Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6-dev, 2017-09-30

| Channels | #perl6-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:01 Zoffix Filed as https://rt.perl.org/Ticket/Display.html?id=132186
00:12 BenGoldberg m: my $a = :foo<bar>; my $b = :foo<bar>; .say for $a.WHERE, $b.WHERE;
00:12 camelia rakudo-moar b9c985: OUTPUT: «140432301836296␤140432301836440␤»
00:15 BenGoldberg m: :foo<bar>.WHERE.say for 1..2
00:15 camelia rakudo-moar b9c985: OUTPUT: «140243756639768␤140243756649696␤»
00:15 BenGoldberg m: 42.WHERE.say for 1..2
00:15 camelia rakudo-moar b9c985: OUTPUT: «140332874332240␤140332874332240␤»
00:15 evalable6 joined #perl6-dev
00:16 BenGoldberg m: "perl".WHERE.say for 1..2
00:16 camelia rakudo-moar b9c985: OUTPUT: «140147600526720␤140147600526720␤»
00:16 BenGoldberg m: pi.WHERE.say for 1..2
00:16 camelia rakudo-moar b9c985: OUTPUT: «47915808␤47915808␤»
00:16 * BenGoldberg blinks.
00:17 BenGoldberg Those don't looks like addresses.
00:19 Zoffix Man, I miss Geth
00:22 * BenGoldberg was looking through the irc log, read lizmat's two commits, one changing QAST::SVal.new( :value<new> ) expressions with $new variables initialized outside the loop (to reduce memory consumption), then a second where $new variables were replaced with :name<new> expressions, and wondered whether such things as :name<new> were interned somehow, thus saving memory.
00:25 * BenGoldberg wonders how hard it would be for literal pairs to be memoized/interned.
00:26 astj joined #perl6-dev
00:27 Zoffix What does "interned" mean?
00:28 geekosaur cached for quick lookup and memory reuse, basically. comes from lisp
00:29 Zoffix Ah
00:30 geekosaur in this case it would imply that the variable is being passed by name and instead of making a new container for each time it's passed, the existing one is pointed to instead
00:32 P6Bot joined #perl6-dev
00:32 Zoffix P6Bot: hello
00:33 Zoffix P6Bot: hello
00:33 Zoffix something's broken with IRC::Client :\ don't wanna subscribe my plugins :\
00:36 Zoffix and even examples no longer work :/
00:42 MasterDuke hm, AlexDaniel had some trouble with committable6 earlier
00:42 AlexDaniel that's probably not related
00:43 AlexDaniel what I'm seeing is that the bots refuse to be started by systemd for some reason
00:43 AlexDaniel but if I start them from ssh normally, then immediatelly they go up from systemd just fine
00:44 AlexDaniel it's some sort of permission issue and precomp, maybe
00:44 AlexDaniel or something. You see, now $HOME is no longer writable, only .precomp folder is
00:44 AlexDaniel so anything attempting to write a file anywhere is going to fail
00:44 * AlexDaniel actually has no idea what exactly precomp does
00:45 AlexDaniel bisectable6: uptime
00:45 bisectable6 AlexDaniel, 16 hours, 51 minutes, and 29 seconds, 313.628KB maxrss. This is Rakudo version 2017.09-27-gda5c36c13 built on MoarVM version 2017.09.1-7-geeb664ea6 implementing Perl 6.c.
00:46 AlexDaniel ↑ this is also a little bit outdated
00:50 Zoffix AlexDaniel: are *ables run on latest rakudo/IRC::Client?
00:50 Zoffix Having trouble even making this basic bot respond :S https://github.com/zoffixznet/perl6-IRC-Client/blob/master/examples/01-uppercase-bot.p6
00:50 AlexDaniel Identity: IRC::Client:ver<3.007003>
00:51 Zoffix yeah, that's latest :\
00:51 Zoffix wtf
00:51 Zoffix I'm not even seeing PRIVMSG show up :\
00:52 Zoffix This is Rakudo version 2017.09-169-gb9c9853
00:52 AlexDaniel .oO( immediately*… one day I'll learn to spell that right )
01:01 P6Bot joined #perl6-dev
01:01 Zoffix P6Bot: hello
01:01 P6Bot Zoffix, Hello to you too!
01:02 Zoffix well shit, it's possible it's latest rakudos that bustified IRC::Client
01:08 Zoffix nope.
01:08 Zoffix Something with my local box. Still broken on the same rakudo version that I got on perl6.party
01:09 Zoffix m: use nqp; nqp::getcomp('perl6').version
01:09 camelia rakudo-moar b9c985: OUTPUT: «This is Rakudo version 2017.09-169-gb9c98531c built on MoarVM version 2017.09.1-49-gb3dd812aa␤implementing Perl 6.c.␤»
01:09 Zoffix oh, my downgrade didn't downgrade MoarVM
01:09 MasterDuke oh, btw, that io::socket::async IPv6 test fails on my machine also
01:10 Zoffix It no longer fails on my VM, though I didn't even attempt at fixing it
01:11 MasterDuke it's fine on my laptop (mostly unmodified kubuntu 17.04), but fails on my desktop (arch linux with IPv6 disabled)
01:18 MasterDuke .tell eater some comments about the IPv6 io::socket::async test https://irclog.perlgeek.de/perl6-dev/2017-09-30#i_15237813
01:18 yoleaux MasterDuke: I'll pass your message to eater.
01:22 Zoffix Yup. Rakudo
01:23 Zoffix works fine on 2017.09-23-g476741e, broken on 2017.09-169-gb9c98531c
01:24 Zoffix .tell AlexDaniel FWIW basic IRC::Client bot works fine on 2017.09-23-g476741e, but is broken on 2017.09-169-gb9c98531c. Events don't even show up in debug output. No errors and I'm suspecting they get lost in one of the async constructs that don't catch anything: https://github.com/zoffixznet/perl6-IRC-Client/blob/master/examples/01-uppercase-bot.p6
01:24 yoleaux Zoffix: I'll pass your message to AlexDaniel.
