Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2015-08-27

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:16 BenGoldberg joined #perl6
00:17 BenGoldberg joined #perl6
00:26 llfourn joined #perl6
00:27 xzenfar joined #perl6
00:59 aborazmeh joined #perl6
00:59 aborazmeh joined #perl6
01:12 Sgeo joined #perl6
01:49 dayangkun joined #perl6
01:56 kaare__ joined #perl6
01:59 retupmoca YuviPanda: run: "panda gen-meta" from the root directory of your module to get a META.info created
01:59 retupmoca YuviPanda: then you just need to add the link to that META.info to the META.list file in this repo: https://github.com/perl6/ecosystem
02:14 mdinger joined #perl6
02:18 noganex_ joined #perl6
02:18 yeahnoob joined #perl6
02:27 llfourn joined #perl6
03:40 pierrot joined #perl6
03:41 xzenfar joined #perl6
03:45 mr-foobar joined #perl6
04:22 llfourn joined #perl6
04:34 telex joined #perl6
04:35 cognominal joined #perl6
05:08 davido_ joined #perl6
05:18 [Sno] joined #perl6
05:20 * colomon needs to get moving so he can catch the 8:48 train to Olten.
05:29 breinbaas joined #perl6
05:30 davercc` joined #perl6
05:33 dha joined #perl6
05:47 brrt joined #perl6
05:51 Sqirrel joined #perl6
06:03 colomon is it difficult to find the hackathon room from the front door of the venue?  I’ve got the venue starred in google maps so I can find it, but I won’t have an easy way to communicate with IRC once I’m there....
06:05 diana_olhovik_ joined #perl6
06:06 timotimo *yawn*
06:06 maettu timotimo: indeed.. :-)
06:07 llfourn joined #perl6
06:08 timotimo i forgot what people said when they were going to get breakfast, so i just guessed "getting up at 0800, then showering, then breaking fast shall be good enough"
06:08 maettu colomon: no, that's easy.
06:08 colomon maettu: yay!
06:09 maettu we're opening the door @ 9 am
06:09 colomon I should be there around 11
06:11 bioexpress joined #perl6
06:18 masak morning, #perl6
06:18 masak hope everyone had a good night's sleep
06:18 * masak heads down for breakfast
06:18 domidumont joined #perl6
06:20 bioexpress "doesnotexist" does not exist. Why dies
06:20 bioexpress p6: my $contents = "doesnotexist".IO.slurp;
06:20 GLRelia rakudo-moar a11e8f: OUTPUT«Failed to open file /home/steve/doesnotexist: no such file or directory␤  in block <unit> at /tmp/tmpfile:1␤␤»
06:20 camelia rakudo-moar a7c845: OUTPUT«Failed to open file /home/camelia/doesnotexist: no such file or directory␤  in block <unit> at /tmp/tmpfile:1␤␤»
06:20 bioexpress and
06:20 bioexpress p6: rmdir "dasnotexist";
06:20 GLRelia rakudo-moar a11e8f: OUTPUT«rmdir is disallowed in restricted setting␤  in sub restricted at src/RESTRICTED.setting:1␤  in sub rmdir at src/RESTRICTED.setting:13␤  in block <unit> at /tmp/tmpfile:1␤␤»
06:20 camelia rakudo-moar a7c845: OUTPUT«rmdir is disallowed in restricted setting␤  in sub restricted at src/RESTRICTED.setting:1␤  in sub rmdir at src/RESTRICTED.setting:13␤  in block <unit> at /tmp/tmpfile:1␤␤»
06:21 bioexpress does not die?
06:21 nine .tell skids Ah, yes thanks! I'll have a look at it after breakfast. Now that I understand the problem it solves, I think it's a shame, your PR has not been merged yet.
06:21 yoleaux nine: I'll pass your message to skids.
06:22 masak bioexpress: you seem surprised that `slurp` on something that doesn't exist dies. I'm not surprised at that. why are you? :)
06:22 masak bioexpress: also, removing files tends to not die if they're not there. kind of a vacuous success.
06:22 masak bioexpress: not entirely evident that this goes for directories as well as normal files, but...
06:24 bioexpress Thx!
06:27 abaugher joined #perl6
06:38 dalek rakudo/glr: d44cd4b | skids++ | src/core/array_slice.pm:
06:38 dalek rakudo/glr: Fix Whatevercode in array slice indices past the first one.
06:38 dalek rakudo/glr:
06:38 dalek rakudo/glr: Lets us pass /S32-array/delete.t which used to fail with
06:38 dalek rakudo/glr: "Cannot call Real(WhateverCode: )"
06:38 dalek rakudo/glr:
06:38 dalek rakudo/glr: (Surprisingly this is the only file containing a test that uses them,
06:38 dalek rakudo/glr: unless they are skipped)
06:38 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/d44cd4b38d
06:38 dalek rakudo/glr: 91d2127 | niner++ | src/core/array_slice.pm:
06:38 dalek rakudo/glr: Merge pull request #515 from skids/glr3
06:38 dalek rakudo/glr:
06:38 dalek rakudo/glr: Fix Whatevercode in array slice indices past the first one.
06:38 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/91d2127c6d
06:39 nine .tell skids thank you for fixing @a[1, *-1]:delete :) I've had a look at this yesterday but my train arrived before I came to writing code. I really wonder how this worked in nom though. Couldn't find any code handling that case there.
06:39 yoleaux nine: I'll pass your message to skids.
06:40 dha nine - have the 5to6 docs been of any use to you?
06:41 dha (it was you that was asking about them a while back, right?)
06:41 laben joined #perl6
06:42 laben hello p6'ers!
06:48 timotimo hello laben!
06:50 domidumont joined #perl6
06:52 Woodi joined #perl6
06:53 zakharyas joined #perl6
06:54 domidumont joined #perl6
07:03 * timotimo heads towards the breakfast area
07:04 moritz down to 84 failing test files, it seems
07:06 laben so is it under nine% ?
07:09 rurban joined #perl6
07:11 brrt joined #perl6
07:20 darutoko joined #perl6
07:28 Sqirrel joined #perl6
07:28 laouji joined #perl6
07:35 * TimToady waves from hackython
07:39 nine Actually it's 83 here
07:42 espadrine joined #perl6
07:45 moritz might be one of the flappers
07:48 timotimo o/
07:49 lizmat joined #perl6
07:50 * masak waves too
07:50 hoelzro o/
07:51 hoelzro links from breakfast (not sausage links): https://www.youtube.com/watch?v=WIKqgE4BwAY
07:51 masak plan for the day: help someone else with something. newbies take precedence. :)
07:52 moritz I've been mulling an idea for another Perl 6 coding contest
07:53 moritz but learning from masak's mistakes, crowd-source the reviews
07:53 FROGGS joined #perl6
07:53 masak ooh
07:53 masak yeah, doing the reviews all by yourself is fine, until $life intervenes.
07:54 masak crowd-sourcing the reviews might actually work really well if pulled off correctly.
07:54 moritz and then it would still be the one holding the contest (probably me, maybe a second confederate) who selects the winner(s)
07:55 nine m: my @a = [$[1, 2, 3], $[4]]; @a>>.say;
07:55 GLRelia rakudo-moar a11e8f: OUTPUT«1␤2␤3␤4␤»
07:55 camelia rakudo-moar a7c845: OUTPUT«(Any)␤(Any)␤(Any)␤(Any)␤»
07:55 nine m: my @a = [[1, 2, 3], [4]]; @a>>.say;
07:55 GLRelia rakudo-moar a11e8f: OUTPUT«1␤2␤3␤4␤»
07:55 camelia rakudo-moar a7c845: OUTPUT«4␤3␤1␤2␤»
07:56 moritz masak: any ideas on how to pull of the crowed-sourced reviews?
07:56 FROGGS a huperbug /o/
07:56 moritz *crowd
07:56 nine a plain normal hyper bug
07:56 moritz .oO( cloud-sourced reviews )
07:56 moritz m: say () <<+>> ()
07:56 GLRelia rakudo-moar a11e8f: OUTPUT«␤»
07:56 camelia rakudo-moar a7c845: OUTPUT«␤»
07:56 moritz m: say () <<+<< ()
07:56 GLRelia rakudo-moar a11e8f: OUTPUT«␤»
07:56 camelia rakudo-moar a7c845: OUTPUT«␤»
07:57 moritz m: say () >>+<< ()
07:57 GLRelia rakudo-moar a11e8f: OUTPUT«␤»
07:57 camelia rakudo-moar a7c845: OUTPUT«␤»
07:57 moritz didn't I see one of those fail?
07:57 moritz maybe fixed in the mean time
07:57 laben m: say (1..3) >>+<< (1..4)
07:57 GLRelia rakudo-moar a11e8f: OUTPUT«2 4 6 5␤»
07:57 camelia rakudo-moar a7c845: OUTPUT«Lists on either side of non-dwimmy hyperop of infix:<+> are not of the same length␤left: 3 elements, right: 4 elements␤  in block <unit> at /tmp/lsJZAZiUg2:1␤␤»
07:57 moritz GLR++
07:58 laben nine: ^^^ want me to send a PR for this?
07:58 nine laben: please go ahead :)
07:59 moritz oh, I mis-read
07:59 moritz nom++
07:59 ShimmerFairy moritz: first idea that comes to mind is some sort of anonymous reviewing thing, perhaps some kind of website that presented people a couple of anonymized tests, and likely some guidelines on what's desired.
07:59 moritz ShimmerFairy: I'd explicitly not make it anonymous :-)
07:59 dalek roast/glr: e960b57 | (David Warring)++ | / (4 files):
07:59 dalek roast/glr: utilize does-ok test assertion
07:59 dalek roast/glr: review: https://github.com/perl6/roast/commit/e960b57767
07:59 dalek roast/glr: 67b2039 | (David Warring david.warring@gmail.com)++ | / (4 files):
07:59 dalek roast/glr: Merge branch 'does-ok'
07:59 dalek roast/glr: review: https://github.com/perl6/roast/commit/67b2039297
07:59 dalek roast/glr: 1419d17 | moritz++ | / (4 files):
07:59 dalek roast/glr: Merge remote-tracking branch 'origin/master' into glr
07:59 dalek roast/glr:
07:59 dalek roast/glr: Conflicts:
07:59 dalek roast/glr: S02-types/sigils-and-types.t
07:59 dalek roast/glr: review: https://github.com/perl6/roast/commit/1419d17ee5
08:00 moritz fwiw that was merging master into nom in roast
08:00 timotimo ah
08:00 moritz ShimmerFairy: allow everybody to submit a review to other people's solutions
08:00 moritz does anybody know what's up with the index and rindex tests?
08:01 moritz I think something changed in nom, but both nom and the tests have been merged into the glr branches
08:01 ShimmerFairy moritz: sounds good, my idea was just to eliminate knowing anything about the coder beforehand :) . I definitely think the goal for crowd-sourced reviewing would be to have multiple reviews per entry.
08:02 moritz ShimmerFairy: aye; and then some kind of summary
08:03 [Sno] joined #perl6
08:05 elohmrow joined #perl6
08:06 timotimo oh, is this about a p6cc?
08:07 moritz yes
08:08 RabidGravy joined #perl6
08:09 ShimmerFairy moritz: I don't know if aggregating reviews would be easier for the contest runner(s) than making the reviews, but I would hope that it is the case :)
08:09 laben nine: https://github.com/rakudo/rakudo/pull/516
08:10 moritz ShimmerFairy: maybe aggregating the reviews can also be crowd-sourced
08:11 nine m: my @a = [$[1, 2, 3], $[4]]; @a>>.say;
08:11 GLRelia rakudo-moar a11e8f: OUTPUT«1␤2␤3␤4␤»
08:11 camelia rakudo-moar a7c845: OUTPUT«(Any)␤(Any)␤(Any)␤(Any)␤»
08:12 * btyler_ waves from SPW
08:12 nine I'm no longer sure this really _is_ a bug. The hyper is supposed to apply the op recursively to the whole list. So .say is called on every item in the nested arrays.
08:12 btyler_ need a bigger hackathon room :)
08:12 nine TimToady: am I correct?
08:12 moritz nine: looks correct to me
08:13 moritz and, I mean, where would the Any even come from?
08:13 moritz [$[1, 2, 3], $[4]]
08:13 dakkar joined #perl6
08:14 FROGGS my expectation was that "my @a = [$[1, 2, 3], $[4]]; @a>>.say;" results in "1 2 3␤4␤"
08:14 masak m: my @a = 1, 2, 3; for @a X @a { .say }
08:15 GLRelia rakudo-moar a11e8f: OUTPUT«1 1␤1 2␤1 3␤2 1␤2 2␤2 3␤3 1␤3 2␤3 3␤»
08:15 camelia rakudo-moar a7c845: OUTPUT«1 1␤1 2␤1 3␤2 1␤2 2␤2 3␤3 1␤3 2␤3 3␤»
08:15 moritz FROGGS: ... or any permutation thereof
08:15 dalek rakudo/glr: 8606169 | lizmat++ | src/core/Supply.pm:
08:15 dalek rakudo/glr: Fix some flattening issues with Ranges and emit
08:15 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/8606169aa2
08:16 nine FROGGS: from S03: Non-structural unary hyper operators produce a hash or array of exactly the same shape as the single argument. The hyper will descend into nested lists and hashes to distribute over the lower-level values just as they distribute over the top-level values that are leaves in the tree.
08:16 [Tux] good morning all
08:16 [Tux] having most of you in Switserland, makes "morning" a well educated guess :)
08:17 FROGGS nine: okay, then it seems correct when using .say
08:17 FROGGS m: my @a = [$[1, 2, 3], $[4]]; my @b = @a>>.++; say @b
08:17 GLRelia rakudo-moar a11e8f: OUTPUT«1 2 3 4␤»
08:17 camelia rakudo-moar a7c845: OUTPUT«0 0 0 0␤»
08:17 [Tux] laben, did you fix anything in glr ot in T::C?
08:17 nine m: my @a = [$[1, 2, 3], $[4]]; my @b = @a>>.++; say @b; say @a;
08:17 GLRelia rakudo-moar a11e8f: OUTPUT«1 2 3 4␤2 3 4 5␤»
08:17 camelia rakudo-moar a7c845: OUTPUT«0 0 0 0␤1 1 1 1␤»
08:18 FROGGS \o/
08:18 FROGGS m: my @a = [$[1, 2, 3], $[4]]; say @a.perl; my @b = @a>>.++; say @b.perl; say @a.perl;
08:18 GLRelia rakudo-moar a11e8f: OUTPUT«[$[1, 2, 3], $[4]]␤[$[1, 2, 3], $[4]]␤[$[2, 3, 4], $[5]]␤»
08:18 camelia rakudo-moar a7c845: OUTPUT«[[Any, Any, Any, Any]]<>␤[[0, 0, 0, 0]]<>␤[[1, 1, 1, 1]]<>␤»
08:19 FROGGS GLR++
08:19 laben [Tux]: i did send you some msg to |Tux|. i fixed tests up to 55.t excluded, which needs an input on your part since i am not sure and fixed test 60.t and part of 77.t
08:20 [Tux] ah, I have to ssh to the other box to check the log then: [Tux] is at home, |Tux| is at work
08:20 laben you can check irclog
08:22 Sqirrel joined #perl6
08:22 dalek rakudo/glr: 3cd8dd7 | FROGGS++ | src/core/Cool.pm:
08:22 dalek rakudo/glr: fix Cool.{index|rindex} when passing insane offsets
08:22 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/3cd8dd73a1
08:22 aborazmeh joined #perl6
08:22 aborazmeh joined #perl6
08:22 FROGGS this ^^^ is also a fix for two tests failing on nom, but I guess we can wait until glr gets merged into nom
08:24 nine laben: I experience some kind of deja vu. Haven't you sent that pull request before?
08:24 laben nine: no, i just posted it on a gist
08:24 nine aah, that was it
08:25 dalek rakudo/glr: fadd434 | laben++ | src/core/metaops.pm:
08:25 dalek rakudo/glr: Fix dwimmy hyper code to correctly consider throw condition
08:25 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/fadd434708
08:25 dalek rakudo/glr: fca2873 | niner++ | src/core/metaops.pm:
08:25 dalek rakudo/glr: Merge pull request #516 from laben/myglr
08:25 dalek rakudo/glr:
08:25 dalek rakudo/glr: Fix dwimmy hyper code to correctly consider throw condition
08:25 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/fca287328b
08:26 masak m: my @c = <r k d>.pick(*); my @v = <a u o>.pick(*); for ^3 -> $i { print @c[$i]; print @v[$i] }; say "";
08:26 camelia rakudo-moar a7c845: OUTPUT«rudako␤»
08:26 GLRelia rakudo-moar a11e8f: OUTPUT«daroku␤»
08:26 masak :D
08:28 moritz m: say flat <r k d>.pick(*) Z <a u o>.pick(*)
08:28 GLRelia rakudo-moar a11e8f: OUTPUT«k u d a r o␤»
08:28 camelia rakudo-moar a7c845: OUTPUT«d o r u k a␤»
08:28 moritz m: say join '', flat <r k d>.pick(*) Z <a u o>.pick(*)
08:28 GLRelia rakudo-moar a11e8f: OUTPUT«dakoru␤»
08:28 camelia rakudo-moar a7c845: OUTPUT«rodaku␤»
08:30 masak moritz++ # functional to my imperative
08:30 masak m: say join '', <r k d>.pick(*) Z <a u o>.pick(*)
08:30 GLRelia rakudo-moar a11e8f: OUTPUT«␤»
08:30 camelia rakudo-moar a7c845: OUTPUT«␤»
08:30 larion joined #perl6
08:30 * masak realizes he has to learn to like "flat" from now on :)
08:31 laben m: say (1..3) >>+<< (1..4)
08:31 GLRelia rakudo-moar a11e8f: OUTPUT«2 4 6 5␤»
08:31 camelia rakudo-moar a7c845: OUTPUT«Lists on either side of non-dwimmy hyperop of infix:<+> are not of the same length␤left: 3 elements, right: 4 elements␤  in block <unit> at /tmp/ywxB3T6p7x:1␤␤»
08:31 laben still not updated
08:31 lizmat m: sub a(*@a) { say @a.perl }; a "E"..."A","e"..."a"
08:31 GLRelia rakudo-moar a11e8f: OUTPUT«Method 'splice' not found for invocant of class 'List'␤  in block <unit> at /tmp/54J6KyIye9:1␤␤»
08:31 camelia rakudo-moar a7c845: OUTPUT«["E", "D", "C", "B", "A", "e", "d", "c", "b", "a"]<>␤»
08:31 nine I though GLRelia is updating every 5 minutes?
08:32 moritz nine: well, I guess it triggers an updated every 5 minutes, unless a rebuild is currently in progress
08:33 moritz at least that's how I'd do it
08:33 laben the commit hash was not updated, hence i took it as not updated ...
08:35 xfix joined #perl6
08:35 masak for those listening in from elsewhere:
08:35 masak we just have a Big Table Refactor, because there were more hackathoners than seats around the table
08:35 masak great fun
08:35 laben masak: did you implement a SeatIterator?
