :-)
s/such/such as/ ?
Zoffix Ah right. Fixed 18:57
Also, I didn't add periods at the end of bulleted items on purpose :)
lizmat yeah, I saw what you did (not) do there :-)
cool... I guess we will try to find some printer that can print use a few copies for this weekend's T-Dose 18:58
Zoffix OK. I'll wait a bit for more typos to be spotted (and look over it myself some more) and then generate the print-quality PDF then 18:59
lizmat Zoffix++ 19:00
so, if I read correctly: news.perlfoundation.org/2017/11/cal...er-20.html 19:03
there is not going to be a November 2017 round, right ? 19:04
[Coke]: ^^^ ?
[Coke] Correct.
Hopefully we can start back up in January
lizmat then Perl Weekly didn't get the message: twitter.com/PerlWeekly/status/9298...1923024897
[Coke] lizmat: replied. thanks. 19:08
lizmat wonders whether [Coke] needs a nudge wrt to posting a grant report 19:56
sorry [Coke]: I just hear it was someone else who needed nudging first :-) 19:57
ufobat why do some parts in rakudo use (a regular) if where other parts use nqp::if? 20:04
moritz mostly a question of performance 20:08
Perl 6 "if" coerces the condition to a Bool
ufobat ty :) 20:11
Zoffix huh? What do you mean? Doesn't `if ... {}` compile to nqp::if? 20:19
jnthn Yes. The performance differences are more in the block/introduction of an extra scope 20:20
Zoffix Yeah. Also, in some places an `nqp::if` instead of an `if {}` is used likely 'cause there's other nqp code used. It keeps it consistent and looks prettier to have just pure NQP nest 20:21
m: for ^1000_000 { if rand { $ = $ } }; say now - INIT now
camelia 0.67543181
Zoffix m: for ^1000_000 { $ = $ if rand }; say now - INIT now
camelia 0.285267
timotimo m: use nqp; for ^1000_000 { nqp::if(rand, $ = $) }; say now - INIT now 20:22
camelia 0.2551589
Zoffix m: use nqp; nqp::if(class {method Bool { say "booliefied" } }, nqp::null) 20:24
camelia booliefied
lizmat m: use nqp; nqp::if(rand, $ = $) for ^1000_000; say now - INIT now
camelia 0.30504163
lizmat hmmm...
m: use nqp; nqp::if(rand, $ = $) for ^1000_000; say now - INIT now 20:25
camelia 0.22861110
Geth nqp: 2e12d4750b | (Zoffix Znet)++ | tools/build/MOAR_REVISION
Bump MoarVM
20:30
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...0-gb379050
rakudo: b2fbf893db | (Zoffix Znet)++ | tools/build/NQP_REVISION
Bump NQP
Zoffix ZOFVM: Files=1283, Tests=152837, 148 wallclock secs (20.85 usr 2.96 sys + 3195.93 cusr 160.20 csys = 3379.94 CPU)
Also, a ton of passing TODOs in sprintf tests: gist.github.com/zoffixznet/b7edfc2...dd5c45351f 20:31
Geth rakudo: 88a9cb259e | (Elizabeth Mattijsen)++ | 2 files
'make test' should also do Telemetry tests
21:09
rakudo: 88d6751664 | (Elizabeth Mattijsen)++ | src/Perl6/Grammar.nqp
Fix typo in error message
21:10
lizmat nudges [Coke] again 21:31
Geth rakudo: ufobat++ created pull request #1249:
RFC && 3rd attempt: Fixed resolve of IO::Path for Windows
22:01
synopsebot RAKUDO#1249 [open]: github.com/rakudo/rakudo/pull/1249 RFC && 3rd attempt: Fixed resolve of IO::Path for Windows
ufobat !! Zoffix ;-)
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2017/11/13/...explained/ 22:11
teatime yay
Zoffix \o/ 22:12
teatime "concatenation of strings from O² to O¹" interesting notation 22:13
does that mean the same thing as O(n²) to O(1) ?
or O(n) maybe
lizmat teatime: changed :-) 22:15
timotimo lizmat: last paragraph has "rakudo" rather than "Rakudo" 22:16
teatime looks the same here. did you markdown converter eat parts of it?
s/markdown/markup/ 22:17
lizmat timotimo: argh, yeah, had seen that, forgotten to fix: now fixed # timotimo++
timotimo :) 22:18
thundergnat .tell Zoffix Typo in brochure, page 9 under "Perl 6 is a brand new language" - "such us Python, C, and Perl 5", should be as 22:25
yoleaux thundergnat: I'll pass your message to Zoffix.