01:24 AlexDaniel .
01:24 yoleaux 01:24Z <Zoffix> AlexDaniel: FWIW basic IRC::Client bot works fine on 2017.09-23-g476741e, but is broken on 2017.09-169-gb9c98531c. Events don't even show up in debug output. No errors and I'm suspecting they get lost in one of the async constructs that don't catch anything: https://github.com/zoffixznet/perl6-IRC-Client/blob/master/examples/01-uppercase-bot.p6
01:25 AlexDaniel let's bisect it?
01:25 * Zoffix has no idea how
01:25 AlexDaniel Zoffix: what happens on broken revisions?
01:26 AlexDaniel irc-to-me does not fire?
01:27 Zoffix AlexDaniel: PRIVMSG events don't show up in debug log (and no, they don't fire): broken https://gist.github.com/zoffixznet/38d4fb24ca43a561a0fae115a13f5bbc  working (notice the PRIVMSG event shows up at the bottom): https://gist.github.com/zoffixznet/694aa5a318422f4f85afc96030b6ed42
01:29 AlexDaniel something like this should work https://gist.github.com/AlexDaniel/4c1dbf4de7ceade927d6856c2cfaa693
01:29 AlexDaniel I'll move to #zofbot to try it
01:29 AlexDaniel (yes, you'll have to talk to the bot when he comes up, about 10 times)
01:29 P6Bot joined #perl6-dev
01:29 AlexDaniel but that's ok
01:34 committable6 joined #perl6-dev
01:34 committable6 joined #perl6-dev
01:36 committable6 joined #perl6-dev
01:37 bisectable6 joined #perl6-dev
01:40 bisectable6 joined #perl6-dev
01:55 ilbot3 joined #perl6-dev
01:55 Topic for #perl6-dev is now Perl 6 language and compiler development 2.0 | Logs at http://irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
01:55 Geth_ ¦ synopsebot: 39f0c93f46 | (Zoffix Znet)++ | bin/synopsebot.p6
01:55 Geth_ ¦ synopsebot: Add #zofbot to channels
01:55 Geth_ ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/39f0c93f46
01:56 synopsebot joined #perl6-dev
01:56 Zoffix blah blah blah RT#132186 #131417
01:56 synopsebot6 Link:  https://rt.perl.org/rt3/Public/Bug/Display.html?id=132186
01:56 synopsebot6 Link:  https://rt.perl.org/rt3/Public/Bug/Display.html?id=131417
01:56 synopsebot RT#132186 [new]: https://rt.perl.org/Ticket/Display.html?id=132186 last updated Fri, 29 Sep 2017 17:01:41 -0700: [REGEX][PRECOMP] "getlex: outer index out of range" with precomped `constant` Regex
01:56 synopsebot6 Link:  https://rt.perl.org/rt3/Public/Bug/Display.html?id=132186
01:56 synopsebot RT#131417 [rejected]: https://rt.perl.org/Ticket/Display.html?id=131417 last updated Fri, 14 Jul 2017 07:29:49 -0700: [BUG] Incorrect useless use warning in `so try await`
01:56 synopsebot6 Link:  https://rt.perl.org/rt3/Public/Bug/Display.html?id=131417
01:56 Zoffix synopsebot6: shush
01:57 AlexDaniel Zoffix: fwiw these colors don't look very nice
01:57 Zoffix ZofBot: phhhh! Humans
01:57 ZofBot Zoffix, So cur_frame->args might actually be the better place to look, as ncinvoke won't set up a new call frame and thus not initialize cur_frame->params.args
01:57 Zoffix They look awesome on my terminal
01:58 Zoffix https://i.imgur.com/UPrruty.png
02:01 AlexDaniel https://user-images.githubusercontent.com/5507503/31041347-5c20590e-a59c-11e7-93d0-9a27bd8a5951.png
02:02 AlexDaniel m: error
02:02 camelia rakudo-moar b9c985: OUTPUT: «5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤    error used at line 1␤␤»
02:02 AlexDaniel e: error
02:02 evalable6 AlexDaniel, rakudo-moar b9c98531c: OUTPUT: «(exit code 1) ===SORRY!=== Error while compiling /tmp/qliXfAbLz_␤Undeclared routine:␤    error used at line 1␤»
02:02 Zoffix AlexDaniel: wtf
02:02 Zoffix u:  29
02:02 unicodable6 Zoffix, U+00A0 NO-BREAK SPACE [Zs] ( )
02:02 unicodable6 Zoffix, U+0032 DIGIT TWO [Nd] (2)
02:02 unicodable6 Zoffix, 4 characters in total ( 29 ): https://gist.github.com/126036329ffd0ca24b5f99a499935a5c
02:02 Zoffix Ahhh
02:02 AlexDaniel Zoffix: camelia is using nicer colors too
02:03 AlexDaniel that's OK, nevermind nbsp
02:04 Geth_ ¦ synopsebot: f0d9373c2f | (Zoffix Znet)++ | lib/SB/Plugin/RT.pm6
02:04 Geth_ ¦ synopsebot: Tweakerize
02:04 Geth_ ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/f0d9373c2f
02:04 synopsebot joined #perl6-dev
02:04 Zoffix blah blah blah RT#132186 #131417
02:04 synopsebot RT#132186 [new]: https://rt.perl.org/Ticket/Display.html?id=132186 last updated Fri, 29 Sep 2017 17:01:41 -0700: [REGEX][PRECOMP] "getlex: outer index out of range" with precomped `constant` Regex
02:04 synopsebot RT#131417 [rejected]: https://rt.perl.org/Ticket/Display.html?id=131417 last updated Fri, 14 Jul 2017 07:29:49 -0700: [BUG] Incorrect useless use warning in `so try await`
02:04 Zoffix blah blah blah RT#132186 #131417
02:04 Zoffix blah blah blah RT#132186 #131417
02:05 Zoffix blah blah blah RT#132186 #131417
02:05 Zoffix <3
02:05 MasterDuke nice
02:07 AlexDaniel for some reason blue color looks horrible everywhere. That's a screenshot from Riot: https://user-images.githubusercontent.com/5507503/31041387-31a37926-a59d-11e7-95de-37bc7e5be19b.png
02:08 unicodable6 joined #perl6-dev
02:09 AlexDaniel unicodable, what the crap was that? I appreciate your new ability to reconnect automatically, but can't you just stay online?