08:36 btyler_ there's also a spillover hackathon room, if it gets too cozy there
08:36 masak no, we hand-coded everything for maximum speed
08:37 FROGGS my spectest hangs :o(
08:37 moritz mine too
08:37 moritz after t/spec/rosettacode/sierpinski_triangle.t
08:37 FROGGS t/spec/S17-supply/start.t and t/spec/S32-io/IO-Socket-Async.rakudo.moar
08:37 lizmat m: sub a(*@a) { say @a.perl }; a 0..10,0...10
08:37 GLRelia rakudo-moar a11e8f: OUTPUT«[0, 1, 2, 3, 4, 5, 6, 7, 8, Failure.new(exception => X::Sequence::Deduction.new(from => "9,10,0"), backtrace => Backtrace.new)]␤»
08:37 camelia rakudo-moar a7c845: OUTPUT«[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]<>␤»
08:38 tadzik hello hackathon here too!
08:38 FROGGS and we regressed 8 test files :o(
08:38 virtualsue joined #perl6
08:38 hoelzro masak: and good thing no one got flattened
08:38 moritz huh :(
08:39 FROGGS lizmat: did you spectest before&&after your patch?
08:39 FROGGS I guess I'll re-spectest the last two commits
08:39 lizmat I did not spectest, I only tested S17-supply/minmax.t
08:40 lizmat which now has fewer failing tests
08:40 moritz fwiw I typically capture the spectest output in a file, for example glr-1
08:40 lizmat I'll run a spectest now
08:40 moritz and then I run grep 'ok$' glr-1 | |cut -d\  -f 1|perl -plE 's/\.rakudo\.moar$/.t/' > t/localtest
08:40 FROGGS this is the fallout of the last (two?) commits: https://gist.github.com/FR​OGGS/2f06ed42f1755b1235bb
08:41 moritz and then a 'make localtest' after a change to make sure there are no regressions
08:41 lizmat so that's good ?
08:41 lizmat ah no...  sorry  :-)
08:42 * itz waves from room 2
08:42 FROGGS lizmat: t/spec/S17-supply/syntax.t, t/spec/S17-supply/unique.t and t/spec/S17-supply/words.t did not fail before
08:42 * lizmat looks
08:42 masak moritz: if you do do a p6cc, maybe I can help with finding tasks. the tasks are very important to get right -- and that's one of the things I believe I got right all three times.
08:42 ShimmerFairy except for the commandline magic, that's pretty much how I do it :)  (take before and after results, and put in localtest.data any of the interesting files to investigate)
08:42 moritz masak: that would be very much appreciated
08:42 laben itz: do you have time to check why GLRelia isn't updating anymore?
08:43 moritz (the disadvantage of organizing a p6cc is that you can can't participate yourself :-) )
08:43 dalek perl6-roast-data: ad9443b | coke++ | / (2 files):
08:43 dalek perl6-roast-data: another glr run for 'today'
08:43 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/ad9443bef5
08:43 ShimmerFairy masak: task 1: freaking hexslide :P
08:44 masak ShimmerFairy: is that a variant of ordinary hexslide? :P
08:44 lizmat FROGGS: think I found it
08:44 FROGGS lizmat++
08:45 virtualsue joined #perl6
08:46 [Coke] ^^ that commit also shows a slight regression, if that helps track down versions.
08:47 FROGGS lizmat / others: you can always compare against: https://github.com/perl6/glr-test-failure​s-annotated/blob/master/test-failures.txt
08:48 FROGGS it is usually almost up-to-date :o)
08:48 dalek rakudo/glr: 0f43fb6 | lizmat++ | src/core/Supply.pm:
08:48 dalek rakudo/glr: Unbreak some supply tests, break others again :-(
08:48 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/0f43fb6bbf
08:48 dalek roast/glr: 5f85245 | (Stefan Seifert)++ | S17-supply/on.t:
08:48 dalek roast/glr: Fix S17-supply/on.t misusing >> when it wants map
08:48 dalek roast/glr:
08:48 dalek roast/glr: >> is supposed to recurse into nested arrays. It seems like some bug on
08:48 dalek roast/glr: nom prevented that. Now that that's fixe on glr, we have to use map
08:48 dalek roast/glr: instead.
08:48 dalek roast/glr: review: https://github.com/perl6/roast/commit/5f852454f2
08:50 zakharyas joined #perl6
08:51 itz laben: yes I've have a look now
08:52 lizmat my @tail := @lol[$m].list.splice(1); # trailing elems of last list added back later    # definitely Wrong(TM)
08:52 itz m: say 2+3
08:52 GLRelia rakudo-moar a11e8f: OUTPUT«5␤»
08:52 camelia rakudo-moar a7c845: OUTPUT«5␤»
08:52 masak m: sub foo { my $x = 7; class C { method bar { say $x } } }; C.new.bar
08:52 GLRelia rakudo-moar a11e8f: OUTPUT«(Any)␤»
08:52 camelia rakudo-moar a7c845: OUTPUT«(Any)␤»
08:53 masak should we have a warning when an our-scoped thing sits in a my-scoped thing -- like Perl 5 does?
08:53 FROGGS ahh, that ol' bug :o)
08:53 nine m: say [[1..5],[6..10],[11..14]].perl
08:53 camelia rakudo-moar a7c845: OUTPUT«[[1, 2, 3, 4, 5], [6, 7, 8, 9, 10], [11, 12, 13, 14]]␤»
08:53 GLRelia rakudo-moar a11e8f: OUTPUT«[$[1, 2, 3, 4, 5], $[6, 7, 8, 9, 10], $[11, 12, 13, 14]]␤»
08:54 nine m: use Test; is-deeply [[1..5],[6..10],[11..14]], [$[1, 2, 3, 4, 5], $[6, 7, 8, 9, 10], $[11, 12, 13, 14]]
08:54 GLRelia rakudo-moar a11e8f: OUTPUT«ok 1 - ␤»
08:54 camelia rakudo-moar a7c845: OUTPUT«not ok 1 - ␤␤# Failed test at /tmp/xgsYOcYtmI line 1␤# expected: [Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any, Any]␤#      got: [[1, 2, 3, 4, 5], [6, 7, 8, 9, 10], [11, 12, 13, 14]]␤»
08:57 masak does anyone want to pair up on something moderately tricky and semi-secret?
08:59 masak on a completely unrelated note, has anyone ever done `take` inside of a grammar? :P
08:59 FROGGS m: my $py = 0 && try { my $py = 42; $py.bla() }
08:59 camelia rakudo-moar a7c845: OUTPUT«5===SORRY!5=== Error while compiling /tmp/GaBH5Z1ixo␤Cannot use variable $py in declaration to initialize itself␤at /tmp/GaBH5Z1ixo:1␤------> 3my $py = 0 && try { my $py = 42; $7⏏5py.bla() }␤    expecting any of:␤        term␤»
09:00 FROGGS m: my $py = 0 && try { { my $py = 42; $py.bla()}() }
09:00 camelia rakudo-moar a7c845: ( no output )
09:00 * moritz should probably start the nom release process
09:01 abraxxa joined #perl6
09:02 itz Makefile:422: recipe for target 'CORE.setting.moarvm' failed
09:03 FROGGS itz: does it reconfigure on each build?
09:03 itz maybe cruft left in my env unless others can't compile GLR
09:03 FROGGS because lately a new file was added to the setting build (SomethingIterator.pm or so)
09:04 itz FROGGS: ah thanks
09:07 FROGGS [ptc]: it is now rakudobugged as RT #125921
09:07 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=125921
09:07 brrt joined #perl6
09:08 masak ++moritz # nom release process
09:08 masak by the way, colomon++ has now joined the hackathon too! \o/
09:09 colomon joined #perl6
09:09 colomon hello hackathon!
09:09 masak colomon! \o/
09:09 FROGGS \o\
09:09 * masak starts and looks over his shoulder
09:09 masak aie!
09:09 masak colomon: don't sneak up on me like that!
09:10 masak :P
09:10 FROGGS hoelzro: this is ven in case you've not seen him yet
09:10 dalek rakudo/glr: 04a8f10 | (Stefan Seifert)++ | src/core/Supply.pm:
09:10 dalek rakudo/glr: Fix Supply.batch flattening the results too much
09:10 dalek rakudo/glr:
09:10 dalek rakudo/glr: batch is supposed to pass an array to emit. Instead it passed a list of
09:10 dalek rakudo/glr: arguments due to the single arg rule.
09:10 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/04a8f102dc
09:10 masak ooh, and now vendethiel arrived! \o/
09:10 hoelzro ahhhhh
09:10 FROGGS ahh, he has a badge... nvm :o)
09:11 masak I hope you all realize...
09:11 masak ...me and vendethiel in the same room...
09:11 masak ...there will be macros. :)
09:11 [ptc] FROGGS: sweet, thanks
09:11 yoleaux 20 Aug 2015 16:27Z <moritz> [ptc]: a perl6-examples job on hack was running for 72 CPU hours before I killed it; please investigate (and add a CPU limit to all cron jobs)
09:11 espadrine joined #perl6
09:11 yoleaux 21 Aug 2015 14:42Z <moritz> [ptc]: I've set some ulimits in perl6-examples' update-and-sync, and updated the checkout on hack; that was in a detached HEAD previously. If master sounds wrong to you, please use "git reflog" to restore the old state (and fix the issue otherwise)
09:11 colomon NOOOOOOOO!
09:11 yoleaux 25 Aug 2015 10:33Z <_itz> [ptc]: examples.perl6.org appears to have stopped updating
09:11 masak colomon: mwhahaha
09:13 ShimmerFairy .oO(rows of apple comps)
09:13 [ptc] moritz: sorry about the perl6-examples jobs!  I've been only a lurker for a while ($work has been pretty busy) and hadn't seen your message
09:13 tadzik and sorted thinkpads :P
09:13 nine masak: yeah we hope so :)
09:13 [ptc] itz: thanks for the message.  Will look into it over the next couple of days :-)
09:14 itz [ptc]: are in .ch .. I was planning on duplicating the examples.perl6.org update process elsewhere as a dev instance?
09:14 * [ptc] waves from the next room
09:14 itz ah right
09:14 lizmat m: sub a(*@a) { say @a.perl }; a 0..10,0...10
09:14 camelia rakudo-moar a7c845: OUTPUT«[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]<>␤»
09:14 lizmat m: sub a(*@a) { say @a.perl }; a 0..10,0...10
09:14 camelia rakudo-moar a7c845: OUTPUT«[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]<>␤»
09:15 lizmat m: sub a(*@a) { say @a.perl }; a 0...10
09:15 camelia rakudo-moar a7c845: OUTPUT«[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]<>␤»
09:15 lizmat m: sub a(*@a) { say @a.perl }; a 0..10
09:15 camelia rakudo-moar a7c845: OUTPUT«[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]<>␤»
09:15 lizmat m: sub a(*@a) { say @a.perl }; a 0..5,6...10
09:15 camelia rakudo-moar a7c845: OUTPUT«[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]<>␤»
09:15 lizmat m: sub a(*@a) { say @a.perl }; a 0..5,0...10
09:15 camelia rakudo-moar a7c845: OUTPUT«Unable to deduce arithmetic or geometric sequence from 4,5,0 (or did you really mean '..'?)␤  in sub a at /tmp/aIraUSs47h:1␤  in block <unit> at /tmp/aIraUSs47h:1␤␤Actually thrown at:␤  in sub a at /tmp/aIraUSs47h:1␤  in block <unit> at /tmp/aI…»
09:17 colomon so what’s the plan of attack here?  should I grab a random hunk of failing roast-glr and try to make it work?  should I ignore glr and try to improve the smoker?  fix modules?
09:19 masak colomon: I think people would be happy if you jumped into GLR.
09:19 masak the other stuff dosuns nice too, though.
09:20 * colomon thinks he will have more time to hack on p6 during this hackathon than he did in the entire rest of the year so far...
09:20 nine colomon: yeah, just grab a failing test :)
09:21 nine Maybe we should start coordinating a bit on the channel though with so many people picking tests and failures becoming fewer and fewer
09:21 dalek roast/glr: 18f810a | lizmat++ | S17-supply/minmax.t:
09:21 dalek roast/glr: Fix test by not using sequences
09:21 dalek roast/glr: review: https://github.com/perl6/roast/commit/18f810a739
09:21 colomon sounds reasonable
09:21 nine Only 81 failing now!
09:21 colomon nine: running a spectest already
09:21 lizmat that's 80 now!
09:22 GLRelia joined #perl6
09:22 itz m: say 2+3
09:22 camelia rakudo-moar a7c845: OUTPUT«5␤»
09:22 GLRelia rakudo-moar a11e8f: OUTPUT«5␤»
09:22 masak yay, I just helped btyler_++ fix his code! \o/
09:22 donaldh joined #perl6
09:23 * masak feels warm an fuzzy inside
09:23 * FROGGS .oO( Fix test by avoiding code that exposes the error )
09:23 lizmat well, the error is with sequences, not with supplies
09:23 FROGGS lizmat: are there other tests for that reverse() error?
09:23 lizmat it's not reverse() that is a problem, it's sequences
09:23 lizmat m: say 1...10,1...10
09:23 camelia rakudo-moar a7c845: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
09:23 GLRelia rakudo-moar a11e8f: OUTPUT«Method 'splice' not found for invocant of class 'List'␤  in block <unit> at /tmp/EyF2HOltok:1␤␤»
09:24 FROGGS lizmat: are there tests for that?
09:24 nine m: say 1..10,1...10
09:24 camelia rakudo-moar a7c845: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
09:24 GLRelia rakudo-moar a11e8f: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
09:24 nine m: say 1..10...10
09:24 camelia rakudo-moar a7c845: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
09:24 GLRelia rakudo-moar a11e8f: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
09:24 nine m: say 1..10,0...10
09:24 camelia rakudo-moar a7c845: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
09:24 GLRelia rakudo-moar a11e8f: OUTPUT«1 2 3 4 5 6 7 8 9 10␤»
09:25 laben m: dd 1...10,1...10
09:25 camelia rakudo-moar a7c845: OUTPUT«(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
09:25 GLRelia rakudo-moar a11e8f: OUTPUT«Method 'splice' not found for invocant of class 'List'␤  in block <unit> at /tmp/v5UaX9Eq09:1␤␤»
09:25 laben m: dd 1..10,1...10
09:25 camelia rakudo-moar a7c845: OUTPUT«(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
09:25 GLRelia rakudo-moar 11e8f: OUTPUT«(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
09:25 laben m: dd ,1...10
09:25 camelia rakudo-moar a7c845: OUTPUT«5===SORRY!5=== Error while compiling /tmp/dwW79VhrtQ␤Preceding context expects a term, but found infix , instead␤at /tmp/dwW79VhrtQ:1␤------> 3dd ,7⏏1...10␤»
09:25 GLRelia rakudo-moar 11e8f: OUTPUT«5===SORRY!5=== Error while compiling /tmp/GuUZkO5nXx␤Preceding context expects a term, but found infix , instead␤at /tmp/GuUZkO5nXx:1␤------> 3dd ,7⏏1...10␤»
09:25 laben m: dd 1...10,
09:25 camelia rakudo-moar a7c845: OUTPUT«(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
09:25 GLRelia rakudo-moar 11e8f: OUTPUT«(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
09:25 laben m: dd 1...10
09:25 camelia rakudo-moar a7c845: OUTPUT«(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
09:25 GLRelia rakudo-moar 11e8f: OUTPUT«(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
09:26 tadzik ..wth is "Bytecode stream version too low"? :O
09:26 TimToady m: say (1,2,3 ... 2)
09:26 camelia rakudo-moar a7c845: OUTPUT«1 2␤»
09:26 GLRelia rakudo-moar 11e8f: OUTPUT«1 2␤»
09:26 TimToady m: my @x; @x[0,2,4...$end]
09:26 camelia rakudo-moar a7c845: OUTPUT«5===SORRY!5=== Error while compiling /tmp/_SokNrE8UA␤Variable '$end' is not declared. Did you mean '&end'?␤at /tmp/_SokNrE8UA:1␤------> 3my @x; @x[0,2,4...7⏏5$end]␤»
09:26 GLRelia rakudo-moar 11e8f: OUTPUT«5===SORRY!5=== Error while compiling /tmp/GIGORVrSkg␤Variable '$end' is not declared. Did you mean '&end'?␤at /tmp/GIGORVrSkg:1␤------> 3my @x; @x[0,2,4...7⏏5$end]␤»
09:27 jnthn ohhai o/
09:27 * jnthn is at l'hackathon now also :)
09:27 lizmat jnthn o/
09:27 itz the GLRelia git sha should correct shortly
09:28 * colomon is going to look at S32-list/pick.t
09:28 itz m: say "sha hash"
09:28 camelia rakudo-moar a7c845: OUTPUT«sha hash␤»
09:28 GLRelia rakudo-moar 04a8f1: OUTPUT«sha hash␤»
09:28 tadzik FROGGS: https://github.com/tadzik/panda/issues/178 is there a reason bundle() shouldn't create source-url instead?
09:28 dalek rakudo/glr: 30161a5 | FROGGS++ | src/Perl6/Grammar.nqp:
09:28 dalek rakudo/glr: reset to-be-initialied variable name in blockoid
09:28 dalek rakudo/glr:
09:28 dalek rakudo/glr: Before we did that in pblock, which means that e.g. blorsts did not get handled
09:28 dalek rakudo/glr: correctly. Fixes RT #125921.
09:28 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/30161a5f01
09:28 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=125921
09:28 masak m: say (flat 0..9, 'a'..'f').roll(40).join
09:28 camelia rakudo-moar a7c845: OUTPUT«30a840c3c9475c53db38dcd25d934dd6ffc6afb8␤»
09:28 GLRelia rakudo-moar 04a8f1: OUTPUT«fd407e3a43cb55d696235b75b5b39189fde0af01␤»
09:29 laben m: say (1..3) >>+<< (1..4)
09:29 camelia rakudo-moar a7c845: OUTPUT«Lists on either side of non-dwimmy hyperop of infix:<+> are not of the same length␤left: 3 elements, right: 4 elements␤  in block <unit> at /tmp/sz3CKfl93d:1␤␤»
09:29 GLRelia rakudo-moar 04a8f1: OUTPUT«Lists on either side of non-dwimmy hyperop of infix:<+> are not of the same length␤left: 3 elements, right: 4 elements␤  in block <unit> at /tmp/jhRdFT93QM:1␤␤»
09:29 laben finally ^^^
09:29 cognominal joined #perl6
09:29 FROGGS tadzik: I guess that is just an oversight, though, please also look at S22, perhaps it is called 'source' there
09:29 dalek ecosystem: f720f31 | tadzik++ | META.list:
09:29 dalek ecosystem: Retire MuEvent
09:29 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/f720f31937
09:30 domidumont joined #perl6
09:30 tadzik FROGGS: ah, it's support.source in S22, and panda just relies on non-S22 field here
09:31 tadzik S22 doesn't seem to know about "here's where you probably downloaded this from" :)
09:31 tadzik and it really only makes sense for the makeshit CPAN we created
09:31 FROGGS tadzik: hmmm... can we make this sane and e.g. let panda understand both?