Zoffix lizmat++ # good weekly 22:27
yoleaux 22:25Z <thundergnat> Zoffix: Typo in brochure, page 9 under "Perl 6 is a brand new language" - "such us Python, C, and Perl 5", should be as
timotimo yes, good weekly! 22:28
Zoffix I've just deleted the community tweet tho 22:29
thundergnat Zoffix Also, bikeshedding a bit, "end up as much as twice shorter" reads oddly to me, perhaps "end up substantially shorter" or "need substantially less code" Otherwise Zoffix++!
Zoffix It was just a nasty experience and don't want even more people telling me crap about it
lizmat Zoffix: understandable 22:30
anyways, Tommy Stanton's tweet stands well on its own
timotimo where is that tweet? i can't find it 22:32
oh, lizmat, the "meanwhile on twitter" heading is part of the previous link
oh, oops
i was looking at last week's post
lizmat ? 22:33
ah
Geth rakudo: d6a3a7a1aa | (Elizabeth Mattijsen)++ | 2 files
Add basic testing for T:I:Thread/ThreadPool
22:36
Zoffix thundergnat: thanks. Typo fixed in sources; will update repo later. For bikeshedding: "substantially shorter" is too vague a quantity. "Half the size" is a measure I heard several times and matches my most scientific measurement of converting my own code. 22:37
thundergnat Zoffix Yeah, substantially does sound kind-of like a weasel word. "Half the size" reads much better than "twice shorter" 22:40
Zoffix thundergnat: changed it to: "Many programmers report their programs become up to half the size, when converted from other languages to Perl 6, without sacrificing legibility." 22:42
thundergnat Zoffix++ 22:43
Zoffix ufobat: won't have time to review it today, but briefly looking at the code, I wonder two things: (1) why is the volume being checked on every loop iteration; shouldn't it be just set by shifting from the list (while checking there's stuff to shift) before the loop? Also: what happens with the $!CWD attribute of the resultant path? Looks like here it's just being set to `$sep`, which would end up as just 22:45
`\` without a volume on Windows, won't it? github.com/rakudo/rakudo/pull/1249...4ea969R332
Zoffix calls it a night 22:46
lizmat good night, Zoffix!
ufobat Zoffix, n8 :-) 22:47
Zoffix ufobat: P.S.: I said "shift" above, but I didn't read the code enough to know if it's shift that's needed or just an assignment of the first element. 22:54
So check which one is right :) 22:55
wander do we ought to add `use v6;` in every .t file of roast? is it for historical reason or for explicit? 22:57
lizmat also calls it a night 23:02
timotimo nite lizmat
bloatable6: releases 23:04
bloatable6 timotimo, gist.github.com/aead9fe84f8cd0a364...6d9ee9c0bb 23:05
timotimo statisfiable6: core0 23:08
statisfiable6 timotimo, OK! Working on it…
timotimo, gist.github.com/321b00595afea91003...6724b8a8f5 23:09
timotimo i wonder if we'll come up with something good to make our core setting actually smaller 23:10
the core setting has hardly any regexes in it, so getting NFA to be represented better won't do much for that 23:11
AlexDaniel oh statisfiable6, what a good boy :)
(or girl?) 23:12
AlexDaniel is happy that rarely used bots still work as expected
statisfiable6: libmoar0 23:13
statisfiable6 AlexDaniel, OK! Working on it…
AlexDaniel timotimo: but what is this sudden bump? Seems to be github.com/rakudo/rakudo/commit/98...a0c7faa6bc 23:15
timotimo yeah, that makes sense 23:16
oh my
AlexDaniel statisfiable6: libmoar0
statisfiable6 AlexDaniel, OK! Working on it…
AlexDaniel, gist.github.com/bd8e0c7c2e87447669...ad43f06297 23:17
AlexDaniel one of them is github.com/rakudo/rakudo/commit/64...9ace551468 23:18
then there's also 23:19
Geth: ver github.com/rakudo/rakudo/commit/27...df27d49f1e
Geth AlexDaniel, version bump brought in these changes: github.com/perl6/nqp/compare/2017....0-ga6a1aa0
AlexDaniel Geth: ver github.com/perl6/nqp/commit/a6a1aa...eee98b028a 23:20
Geth AlexDaniel, version bump brought in these changes: github.com/MoarVM/MoarVM/compare/2...3-ga4fef0b
AlexDaniel so unicode 10 and JIT
timotimo that's not so surprising 23:22
Geth roast: 8182813960 | (Elizabeth Mattijsen)++ | S32-str/sprintf-b.t
Unfudge now passing tests
23:25
lizmat really bed& 23:26
AlexDaniel statisfiable6: install0
statisfiable6 AlexDaniel, OK! Working on it…
AlexDaniel statisfiable6: install0 23:27
statisfiable6 AlexDaniel, OK! Working on it…
AlexDaniel, gist.github.com/170c5a3304dc9eca48...74b7b3366d 23:28
AlexDaniel (it caches the numbers, so there are less problems if you ask for these graphs more often…)
jnthn lizmat++ # weekly 23:48