02:09 unicodable6 AlexDaniel, U+0077 LATIN SMALL LETTER W [Ll] (w)
02:09 unicodable6 AlexDaniel, U+0068 LATIN SMALL LETTER H [Ll] (h)
02:09 unicodable6 AlexDaniel, 113 characters in total: https://gist.github.com/31142d721cbad7e688b8fc01f2a68f01
02:09 AlexDaniel
02:10 AlexDaniel MasterDuke: I know what's the problem with it!!
02:10 AlexDaniel MasterDuke: there's a .precomp file in lib/ folder
02:10 AlexDaniel so ReadWritePaths=/home/bisectable/git/whateverable/lib/.precomp
02:10 AlexDaniel that's it
02:11 MasterDuke good catch
02:28 Ven`` joined #perl6-dev
02:31 japhdc joined #perl6-dev
02:34 vendethiel- joined #perl6-dev
03:02 Geth_ ¦ synopsebot: b56a5ae89e | (Zoffix Znet)++ | 6 files
03:02 Geth_ ¦ synopsebot: Complete old synbot feature transfer
03:02 Geth_ ¦ synopsebot:
03:02 Geth_ ¦ synopsebot: - Fix Doclinks
03:02 Geth_ ¦ synopsebot: - Impl synopses linker
03:02 Geth_ ¦ synopsebot: - Add uptime/help/bostnack plugin
03:02 Geth_ ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/b56a5ae89e
03:06 Geth_ ¦ synopsebot: f09b220aa6 | (Zoffix Znet)++ | 3 files
03:06 Geth_ ¦ synopsebot: Make file layout closer to old bot
03:06 Geth_ ¦ synopsebot:
03:06 Geth_ ¦ synopsebot: So hopefully I won't have to change the service file
03:06 Geth_ ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/f09b220aa6
03:09 Zoffix rakudobrew@hack:~$ rakudobrew build-zef
03:09 Zoffix Your branch is up-to-date with 'origin/master'.
03:09 Zoffix ===> Testing: zef:ver<0.1.29>:auth<github:ugexe>
03:09 Zoffix MoarVM panic: use of invalid eventloop work item index -1
03:09 Zoffix Failed running /home/rakudobrew/rakudobrew/bin/../moar-nom/install/bin/perl6 -Ilib bin/zef test . at /home/rakudobrew/rakudobrew/bin/rakudobrew line 58.
03:09 Zoffix main::run("/home/rakudobrew/rakudobrew/bin/../moar-nom/install/bin/perl6"...) called at /home/rakudobrew/rakudobrew/bin/rakudobrew line 548
03:09 Zoffix main::build_zef() called at /home/rakudobrew/rakudobrew/bin/rakudobrew line 126
03:09 Zoffix >_<
03:09 Zoffix holy crap... it's stil on 2017.01-206-gaac9efc
03:10 * Zoffix updates
03:12 evalable6 joined #perl6-dev
03:18 japhdc joined #perl6-dev
03:19 japhdc joined #perl6-dev
03:48 synopsebot joined #perl6-dev
03:48 geth ¦ doc: 8f9443c3ac | (Meow)++ (committed using GitHub Web editor) | doc/Language/5to6-perlfunc.pod6
03:48 synopsebot Link: https://doc.perl6.org/language/5to6-perlfunc
03:49 Zoffix S22:30
03:49 synopsebot Link: https://design.perl6.org/S22.html#line_30
03:49 Zoffix RT#132186 #131417 RT132148
03:49 synopsebot RT#132186 [new]: https://rt.perl.org/Ticket/Display.html?id=132186 [REGEX][PRECOMP] "getlex: outer index out of range" with precomped `constant` Regex last updated Fri, 29 Sep 2017 17:01:41 -0700
03:49 synopsebot RT#131417 [rejected]: https://rt.perl.org/Ticket/Display.html?id=131417 [BUG] Incorrect useless use warning in `so try await` last updated Fri, 14 Jul 2017 07:29:49 -0700
03:50 synopsebot RT#132148 [open]: https://rt.perl.org/Ticket/Display.html?id=132148 [RFC] native 'str' type unspecced, undocumented, and ill-defined last updated Thu, 28 Sep 2017 10:56:47 -0700
03:50 Zoffix RT#132186 #131417 RT132148
03:50 * Zoffix meow RT#132015
03:50 synopsebot RT#132015 [new]: https://rt.perl.org/Ticket/Display.html?id=132015 [LTA] Backtraces for errors in EVAL print nonexistent paths (‘foo’.EVAL) last updated Fri, 01 Sep 2017 15:00:44 -0700
03:50 Zoffix Perfect.
03:50 Zoffix Transformation complete.