09:31 tadzik sounds good
09:32 CIAvash joined #perl6
09:33 colomon m: say @a.pick(1).WHAT
09:33 camelia rakudo-moar a7c845: OUTPUT«5===SORRY!5=== Error while compiling /tmp/A55Kbj_K3M␤Variable '@a' is not declared␤at /tmp/A55Kbj_K3M:1␤------> 3say 7⏏5@a.pick(1).WHAT␤»
09:33 GLRelia rakudo-moar 04a8f1: OUTPUT«5===SORRY!5=== Error while compiling /tmp/cNRQL5qAxP␤Variable '@a' is not declared␤at /tmp/cNRQL5qAxP:1␤------> 3say 7⏏5@a.pick(1).WHAT␤»
09:33 colomon m: say (1..100).pick(1).WHAT
09:33 camelia rakudo-moar a7c845: OUTPUT«(List)␤»
09:33 GLRelia rakudo-moar 04a8f1: OUTPUT«(Seq)␤»
09:33 colomon Seq is correct for GLR, no?
09:34 jnthn colomon: yeah, correct
09:34 colomon okay, patching roast-glr
09:34 Ven joined #perl6
09:34 masak Ven! \o/
09:35 Ven \o/, spw2015 !
09:36 colomon \o/
09:36 dalek roast/glr: 3bddd39 | colomon++ | S32-list/pick.t:
09:36 dalek roast/glr: pick returns a Seq now.
09:36 dalek roast/glr: review: https://github.com/perl6/roast/commit/3bddd39369
09:36 colomon okay, hmmm … S32-list/classify.t ?
09:38 [Coke] m: say ~3.141592653589793238462643​38327950288419716939937510
09:38 camelia rakudo-moar a7c845: OUTPUT«3.1415926535897933824514​8937858175486326217651367188␤»
09:38 GLRelia rakudo-moar 30161a: OUTPUT«3.1415926535897933824514​8937858175486326217651367188␤»
09:38 colomon m: say classify( {.comb}, 100 .. 119, 104, 119 ).perl
09:38 camelia rakudo-moar a7c845: OUTPUT«Hash[Any,Any].new("1" => Hash[Any,Any].new("0" => Hash[Any,Any].new("0" => [100], "1" => [101], "2" => [102], "3" => [103], "4" => [104, 104], "5" => [105], "6" => [106], "7" => [107], "8" => [108], "9" => [109]), "1" => Hash[Any,Any].new("0" => [110], "1"…»
09:38 GLRelia rakudo-moar 30161a: OUTPUT«Hash[Any,Any].new()␤»
09:38 colomon that looks like a very real bug
09:38 dalek roast/glr: c86bb56 | FROGGS++ | S04-declarations/my.t:
09:38 dalek roast/glr: add test for RT #125371
09:38 dalek roast/glr: review: https://github.com/perl6/roast/commit/c86bb568c2
09:38 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=125371
09:39 Sqirrel joined #perl6
09:39 FROGGS [ptc]: it is fixed in rakudo/glr now
09:39 [Coke] m: say ~3.14159265358979323
09:39 camelia rakudo-moar a7c845: OUTPUT«3.14159265358979323␤»
09:39 GLRelia rakudo-moar 30161a: OUTPUT«3.14159265358979323␤»
09:40 timotimo ohai colomon, you're the gentleman in the corner with the apple laptop?
09:41 colomon indeed
09:41 timotimo yay
09:41 dalek rakudo/glr: 8f5dd0d | lizmat++ | src/core/io_operators.pm:
09:41 dalek rakudo/glr: Remove say(\x) / note(\x) candidates
09:41 dalek rakudo/glr:
09:41 dalek rakudo/glr: These seem like premature optimizations in the GLR, and they break
09:41 dalek rakudo/glr: behaviour.  They're just commented out for future reference / complete
09:41 dalek rakudo/glr: removal when we feel certain that's the way to go
09:41 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/8f5dd0dabd
09:41 itz "with the apple laptop" is hardly a unique identifier :)
09:41 colomon itz: sitting in the corner is
09:41 colomon ;)
09:41 nine colomon: we do have a free seat at the table here
09:41 TimToady not sure how many gentlemen there are here either...
09:42 colomon nine: right now I’m enjoying the breeze from the window I’m next to.  ;)
09:43 nine colomon: good point :)
09:43 * nine is looking at t/spec/S17-supply/stable.t
09:43 timotimo jnthn: i have an early implementation of nfa_to_statelist; i'd like you to take a look at it just to verify that what i'm doing with the MVMROOTs and such is sane and not totally wrong; would you like a diff?
09:44 tadzik moritz: ecocystem-api's projects.json actually generated from https://github.com/perl6/ecosys​tem/blob/master/updatelist.pl?
09:44 moritz tadzik: yes
09:45 tadzik ok, cool
09:45 colomon hmmm… classify-list is calling loads of nqp routines.
09:45 tadzik the "New run at" line from update.log made me wonder
09:45 colomon m: say List ~~ Seq
09:45 camelia rakudo-moar a7c845: OUTPUT«5===SORRY!5=== Error while compiling /tmp/JPpt_1Jn0f␤Undeclared name:␤    Seq used at line 1. Did you mean 'Set'?␤␤»
09:45 GLRelia rakudo-moar 8f5dd0: OUTPUT«False␤»
09:46 colomon m: say Seq ~~ List
09:46 camelia rakudo-moar a7c845: OUTPUT«5===SORRY!5=== Error while compiling /tmp/e9iLq7g7lh␤Expected a term, but found either infix ~~ or redundant prefix ~␤  (to suppress this message, please use a space like ~ ~)␤at /tmp/e9iLq7g7lh:1␤------> 3say Seq ~~7⏏5 List␤»
09:46 GLRelia rakudo-moar 8f5dd0: OUTPUT«False␤»
09:46 timotimo https://gist.github.com/timo/04a201f9c5fe9b5bd752 ← jnthn, this is the diff
09:46 moritz colomon: both are Iterable
09:46 colomon moritz: Aha!  That may be what we need here
09:47 colomon m: say “here”.comb.WHAT
09:47 camelia rakudo-moar a7c845: OUTPUT«(List)␤»
09:47 GLRelia rakudo-moar 8f5dd0: OUTPUT«Invocant requires an instance of type Seq, but a type object was passed.  Did you forget a .new?␤  in block <unit> at /tmp/yHFdJf4qYi:1␤␤»
09:47 colomon ???
09:47 timotimo er, huh? :)
09:47 colomon m: say “here”.comb ~~ Iterable
09:47 camelia rakudo-moar a7c845: OUTPUT«True␤»
09:47 GLRelia rakudo-moar 8f5dd0: OUTPUT«True␤»
09:47 colomon m: say “here”.comb ~~ Seq
09:47 camelia rakudo-moar a7c845: OUTPUT«5===SORRY!5=== Error while compiling /tmp/w7DGW_RUBf␤Undeclared name:␤    Seq used at line 1. Did you mean 'Set'?␤␤»
09:47 GLRelia rakudo-moar 8f5dd0: OUTPUT«True␤»
09:47 colomon m: say “here”.comb ~~ List
09:47 camelia rakudo-moar a7c845: OUTPUT«True␤»
09:47 GLRelia rakudo-moar 8f5dd0: OUTPUT«False␤»
09:48 colomon m: say nqp::istype(“here”.comb, Iterable)
09:48 camelia rakudo-moar a7c845: OUTPUT«======================================​=========================================␤The use of nqp::operations has been deprecated for non-CORE code.  Please␤change your code to not use these non-portable functions.  If you really want␤to keep using nqp:…»
09:48 GLRelia rakudo-moar 8f5dd0: OUTPUT«======================================​=========================================␤The use of nqp::operations has been deprecated for non-CORE code.  Please␤change your code to not use these non-portable functions.  If you really want␤to keep using nqp:…»
09:48 colomon bother
09:49 rindolf joined #perl6
09:49 colomon will nqp::istype work for subtypes?
09:49 dalek rakudo/nom: eab03c3 | coke++ | src/core/ (2 files):
09:49 dalek rakudo/nom: RT #123760 - add Str matchers for comb
09:49 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/eab03c3dbf
09:49 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=123760
09:49 timotimo you just have to "use nqp" in front, colomon
09:49 timotimo that'll shut it up
09:49 laben m: use nqp; say nqp::istype(“here”.comb, Iterable)
09:49 dalek roast: 3871847 | coke++ | S32-str/comb.t:
09:49 dalek roast: RT #123760 update/add tests for Str-based comb
09:49 dalek roast: review: https://github.com/perl6/roast/commit/3871847456
09:49 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=123760
09:49 camelia rakudo-moar a7c845: OUTPUT«1␤»
09:49 GLRelia rakudo-moar 8f5dd0: OUTPUT«1␤»
09:49 colomon timotimo++
09:49 timotimo sadly, it puts the "…" right in front of that hint
09:49 timotimo because of all those "======"
09:51 colomon timotimo: extra sadly, when I got the same message on my local perl6, I didn’t read the second half (with use nqp) because the first half of the message was identical to what I got on GLRelia.
09:51 laben *elia runs on RESTRICTED, no? we could reduce the number of === in that
09:51 andreoss joined #perl6
09:51 sno joined #perl6
09:51 timotimo heh. :)
09:51 timotimo that seems more like a hack than an elegant solution
09:52 timotimo we could just have a CAMELIA.setting instead :)
09:52 colomon so, my first rakudo-glr patch coming up in a min.  (fixed classify, timotimo++)
09:52 timotimo very cool!
09:52 colomon errr, more than a minute, spectesting is slow.
09:52 laben make number of === configurable at runtime/builtime?
09:52 dalek ecosystem: 058a95d | tadzik++ | META.list:
09:52 dalek ecosystem: Retire pandabrew
09:52 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/058a95d821
09:53 laben i think having that at runtime with a flag or a $*VAR could be the best solution
09:53 [Coke] r: constant @t = [\+] 1..5; say (0,@t Z^.. @t).perl;
09:53 GLRelia rakudo-moar 8f5dd0: OUTPUT«This Seq has already been iterated, and its values consumed␤  in block <unit> at /tmp/tmpfile:1␤␤»
09:53 GLRelia ..rakudo-jvm 8f5dd0: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
09:53 camelia rakudo-{moar,jvm} a7c845: OUTPUT«(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)␤»
09:53 ShimmerFairy I think that's actually a hard-coded ===== string for the nqp message, IIRC.
09:53 andreoss left #perl6
09:53 andreoss joined #perl6
09:53 colomon JOBS=4?
09:54 andreoss left #perl6
09:55 dalek doc: ba14f7b | paultcochrane++ | htmlify.p6:
09:55 dalek doc: Work around rakudobug RT #125921
09:55 dalek doc:
09:55 dalek doc: FROGGS++ for adding this to rakudobug.  This is also now fixed in
09:55 dalek doc: rakudo/glr, so we can change the code back if desired after the glr merge.
09:55 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=125921
09:55 dalek doc: review: https://github.com/perl6/doc/commit/ba14f7b273
09:56 colomon will I get extra spectest failures because of TEST_JOBS=4 ?
09:57 timotimo usually not
09:57 colomon \o/
09:57 colomon I must say, running with TEST_JOBS=4 is making my laptop heat up….  :)
09:57 timotimo but nine has found a case where it does break a test on his machine
09:58 timotimo yeah, i can imagine
09:58 [Coke] started adding [GLR] to tickets that are/should be/maybe GLR related. - https://rt.perl.org/Ticket/Display.html?id=113870
09:58 timotimo my laptop has a derf where it odesn't properly turn up the fan speed
09:58 ShimmerFairy It tends to break lock.t for me, in S17
09:58 timotimo so when i'm alone i just tend to "echo 'level disengaged' | sude tee fan"
09:59 timotimo ("level 7" would be the highest speed the fan can turn on via regulation, but "level disengaged" just turns off the regulator and it just goes another 2x faster than "fastest speed")
09:59 araujo joined #perl6
10:00 [Coke] m: <a a>.squish(:with(-> $a, $b {1})).perl.say
10:00 camelia rakudo-moar a7c845: OUTPUT«()␤»
10:00 GLRelia rakudo-moar 8f5dd0: OUTPUT«().Seq␤»
10:01 moritz uhm, why doesn't that return ('a') or ('a').Seq ?
10:02 * colomon wouldn’t be surprised if squish is broken in the same way classify was (well, is, as I haven’t pushed yet)
10:03 dalek perl6-examples: 5e42f33 | (Steve Mynott)++ | categories/cookbook/16proce​sses/16-14-send-signal.pl:
10:03 dalek perl6-examples: simple send signal example using nativecall
10:03 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/5e42f33ef6
10:04 moritz colomon: push, push, push! (https://www.youtube.com/watch?v=ICTmiUhea3g :-)
10:04 colomon moritz: I’m spectesting first, and that’s slow
10:04 timotimo maybe we should have brought a very beefy machine to put into some corner where no ears are close by to do all our spec testing for us
10:05 moritz colomon: I tend to spectest my GLR changes on hack, because it's faster than my laptop :-)
10:05 timotimo like a 32-core machine local to the venue
10:05 colomon timotimo, moritz: yeah, I’m spinning on spectests on my big box back in Michigan now…  ;)
10:05 colomon my TEST_JOBS=4 run on OS X seems like it might be wedged on t/spec/S17-procasync/no-runaway-file-limit.t  ?
10:08 FROGGS m: sub foo(:xen(:$xin)) { say $xin }; foo xin => 23
10:08 camelia rakudo-moar a7c845: OUTPUT«23␤»
10:08 GLRelia rakudo-moar 8f5dd0: OUTPUT«23␤»
10:08 dalek perl6-examples: 7c1187a | (Sue Spence)++ | categories/cookbook/01stri​ngs/01-00introduction.pl:
10:08 dalek perl6-examples: Add examples for $obj.say. Fix minor typo.
10:08 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/7c1187a95c
10:08 dalek perl6-examples: db65979 | (Sue Spence)++ | categories/cookbook/01strings/01-01substrings.pl:
10:08 dalek perl6-examples: Pugs -> Rakudo
10:08 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/db65979dbe
10:08 dalek perl6-examples: 234f01b | (Sue Spence)++ | categories/cookbook/01string​s/01-04converting-values.pl:
10:08 dalek perl6-examples: Answered some commenty questions, added a unicode char
10:08 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/234f01be4a
10:08 dalek perl6-examples: 2049f7e | (Steve Mynott)++ | categories/cookbook/01strings/01-0 (3 files):
10:08 dalek perl6-examples: Merge pull request #27 from virtualsue/master
10:08 dalek perl6-examples:
10:08 dalek perl6-examples: Minor updates
10:08 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/2049f7ebf1
10:09 aborazmeh joined #perl6
10:09 aborazmeh joined #perl6
10:09 [ptc] itz: you beat me to it :-)
10:10 masak TimToady, jnthn: Ven and I think that maybe we are too attached to push's current semantics.
10:10 Sqirrel joined #perl6
10:10 masak and that it's not so unreasonable to do `@a.push(|@b)`
10:11 tadzik masak: I reckon you'll like that: https://github.com/tadzik/panda/commit/cfe6ccb8c54 :)
10:11 masak (not saying we're right, just that it's something worth considering)
10:11 masak tadzik: yaaaaay
10:11 [Coke] m: say ~('Y'..'z')
10:11 camelia rakudo-moar a7c845: OUTPUT«Y Z [ \ ] ^ _ ` a b c d e f g h i j k l m n o p q r s t u v w x y z␤»
10:11 GLRelia rakudo-moar 8f5dd0: OUTPUT«Y Z [ \ ] ^ _ ` a b c d e f g h i j k l m n o p q r s t u v w x y z␤»
10:11 itz [ptc]: she was sitting opposite :)
10:11 ShimmerFairy masak: Looking at it, it feels natural at least, considering the original meaning of prefix:<|> :)
10:11 tadzik it's not even *that* drastically slow
10:11 tadzik loading HTTP::UA goes from 0.8 secs to 5.4 on my machine
10:13 [Coke] m: say ~('Y'..'_')
10:13 camelia rakudo-moar a7c845: OUTPUT«Y Z [ \ ] ^ _␤»
10:13 GLRelia rakudo-moar 8f5dd0: OUTPUT«Y Z [ \ ] ^ _␤»
10:13 [Coke] m: say ~('Y'..'AB')
10:13 camelia rakudo-moar a7c845: OUTPUT«␤»
10:13 GLRelia rakudo-moar 8f5dd0: OUTPUT«␤»
10:14 jnthn masak: |@b is going to suck if the thing is 100s of items long
10:14 dalek rakudo/glr: ca6b9e6 | colomon++ | src/core/Hash.pm:
10:14 dalek rakudo/glr: Switch List to Iterable.
10:14 dalek rakudo/glr:
10:14 dalek rakudo/glr: Pretty sure this was what was originally meant, and Iterable
10:14 dalek rakudo/glr: treats Seq in the same fashion.  Fixes the spectest fail in
10:14 dalek rakudo/glr: S32-list/classify.t.
10:14 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/ca6b9e6542
10:14 * colomon never did get a useful spectest run, but is pretty sure the patch is correct.
10:14 jnthn masak: Of note, on JVM it'll actually break if you make the thing more than about 255 items long.
10:15 Ven joined #perl6
10:15 jnthn masak: Because that is the maximum number of args you can pass on the JVM
10:15 nine colomon: +1
10:15 dalek perl6-examples: 8dd4a70 | (Steve Mynott)++ | categories/cookbook/16process​es/16-18-catch-control-c.pl:
10:15 dalek perl6-examples: Catch ctrl-c to ignore or execute code example
10:15 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/8dd4a7098b
10:15 dalek perl6-examples: 8f84888 | (Steve Mynott)++ | categories/cookbook/01strings/01-0 (3 files):
10:15 dalek perl6-examples: Merge branch 'master' of github.com:perl6/perl6-examples
10:15 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/8f848883fb
10:15 dalek roast: 646e60f | coke++ | S03-operators/range.t:
10:15 dalek roast: RT #124542 #124543 - remove/update dodgy tests.
10:15 dalek roast: review: https://github.com/perl6/roast/commit/646e60fd35
10:15 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=124542
10:16 nine colomon: similar bugs may be in Baggy.pm
10:16 colomon nine: I will investigate
10:16 nine :)
10:17 nine Haven't really looked at Bag/Mix/... bugs, because I don't yet know what they are :)
10:17 ab6tract joined #perl6
10:17 colomon nine++: looks like exactly the same code that I just changed, good call
10:18 colomon not sure if this is triggering a spectest failure, which suggests we need a new test first...
10:18 [Coke] Down to 1.042K tickets.
10:18 nine colomon: t/spec/S02-types/bag.rakudo.moar and t/spec/S02-types/baghash.rakudo.moar are failing right now
10:18 nine colomon: maybe your fix helps there
10:18 colomon oooo!
10:18 colomon nine++
10:19 nine colomon++ # fixing things :)
10:19 colomon nine: it feels really good to be digging back into p6 in a useful way.  :)
10:19 jkva joined #perl6
10:19 FROGGS m: say <foo bar baz> Z=> 1 xx *
10:19 camelia rakudo-moar a7c845: OUTPUT«foo => 1 bar => 1 baz => 1␤»
10:19 GLRelia rakudo-moar 8f5dd0: OUTPUT«foo => 1bar => 1baz => 1␤»
10:19 FROGGS m: say (<foo bar baz> Z=> 1 xx *).perl
10:19 camelia rakudo-moar a7c845: OUTPUT«(:foo(1), :bar(1), :baz(1))␤»
10:19 GLRelia rakudo-moar 8f5dd0: OUTPUT«(:foo(1), :bar(1), :baz(1)).Seq␤»
10:19 timotimo nqp-m: say NQPList
10:19 camelia nqp-moarvm: OUTPUT«Confused at line 2, near "say NQPLis"␤   at gen/moar/stage2/NQPHLL.nqp:516  (/home/camelia/rakudo-inst-2/share​/nqp/lib/NQPHLL.moarvm:panic:105)␤ from gen/moar/stage2/NQP.nqp:921  (/home/camelia/rakudo-inst-2/share/​nqp/lib/nqp.moarvm:comp_unit:872)␤ from gen/…»
10:19 timotimo nqp-m: say(NQPList)
10:19 camelia nqp-moarvm: OUTPUT«␤»
10:20 timotimo nqp-m: say(NQPMu)
10:20 camelia nqp-moarvm: OUTPUT«␤»
10:20 virtualsue joined #perl6
10:20 lizmat m: say "foo".rindex("f",1e35)   # FROGGS, shouldn't that die now ?
10:20 FROGGS m: my %h = <foo bar baz> Z=> 1 xx *; for %h.keys { .say }
10:20 camelia rakudo-moar a7c845: OUTPUT«Nil␤»
10:20 GLRelia rakudo-moar 8f5dd0: OUTPUT«0␤»
10:20 camelia rakudo-moar a7c845: OUTPUT«bar␤baz␤foo␤»
10:20 GLRelia rakudo-moar 8f5dd0: OUTPUT«bar␤baz␤foo␤»
10:20 timotimo ah, needs to be NQPArray
10:20 timotimo nqp-m: say(NQPArray)
10:20 camelia nqp-moarvm: OUTPUT«␤»
10:20 timotimo that doesn't tell me anything, though
10:20 FROGGS lizmat: I don't think so
10:21 FROGGS lizmat: the offset is longer than the string, so you get Nil... err, supposed to get Nil
10:21 colomon nine: hmm, I think the /S02-types/bag.t are likely to be glitches in the bag constructor.  I will look at these once I look at classify.  :)
10:21 masak FROGGS: Ven and I are discussing https://gist.github.com/masak/3438222 now
10:22 FROGGS masak: ool
10:22 FROGGS cool*
10:22 lizmat FROGGS: ok
10:23 FROGGS lizmat: so it is a bug that it does not return Nil
10:23 lizmat yup
10:23 tadzik masak: I feel like both are just avoiding the need of a C parser that'll automatically generate this stuff :)
10:23 FROGGS m: use nqp; say nqp::unbox_i 1e35.Int
10:23 camelia rakudo-moar eab03c: OUTPUT«0␤»
10:23 GLRelia rakudo-moar ca6b9e: OUTPUT«0␤»
10:23 FROGGS :o(
10:23 lizmat yeah :-(
10:24 jnthn FROGGS: Hm, didn't you commit the check for that overflow?
10:24 FROGGS potentially it fits into an MVMuint64, and will later be casted into an MVMint64
10:24 lizmat FROGGS: so *that*'s the bug, right ?
10:24 FROGGS lizmat: aye
10:24 jnthn Oh, 1e35 easily fits, no?
10:24 colomon lizmat: do I trigger the Baggy classify-list routine by using into=>Bag.new when I call classify?
10:24 [Coke] Is this related to the failure on os x about the overflow check?
10:24 jnthn Oh, maybe not :)
10:24 muraiki joined #perl6
10:25 lizmat colomon: without looking at the code, I think it does
10:25 FROGGS m: say 1e35.Int.fmt('%b').chars
10:25 camelia rakudo-moar eab03c: OUTPUT«117␤»
10:25 GLRelia rakudo-moar ca6b9e: OUTPUT«117␤»
10:25 FROGGS m: say 1e35.Int.fmt('%b')
10:25 camelia rakudo-moar eab03c: OUTPUT«10011010000100110000101110010110001​110100110110000010000000000000000000000000​0000000000000000000000000000000000000000␤»
10:25 GLRelia rakudo-moar ca6b9e: OUTPUT«10011010000100110000101110010110001​110100110110000010000000000000000000000000​0000000000000000000000000000000000000000␤»
10:25 [Coke] t/spec/S02-types/declare.t - fails test #19
10:25 lizmat [Coke]: is that a new fail ?
10:25 FROGGS m: use nqp; say nqp::unbox_i 1e36.Int
10:25 camelia rakudo-moar eab03c: OUTPUT«0␤»
10:25 GLRelia rakudo-moar ca6b9e: OUTPUT«0␤»
10:25 FROGGS m: use nqp; say nqp::unbox_i 1e34.Int
10:25 camelia rakudo-moar eab03c: OUTPUT«3458764513820540928␤»
10:25 GLRelia rakudo-moar ca6b9e: OUTPUT«3458764513820540928␤»
10:25 FROGGS m: use nqp; say nqp::unbox_i (-1e35).Int
10:25 camelia rakudo-moar eab03c: OUTPUT«0␤»
10:25 GLRelia rakudo-moar ca6b9e: OUTPUT«0␤»
10:26 FROGGS ??
10:26 [Coke] been failing since the test was added.
10:26 colomon lizmat: "... a large array goes into a bar - I mean bag - with 50k elems and lives"  :)
10:26 FROGGS this was working just the other day
10:26 FROGGS $ perl6 -e 'use nqp; say nqp::unbox_i 1e36.Int'
10:26 FROGGS Cannot unbox 120 bit wide bigint into native integer
10:26 [Coke] I mentioned it previously in backlog, but haven't opened a ticket, my bad.
10:26 FROGGS maybe we are on an old moarvm?
10:27 lizmat $ 6 'use nqp; say nqp::unbox_i 1e36.Int'
10:27 lizmat 0
10:27 lizmat that's the latest rakudo on GLR and OS X
10:27 timotimo jnthn: how can i be segfaulting on https://gist.github.com/timo/04a201f9​c5fe9b5bd752#file-gistfile1-txt-L234 - this line? :o
10:27 FROGGS I'll bump moar revision for nqp+glr
10:28 dalek perl6-examples: 437a6c4 | paultcochrane++ | categories/euler/prob (4 files):
10:28 dalek perl6-examples: Purge hard tabs in examples
10:28 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/437a6c4b8e
10:28 dalek nqp: 02910ff | FROGGS++ | tools/build/MOAR_REVISION:
10:28 dalek nqp: bump moar revision
10:28 dalek nqp: review: https://github.com/perl6/nqp/commit/02910ff48e
10:29 dalek rakudo/glr: 23b3d29 | FROGGS++ | tools/build/NQP_REVISION:
10:29 dalek rakudo/glr: bump nqp/moar revision
10:29 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/23b3d29da6
10:29 FROGGS lizmat / [Coke]: this should do I hope
10:29 JimmyZ timotimo: thre is a  MVMROOT, which cause the line is wrong.
10:29 |Tux| can I see from perl6 --version if it is nom or glr?
10:29 yoleaux 26 Aug 2015 17:43Z <laben> |Tux|: i fixed up to test file 50_utf8.t, i'm now at 55_combi.t where the problem is trying to assign a list of chars (instead of Str) to both $!esc and $!quo. what did you mean to do?
10:29 yoleaux 26 Aug 2015 21:27Z <laben> |Tux|: i fixed 60_samples.t after skipping 55_combi.t, i got up to 90/114 tests of 77_getall.t passing, still needs a lot of work
10:30 lizmat FROGGS: just pulled, building now
10:30 timotimo JimmyZ: so in reality the problem must be on a different line?
10:30 timotimo so ... how do i find the right line? turn on disassembly? X_x
10:30 JimmyZ timotimo: yeah, change it to the function ...
10:30 FROGGS |Tux|: not reliably I guess
10:31 timotimo you mean MVM_gc_temp_roots_push or what it's called?
10:31 JimmyZ yes
10:31 dalek perl6-examples: 6e50a73 | paultcochrane++ | bin/ (2 files):
10:31 dalek perl6-examples: Remove now outdated utility scripts
10:31 dalek perl6-examples:
10:31 dalek perl6-examples: Adding a collaborator can happen either via the GitHub website, or if it has
10:31 dalek perl6-examples: to be scripted, via the GitHub api.  The twitter2irc script doesn't seem to
10:31 dalek perl6-examples: be used at all.
10:31 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/6e50a73aa0
10:31 [Coke] FROGGS: my failure is on nom.
10:31 FROGGS [Coke]: I can bump nom too, hold on
10:32 timotimo i'll have a look
10:32 JimmyZ timotimo: or use MSVC, which seems to be a little friendly to macro
10:32 JimmyZ timotimo: jnthn may know :P
10:33 dalek rakudo/nom: a46b09c | FROGGS++ | tools/build/NQP_REVISION:
10:33 dalek rakudo/nom: bump nqp/moar revision
10:33 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/a46b09c491
10:33 FROGGS [Coke]: ^^
10:34 JimmyZ anyone could merge https://github.com/rakudo/rakudo/pull/506?
10:34 JimmyZ t is safe now
10:36 [Coke] 1) build rakudo 2) install it. 3) panda rebootstrap 4) install linenoise ... 5) build directory's ./perl6-m now supports linenoise?
10:36 timotimo JimmyZ: now the segfault comes from the next MVMROOT …
10:36 JimmyZ timotimo: you need changed all the MVMROOT of this  file, as far as I know
10:36 [Coke] ah, yes, because the ./perl6-m handler is using the installed prefix as a libpath.
10:37 timotimo $2 = {fates = 0xc30180, num_states = 0, num_state_edges = 0x38000000000001, states = 0x6075b8}
10:37 timotimo that looks sane!
10:37 timotimo no problems here!
10:38 Ulti /win 2
10:38 Ulti erk
10:38 * colomon cannot find any sign that roast actually tests Baggy.classify-list (even indirectly)
10:38 [ptc] itz: found lots of examples-related processes were still running from 22nd of August.  Have just killed them all.  Running update-and-sync by hand should hopefully work again now.
10:38 moritz colomon: then that should be corrected :-)
10:39 colomon moritz: I’m working on it.  but it’s tricky, I need to figure out what it is supposed to do.  ;)
10:39 * timotimo grabs one of these bites
10:39 moritz colomon: :-)
10:39 colomon aha!
10:39 jnthn timotimo: Is the debugger pointing at the wrong line 'cus it's confused by the macro?
10:42 colomon m: say (1..10).classify(* %% 2)
10:42 camelia rakudo-moar eab03c: OUTPUT«False => 1 3 5 7 9, True => 2 4 6 8 10␤»
10:42 GLRelia rakudo-moar 23b3d2: OUTPUT«True => 2 4 6 8 10False => 1 3 5 7 9␤»
10:44 colomon m: my %b := BagHash.new; say (1..10).classify(* %% 2, :into(%b))
10:44 camelia rakudo-moar eab03c: OUTPUT«BagHash.new(False(5), True(5))␤»
10:44 GLRelia rakudo-moar 23b3d2: OUTPUT«BagHash.new(True(5), False(5))␤»
10:44 colomon lizmat: do those results look correct to you?  :)
10:44 * colomon did not change anything to get them.
10:44 timotimo jnthn: jimmyz suggested that and i've removed the macros, now i get an nfa with a num_state_edges that's set to an insane number d|
10:44 timotimo :|
10:45 lizmat colomon: looks sane to me?
10:45 colomon lizmat: groovy.
10:45 colomon # expected: ("even"=>2,"odd"=>2).BagHash
10:45 colomon #      got: ("even"=>2,"odd"=>2).BagHash
10:45 timotimo oh
10:46 timotimo num_state_edges is actually supposed to be a pointer to an array
10:46 colomon is-deeply maybe doesn’t handle BagHashes?
10:46 timotimo a pointer ending in a "1" doesn't seem like a sane pointer :|
10:46 lizmat does is-deeply handle different hash orders at all?
10:47 lizmat lunch&
10:49 timotimo is-deeply lunch
10:51 ShimmerFairy colomon: is-deeply does eqv, unless I'm mistaken, so maybe check that?
10:51 btyler_ I think I found something memory-leaky: https://gist.github.com/kanat​ohodets/b3692b4e41a3e47620a1
10:52 FROGGS m: my @a = [$["foo", "bar"], "baz"]; say @a X=> 1 # how do I make that outputting "foo => 1bar => 1baz => 1" ?
10:52 camelia rakudo-moar a46b09: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '⏏foo' (indicated by ⏏)␤  in block <unit> at /tmp/heKhShyIC3:1␤␤Actually thrown at:␤  in block <unit> at /tmp/heKhShyIC3:1␤␤»
10:52 GLRelia rakudo-moar 23b3d2: OUTPUT«foo bar => 1baz => 1␤»
10:52 ShimmerFairy m: my @a = [$["foo", "bar"], "baz"]; say (@a X=> 1).perl
10:52 camelia rakudo-moar a46b09: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '⏏foo' (indicated by ⏏)␤  in block <unit> at /tmp/yQMaBQSapr:1␤␤Actually thrown at:␤  in block <unit> at /tmp/yQMaBQSapr:1␤␤»
10:52 GLRelia rakudo-moar 23b3d2: OUTPUT«($["foo", "bar"] => 1, :baz(1)).Seq␤»
10:53 jnthn timotimo: Sounds like memory corruption of some kind then... :S
10:53 |Tux| laben, /me is ready for some digging while a heavy process is running.
10:53 colomon timotimo: a pointer ending in 1 is just a bit odd, that’s all.
10:53 FROGGS hehe
10:53 FROGGS even that, aye
10:53 btyler_ jnthn: FROGGS pointed me your way on that leaky thing -- anything I can do to narrow down the hunt?
10:53 [Coke] btyler_: verified, that seems to chew memory on my OS X box.
10:54 FROGGS jnthn: btw, he says it does not leak when $count is a constant... leaking code is: my $count = 0; loop { 1 if $count %% 1 }
10:54 FROGGS so it can't be the %% op
10:55 nine colomon: BagHash or one of its base classes may need a WHICH
10:56 Sqirrel joined #perl6
10:56 jnthn Well, 0 %% 1 probably gets constant-folded
10:56 jnthn m say 0 %% 1
10:56 |Tux| does «class C does Iterable does Positional {» in glr need additional roles to also do Hash lookups with AT-KEY ?
10:56 jnthn m: say 0 %% 1
10:56 camelia rakudo-moar a46b09: OUTPUT«True␤»
10:56 GLRelia rakudo-moar 23b3d2: OUTPUT«True␤»
10:57 jnthn |Tux|: Things you can hash-index into normally does Associative
10:58 timotimo i see what you did there, colomon :)
10:58 jnthn I'd probably valgrind it
10:59 |Tux| jnthn, and what should method "of" be implementing in that case?
10:59 colomon timotimo: I was disturbingly pleased by that one.  ;)
10:59 jnthn |Tux|: What type constraint anything you store in it has; return Mu if you aren't implementing a typed data structure
11:00 laben |Tux|: i can send you a gist with the changes, too busy to investigate right now
11:00 |Tux| laben, yes please
11:00 rurban joined #perl6
11:04 laben |Tux|: here is it, they're all together https://gist.github.com/e3ab90806f82cd349493
11:04 |Tux| thanks
11:04 laben some seems precedence problems, maybe due to the slang
11:05 Ven joined #perl6
11:06 colomon m: %b := BagHash.new(); %b.classify-list( {.comb}, 20..40 ); say %b.perl
11:06 camelia rakudo-moar a46b09: OUTPUT«5===SORRY!5=== Error while compiling /tmp/5AlhUIubL2␤Variable '%b' is not declared␤at /tmp/5AlhUIubL2:1␤------> 3<BOL>7⏏5%b := BagHash.new(); %b.classify-list( {␤»
11:06 GLRelia rakudo-moar 23b3d2: OUTPUT«5===SORRY!5=== Error while compiling /tmp/z7RZQ2QFmg␤Variable '%b' is not declared␤at /tmp/z7RZQ2QFmg:1␤------> 3<BOL>7⏏5%b := BagHash.new(); %b.classify-list( {␤»
11:06 colomon m: my %b := BagHash.new(); %b.classify-list( {.comb}, 20..40 ); say %b.perl
11:06 camelia rakudo-moar a46b09: OUTPUT«postcircumfix:<{ }> not defined for type Int␤  in block <unit> at /tmp/jOo0Ksq1et:1␤␤Actually thrown at:␤  in any  at src/gen/m-Metamodel.nqp:2867␤  in block <unit> at /tmp/jOo0Ksq1et:1␤␤»
11:06 GLRelia rakudo-moar 23b3d2: OUTPUT«Method 'succ' not found for invocant of class 'List'␤  in block <unit> at /tmp/TaCRMRZ234:1␤␤»
11:06 colomon lizmat: I think this operation is broken in both both nom and glr.
11:07 timotimo oh, duh
11:07 colomon lizmat: but then, I’m not 100% sure what it is supposed to do
11:07 timotimo if num_states is 0, of course the states array isn't defined memory
11:08 dalek rakudo/glr: f761858 | jnthn++ | src/Perl6/Actions.nqp:
11:08 dalek rakudo/glr: (1, 2 ; 3, 4) now makes a List of List.
11:08 dalek rakudo/glr:
11:08 dalek rakudo/glr: Literally. No LoL.
11:08 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/f761858b8b
11:10 JimmyZ I thought LoL is short for List of List ...
11:11 tadzik we python now
11:12 jnthn JimmyZ: Now there's no type LoL though :)
11:12 sergot hi #perl6!
11:12 tadzik oh hai
11:12 llfourn joined #perl6
11:13 colomon joined #perl6
11:14 dalek perl6-roast-data: 74100de | coke++ | / (2 files):
11:14 dalek perl6-roast-data: another glr run for 'today'
11:14 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/74100de4fd
11:14 timotimo i forgot a single "re-grab the NFA pointer from the nfa root if the gc has kicked in since"
11:15 timotimo argh!