03:50 Zoffix ZofBot: nice working with you
03:50 ZofBot Zoffix, "is assoc" Specifies the associativity of an operator explicitly
03:52 Geth_ ¦ synopsebot: 244db08dbb | (Zoffix Znet)++ | lib/SB/Plugin/RT.pm6
03:52 Geth_ ¦ synopsebot: Remove "last updated" info from RT tickets
03:52 Geth_ ¦ synopsebot:
03:52 Geth_ ¦ synopsebot: Too much info and too little use for last updated
03:52 Geth_ ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/244db08dbb
03:52 synopsebot joined #perl6-dev
03:53 * Zoffix meow RT#132015
03:53 synopsebot RT#132015 [new]: https://rt.perl.org/Ticket/Display.html?id=132015 [LTA] Backtraces for errors in EVAL print nonexistent paths (‘foo’.EVAL)
03:53 AlexDaniel blah #130442
03:53 synopsebot RT#130442 [new]: https://rt.perl.org/Ticket/Display.html?id=130442 [REGRESSION] [LTA] redo without a loop no longer prints the line number, also claims it is a compile-time error (redo)
03:54 AlexDaniel oh noes, I no longer have a favorite open ticket
04:04 lizmat joined #perl6-dev
04:14 Geth joined #perl6-dev
04:18 nine .tell Zoffix added your user to the camelia VM where geth is running
04:18 yoleaux 29 Sep 2017 21:40Z <AlexDaniel> nine: Geth is down
04:18 yoleaux nine: I'll pass your message to Zoffix.
04:18 yoleaux 29 Sep 2017 22:48Z <Zoffix> nine: would I have the option to getting keys to geth so I could restart it, or pull new changes to it, if needed?
04:38 japhdc joined #perl6-dev
04:42 japhdc joined #perl6-dev
06:21 llfourn joined #perl6-dev
07:53 Geth ¦ rakudo/nom: 415e9adb96 | (Tom Browder)++ | 9 files
07:53 Geth ¦ rakudo/nom: changes for Texas => ASCII
07:53 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/415e9adb96
07:53 Geth ¦ rakudo/nom: 9bbfed162b | lizmat++ (committed using GitHub Web editor) | 9 files
07:53 Geth ¦ rakudo/nom: Merge pull request #1176 from tbrowder/texas
07:53 Geth ¦ rakudo/nom:
07:53 Geth ¦ rakudo/nom: changes for Texas => ASCII
07:53 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/9bbfed162b
08:19 hankache joined #perl6-dev
09:10 [Tux] This is Rakudo version 2017.09-171-g9bbfed162 built on MoarVM version 2017.09.1-50-g3059ba28
09:10 [Tux] csv-ip5xs        1.318 -  1.329
09:10 [Tux] test             9.816 -  9.902
09:10 [Tux] test-t           3.198 -  3.251
09:10 [Tux] csv-parser      12.367 - 12.640
09:17 Ven`` joined #perl6-dev
09:39 evalable6 joined #perl6-dev
09:53 lizmat afk for rest of the day&
09:54 timotimo buggable: speed 10
09:54 buggable timotimo, ▅↑█▆▃▂▁▁▂▁ dates: 2017-09-28–2017-09-30; range: 3.183s–4.010s; speed: 12% faster
10:27 evalable6 joined #perl6-dev
10:49 evalable6 joined #perl6-dev
11:06 TimToady joined #perl6-dev
11:16 Geth_ ¦ synopsebot: 895a6adf03 | (Zoffix Znet)++ | 2 files
11:16 Geth_ ¦ synopsebot: Reword %seen business
11:16 Geth_ ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/895a6adf03
11:17 Geth ¦ synopsebot: 895a6adf03 | (Zoffix Znet)++ | 2 files
11:17 Geth ¦ synopsebot: Reword %seen business
11:17 Geth ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/895a6adf03
11:19 synopsebot joined #perl6-dev
11:27 AlexDaniel squashable6: status
11:28 squashable6 AlexDaniel, Next SQUASHathon in 5 days and ≈22 hours (2017-10-07 UTC-12⌁UTC+14)
11:31 timotimo releasable6: status
11:31 releasable6 timotimo, Next release in 21 days and ≈7 hours. No blockers. Changelog for this release was not started yet
11:31 releasable6 timotimo, Details: https://gist.github.com/46a441880f2737503a886fa4ed11633f
11:38 Geth ¦ roast: tbrowder++ created pull request #338: changes for Texas => ASCII
11:38 Geth ¦ roast: review: https://github.com/perl6/roast/pull/338
11:38 Geth ¦ roast: 19a273a224 | (Tom Browder)++ (committed using GitHub Web editor) | 15 files
11:38 Geth ¦ roast: changes for Texas => ASCII (#338)
11:38 Geth ¦ roast: review: https://github.com/perl6/roast/commit/19a273a224
11:55 evalable6 joined #perl6-dev
11:57 pmurias joined #perl6-dev
11:58 Zoffix This is weird:
11:58 Zoffix m: with (Seq.new: class :: does Iterator { method pull-one { say "pull"; $++ < 3 ?? $++ !! IterationEnd } }.new,).map(|*).skip { .sink; .say }
11:58 camelia rakudo-moar 9bbfed: OUTPUT: «pull␤This Seq has already been iterated, and its values consumed␤(you might solve this by adding .cache on usages of the Seq, or␤by assigning the Seq into an array)␤  in block <unit> at <tmp> line 1␤␤pull␤pull␤pull␤»
11:58 Zoffix How come the remaining 3 pulls are after the error?
12:03 Zoffix m: with (Seq.new: class :: does Iterator { method pull-one { say "pull"; $++ < 3 ?? $++ !! IterationEnd } }.new,).map(|*) { .iterator.sink-all; .perl.say  }
12:03 camelia rakudo-moar 9bbfed: OUTPUT: «pull␤pull␤pull␤pull␤Seq.new-consumed()␤»
12:03 Zoffix c: 2017.07 with (Seq.new: class :: does Iterator { method pull-one { say "pull"; $++ < 3 ?? $++ !! IterationEnd } }.new,).map(|*).skip { .sink; .say }
12:03 committable6 Zoffix, ¦2017.07: «pull␤pull␤pull␤pull␤This Seq has already been iterated, and its values consumed␤(you might solve this by adding .cache on usages of the Seq, or␤by assigning the Seq into an array)␤  in block <unit> at /tmp/SEq04fO07k line 1␤ «exit code = 1»»
12:03 Zoffix I guess something with buffering and STDERR/STDOUT
12:05 Zoffix .tell AlexDaniel (for release review): don't know if it's a problem or not, but noticed on HEAD this code consistently outputs 3 "pull"s after the error, while in the past all 4 "pulls" came before it: https://irclog.perlgeek.de/perl6-dev/2017-09-30#i_15239023
12:05 yoleaux Zoffix: I'll pass your message to AlexDaniel.