11:16 [Coke] m: 118901 - 118302 # of tests moar-jit is passing vs. moar-glr
11:16 camelia rakudo-moar a46b09: OUTPUT«WARNINGS:␤Useless use of "-" in expression "118901 - 118302" in sink context (line 1)␤»
11:16 GLRelia rakudo-moar 23b3d2: OUTPUT«WARNINGS:␤Useless use of "-" in expression "118901 - 118302" in sink context (line 1)␤»
11:16 [Coke] m: say 118901 - 118302 # of tests moar-jit is passing vs. moar-glr
11:16 camelia rakudo-moar a46b09: OUTPUT«599␤»
11:16 GLRelia rakudo-moar 23b3d2: OUTPUT«599␤»
11:16 timotimo the code to do that was actually totally bogus %)
11:19 [Coke] m: for (9...1) X (9...1) X (9...1) X (9...1) X (9...1) X (9,7,3) -> $a,$b,$c,$d,$e,$f {}
11:19 GLRelia rakudo-moar f76185: OUTPUT«Method 'pull-one' not found for invocant of class 'List'␤  in block <unit> at /tmp/P1lB7Eam_c:1␤␤»
11:19 camelia rakudo-moar a46b09: OUTPUT«Too few positionals passed; expected 6 arguments but got 3␤  in block <unit> at /tmp/z2z1NpgXqB:1␤␤»
11:19 dalek rakudo/nom: 5d3fe6a | moritz++ | t/spectest.data:
11:19 dalek rakudo/nom: Comment out IO-Socket-Async.t for the release
11:19 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5d3fe6a40c
11:19 [Coke] ^^ jnthn - that's from your RT #114918
11:19 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=114918
11:20 colomon m: my %b := BagHash.new(); %b.classify-list( {.comb}, 20..40 ); say %b.perl
11:20 camelia rakudo-moar a46b09: OUTPUT«postcircumfix:<{ }> not defined for type Int␤  in block <unit> at /tmp/Oojcn28zVs:1␤␤Actually thrown at:␤  in any  at src/gen/m-Metamodel.nqp:2867␤  in block <unit> at /tmp/Oojcn28zVs:1␤␤»
11:20 GLRelia rakudo-moar f76185: OUTPUT«Method 'succ' not found for invocant of class 'List'␤  in block <unit> at /tmp/oxvcsh2p4W:1␤␤»
11:20 colomon masak: ^^
11:20 * masak submits rakudobug
11:21 colomon m: my %b; %b.classify-list( {.comb}, 20..40 ); say %b.perl
11:21 camelia rakudo-moar a46b09: OUTPUT«{"2" => {"0" => [20], "1" => [21], "2" => [22], "3" => [23], "4" => [24], "5" => [25], "6" => [26], "7" => [27], "8" => [28], "9" => [29]}, "3" => {"0" => [30], "1" => [31], "2" => [32], "3" => [33], "4" => [34], "5" => [35], "6" => [36], "7" => [37], "8" …»
11:21 GLRelia rakudo-moar f76185: OUTPUT«{"2" => ${"0" => $[20], "1" => $[21], "2" => $[22], "3" => $[23], "4" => $[24], "5" => $[25], "6" => $[26], "7" => $[27], "8" => $[28], "9" => $[29]}, "3" => ${"0" => $[30], "1" => $[31], "2" => $[32], "3" => $[33], "4" => $[34], "5" => $[35], "6" => $[36]…»
11:22 [Coke] r: []=5; say "alive"
11:22 GLRelia rakudo-moar f76185: OUTPUT«alive␤»
11:22 GLRelia ..rakudo-jvm f76185: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
11:22 camelia rakudo-{moar,jvm} a46b09: OUTPUT«alive␤»
11:22 Ven joined #perl6
11:24 masak m: my %b := BagHash.new(); say %b{'foo'}
11:24 camelia rakudo-moar a46b09: OUTPUT«0␤»
11:24 GLRelia rakudo-moar f76185: OUTPUT«0␤»
11:24 FROGGS m: say (42, 3).WHAT
11:24 camelia rakudo-moar a46b09: OUTPUT«(Parcel)␤»
11:24 GLRelia rakudo-moar f76185: OUTPUT«Invocant requires an instance of type List, but a type object was passed.  Did you forget a .new?␤  in block <unit> at /tmp/7_bfpx7ZdU:1␤␤»
11:25 moritz huh.
11:25 [Coke] my @RT120506-bind := Array[Array[Bool]].new($(Array[Bool].new(True, False, True)), $(Array[Bool].new(True))); use Test; is-deeply @RT120506-bind[0, 1]».Parcel, ((True, False, True), (True,)), "urk?"
11:25 [Coke] m: my @RT120506-bind := Array[Array[Bool]].new($(Array[Bool].new(True, False, True)), $(Array[Bool].new(True))); use Test; is-deeply @RT120506-bind[0, 1]».Parcel, ((True, False, True), (True,)), "urk?"
11:25 camelia rakudo-moar a46b09: OUTPUT«not ok 1 - urk?␤␤# Failed test 'urk?'␤# at /tmp/RUm0uDPc9K line 1␤# expected: $((Bool::True, Bool::False, Bool::True), (Bool::True,))␤#      got: $(Array[Bool].new(Bool::True, Bool::False, Bool::True), Array[Bool].new(Bool::True))␤»
11:25 GLRelia rakudo-moar f76185: OUTPUT«Method 'Parcel' not found for invocant of class 'Bool'␤  in block <unit> at /tmp/DwdNCPC3u2:1␤␤»
11:25 laben m: for (1..2) X (1..2) -> $a, $b {}; say "OK"; for (1..9) X (1..9) -> $a, $b {}
11:25 camelia rakudo-moar a46b09: OUTPUT«OK␤»
11:25 GLRelia rakudo-moar f76185: OUTPUT«OK␤Too few positionals passed; expected 2 arguments but got 1␤  in block <unit> at /tmp/Lw8XlPGktm:1␤␤»
11:25 laben why ^^^ is it different between 1..2 and 1..9?
11:26 colomon what does 1..3 do?
11:26 masak laben: even vs odd number of elements
11:27 masak laben: you're looping over the tuples themselves, and there are 4 and 81
11:27 laben masak: why does it even matter, shouldn't there be a single list created by X?
11:27 [Coke] m: my @m <== sort <== ("c", "b", "a") ; say @m;
11:27 camelia rakudo-moar a46b09: OUTPUT«5===SORRY!5=== Error while compiling /tmp/RbxmL4VmsT␤Unable to parse expression in quote words; couldn't find final '>'␤        ␤at /tmp/RbxmL4VmsT:1␤------> 3m <== sort <== ("c", "b", "a") ; say @m;7⏏5<EOL>␤    expecting any of:␤    …»
11:27 GLRelia rakudo-moar f76185: OUTPUT«5===SORRY!5=== Error while compiling /tmp/82aR3aEFpM␤Unable to parse expression in quote words; couldn't find final '>'␤        ␤at /tmp/82aR3aEFpM:1␤------> 3m <== sort <== ("c", "b", "a") ; say @m;7⏏5<EOL>␤    expecting any of:␤    …»
11:27 masak nope, not without `flat`
11:27 laben doesnt X create a list of tuples?
11:27 masak that's what's meant by "learn to love `flat`" :)
11:27 masak laben: yes, it does. that's what you're seeing.
11:28 [Coke] m: my @m <== sort() <== ("c", "b", "a") ; say @m;
11:28 camelia rakudo-moar a46b09: OUTPUT«a b c␤»
11:28 GLRelia rakudo-moar f76185: OUTPUT«abc␤»
11:28 masak laben: $a and $b each get such a *tuple*
11:28 [Coke] ^^ is it reasonable to require the parens there? RT #121843
11:28 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=121843
11:28 masak m: for (1..2) X (1..2) -> $a, $b { say $a }
11:28 camelia rakudo-moar a46b09: OUTPUT«1␤1␤2␤2␤»
11:28 GLRelia rakudo-moar f76185: OUTPUT«1 1␤2 1␤»
11:28 masak oh, it differs between cam and GLR
11:28 laben m: for $((1..2) X (1..2)) -> $a { dd $a };
11:28 camelia rakudo-moar a46b09: OUTPUT«$a = $(1, 1, 1, 2, 2, 1, 2, 2)␤»
11:28 GLRelia rakudo-moar f76185: OUTPUT«$a = $(1, 1)␤$a = $(1, 2)␤$a = $(2, 1)␤$a = $(2, 2)␤»
11:29 masak [Coke]: reasonable to require them? I dunno. they don't look required to me.
11:29 masak [Coke]: <== has very low precedence
11:29 masak certainly lower than comme
11:29 masak comma*
11:29 dalek roast/glr: 5d29822 | colomon++ | S32-list/classify.t:
11:29 dalek roast/glr: Add some very minimal tests for Baggy classify.
11:29 dalek roast/glr:
11:29 dalek roast/glr: As far as I could tell, there weren't any tests for this in roast
11:29 dalek roast/glr: so far.
11:29 dalek roast/glr: review: https://github.com/perl6/roast/commit/5d29822587
11:30 [Coke] masak: required to get the "desired" output. If the original request is borked, please reject the ticket.
11:31 * masak looks at ticket
11:31 dalek perl6-examples: 0c7e431 | paultcochrane++ | / (3 files):
11:31 dalek perl6-examples: Add HTTP::Easy::PSGI as a dependency
11:31 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/0c7e431407
11:31 [Coke] masak++
11:32 timotimo oh yay!
11:32 laben masak: thanks! your example helped me understand
11:32 timotimo edges of 18: 2 Internal error: invalid thread ID -334888944 in GC work pass
11:33 masak [Coke]: oh, the parens on `sort()` ?
11:33 [Coke] masak: Hai.
11:33 masak laben: running code beats most things in terms of explanatory power :)
11:33 masak [Coke]: I... dunno. I usually put those parens in, "because then it works"
11:34 masak [Coke]: I don't actually know if it should work without them. kinda feels like it should.
11:34 masak [Coke]: simply because `sort` === `sort()`
11:34 [Coke] Ditto
11:34 nine m: for (1..2) X (1..2) -> ($a, $b) { say $a } # explicit deconstruction
11:34 camelia rakudo-moar a46b09: OUTPUT«Too few positionals passed; expected 2 arguments but got 0 in sub-signature␤  in block <unit> at /tmp/EM1I8Qz7Ih:1␤␤»
11:34 GLRelia rakudo-moar f76185: OUTPUT«1␤1␤2␤2␤»
11:35 masak nine++
11:37 masak m: say 42.indent(4)
11:37 camelia rakudo-moar a46b09: OUTPUT«Method 'indent' not found for invocant of class 'Int'␤  in block <unit> at /tmp/kCES7V68L9:1␤␤»
11:37 GLRelia rakudo-moar f76185: OUTPUT«Method 'indent' not found for invocant of class 'Int'␤  in block <unit> at /tmp/zmnIOfxTcT:1␤␤»
11:37 dalek rakudo/nom: 47ddca1 | moritz++ | docs/ChangeLog:
11:37 dalek rakudo/nom: Add some more changelog entries
11:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/47ddca117d
11:37 masak should .indent be on Cool?
11:37 timotimo m: say -1073741810.base(16)
11:37 camelia rakudo-moar a46b09: OUTPUT«Cannot convert string to number: trailing characters after number in '3⏏FFFFFF2' (indicated by ⏏)␤  in block <unit> at /tmp/ny19SH5eIL:1␤␤Actually thrown at:␤  in block <unit> at /tmp/ny19SH5eIL:1␤␤»
11:37 GLRelia rakudo-moar f76185: OUTPUT«Cannot convert string to number: trailing characters after number in '3⏏FFFFFF2' (indicated by ⏏)␤  in block <unit> at /tmp/y0jUPFzR_i:1␤␤Actually thrown at:␤  in block <unit> at /tmp/y0jUPFzR_i:1␤␤»
11:37 * [Coke] wouldn't mind if it was string only. anything not cool is just going to be a single line, no?
11:37 moritz masak: that would be consistent with how we do it usually
11:37 timotimo m: say "-1073741810".base(16)
11:37 camelia rakudo-moar a46b09: OUTPUT«Method 'base' not found for invocant of class 'Str'␤  in block <unit> at /tmp/CoNGQxN2Fc:1␤␤»
11:37 GLRelia rakudo-moar f76185: OUTPUT«Method 'base' not found for invocant of class 'Str'␤  in block <unit> at /tmp/lqs6JXuy1N:1␤␤»
11:37 dalek rakudo/glr: f09891f | colomon++ | src/core/Baggy.pm:
11:37 dalek rakudo/glr: Hash.classify-list fix to Baggy.classify-list.
11:37 dalek rakudo/glr:
11:37 dalek rakudo/glr: Unfortunately, even in nom this part of the classify-list
11:37 dalek rakudo/glr: functionality didn't work.  This fix makes it fail like nom
11:37 dalek rakudo/glr: does, rather than the entire new way of failing that glr had
11:37 timotimo ... huh?
11:37 dalek rakudo/glr: discovered.
11:37 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/f09891f9c7
11:38 timotimo m: say (-1073741810).base(16)
11:38 camelia rakudo-moar a46b09: OUTPUT«-3FFFFFF2␤»
11:38 GLRelia rakudo-moar f76185: OUTPUT«-3FFFFFF2␤»
11:38 masak ok, if someone wants an LHF:
11:38 masak spectest and move Str.indent to Cool.indent :)
11:38 colomon m: say "-1073741810”.Int.base(16)
11:38 camelia rakudo-moar a46b09: OUTPUT«5===SORRY!5=== Error while compiling /tmp/1OCZvmdXWF␤Unable to parse expression in double quotes; couldn't find final '"' ␤at /tmp/1OCZvmdXWF:1␤------> 3say "-1073741810”.Int.base(16)7⏏5<EOL>␤    expecting any of:␤        argument lis…»
11:38 GLRelia rakudo-moar f76185: OUTPUT«5===SORRY!5=== Error while compiling /tmp/wmGNaJlTAN␤Unable to parse expression in double quotes; couldn't find final '"' ␤at /tmp/wmGNaJlTAN:1␤------> 3say "-1073741810”.Int.base(16)7⏏5<EOL>␤    expecting any of:␤        argument lis…»
11:38 dalek perl6-examples: 0714a62 | paultcochrane++ | categories/cookbook/19cgi-pro​gramming/19-01cgi-script.pl:
11:38 dalek perl6-examples: Purge trailing whitespace
11:38 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/0714a62d90
11:38 dalek perl6-examples: f87f75e | paultcochrane++ | categories/cookbook/ (2 files):
11:38 dalek perl6-examples: Put description text after AUTHOR
11:38 dalek perl6-examples:
11:38 dalek perl6-examples: Now the parser can find the example author information.
11:38 colomon m: say “-1073741810”.Int.base(16)
11:38 camelia rakudo-moar a46b09: OUTPUT«-3FFFFFF2␤»
11:38 GLRelia rakudo-moar f76185: OUTPUT«-3FFFFFF2␤»
11:38 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/f87f75e238
11:44 * ab6tract is excited to leave for the airport to a couple hours :)
11:45 ab6tract *in
11:46 colomon m: sub subber ($a) { $a - ($a % 10) }; my @list      = 29, 7, 12, 9, 18, 23, 3, 7; say categorize(&subber, @list)
11:46 camelia rakudo-moar a46b09: OUTPUT«0 => 7 9 3 7, 10 => 12 18, 20 => 29 23␤»
11:46 GLRelia rakudo-moar f09891: OUTPUT«Cannot call categorize-list(Hash[Any,Any]: Sub, Array); none of these signatures match:␤    (Hash $: %test, $list, *%_)␤    (Hash $: @test, $list, *%_)␤  in block <unit> at /tmp/yn6E8RbDFm:1␤␤»
11:46 Ven ab6tract: are you going to yapc::eu? :)
11:47 FROGGS nine: besides that I think the Array->List conversion makes sense for Signature.named_names, I just need to .Slip in that map in MAIN_HELPER
11:48 nine FROGGS: sounds good :)
11:48 virtualsue joined #perl6
11:52 brrt joined #perl6
11:53 [ptc] itz: you've got a vim process on hack going nuts.
11:53 [ptc] itz: looks like it's hanging; it's using ~100% cpu and has been for a while
11:53 [ptc] itz: oddly enough, I had such a process just now as well.  Had to kill it and then all was good
11:53 ab6tract Ven: SPW today, YAPC::EU on Tuesday :D
11:54 tadzik is yapc::eu so soon?
11:54 tadzik indeed :)
11:55 muraiki is http://doc.perl6.org/language/containers still up to date? I thought I had seen talk about changing/eliminating parcels or something a while back
11:55 ab6tract ('the hackathon' xx *)>>.++
11:56 moritz muraiki: it's up to date with respect to current rakudo/nom; GLR changes things.
11:56 muraiki moritz: thanks
11:56 ab6tract muraiki: indeed. the very last details seem to be getting hammered out as we speak
11:56 muraiki woot :)
11:57 itz [ptc]: I've just killed all those process I was running a cron which was an old snapshot of the perl6-examples update processes
11:58 moritz is LoL officially dead?
11:58 itz [ptc]: I think I will put a timeout in the code .. perhaps perl6 syntax is sometimes too much for vim syntax hilighting
11:58 dalek rakudo/glr: 285febd | jnthn++ | src/ (4 files):
11:58 dalek rakudo/glr: Stub in new [; ] and {; } handling.
11:58 dalek rakudo/glr:
11:58 dalek rakudo/glr: All mutli-dispatch errors for now.
11:58 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/285febda22
11:58 moritz itz: fwiw I've added a CPU time limit in the updated script, but it applies globally
11:59 Ven joined #perl6
11:59 * Ven absolutely <3 OO::*
11:59 nine moritz: yes
12:04 dalek rakudo/glr: a70367c | jnthn++ | src/core/A (2 files):
12:04 dalek rakudo/glr: Start preparing for multi-dim *-POS.
12:04 dalek rakudo/glr:
12:04 dalek rakudo/glr: We'll pass multiple indices to the [AT|ASSIGN|EXISTS...]-POS methods.
12:04 dalek rakudo/glr: This will allow things that know how many dimensions they have to just
12:04 dalek rakudo/glr: bind them directly to parameters, which will be efficient (and allow
12:04 dalek rakudo/glr: the dynamic optimizer to inline, etc.). We'll be able to do some case
12:04 dalek rakudo/glr: analysis on the indexer itself at optimization time to eliminte the
12:04 dalek rakudo/glr: LoL in the commonest cases also.
12:04 dalek rakudo/glr:
12:04 dalek rakudo/glr: This patch simply adds fallbacks for the multi-indice case that dig
12:04 dalek rakudo/glr: into the data structure a level at a time.
12:04 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/a70367c0cd
12:04 dalek rakudo/glr: 6b75f33 | jnthn++ | src/core/Any.pm:
12:04 dalek rakudo/glr: Fix a missing "is rw".
12:04 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/6b75f33ce3
12:06 FROGGS nine: there are more code paths that expect an Array but not a List sadly :o(
12:06 FROGGS nine: I'll care about this later I think
12:06 dalek roast/glr: 3cdd87a | moritz++ | S04-statements/gather.t:
12:06 dalek roast/glr: Remove .lol from gather.t
12:06 dalek roast/glr: review: https://github.com/perl6/roast/commit/3cdd87a858
12:08 dalek perl6-roast-data: 3cfb0b3 | coke++ | / (2 files):
12:08 dalek perl6-roast-data: another glr run for 'today'
12:08 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/3cfb0b3a43
12:09 itz joined #perl6
12:10 |Tux| if I have a to_list that is implemented as a gather, .flat on that list in glr will take forever
12:10 |Tux| as in it is not lazy at all
12:10 |Tux| laben, that is the basic cause of failure for t/78
12:11 |Tux| and adding .flat is not helping there
12:11 dalek roast: 4584bfe | ShimmerFairy++ | S02-literals/allomorphic.t:
12:11 dalek roast: Add tests for allomorphic types (IntStr etc.)
12:11 dalek roast:
12:11 dalek roast: Unfortunately they all have to be fudged for rakudo for the time being.
12:11 dalek roast: review: https://github.com/perl6/roast/commit/4584bfed21
12:12 colomon in GLR,  are there gotchas about using push?