12:07 BenGoldberg joined #perl6-dev
12:08 Zoffix .tell AlexDaniel nevermind. I need more coffee >_< one version does .skip and the other doesn't
12:08 yoleaux Zoffix: I'll pass your message to AlexDaniel.
12:11 Ben_Goldberg joined #perl6-dev
12:13 Zoffix Would be cool to have a bool version of `andthen` that aliases to $_
12:13 Zoffix Second time today finding myself wanting to do `$foo and return $foo`
12:13 Zoffix .oO( $foo andso .return )
12:14 jnthn .return if .so given $foo
12:14 Zoffix Thanks
12:14 Zoffix Tried making a module, but no idea how to make thunks in userland https://github.com/zoffixznet/perl6-AndSo
12:14 jnthn Not quite as cute, but gets to one mention of $foo
12:15 AlexDaniel c: 2017.07 with (Seq.new: class :: does Iterator { method pull-one { note "pull"; $++ < 3 ?? $++ !! IterationEnd } }.new,).map(|*).skip { .sink; .say }
12:15 yoleaux 12:05Z <Zoffix> AlexDaniel: (for release review): don't know if it's a problem or not, but noticed on HEAD this code consistently outputs 3 "pull"s after the error, while in the past all 4 "pulls" came before it: https://irclog.perlgeek.de/perl6-dev/2017-09-30#i_15239023
12:15 committable6 AlexDaniel, ¦2017.07: «pull␤pull␤pull␤pull␤This Seq has already been iterated, and its values consumed␤(you might solve this by adding .cache on usages of the Seq, or␤by assigning the Seq into an array)␤  in block <unit> at /tmp/LG_6wuI5A1 line 1␤ «exit code = 1»»
12:15 yoleaux 12:08Z <Zoffix> AlexDaniel: nevermind. I need more coffee >_< one version does .skip and the other doesn't
12:15 AlexDaniel c: 2017.07,HEAD with (Seq.new: class :: does Iterator { method pull-one { note "pull"; $++ < 3 ?? $++ !! IterationEnd } }.new,).map(|*).skip { .sink; .say }
12:15 committable6 AlexDaniel, https://gist.github.com/c3c03a20aaab432b06fa1534e7d1f787
12:17 AlexDaniel Zoffix: Isn't it just that the stdout/stderr merging happens a little bit differently (which is ok)? https://gist.github.com/Whateverable/d0651456b85a8179d0aa4d1c2b7f87a8
12:17 AlexDaniel either way coffee that is
12:18 Zoffix No, the output is the same with same code. I just mistakenly used different code
12:18 AlexDaniel ah right, you said it. Should backlog till the end before writing any of my stupid answers :D
12:18 Zoffix :)
12:19 AlexDaniel and yea, I did already bisect it to 4b02b8aadcb47… :)
12:20 AlexDaniel ah, that's in the gist even… OK, more coffee for me please /o\
12:44 AlexDaniel thinking about it, bisectable does not pretend to be a TTY so it will give a different result than what most people will see in their terminal :S
12:45 * Zoffix calls dibs on RT132109
12:45 synopsebot RT#132109 [open]: https://rt.perl.org/Ticket/Display.html?id=132109 [BUG] `.skip` on a sequence starting with a `Slip`, returns a corrupted `Seq`
12:45 pmurias joined #perl6-dev
12:45 Zoffix Already fixed it, but the bug exists in half a dozen other cases and I need to hunt them down and write tests, but gotta $do-other-things first
12:46 Zoffix will push in 5hr
12:46 * Zoffix &
12:46 AlexDaniel (v6.d prep?)
12:46 AlexDaniel .oO*
12:46 Zoffix ?
12:46 AlexDaniel Zoffix: just kidding, but I thought you were in v6.d-prep mode
12:47 Zoffix Well, it stalled since I found out there's currently no way to find caller's version and I don't think I know enough to fix that
12:47 Zoffix https://github.com/perl6/6.d-prep/blob/master/TODO/FEATURES.md#need-a-way-to-know-callers-language
13:09 evalable6 joined #perl6-dev
13:19 Zoffix I guess not quite stalled. There's still 1000+ roast commits to review :D
13:33 skids joined #perl6-dev
13:50 llfourn joined #perl6-dev
13:55 [TuxCM] joined #perl6-dev
14:03 Zoffix m: (<a b c>, <d e>).map(-> $a, $b {|$_}).iterator.^name.say
14:03 camelia rakudo-moar 9bbfed: OUTPUT: «Any::IterateTwoWithoutPhasers␤»
14:03 Zoffix m: (<a b c>, <d e>).map(-> $a, $b {NEXT $++; |$_}).iterator.^name.say
14:03 camelia rakudo-moar 9bbfed: OUTPUT: «Any::IterateMoreWithPhasers␤»
14:04 Zoffix oh, disregard. I thought there was a TwoWithPhasers
14:06 yoleaux Zoffix: make separate t/spectest.data for each language version, bruh
14:06 Zoffix yoleaux: already done, bruh
15:23 Geth ¦ rakudo/nom: 41896b7bbf | (Zoffix Znet)++ | src/core/Any-iterable-methods.pm
15:23 Geth ¦ rakudo/nom: Fix .push-all/.skip-all on SlippyIterators
15:23 Geth ¦ rakudo/nom:
15:23 Geth ¦ rakudo/nom: Fixes RT#132109: https://rt.perl.org/Ticket/Display.html?id=132109
15:23 Geth ¦ rakudo/nom:
15:23 Geth ¦ rakudo/nom: The implemented methods forget to check whether we're currently
15:23 Geth ¦ rakudo/nom: $!slipping, which causes loss of all the remaining elements in the
15:23 Geth ¦ rakudo/nom: currently slipping Slip.