12:12 lizmat colomon: yes
12:13 lizmat colomon: and we're thinking about them
12:13 * colomon is ready to recieve wisdom
12:13 colomon …. oh
12:13 lizmat but no solution just yet  :-(
12:13 lizmat or resolution, I should say
12:14 domidumont joined #perl6
12:15 moritz why not just git push a **@elems signature, and consider everything on the first level a new element to add?
12:15 moritz m: sub f(**@e) { @e.perl }; say f 1..3
12:15 camelia rakudo-moar 47ddca: OUTPUT«(1, 2, 3)␤»
12:15 GLRelia rakudo-moar 6b75f3: OUTPUT«[1..3]␤»
12:15 moritz m: sub f(**@e) { @e[0].perl }; say f 1..3
12:15 camelia rakudo-moar 47ddca: OUTPUT«1..3␤»
12:16 GLRelia rakudo-moar 6b75f3: OUTPUT«1..3␤»
12:16 [Coke] we have a pending PR for nqp that's a few months old: https://github.com/perl6/nqp/pull/228
12:17 moritz though there are probably implications that I've missed (re .push)
12:17 TimToady that is one of the proposed solutions, but when I tried it, the setting blew up :)
12:17 rarara joined #perl6
12:17 moritz TimToady: that's decidedly LTA :-)
12:20 dalek rakudo/glr: 496300b | lizmat++ | src/core/Backtrace.pm:
12:20 dalek rakudo/glr: Fix RAKUDO_VERBOSE_STACKFRAME option
12:20 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/496300b8b5
12:21 araujo_ joined #perl6
12:23 timotimo as sort of expected, the MVM_nfa_to_statelist seems to be correct semantically, but it's not gc-clean :|
12:24 jnthn timotimo: Give it a GC bath
12:24 * colomon ’s problem didn’t have anything to do with push, it turned out.  :)
12:24 dalek rakudo/glr: f2b993c | jnthn++ | src/core/array_slice.pm:
12:24 dalek rakudo/glr: Make basic ; array indexing work again.
12:24 dalek rakudo/glr:
12:24 dalek rakudo/glr: Just for simple indexes, not slices.
12:24 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/f2b993c790
12:25 jdv79 so, i have a fix for cpan but i'm not sure if the p5/p6 ecosystem sharing thing is well thought out.  is there a consolidated info somewhere about this?
12:25 FROGGS jdv79: probably only in my brain
12:26 jdv79 can that be accessed?
12:26 FROGGS indirectly
12:26 FROGGS there is a brain to text api you can use
12:27 timotimo yeah, you grab froggs at the back of his head and gently push it against the keyboard
12:27 lizmat so those are not freckles, but key impressions ?
12:28 FROGGS really gently please :o)
12:28 FROGGS *g*
12:28 FROGGS good that the L key is not at the middle top
12:30 dalek nqp: 5146561 | coke++ | t/docs/tests.t:
12:30 dalek nqp: Don't use substr2 where substr will do.
12:30 dalek nqp:
12:30 dalek nqp: Close #223
12:30 dalek nqp: review: https://github.com/perl6/nqp/commit/51465619e1
12:36 jdv79 so it appears that most p5 ecosystemm things treat the dist/module/package names as unique names so what was the thinking on how that would be addreseed?
12:36 tadzik we don't :)
12:36 FROGGS lizmat: I can't spot in the clogs that you states that your patch causes regressions
12:37 tadzik panda recently stopped doing that
12:37 tadzik ie we have 2 modules named Foo in the ecosystem
12:37 jdv79 well, if we want to take advantage of the p5 stuff that's a problem
12:37 lizmat FROGGS: but I never stated that...  because I didn't know it caused a regression
12:38 ShimmerFairy We generally don't, we're just using a P5-coded system :)
12:38 FROGGS lizmat: can you please check that? I loose again almost an hour :o(
12:38 lizmat in fact, I thought it fixed several tests and we were *pro*gressing
12:38 lizmat :-(
12:38 FROGGS s/check that/chek that next time/
12:38 nine Are we sure it's really regressions and not just uncovered bugs?
12:39 FROGGS nine: looks like regression to me
12:40 nine FROGGS: I'm curious. Is there a golf for reproducing the new error?
12:41 FROGGS m: my @a = ("array", "of", "params"); note @a
12:41 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:41 GLRelia rakudo-moar f2b993: OUTPUT«arrayofparams␤»
12:41 lizmat m: my @a = ("array", "of", "params"); say @a
12:41 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:41 GLRelia rakudo-moar f2b993: OUTPUT«arrayofparams␤»
12:41 dalek perl6-examples: b1f0b92 | (Steve Mynott)++ | / (2 files):
12:41 dalek perl6-examples: HTTP::Easy::PSGI in ecosystem HTTP::Easy
12:41 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/b1f0b92965
12:41 FROGGS it might be just that the remaining note() candidate needs to be note(**@) instead of note(*@)
12:43 lizmat m: my @a = ("array", "of", "params"); say @a,
12:43 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:43 GLRelia rakudo-moar f2b993: OUTPUT«arrayofparams␤»
12:43 lizmat m: my @a = ("array", "of", "params"); say (@a,)
12:43 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:43 GLRelia rakudo-moar f2b993: OUTPUT«arrayofparams␤»
12:43 nine m: my @ac ("array", "of", "params"); say "@a";
12:43 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/K1pOSCRa8v␤Signatures as constraints on variables not yet implemented. Sorry. ␤at /tmp/K1pOSCRa8v:1␤------> 3my @ac ("array", "of", "params")7⏏5; say "@a";␤»
12:43 GLRelia rakudo-moar f2b993: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ddLirgsRQK␤Signatures as constraints on variables not yet implemented. Sorry. ␤at /tmp/ddLirgsRQK:1␤------> 3my @ac ("array", "of", "params")7⏏5; say "@a";␤»
12:43 nine m: my @a = ("array", "of", "params"); say "@a";
12:43 camelia rakudo-moar 47ddca: OUTPUT«@a␤»
12:43 GLRelia rakudo-moar f2b993: OUTPUT«@a␤»
12:43 nine m: my @a = ("array", "of", "params"); say "{@a}";
12:43 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:43 GLRelia rakudo-moar f2b993: OUTPUT«array of params␤»
12:43 nine m: my @a = ("array", "of", "params"); say @a;
12:43 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:43 GLRelia rakudo-moar f2b993: OUTPUT«arrayofparams␤»
12:43 ShimmerFairy m: my @a = ("array", "of", "params"); say "@a[]";
12:43 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:43 GLRelia rakudo-moar f2b993: OUTPUT«array of params␤»
12:43 nine m: my @a = ("array", "of", "params"); say "array", "of", "params";
12:43 camelia rakudo-moar 47ddca: OUTPUT«arrayofparams␤»
12:43 GLRelia rakudo-moar f2b993: OUTPUT«arrayofparams␤»
12:44 nine m: my @a = ("array", "of", "params"); say $@a;
12:44 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:44 GLRelia rakudo-moar f2b993: OUTPUT«array of params␤»
12:44 jdv79 i only see one foo on modules.perl6.org
12:45 FROGGS jdv79: we need to fix the ecosystem code and panda to support multiple same named dists
12:45 nine m: my @a = ("array", "of", "params"); say @a.gist;
12:45 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:45 GLRelia rakudo-moar f2b993: OUTPUT«array of params␤»
12:45 lizmat m: my @a = ("array", "of", "params"); say @a.gist
12:45 camelia rakudo-moar 47ddca: OUTPUT«array of params␤»
12:45 GLRelia rakudo-moar f2b993: OUTPUT«array of params␤»
12:45 tadzik I just noticed that modules.perl6.org doesn't support it
12:46 FROGGS I was always under the impression that 'say @a' implies 'say @a.gist'
12:47 FROGGS and I vote for that 'say 1, 2' always has been wrong
12:47 nine say @a meaning say @a.gist was probably one of the many rules that we replaced by the single-arg-rule
12:47 FROGGS true, though I want to have a space between the elems
12:49 nine m: say 1, 2, 3;
12:49 camelia rakudo-moar 47ddca: OUTPUT«123␤»
12:49 GLRelia rakudo-moar f2b993: OUTPUT«123␤»
12:49 RabidGravy joined #perl6
12:49 FROGGS TimToady: can we please have a space character in there?
12:49 dalek rakudo/glr: 2ccea71 | lizmat++ | src/core/io_operators.pm:
12:49 dalek rakudo/glr: Revert "Remove say(\x) / note(\x) candidates"
12:49 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/2ccea71e81
12:50 dalek ecosystem: 7f151ba | (Nova Patch)++ | META.list:
12:50 dalek ecosystem: CLDR::List moved on GitHub
12:50 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/7f151ba7c1
12:50 [Coke] fixed the nqp docs tests so they are runnable again. Lots of docs missing +/or untested opcodes.
12:51 ShimmerFairy FROGGS: however,  say "Foo: ", something-to-print($arg)  will end up breaking (by putting extra spaces where you don't mean them)
12:52 [Coke] colomon opened a new RT? aiiigh. :)
12:52 |Tux| GLR: method to_list { gather { take 3 .. Inf }} ← that now returns a Range instead of a lazy list. Is that intended?
12:53 FROGGS m: say 12, 3; say (my @a = 12, 3); # I want bot output "12 3"
12:53 camelia rakudo-moar 47ddca: OUTPUT«123␤12 3␤»
12:53 GLRelia rakudo-moar f2b993: OUTPUT«123␤123␤»
12:53 FROGGS TimToady / lizmat: ^^
12:54 FROGGS both*
12:54 [Coke] m: say 12, 3; say (my @a = 12, 3).perl
12:54 camelia rakudo-moar 47ddca: OUTPUT«123␤[12, 3]<>␤»
12:54 GLRelia rakudo-moar f2b993: OUTPUT«123␤[12, 3]␤»
12:54 [Coke] do you want default join() changed or default gist?
12:55 lizmat nothing to do with join
12:55 FROGGS [Coke]: neither
12:55 FROGGS say calls .gist on the elems of the input array, and joins it with ''
12:55 ShimmerFairy like I said, it would break a lot of literal arg lists passed to 'say' that expect nothing to be placed between elements, so I don't think changing that is the right move.
12:55 FROGGS and this is wrong IMO
12:56 FROGGS ShimmerFairy: I'm thinking about it
12:57 ShimmerFairy I think the better solution would be something that preserved the fact that you're passing a list object to say (isn't **@a supposed to do that?)
12:58 |Tux| m: class C{method foo{gather { take 3..Inf}}};C.new.foo[^9].perl.say
12:58 camelia rakudo-moar 47ddca: OUTPUT«(3, 4, 5, 6, 7, 8, 9, 10, 11)␤»
12:58 GLRelia rakudo-moar f2b993: OUTPUT«(3..Inf)␤»
12:58 [Coke] r: my @a=<a a a>;my @b=<b b b>;for (@a »,« @b) »,« @a -> $a, $b,$c {say "$a $b $c"}
12:58 GLRelia rakudo-jvm 2ccea7: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
12:58 GLRelia ..rakudo-moar 2ccea7: OUTPUT«Lists on either side of non-dwimmy hyperop of infix:<,> are not of the same length␤left: 2 elements, right: 1 elements␤  in block <unit> at /tmp/tmpfile:1␤␤»
12:58 timotimo m: say rand 5
12:58 GLRelia rakudo-moar 2ccea7: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ploXygyYbC␤Unsupported use of rand(N); in Perl 6 please use N.rand for Num or (^N).pick for Int result␤at /tmp/ploXygyYbC:1␤------> 3say rand7⏏5 5␤»
12:58 camelia rakudo-{moar,jvm} 47ddca: OUTPUT«Lists on either side of non-dwimmy hyperop of infix:<,> are not of the same length␤left: 6 elements, right: 3 elements␤  in block <unit> at /tmp/tmpfile:1␤␤»
12:58 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/od5piU737C␤Unsupported use of rand(N); in Perl 6 please use N.rand for Num or (^N).pick for Int result␤at /tmp/od5piU737C:1␤------> 3say rand7⏏5 5␤»
12:58 |Tux| is that glr behavior expected or shall nom and glr be the same?
13:00 [Coke] r: my @a := 1..*; @a := @a[1, 3 ... *].list
13:00 GLRelia rakudo-moar 2ccea7: OUTPUT«(timeout)»
13:00 GLRelia ..rakudo-jvm 2ccea7: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
13:00 camelia rakudo-jvm 47ddca: OUTPUT«␤(timeout)␤»
13:00 camelia ..rakudo-moar 47ddca: OUTPUT«(timeout)»
13:01 masak <FROGGS> TimToady: can we please have a space character in there?
13:01 masak FROGGS: I'm pretty sure I wouldn't want that
13:01 FROGGS masak: also not for 'say @a' ?
13:01 masak FROGGS: I use `say thing, thing, thing;` when I want to print things next to each other, no space.
13:01 dalek rakudo/glr: 99f158a | TimToady++ | src/core/Array.pm:
13:01 dalek rakudo/glr: omit redundant $ inside Array.perl
13:01 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/99f158a6ee
13:01 masak FROGGS: I see `say @a` as a different case.
13:01 FROGGS masak: okay, I can live with that
13:01 FROGGS masak: so we want the old behaviour?
13:02 masak FROGGS: if I was the Emperor of Perl 6, `say @a` would actually print something like "[1, 2, 3]". but TimToady has already overruled that.
13:02 masak FROGGS: and so I'll have to contend with "1 2 3"
13:02 FROGGS k
13:02 dalek nqp/nfa_to_statelist: 2e95469 | timotimo++ | src/ (13 files):
13:02 dalek nqp/nfa_to_statelist: test nfatostatelist
13:02 dalek nqp/nfa_to_statelist:
13:02 dalek nqp/nfa_to_statelist: by calling it on every nfa we call nfafromstatelist on.
13:02 dalek nqp/nfa_to_statelist: review: https://github.com/perl6/nqp/commit/2e95469680
13:02 moritz m: my @a = 1 2, 3; say @a
13:02 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/4pb6s5CeRH␤Two terms in a row␤at /tmp/4pb6s5CeRH:1␤------> 3my @a = 17⏏5 2, 3; say @a␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        statement end␤    …»
13:02 GLRelia rakudo-moar 2ccea7: OUTPUT«5===SORRY!5=== Error while compiling /tmp/8Dnazf07Ct␤Two terms in a row␤at /tmp/8Dnazf07Ct:1␤------> 3my @a = 17⏏5 2, 3; say @a␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        statement end␤    …»
13:02 moritz m: my @a = 1, 2, 3; say @a
13:02 camelia rakudo-moar 47ddca: OUTPUT«1 2 3␤»
13:02 GLRelia rakudo-moar 2ccea7: OUTPUT«1 2 3␤»
13:03 FROGGS lizmat / TimToady: so given ShimmerFairy's and masak's opinions, I want to old behaviour back
13:03 moritz I expected that to say 123, because say() has a slurpy arg, no?
13:03 masak you mean you expected that under GLR?
13:03 dalek rakudo/glr: f26f52d | (Stefan Seifert)++ | src/core/Supply.pm:
13:03 dalek rakudo/glr: Fix Supply.rotor on non-lazy lists
13:03 dalek rakudo/glr:
13:03 dalek rakudo/glr: Fixes "Type check failed in binding; expected 'Positional' but got
13:03 dalek rakudo/glr: 'Seq'" in rotor caused by xx returning a Seq now and we try to bind to
13:03 dalek rakudo/glr: an @array. Need to .flat that because the code expects a flattened list
13:03 dalek rakudo/glr: of values instead of a list of lists.
13:03 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/f26f52d26a
13:03 dalek nqp: 34501ae | coke++ | t/docs/tests.t:
13:03 dalek nqp: make t/docs/tests.t runnable again
13:03 dalek nqp: review: https://github.com/perl6/nqp/commit/34501aeab8
13:03 FROGGS lizmat / TimToady: so, space for passing an @a, no space for literal lists
13:03 FROGGS moritz: don't mind the implementation
13:03 rurban joined #perl6
13:03 [Coke] r: role c {}; my @a = 1,2,3,4; my $b = @a.pick(2) but c; say $b.WHAT; say $b;
13:03 GLRelia rakudo-moar 2ccea7: OUTPUT«(Seq+{c})␤2 4␤»
13:03 GLRelia ..rakudo-jvm 2ccea7: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
13:03 camelia rakudo-moar 47ddca: OUTPUT«(List+{c})␤This type does not support elems␤  in block <unit> at /tmp/tmpfile:1␤␤»
13:03 camelia ..rakudo-jvm 47ddca: OUTPUT«(List+{c})␤java.lang.RuntimeException: This type does not support positional operations␤  in block <unit> at /tmp/tmpfile:1␤␤»
13:04 ShimmerFairy masak: there can be brackets at least, so we know how things are nested :)
13:04 laben |Tux|: i'm still at work for some hours, but did you try s/take $thing/take flat $thing/ ?
13:04 FROGGS gah, everything boils down to the fact that I like and dislike the one-arg-rule
13:05 masak ShimmerFairy: aye, that would be a step up
13:05 |Tux| laben. .flat will cause a hang on .inf
13:05 masak ShimmerFairy: "[1 2 3]" then?
13:05 jdv79 mulitple opinion disorder?
13:06 masak jdv79: multiple opinions -- the first step to finding a good solution.
13:06 laben |Tux|: then "take eager $thing" should avoid it
13:06 laben or take $thing.eager
13:06 nine I really don't see why say $@array; is not enough. You will get the spaces and my @array = 1, 2, 3; say @array does exactly the same as say 1, 2, 3;
13:07 FROGGS nine: I'm not saying I cannot be convinced, I just say I don't like it atm...
13:08 ShimmerFairy masak: yeah, particularly for things like  (1 2 (3 4 5))  they'd be nice. Brackets are probably debatable (since it's .gist and not .perl), and I'm half-tempted to suggest some pair of brackets that aren't used for anything in standard P6 code :P (sorta like how we use corner quotes in Matches, except they _are_ valid P6)
13:08 TimToady I think I've already said I'm fine with brackets for .gist
13:09 TimToady at least for sub-gists
13:09 * FROGGS .oO( say quickly: sub-gists sub-gists sub-gists )
13:10 ShimmerFairy TimToady: yeah, that's what I was pointing out, you being fine with .gist brackets :)
13:12 skids joined #perl6
13:14 rmgk_ joined #perl6
13:14 dalek perl6-examples: fb858cb | (Steve Mynott)++ | .travis.yml:
13:14 dalek perl6-examples: fix travis dep
13:14 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/fb858cba40
13:14 telex joined #perl6
13:17 Ven masak: https://github.com/vendethiel/6meta-experiments is my church.pl :)
13:17 btyler_ jnthn: bingo! that was it -- shifting clear_temp_bigints out of the divzero conditional stopped the leak
13:18 tadzik if a sub has a named parameter as a default value, is there any value I can pass so that it chooses the default over it?
13:18 tadzik I'm trying to avoid having two calls in my code, one with the named argument and the other without
13:18 timotimo you can |%(...) to get that
13:19 colomon [Coke]: found code in rakudo that never worked.  I’m not even sure anyone knows what it should do…
13:20 timotimo those things are there sometimes
13:22 masak m: sub foo(:$bar = 42) { say $bar }; foo(bar => Nil)
13:22 camelia rakudo-moar 47ddca: OUTPUT«Nil␤»
13:22 GLRelia rakudo-moar f26f52: OUTPUT«Nil␤»
13:22 masak m: sub foo(:$bar = 42) { say $bar }; foo(bar => Empty)
13:22 camelia rakudo-moar 47ddca: OUTPUT«Empty␤»
13:22 GLRelia rakudo-moar f26f52: OUTPUT«␤»
13:22 masak m: sub foo(:$bar = 42) { say $bar }; foo(bar => Int)
13:22 camelia rakudo-moar 47ddca: OUTPUT«(Int)␤»
13:22 GLRelia rakudo-moar f26f52: OUTPUT«(Int)␤»
13:22 masak tadzik: doesn't look like it
13:23 masak m: sub foo(:$bar = 42) { say $bar }; my %h; foo(|%h)
13:23 camelia rakudo-moar 47ddca: OUTPUT«42␤»
13:23 GLRelia rakudo-moar f26f52: OUTPUT«42␤»
13:23 masak m: sub foo(:$bar = 42) { say $bar }; my %h = False ?? bar => 42 !! (); foo(|%h)
13:23 camelia rakudo-moar 47ddca: OUTPUT«42␤»
13:23 GLRelia rakudo-moar f26f52: OUTPUT«42␤»
13:23 dalek rakudo/glr: 4814073 | (Stefan Seifert)++ | src/core/io_operators.pm:
13:23 dalek rakudo/glr: Revert "Revert "Remove say(\x) / note(\x) candidates""
13:23 dalek rakudo/glr:
13:23 dalek rakudo/glr: This reverts commit 2ccea71e813c98e42c27d0da77f0a46878c197d8.
13:23 dalek rakudo/glr:
13:23 dalek rakudo/glr: The original change actually is correct and consistent with the
13:23 dalek rakudo/glr: single-arg-rule of GLR.
13:23 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/4814073cca
13:23 dalek rakudo/glr: 041bf0f | (Stefan Seifert)++ | src/core/io_operators.pm:
13:23 dalek rakudo/glr: Avoid flattening of say's arguments
13:24 dalek rakudo/glr:
13:24 dalek rakudo/glr: This makes say @a, @b; mean say @a.gist, @b.gist; which would probably
13:24 dalek rakudo/glr: be what users expect.
13:24 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/041bf0fa8d
13:24 masak m: sub foo(:$bar = 42) { say $bar }; my %h = True ?? bar => 5 !! (); foo(|%h)
13:24 camelia rakudo-moar 47ddca: OUTPUT«5␤»
13:24 GLRelia rakudo-moar f26f52: OUTPUT«5␤»
13:24 masak tadzik: that seems to work. timotimo++
13:25 dalek rakudo/glr: 965539c | FROGGS++ | src/core/Main.pm:
13:25 dalek rakudo/glr: handle param aliases in MAIN_HELPER
13:25 dalek rakudo/glr:
13:25 dalek rakudo/glr: This fixes the case where an alias to a named param existed, and the
13:25 dalek rakudo/glr: usage message was printed as a false positive when passing either the
13:25 dalek rakudo/glr: param or the alias.
13:25 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/965539c0b1
13:27 FROGGS okay, now I was able to check that my path does not do any harm
13:27 nine :)
13:27 FROGGS :o)
13:28 FROGGS nine: that means we will adjust note.t and say.t?
13:29 nine FROGGS: I'll have a look at them, yes.
13:30 FROGGS nine: I currently do
13:30 colomon when you guys have been counting the number of GLR test failures, was that counting categorize.t (which crashed completely, rather than printing not oks)?
13:30 timotimo masak: you know what confused me for a moment there? :)
13:32 dalek roast/glr: cb71834 | FROGGS++ | S32-io/note.t:
13:32 dalek roast/glr: adjust expectation of "note @a"
13:32 dalek roast/glr: review: https://github.com/perl6/roast/commit/cb718344bc
13:34 dalek perl6-examples: 29ea342 | (Steve Mynott)++ | / (4 files):
13:34 dalek perl6-examples: remove one list of dependencies and doc the others
13:34 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/29ea342817
13:34 FROGGS m: say "o", "k 2 - say with multiple parame", "ters (1)"
13:34 camelia rakudo-moar 47ddca: OUTPUT«ok 2 - say with multiple parameters (1)␤»
13:34 GLRelia rakudo-moar 965539: OUTPUT«ok 2 - say with multiple parameters (1)␤»
13:34 nine FROGGS: please hold your horses!
13:34 ugexe RecommendationManager is such a long name :(
13:34 nine My **@args change made it regress to before lizmat++'s change
13:34 FROGGS ohh!
13:35 jdv79 RecoMan?
13:35 FROGGS boy! you! :D
13:35 colomon m: my @list      = 29, 7, 12, 9, 18, 23, 3, 7; say categorize( [ 0 xx 10, 10 xx 10, 20 xx 10 ], @list).perl
13:35 camelia rakudo-moar 47ddca: OUTPUT«Hash[Any,Any].new(0 => [7, 9, 3, 7], 10 => [12, 18], 20 => [29, 23])␤»
13:35 GLRelia rakudo-moar 965539: OUTPUT«Cannot call categorize-list(Hash[Any,Any]: Block, Array); none of these signatures match:␤    (Hash $: %test, $list, *%_)␤    (Hash $: @test, $list, *%_)␤  in block <unit> at /tmp/LfBz7gVKSN:1␤␤»
13:38 tadzik ugexe: yt?
13:38 ugexe yes
13:38 tadzik awesome
13:38 tadzik want to test a fix for https://github.com/tadzik/panda/issues/142?
13:39 tadzik ugexe: it's at https://github.com/tadzik/panda/tree/gh-142
13:42 FROGGS m: say 42 for Nil # everybody agree that this should iterate onec?
13:42 camelia rakudo-moar 47ddca: OUTPUT«42␤»
13:42 GLRelia rakudo-moar 965539: ( no output )
13:42 ugexe i will, but im not sure where i'll have something to smoke with
13:42 FROGGS once, even
13:42 ugexe when rather
13:42 tadzik you mean a module, or hardware?
13:43 ugexe hardware. which i have, its just working atm heh
13:43 tadzik oh, okay :)
13:44 FROGGS m: say 42 for Int; say Int.elems; # ohh!
13:44 camelia rakudo-moar 47ddca: OUTPUT«42␤0␤»
13:44 GLRelia rakudo-moar 965539: OUTPUT«0␤»
13:44 tadzik FROGGS: http://reactionimage.org/i​mg/gallery/1968804931.jpg
13:44 FROGGS jnthn: can you tell me that this is correct or not?
13:44 tadzik s/65/42/
13:45 FROGGS tadzik++
13:45 masak m: say 42 for Int
13:45 camelia rakudo-moar 47ddca: OUTPUT«42␤»
13:45 GLRelia rakudo-moar 965539: ( no output )
13:45 masak m: Int.map({ say 42 })
13:45 camelia rakudo-moar 47ddca: OUTPUT«42␤»
13:45 GLRelia rakudo-moar 965539: ( no output )
13:45 masak Int-eresting...
13:45 FROGGS masak: its .elems is zero, so there is no iteration
13:46 masak yes, but whyyyyyyyyyyyyy
13:46 FROGGS though I wonder... yes, exactly
13:46 masak m: say Int.elems
13:46 camelia rakudo-moar 47ddca: OUTPUT«0␤»
13:46 GLRelia rakudo-moar 965539: OUTPUT«0␤»
13:46 TimToady m: say Any.elems
13:46 camelia rakudo-moar 47ddca: OUTPUT«0␤»
13:46 GLRelia rakudo-moar 965539: OUTPUT«0␤»
13:46 masak maybe the only way to get an answer to that is to spectest a change to it
13:48 laben m: say 42 for Empty
13:48 camelia rakudo-moar 47ddca: ( no output )
13:48 GLRelia rakudo-moar 965539: ( no output )
13:48 laben m: say 42 for ()
13:48 camelia rakudo-moar 47ddca: ( no output )
13:48 GLRelia rakudo-moar 965539: ( no output )
13:48 jnthn I'm pretty sure Int.elems should say 1
13:48 laben m: say 42 for (Any,)
13:48 camelia rakudo-moar 47ddca: OUTPUT«42␤»
13:48 GLRelia rakudo-moar 965539: OUTPUT«42␤»
13:48 jnthn Because it should be equivalent to Int.list.elems
13:48 llfourn joined #perl6
13:48 jnthn And Int.list should be (Int,)
13:48 masak ooh, maybe a blame can answer it?
13:48 timotimo moritz:    + with/orwith/without conditionals allow you to check for .defined but
13:49 FROGGS /home/froggs/dev/rakudo/src/core/Any.pm:93:    multi method elems(Any:U:) { 0 }
13:49 FROGGS /home/froggs/dev/rakudo/src/core/Any.pm:94:    multi method elems(Any:D:) { self.list.elems }
13:49 timotimo it looks like the supply/whenever line slipped in between
13:50 masak yep
13:50 masak 6a3fc316b185057eee161d210493e8a442ec0a22
13:50 masak https://github.com/rakudo/rakudo/commit/​6a3fc316b185057eee161d210493e8a442ec0a22
13:50 masak "Fix Any.end/elems"
13:50 FROGGS hmmm
13:51 masak that's when it changed from `self.list.elems` to `0` for the type object
13:51 araujo__ joined #perl6
13:51 masak I think it might be a thinko
13:51 jnthn quite possibly, yes
13:51 FROGGS I'll spectest a change
13:51 masak something like "well, the type object has no elements"
13:51 [Coke] colomon: (counting) yes, as long as it has a plan.
13:51 domidumont joined #perl6
13:51 tadzik FROGGS: is https://github.com/tadzik/panda/issues/112 fixed?
13:52 TimToady I think that was an attempt to fix index() before we had with/without
13:52 colomon [Coke]++
13:53 dalek roast/glr: 93eff8e | (Stefan Seifert)++ | S16-io/say.t:
13:53 dalek roast/glr: Tests for GLRified semantics of say
13:53 dalek roast/glr: review: https://github.com/perl6/roast/commit/93eff8e2c9
13:53 dalek rakudo/glr: bb736ad | (Stefan Seifert)++ | src/core/io_operators.pm:
13:53 dalek rakudo/glr: Fix say @array not iterating over the arguments items
13:53 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/bb736ad0e3
13:54 masak the GLRification of say
13:54 [Coke] r: use MONKEY_TYPING; augment slang Regex { token numish { \d+ } };
13:54 GLRelia rakudo-moar 965539: OUTPUT«===SORRY!===␤QAST::Block with cuid cuid_1_1440683686.69991 has not appeared␤Saw 1 occurrence of deprecated code.␤========================================​========================================␤'use MONKEY_TYPING' seen at:␤  /tmp/tmpfile, line 1…»
13:54 GLRelia ..rakudo-jvm 965539: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
13:54 camelia rakudo-moar 47ddca: OUTPUT«===SORRY!===␤QAST::Block with cuid cuid_1_1440683686.63467 has not appeared␤Saw 1 occurrence of deprecated code.␤========================================​========================================␤'use MONKEY_TYPING' seen at:␤  /tmp/tmpfile, line 1…»
13:54 camelia ..rakudo-jvm 47ddca: OUTPUT«===SORRY!===␤setcodeobj can only be used with a CodeRef␤Unhandled exception: java.lang.NullPointerException␤  in print_exception (gen/jvm/CORE.setting:15481)␤  in <anon> (gen/jvm/CORE.setting:15522)␤  in command_eval (gen/jvm/stage2/NQPHLL.nqp:130…»
13:55 masak whoa
13:55 FROGGS r: use MONKEY-TYPING; augment slang Regex { token numish { \d+ } };
13:55 GLRelia rakudo-jvm 965539: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
13:55 GLRelia ..rakudo-moar 965539: OUTPUT«===SORRY!===␤QAST::Block with cuid cuid_1_1440683724.15141 has not appeared␤»
13:55 camelia rakudo-jvm 47ddca: OUTPUT«===SORRY!===␤setcodeobj can only be used with a CodeRef␤»
13:55 camelia ..rakudo-moar 47ddca: OUTPUT«===SORRY!===␤QAST::Block with cuid cuid_1_1440683724.08868 has not appeared␤»
13:55 masak whatever that was, it hit *both* Moar *and* JVM
13:55 masak [Coke]: could you please file that one?
13:56 [Coke] m: my Str constant foo = 1 ;
13:56 camelia rakudo-moar 47ddca: ( no output )
13:56 GLRelia rakudo-moar 965539: ( no output )
13:56 [Coke] masak: It's already filed.
13:56 FROGGS [Coke]: you're checking old tickets?
13:56 [Coke] FROGGS: yes.
13:56 FROGGS [Coke]++
13:56 masak [Coke]: the 'QAST::Block with cuid cuid_1_1440683686.69991 has not appeared' thing?
13:57 masak [Coke]: because that seems it should never ever happen.
14:00 tadzik FROGGS: is there something in existence that parses JSON::Slow:auth<cpan:*>:ver(1..*)-like things?
14:00 [Coke] r: https://gist.github.com/diakopter/2830046
14:00 GLRelia rakudo-moar bb736a: OUTPUT«5===SORRY!5=== Error while compiling /tmp/tmpfile␤Shaped variable declarations not yet implemented. Sorry. ␤at /tmp/tmpfile:1␤------> 3my @d[4;4]7⏏5;␤»
14:00 GLRelia ..rakudo-jvm bb736a: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
14:00 camelia rakudo-{moar,jvm} 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/tmpfile␤Shaped variable declarations not yet implemented. Sorry. ␤at /tmp/tmpfile:1␤------> 3my @d[4;4]7⏏5;␤»
14:00 [Coke] masak: yes, that's ticketed.
14:00 breinbaas joined #perl6
14:01 TimToady m: say [[1,2,3],<a b c>].perl
14:01 camelia rakudo-moar 47ddca: OUTPUT«[[1, 2, 3], "a", "b", "c"]␤»
14:01 GLRelia rakudo-moar bb736a: OUTPUT«[[1, 2, 3], ("a", "b", "c")]␤»
14:01 masak [Coke]: ok, excellent.
14:02 [Coke] https://rt.perl.org/Ticket/Display.html?id=117193 - opened by masak.
14:02 masak don't know who that is
14:03 masak but great that there's a ticket
14:04 [Coke] m: say (1, 2, *) Z (1, 2, 3, *)
14:04 camelia rakudo-moar 47ddca: OUTPUT«1 1 2 2 * 3␤»
14:05 GLRelia rakudo-moar bb736a: OUTPUT«1 12 2* 3␤»
14:08 lizmat 「ok 7 - Match.note
14:08 lizmat
14:09 FROGGS tadzik: ask the one to your left... when he is back
14:10 tadzik ok :)
14:12 dalek roast: 9208bf0 | coke++ | S32-io/IO-Socket-Async.t:
14:12 dalek roast: This test passing on mac os x now
14:12 dalek roast: review: https://github.com/perl6/roast/commit/9208bf0f56
14:12 itz joined #perl6
14:14 [Coke] m: (1..*).eager
14:14 camelia rakudo-moar 47ddca: ( no output )
14:14 GLRelia rakudo-moar bb736a: OUTPUT«(timeout)»
14:14 FROGGS star-m: my $a; say $a.elems
14:14 camelia star-m 2015.03: OUTPUT«0␤»
14:14 FROGGS m: my $a; say $a.elems
14:14 camelia rakudo-moar 47ddca: OUTPUT«0␤»
14:15 abaugher joined #perl6
14:15 laben m: (1..*).eager.say
14:15 camelia rakudo-moar 47ddca: OUTPUT«1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 8…»
14:15 GLRelia joined #perl6
14:15 FROGGS m: my $a; say $a.elems
14:15 camelia rakudo-moar 47ddca: OUTPUT«0␤»
14:15 GLRelia star-m : OUTPUT«Can't chdir to '/home/steve/star/': No such file or directory at lib/EvalbotExecuter.pm line 171.␤      EvalbotExecuter::_auto_execute(HASH(0x2070390), "my \$a; say \$a.elems", GLOB(0x23e3288), "/tmp/Mm28zjppdC", "star-m") called at lib/EvalbotExecuter.pm line 119␤       Ev…»
14:16 GLRelia rakudo-moar bb736a: OUTPUT«0␤»
14:16 colomon m: [ 0 xx 10, 10 xx 10, 20 xx 10 ].perl.say
14:16 camelia rakudo-moar 47ddca: OUTPUT«[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20]␤»
14:16 GLRelia rakudo-moar bb736a: OUTPUT«[(0, 0, 0, 0, 0, 0, 0, 0, 0, 0), (10, 10, 10, 10, 10, 10, 10, 10, 10, 10), (20, 20, 20, 20, 20, 20, 20, 20, 20, 20)]␤»
14:16 [Coke] r:  say &infix:<max>(|(1..*))
14:16 camelia rakudo-{moar,jvm} 47ddca: OUTPUT«-Inf␤»
14:16 GLRelia rakudo-moar bb736a: OUTPUT«(timeout)»
14:16 GLRelia ..rakudo-jvm bb736a: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
14:16 dalek roast/glr: 2c53e61 | lizmat++ | S16-io/note.t:
14:16 dalek roast/glr: Make note test work
14:16 dalek roast/glr: review: https://github.com/perl6/roast/commit/2c53e61ee8
14:16 tadzik timotimo++ # JSON::Fast
14:17 colomon Any easy way to get the flat version back in GLR?  of  [ 0 xx 10, 10 xx 10, 20 xx 10 ]
14:17 FROGGS m: my $a; say $a.elems # jnthn: this will also report '1' with my patch
14:17 nine m: say [flat 0 xx 10, 10 xx 10, 20 xx 10 ].perl
14:17 camelia rakudo-moar 47ddca: OUTPUT«0␤»
14:17 GLRelia rakudo-moar bb736a: OUTPUT«0␤»
14:17 GLRelia rakudo-moar bb736a: OUTPUT«[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20]␤»
14:17 camelia rakudo-moar 47ddca: OUTPUT«[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20]␤»
14:17 colomon nine++
14:19 [Coke] jnthn: can you verify that RT #122948 is closable given glr? I am pretty sure it is, but it's written using old ".infinite" syntax. the bug that it was the root cause of is already fixed in GLR...
14:19 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=122948
14:22 dalek rakudo/glr: 2023f02 | TimToady++ | src/core/ (2 files):
14:22 dalek rakudo/glr: try out gist bracketing
14:22 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/2023f0218f
14:23 TimToady ^^^ is likely to break a few spectests
14:24 [Coke] # XXX -- Lists are not real datatypes, but I haven't found a better location   ---- This comment seems deletable. S02-types/list.t
14:24 colomon [Coke]: :)
14:25 laouji joined #perl6
14:25 dalek roast: edea297 | coke++ | S02-types/lists.t:
14:25 dalek roast: Remove very old comment.
14:25 dalek roast: review: https://github.com/perl6/roast/commit/edea297df8
14:26 [Coke] .ell-ems
14:26 dalek roast/glr: 28feb5b | colomon++ | S32-list/categorize.t:
14:26 dalek roast/glr: Make $arrayer do the right thing under GLR.