15:23 Geth ¦ rakudo/nom:
15:23 Geth ¦ rakudo/nom: Fix by add ing a check for $!slipping and doing self.skip-one and
15:23 Geth ¦ rakudo/nom: pushing (or skipping) those values before we go into the main
15:23 Geth ¦ rakudo/nom: loop of pulling more from the source iterator.
15:23 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/41896b7bbf
15:24 timotimo good catch
15:24 Zoffix <synopsebot> RT#132109 [open]: https://rt.perl.org/Ticket/Display.html?id=132109 [BUG] `.skip` on a sequence starting with a `Slip`, returns a corrupted `Seq`
15:24 timotimo why didn't synopsebot write that?
15:24 Zoffix 'cause I tested whether the semicolon works in #zofbot first and the %seen check is currently broken in that it doesn't consider the channel where the RT was seen :P
15:25 timotimo ah
15:25 Geth ¦ roast: 2e23965fa1 | (Zoffix Znet)++ | S32-list/skip.t
15:25 Geth ¦ roast: Test SlippyIterators correctly .push-all/.skip-all
15:25 Geth ¦ roast:
15:25 Geth ¦ roast: RT#132109: https://rt.perl.org/Ticket/Display.html?id=132109
15:25 Geth ¦ roast: Rakudo fix: https://github.com/rakudo/rakudo/commit/41896b7bbf
15:25 Geth ¦ roast: review: https://github.com/perl6/roast/commit/2e23965fa1
15:26 timotimo i wonder if synopsebot should even still be called synopsebot. when did we last use it to link to the design docs?
15:26 timotimo quotable6: synopsebot.*design.perl6.org
15:26 timotimo that's not how that bot works?
15:27 Zoffix quotable6: help
15:27 quotable6 Zoffix, Like this: quotable6: /^ ‘bisect: ’ / # See wiki for more examples: https://github.com/perl6/whateverable/wiki/Quotable
15:27 quotable6 timotimo, https://gist.github.com/83b13b98f160c958826c8837a3a428d9
15:27 Zoffix quotable6: /synopsebot.*design.perl6.org/
15:28 Zoffix Oh, it did respond
15:29 quotable6 Zoffix, https://gist.github.com/a69cb8c2e97b475c22715a9f3bce5fb1
15:29 timotimo hah
15:29 timotimo S99:foo
15:29 synopsebot Link: https://design.perl6.org/S99.html#foo
15:30 timotimo it does give links to design.perl6.org now, but it doesn't seem like it did that in a long time
15:30 timotimo wait ... it never did that?
15:30 timotimo quotable6: synopsebot.*perlcabal
15:30 Geth ¦ synopsebot: 03acd9f59a | (Zoffix Znet)++ | lib/SB/Plugin/RT.pm6
15:30 Geth ¦ synopsebot: Make RT seen-er see per-channel
15:30 Geth ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/03acd9f59a
15:31 Zoffix timotimo: pretty sure I copied the text it gives from the old guts
15:31 quotable6 timotimo, https://gist.github.com/79e59ab3826865e8ea625e9794d6017d
15:32 synopsebot joined #perl6-dev
15:32 timotimo that's not the problem :)
15:32 Zoffix RT#132109
15:32 synopsebot RT#132109 [resolved]: https://rt.perl.org/Ticket/Display.html?id=132109 [BUG] `.skip` on a sequence starting with a `Slip`, returns a corrupted `Seq`
15:32 timotimo i'm just wondering, with the synopses not being as interesting any more ...
15:32 Zoffix RT#132109
15:32 Zoffix k, works correctly now
15:39 timotimo i think we should remove the debug output for when a class fails to get a generated BUILDALL
15:40 Geth ¦ rakudo/nom: 4d0ead240a | (Timo Paulssen)++ | src/Perl6/Metamodel/ClassHOW.nqp
15:40 Geth ¦ rakudo/nom: don't output when add_method for BUILDALL fails.
15:40 Geth ¦ rakudo/nom: review: https://github.com/rakudo/rakudo/commit/4d0ead240a
15:40 timotimo i triggered it with zef, but that's probably because my zef still has its own BUILDALL
15:40 timotimo m: class test { submethod BUILDALL { } }
15:40 camelia rakudo-moar 41896b: ( no output )
15:40 timotimo m: class test { method BUILDALL { } }
15:40 camelia rakudo-moar 41896b: ( no output )
15:40 timotimo m: class test { method BUILDALL(|) { } }
15:40 camelia rakudo-moar 41896b: ( no output )
15:40 timotimo m: class test { submethod BUILDALL(|) { } }
15:40 camelia rakudo-moar 41896b: ( no output )
15:40 timotimo hmm
15:41 timotimo m: class test { has $.foo; submethod BUILDALL(|) { } }
15:41 camelia rakudo-moar 41896b: OUTPUT: «test failed to add a BUILDALL␤»
15:41 timotimo there it is
15:49 timotimo something bad is happening in Data::Dump::Tree where rakudo gets confused parsing the pod in the beginning of the file
15:50 timotimo in all files actually
15:51 timotimo well, multiple files anyway
16:02 nine Odd. I now got it to compile optimized function bodies for native calls at CHECK time but when I load a precompiled Inline::Perl5, they are not there ($!optimized-body of the Routine) anymore and it compiles again.
16:03 nine Ah, I probably need to do something with the deserialize_ast
16:13 AlexDaniel err yeah quotable doesn't really include the nickname of the sender
16:13 AlexDaniel also no timestamp and no channel :(
16:15 timo joined #perl6-dev
16:19 ugexe should i be able to use $*USAGE outside of a MAIN block? e.g. to modify the usage output shown when a bad command is entered
16:20 ugexe `perl6 -e '$*USAGE.subst(/Usage/, "Foosage"); sub MAIN(Int $a) { }' 1` doesn't work for instance
16:33 ugexe `multi sub MAIN(Int $a) { }; multi sub MAIN(|) { say $*USAGE.subst(/Usage/, "Foosage"); };` works I guess
16:43 Zoffix ugexe: no, it's dynamically scoped to sub MAIN and sub USAGE.
16:49 Zoffix (more accurately: to MAIN_HELPER that then invokes MAIN or USAGE)
17:01 evalable6 joined #perl6-dev
17:02 pmurias joined #perl6-dev
17:08 AlexDaniel evalable6 is on the strike :S
17:10 timotimo who's on first?
17:15 timotimo uh oh
17:15 timotimo nativecall arrays.t crashed in one of the travis jobs
17:16 timotimo quotable6: ZOFFLOP
17:17 quotable6 timotimo, https://gist.github.com/9ba33f043ccc6fcf525905015a48d5f2
17:21 nine Odd again is that finish_code_object doesn't need to do anything in the deserialize_ast
17:22 nine timotimo: any error message?
17:22 timotimo looks like it crashes, but there was no output because prove wasn't told to output anything
17:22 timotimo hm. perhaps we should do something like "make test or make test VERBOSE=1" so that if the test fails it'll run again but with output?
17:24 Zoffix why not run it verbose from the start?
17:24 timotimo hm, let's see how much output it is
17:24 Zoffix Oh, you weren't talking about travis. Neveermind...
17:25 timotimo well, VERBOSE=1 isn't the right one
17:28 timotimo it's a bunch of output, but not gigantic
17:28 timotimo and --verbosity=1 is not compatible with any --jobs but 1
17:28 timotimo oh no, the timezone on this shellhost is ... not what i was expecting
17:29 Zoffix There'll be a bunch more output soon. There's a swath of tests that isn't running ATM. I meant it would be nice to enable verbose output by default on travis so that it'd showed what the failure was
17:30 timo joined #perl6-dev
17:30 timo praise be the /upgrade command in weechat
17:31 Zoffix "Upgrade done (0.58 seconds)" that was too fast :/
17:32 Zoffix "WeeChat 1.0.1" ... is that the latest version?
17:32 Zoffix guess not, since timotimo is on 1.9 :P
17:33 timotimo oh, /upgrade is just so it execs itself but keeps buffers and stuff around
17:34 Zoffix AH
17:34 timotimo i used it after setting the TZ env var from inside weechat with "/set env ..."
17:36 AlexDaniel timotimo: fwiw I don't think it's the latest version (so it does not show the most recent flops)
17:36 AlexDaniel I'll work on it now
17:37 timotimo latest version of the irc logs?
17:38 AlexDaniel yea
17:44 releasable6 joined #perl6-dev
17:46 timotimo i didn't know we version our logs :)
17:48 AlexDaniel … I do :) When talking about quotable there are two versions: “outdated” and “up-to-date”
17:50 AlexDaniel .oO( we will get quotable with up-to-date logs on Christmas, I just don't know which one )
17:51 timotimo oh no ... you didn't put question marks before and after those words, right?
17:51 timotimo is something in my weechat-to-terminal chain broken again? >_>
17:52 AlexDaniel timotimo: no, these are curly quotes
17:53 timotimo oh, did i have to set LC_ALL ...
17:53 AlexDaniel lol, it probably renders ¿ the wrong way too :D
17:57 Ben_Goldberg joined #perl6-dev
18:01 quotable6 joined #perl6-dev
18:25 evalable6 joined #perl6-dev
18:34 Zoffix joined #perl6-dev
18:40 yoleaux Zoffix: so, what's happening with $?USAGE?
18:40 Zoffix yoleaux: it's completed. Now known as $*USAGE
18:43 nine Ha! The else block in src/NQP/World.nqp:353 is dead code. There hasn't been a replacecoderef op since parrot went away.
18:47 Zoffix joined #perl6-dev
19:06 BenGoldberg m: dd $*USAGE
19:06 camelia rakudo-moar 4d0ead: OUTPUT: «Failure.new(exception => X::Dynamic::NotFound.new(name => "\$*USAGE"), backtrace => Backtrace.new)␤»
19:07 Zoffix It's dynamically scoped to sub MAIN and sub USAGE.
19:08 Zoffix m: sub MAIN($x, $y, $z) {}; sub USAGE { $*USAGE.flip.say }
19:08 camelia rakudo-moar 4d0ead: OUTPUT: « >z< >y< >x< odFq23jB6O-elif-toblave/pmt/  ␤:egasU␤»
19:33 geekosaur joined #perl6-dev
19:45 Geth ¦ synopsebot: 18e6b39407 | (Zoffix Znet)++ | META6.json
19:45 Geth ¦ synopsebot: Add missing modules to META
19:45 Geth ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/18e6b39407
19:52 Geth ¦ synopsebot: 8b893d6153 | (Zoffix Znet)++ | 5 files
19:52 Geth ¦ synopsebot: Seen-protect doclinks and synlinks too
19:52 Geth ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/8b893d6153
19:53 synopsebot joined #perl6-dev
19:55 Geth ¦ synopsebot: 21379e6d42 | (Zoffix Znet)++ | lib/SB/Plugin/RT.pm6
19:55 Geth ¦ synopsebot: Remove dead code
19:55 Geth ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/21379e6d42
20:05 Geth ¦ synopsebot: 5531a83267 | (Zoffix Znet)++ | META6.json
20:05 Geth ¦ synopsebot: Fix meta module path
20:05 Geth ¦ synopsebot: review: https://github.com/perl6/synopsebot/commit/5531a83267
20:46 sjn is there a way to make MAIN accept a string as the first argument (e.g. "multi MAIN ("save", Str :$filename) { ... }" )?
20:48 sjn ...in such a way that one can run that script in a similar way that git works? (e.g. "myprogram save --filename=foo")
20:50 Zoffix AFAIK named params have to come before positioonals. There's a stale PR implementing the git-way
20:57 geekosaur joined #perl6-dev
21:02 * sjn tries to find that PR
21:03 sjn https://github.com/rakudo/rakudo/pull/688
21:06 Zoffix Wonder what sort of exception to make for runtime errors with `**` quantier, like this one:
21:06 Zoffix m: say "foo" ~~ /. ** { "a" .. "c" }/
21:06 camelia rakudo-moar 4d0ead: OUTPUT: «Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5a' (indicated by ⏏)␤  in block <unit> at <tmp> line 1␤␤»
21:06 Zoffix All the `Regex` errors are subclasses of X::Syntax, but this doesn't seem to be about syntax, but rather about runtime value
21:07 Zoffix m: try "3".parse-base: 42; $!.^name.say
21:07 camelia rakudo-moar 4d0ead: OUTPUT: «X::Syntax::Number::RadixOutOfRange␤»
21:08 Zoffix This one is syntax too .... though maybe that's just also a poor exception type
21:09 timotimo we do have a variable that lets you configure the argument parser and you can tell it to allow named arguments anywhere, i.e. also after positionals
21:10 Zoffix oh right
21:11 Zoffix m: @*ARGS = <foo --meow=42>; (my %*SUB-MAIN-OPTS)<named-anywhere> := 1; sub MAIN("foo", :$meow) { dd $meow }
21:11 camelia rakudo-moar 4d0ead: OUTPUT: «IntStr $meow = IntStr.new(42, "42")␤»
21:11 Zoffix sjn: ^
21:19 Ven`` joined #perl6-dev
21:21 sjn Zoffix: thanks. I'd prefer that the literal string would be allowed by default, though.
21:21 * sjn thinks that PR is still worth fixing and accepting
21:22 * AlexDaniel hears “core rakudo regression” and is ready to bisect
21:23 Zoffix sjn: would be worth find it out why it wasn't allowed originally
21:24 Zoffix s/find it/finding/;
21:24 * sjn would guess it's because of "this was the way to do things when the spec was written"
21:26 * Zoffix would guess the <named-anywhere> has ambiguities the current way doesn't
21:27 sjn ETOOMUCHBENDINGOVERBACKWARDS
21:36 geekosaur that can go both ways; sometimes things get punted to the ecosystem because supporting them in core is too much bending over backwards
21:37 Zoffix well, it *is* currently supported in core :D
21:37 AlexDaniel Zoffix: fwiw https://gist.github.com/Whateverable/b38da9bff3eb67d6494dc223de648cca
21:37 Zoffix m: say q|(my %*SUB-MAIN-OPTS)<named-anywhere> = 1|.chars
21:37 camelia rakudo-moar 4d0ead: OUTPUT: «40␤»
21:37 Zoffix half a line of code to enable it
21:37 geekosaur or, otherwise put: do you want your perl script to take over a minute to start because there's a half-sentient arg parser in play instead of a simple/works for most uses one?
21:38 * sjn doesn't mind a smart arg parser (and consequently, a smart usage text generator)
21:39 sjn if it ever should gain sentienence, we have other problems. :)
21:39 Zoffix AlexDaniel: so it does fail on 2017.09?
21:40 AlexDaniel Zoffix: fails everywhere. The string that it is attempting to test is “IO::Path::ChildSecure is a builtin type, not an external module\n in block <unit> at -e line 13\n\n”
21:40 AlexDaniel what does it mean?
21:40 Zoffix sjn: but the problem geekosaur pointed out is performance, not sentience :P
21:40 Zoffix AlexDaniel: ohh
21:41 Zoffix AlexDaniel: this https://rt.perl.org/Ticket/Display.html?id=130200#ticket-history
21:41 Zoffix Oh wait
21:42 Zoffix well, I was gonna say the bot's restricted
21:42 Zoffix But don't see anything obvious in module that'd block it
21:43 Zoffix m: with '/tmp/ZZZ777'.IO.mkdir { .add("Foo.pm6").spurt: "unit class IO::Path::Foo;" }
21:43 camelia rakudo-moar 4d0ead: ( no output )
21:43 AlexDaniel it's not restricted anyway
21:44 AlexDaniel buuut some paths are not writable, yeah
21:44 Zoffix m: with '/tmp/ZZZ777/IO/Path/'.IO.mkdir { .add("Foo.pm6").spurt: "unit class IO::Path::Foo;" }
21:44 camelia rakudo-moar 4d0ead: ( no output )
21:44 Zoffix m: use lib </tmp/ZZZ777/>; use IO::Path::Foo
21:44 camelia rakudo-moar 4d0ead: ( no output )
21:44 Zoffix m: use lib </tmp/ZZZ777/>; use IO::Path::Foo; IO::Path::Foo.new
21:44 camelia rakudo-moar 4d0ead: ( no output )
21:45 Zoffix No idea
21:45 * Zoffix &
21:58 AlexDaniel .oO( bisectable for python can be called pysectable )
22:00 AlexDaniel .oO( “?.oO( … )” ? )
22:01 AlexDaniel huggable: clog
22:01 huggable AlexDaniel, nothing found
22:01 AlexDaniel huggable: irclog
22:01 huggable AlexDaniel, nothing found
22:01 AlexDaniel huggable: log
22:01 huggable AlexDaniel, Comes out 2.41x faster than core .log10 and can handle Ints larger than 1.8e308: https://gist.github.com/zoffixznet/2539e8bebe9c5e19ab38ca43e173fe8f
22:01 AlexDaniel huggable: irc log
22:01 huggable AlexDaniel, nothing found
22:24 evalable6 joined #perl6-dev
23:26 Ven`` joined #perl6-dev

| Channels | #perl6-dev index | Today | | Search | Google Search | Plain-Text | summary