14:26 dalek roast/glr: review: https://github.com/perl6/roast/commit/28feb5ba40
14:27 timotimo https://gist.github.com/timo/f1324355a5ae8dc95d36 - this is what it looks like when you accidentally run rakudo/moar against roast/glr
14:28 muraiki m: say 0o7.WHAT
14:28 GLRelia rakudo-moar bb736a: OUTPUT«␤»
14:28 camelia rakudo-moar 47ddca: OUTPUT«(Int)␤»
14:28 [Coke] m: role sidecat is {} ; my @a = 1,2,3 but sidecat; say @a.gist
14:28 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ewdrdM6NJv␤Invalid name␤at /tmp/ewdrdM6NJv:1␤------> 3role sidecat is7⏏5 {} ; my @a = 1,2,3 but sidecat; say @a.␤»
14:28 GLRelia rakudo-moar bb736a: OUTPUT«5===SORRY!5=== Error while compiling /tmp/PAaQsjh1c7␤Invalid name␤at /tmp/PAaQsjh1c7:1␤------> 3role sidecat is7⏏5 {} ; my @a = 1,2,3 but sidecat; say @a.␤»
14:28 muraiki hrm, is there any way to find out that it's octal? for instance, if I want to use it as a type in a sub parameter
14:29 [Coke] m: role sidecat {} ; my @a = 1,2,3 but sidecat; say @a.gist
14:29 camelia rakudo-moar 47ddca: OUTPUT«1 2 3␤»
14:29 GLRelia rakudo-moar bb736a: OUTPUT«1 2 3␤»
14:29 timotimo rakudo/nom is what i meant to write
14:29 [Coke] m: role sidecat {} ; my @a = 1,2,3 but sidecat; my @b; @b.pick(2); say @b.gist
14:29 camelia rakudo-moar 47ddca: OUTPUT«␤»
14:29 GLRelia rakudo-moar bb736a: OUTPUT«␤»
14:29 [Coke] m: role sidecat {} ; my @a = 1,2,3 ; my @b; @b.pick(2) but sidecat; say @b.gist
14:29 GLRelia rakudo-moar bb736a: OUTPUT«␤»
14:29 camelia rakudo-moar 47ddca: OUTPUT«␤»
14:29 timotimo muraiki: no, something that's octal is turned into an Int immediately
14:29 [Coke] m: role sidecat {} ; my @a = 1,2,3 ; my @b; @b = @a.pick(2) but sidecat; say @b.gist
14:29 GLRelia rakudo-moar bb736a: OUTPUT«3 2␤»
14:29 camelia rakudo-moar 47ddca: OUTPUT«This type does not support elems␤  in block <unit> at /tmp/EtQ8JFL7kd:1␤␤»
14:30 [Coke] m: role sidecat {} ; my @a = 1,2,3 ; my @b; @b = @a.pick(*).sort but sidecat; say @b.gist
14:30 camelia rakudo-moar 47ddca: OUTPUT«1 2 3␤»
14:30 GLRelia rakudo-moar bb736a: OUTPUT«1 2 3␤»
14:30 timotimo but how about <0o7>
14:30 timotimo m: say <0o7>.WHAT
14:30 camelia rakudo-moar 47ddca: OUTPUT«(Str)␤»
14:30 GLRelia rakudo-moar bb736a: OUTPUT«␤»
14:30 timotimo hm.
14:30 rindolf joined #perl6
14:30 dalek rakudo/glr: ab4d41f | colomon++ | src/core/Hash.pm:
14:30 dalek rakudo/glr: Get most of categorize working again on GLR.
14:30 dalek rakudo/glr:
14:30 dalek rakudo/glr: This is somewhat hacky, definitely needs work on the more
14:30 dalek rakudo/glr: complicated case, probably could use some efficiency improvements
14:30 dalek rakudo/glr: to bring it back in line with the nom version.
14:30 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/ab4d41f92a
14:31 colomon that’s 27 additional passing tests for GLR
14:31 FROGGS \o/
14:31 timotimo way cool!
14:31 laben joined #perl6
14:32 colomon I’m not sure if the 27 were included in 81/80/77 counts tody, as categorize.t previously crashed before executing the first test.  ;)
14:32 [Coke] m: role sidecat {} ; my @a = 1,2,3 ; my @b; @b = @a.pick(*).sort.list but sidecat; say @b.gist
14:32 camelia rakudo-moar 47ddca: OUTPUT«This type does not support elems␤  in block <unit> at /tmp/5agqRSwWlw:1␤␤»
14:32 GLRelia rakudo-moar 2023f0: OUTPUT«[1 2 3]␤»
14:32 [Coke] \o/
14:33 nine tests, not test files surely?
14:34 colomon oh, was that 81 test files failing?
14:34 jnthn colomon: The number has been test files, yes
14:34 nine Yes. But I think it was 76 last count
14:34 nine Running another one right now.
14:35 colomon jnthn: ah
14:35 colomon well, then I’m not helping there at all, there is still one failing test
14:35 dalek rakudo/glr: fffbe76 | lizmat++ | src/core/io_operators.pm:
14:35 dalek rakudo/glr: Get note in line with say (with all the quirks)
14:35 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/fffbe767e9
14:35 lizmat this may introduce note breakage again
14:35 lizmat working on that
14:36 yqt joined #perl6
14:36 dalek roast/glr: 00cbae5 | coke++ | S02-types/list.t:
14:36 dalek roast/glr: Add a test for RT #116527
14:36 dalek roast/glr:
14:36 dalek roast/glr: only passes on glr
14:36 dalek roast/glr: review: https://github.com/perl6/roast/commit/00cbae5234
14:36 synbot6 Link:  https://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=116527
14:37 hahainternet is there a more succinct way to substr than .substr(0,10)?
14:38 [Coke] 1041 tickets.
14:39 [Coke] we can go the CF route and add Left($_, 10) !
14:39 muraiki timotimo: thanks!
14:39 jnthn hahainternet: No
14:40 hahainternet jnthn: merci
14:40 colomon oh!
14:40 TimToady m: $_ = "abcdefghijklmnopqrstuvwxyz"; say /^.**10/
14:40 camelia rakudo-moar 47ddca: OUTPUT«/^.**10/␤»
14:40 GLRelia rakudo-moar ab4d41: OUTPUT«/^.**10/␤»
14:40 TimToady m: $_ = "abcdefghijklmnopqrstuvwxyz"; say m/^.**10/
14:40 camelia rakudo-moar 47ddca: OUTPUT«「abcdefghij」␤»
14:40 GLRelia rakudo-moar ab4d41: OUTPUT«␤»
14:40 TimToady m: $_ = "abcdefghijklmnopqrstuvwxyz"; say ~m/^.**10/
14:40 GLRelia rakudo-moar ab4d41: OUTPUT«abcdefghij␤»
14:40 camelia rakudo-moar 47ddca: OUTPUT«abcdefghij␤»
14:41 TimToady that's more succinct :)
14:41 timotimo Visual Basic had "Mid" for substr (the general case)
14:42 jdv79 FROGGS: https://github.com/andk/pause/issues/182 - finally got it done; hopefully
14:43 ShimmerFairy I still think it'd be nice if substr used :end and :chars adverbs to specify the endpoint, so that it was less ambiguous in general
14:43 * ShimmerFairy needs to finish that S32::Stringy rewrite of hers soon...
14:44 dalek perl6-examples: 98e0851 | (Sue Spence)++ | categories/cookbook/01string​s/01-04converting-values.pl:
14:44 dalek perl6-examples: Change ord to ords, might as well convert the entire string
14:44 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/98e085100c
14:44 dalek perl6-examples: a83f65d | (Sue Spence)++ | categories/cookbook/01string​s/01-04converting-values.pl:
14:44 dalek perl6-examples: Fix comment
14:44 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/a83f65d2a7
14:44 dalek perl6-examples: 14178a5 | (Sue Spence)++ | categories/cookbook/01string​s/01-13upper-lower-case.pl:
14:44 dalek perl6-examples: Add example
14:44 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/14178a5e17
14:44 dalek perl6-examples: 61f427f | (Sue Spence)++ | / (3 files):
14:44 dalek perl6-examples: Test fixes
14:44 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/61f427ff4e
14:44 dalek perl6-examples: 936f2a8 | (sue spence)++ | / (5 files):
14:44 [Coke] timotimo: CF has left, mid, and right. :|
14:44 dalek perl6-examples: Merge pull request #28 from virtualsue/master
14:44 dalek perl6-examples:
14:44 dalek perl6-examples: More minor updates
14:44 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/936f2a879b
14:44 dalek rakudo/glr: 7a4bd87 | colomon++ | src/core/Hash.pm:
14:44 dalek rakudo/glr: Turns out the code I had commented out worked.
14:44 dalek rakudo/glr:
14:44 dalek rakudo/glr: I was getting two different problems crossed up.
14:44 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/7a4bd87eef
14:44 TimToady m: say [[1,2,3],<a b c>].gist
14:44 camelia rakudo-moar 47ddca: OUTPUT«1 2 3 a b c␤»
14:44 GLRelia rakudo-moar fffbe7: OUTPUT«[[1 2 3] (a b c)]␤»
14:44 colomon okay, now categorize.t completely passes.
14:44 hahainternet TimToady: that's quite cute :)
14:45 [Coke] m: sub foo { my $s = 0; ($s += $_ for 1..3) }; say foo()
14:45 GLRelia rakudo-moar fffbe7: OUTPUT«666␤»
14:45 camelia rakudo-moar 47ddca: OUTPUT«6 6 6␤»
14:45 hahainternet i was just thinking indexing into string or similar, i recognise that has ambiguity so *shrug*
14:45 hahainternet i'm not a language designer
14:45 hahainternet and i enjoy just writing what i can in p6 atm :)
14:45 [Coke] that's from https://rt.perl.org/Ticket/Display.html?id=77334 which we might want to revisit in the light of GLR.
14:46 timotimo [Coke]: i think that's actually the same as VB, then
14:46 TimToady m: sub foo { my $s = 0; (($s += $_)<> for 1..3) }; say foo()
14:46 camelia rakudo-moar 47ddca: OUTPUT«1 3 6␤»
14:46 GLRelia rakudo-moar fffbe7: OUTPUT«136␤»
14:46 dalek rakudo/glr: 2003eac | colomon++ | src/core/Hash.pm:
14:46 dalek rakudo/glr: Bring comments in line with the code again.
14:46 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/2003eac837
14:46 TimToady the problem is you're returning $s (the variable) three times
14:47 [Coke] m: my $a = 1; my $b = 2; $a = ($a, $b); say $a[0]
14:47 camelia rakudo-moar 47ddca: OUTPUT«Memory allocation failed; could not allocate 29952 bytes␤»
14:47 GLRelia rakudo-moar fffbe7: OUTPUT«(timeout)»
14:48 TimToady m: my $a = 1; my $b = 2; $a = (|$a, $b); say $a[0]
14:48 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/gc6pUtxwbw␤Arg-flattening | is only valid in an argument list␤at /tmp/gc6pUtxwbw:1␤------> 3my $a = 1; my $b = 2; $a = (|7⏏5$a, $b); say $a[0]␤»
14:48 GLRelia rakudo-moar fffbe7: OUTPUT«1␤»
14:48 [Coke] adding that to RT #82424
14:48 synbot6 Link:  https://rt.perl.org/rt3/Publ​ic/Bug/Display.html?id=82424
14:50 FROGGS we are up to 85 spectests failing now, I'm about to fix the tests wrt parens outputted by .gist
14:51 [Coke] m: https://gist.github.com/coke/a19cab478ca24fb0a153
14:51 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/sORoDr0qBh␤Cannot use variable @hamming in declaration to initialize itself␤at /tmp/sORoDr0qBh:11␤------> 3my @hamming := (1, (@7⏏5hamming X* 2) merge (@hamming X* 3) merg␤    expecting any of:␤…»
14:51 GLRelia rakudo-moar 2003ea: OUTPUT«5===SORRY!5=== Error while compiling /tmp/KJVLyL4YdG␤Cannot use variable @hamming in declaration to initialize itself␤at /tmp/KJVLyL4YdG:11␤------> 3my @hamming := (1, (@7⏏5hamming X* 2) merge (@hamming X* 3) merg␤    expecting any of:␤…»
14:53 dalek doc: d3ab2bf | paultcochrane++ | lib/Type/Exception.pod:
14:53 dalek doc: Document die and warn
14:53 dalek doc:
14:53 dalek doc: ... by shamelessly stealing from the specs text.
14:53 dalek doc: review: https://github.com/perl6/doc/commit/d3ab2bfba9
14:53 dalek rakudo/glr: fbe48b8 | FROGGS++ | src/core/Any-iterable-methods.pm:
14:53 dalek rakudo/glr: add label as attribute to MapIterCommon
14:53 dalek rakudo/glr:
14:53 dalek rakudo/glr: Before it was a lexical which just screamed BUG! jnthn++
14:53 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/fbe48b89cd
14:54 timotimo jnthn: in what places do i need to change stuff so that the statelist no longer ends up serialized, btw?
14:54 masak colomon++ # categorically passing
14:54 jnthn timotimo: NFA.nqp or so
14:54 timotimo and at what exact position do i put nfatostatelist back in?
14:54 nine 84 failing right now
14:54 jnthn In NQP
14:54 timotimo that's the file i'm looking at right now :)
14:54 jnthn oh
14:54 timotimo nfa_run and nfa_run_proto is where i've put it right now so that it exercises the nqp:: op
14:54 dalek rakudo/glr: 5e98abb | jnthn++ | src/core/array_slice.pm:
14:54 dalek rakudo/glr: Multi-dim array slicing (.[0,1;1]).
14:54 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/5e98abb2a0
14:55 timotimo that's the place where the state list gets turned into an nfa if it hasn't been done yet
14:55 timotimo so perhaps after turning the statelist into the nfa object, the statelist ought to be cleared out?
14:55 jnthn timotimo: I'd do that and then see where it gets sad.
14:55 timotimo will do
14:55 jnthn timotimo: I thought we pre-computed the NFAs tough
14:55 jnthn *though
14:56 jnthn Look for precompute or so in Curor.nqp maybe
14:56 timotimo ah, indeed
14:57 nine m: say (1 => 'a', 2 => 'b')
14:57 GLRelia rakudo-moar 2003ea: OUTPUT«1 => a2 => b␤»
14:57 camelia rakudo-moar 47ddca: OUTPUT«1 => a 2 => b␤»
14:58 smls joined #perl6
14:59 timotimo so perhaps at the end of optimize would be a possible position where we could clear out the states list
14:59 jnthn perhaps so
14:59 Ven joined #perl6
15:00 timotimo sadly, this would come with a stage0 upgrade to get the nqp::nfatostatelist op available soon enough :|
15:00 dalek roast/glr: fdc2691 | (Stefan Seifert)++ | S02-types/pair.t:
15:00 dalek roast/glr: Fix fallout from rakudo commit 2023f0218f19cd1a625f5a5b6ffb69a532db1053
15:00 dalek roast/glr: review: https://github.com/perl6/roast/commit/fdc2691f77
15:01 timotimo but maybe we can do the stage0 update after we get the size improvements from no longer serializing the NFA's statelists
15:01 timotimo so it'd be a tiny bit less painful
15:01 dalek roast/glr: e0dc90d | (Stefan Seifert)++ | S03-metaops/reduce.t:
15:01 dalek roast/glr: More fallout from rakudo commit 2023f0218f19cd1a625f5a5b6ffb69a532db1053
15:01 dalek roast/glr: review: https://github.com/perl6/roast/commit/e0dc90d7e6
15:02 jnthn timotimo: Do it in a branch, then do the two, do the two rebootstraps, then rebase -i ;)
15:02 dalek roast/glr: a64915a | (Stefan Seifert)++ | integration/advent2013-day08.t:
15:02 dalek roast/glr: More fallout from rakudo commit 2023f0218f19cd1a625f5a5b6ffb69a532db1053
15:02 dalek roast/glr: review: https://github.com/perl6/roast/commit/a64915ac8a
15:03 colomon lizmat: I’m thinking that the multi-layer classify just doesn’t make sense for Baggy
15:03 dalek roast/glr: c53ca4f | FROGGS++ | S32-io/note.t:
15:03 dalek roast/glr: More fallout from rakudo commit 2023f0218f19cd1a625f5a5b6ffb69a532db1053
15:03 dalek roast/glr: review: https://github.com/perl6/roast/commit/c53ca4f7eb
15:04 timotimo ah, !protoregex wants to have the list of fates, but we can just keep only that around, perhaps in @!fates
15:04 timotimo so far it has been in $!states[0]
15:04 dalek roast/glr: 4430626 | (Stefan Seifert)++ | integration/advent2013-day15.t:
15:04 dalek roast/glr: More fallout from rakudo commit 2023f0218f19cd1a625f5a5b6ffb69a532db1053
15:04 dalek roast/glr: review: https://github.com/perl6/roast/commit/4430626df4
15:05 [Coke] colomon: is it expected that ./smoker projects.json would run for 40+ minutes at 100% cpu?
15:05 dalek rakudo/glr: fb323f4 | jnthn++ | src/core/array_slice.pm:
15:05 dalek rakudo/glr: Make * work in multi-dim slices.
15:05 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/fb323f4bb6
15:05 timotimo or we could have a nqp::getnfafates op that pulls it out of the NFA object
15:05 colomon [Coke]: think so
15:05 timotimo that's basically a pointer steal
15:05 colomon [Coke]: well, it takes a long time, anyway.
15:05 timotimo no copying involved, and so we don't need to have it duplicated in memory either
15:05 colomon of course, it could be wedged, too.
15:06 colomon … actually, that’s kind of weird
15:06 masak could someone show a use case of the multi-layer classify? I'm not sure I get it.
15:07 dalek roast/glr: 6fa9955 | (Stefan Seifert)++ | S12-class/mro.t:
15:07 dalek roast/glr: More fallout from rakudo commit 2023f0218f19cd1a625f5a5b6ffb69a532db1053
15:07 dalek roast/glr: review: https://github.com/perl6/roast/commit/6fa995561e
15:07 dalek rakudo/glr: c563ecd | lizmat++ | src/core/IO/Handle.pm:
15:07 dalek rakudo/glr: Add a str candidate for IO::Handle.print
15:07 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/c563ecde66
15:07 colomon [Coke]: just killed it and started over in a terminal, so I can watch what it’s doing
15:08 dalek roast/glr: 6faf49f | FROGGS++ | S32-io/pipe.t:
15:08 dalek roast/glr: correctly say multiple lines
15:08 dalek roast/glr: review: https://github.com/perl6/roast/commit/6faf49fd10
15:08 skids o/
15:08 yoleaux 06:21Z <nine> skids: Ah, yes thanks! I'll have a look at it after breakfast. Now that I understand the problem it solves, I think it's a shame, your PR has not been merged yet.
15:08 yoleaux 06:39Z <nine> skids: thank you for fixing @a[1, *-1]:delete :) I've had a look at this yesterday but my train arrived before I came to writing code. I really wonder how this worked in nom though. Couldn't find any code handling that case there.
15:09 skids nine: np glad to be of help.  Also, glad it wasn't in the slice code :-)
15:09 nine skids: btw. jnthn++ +1ed IterationBuffer.push not returning the pushed value.
15:10 skids Oh nice.
15:15 dalek perl6-roast-data: f357b2c | coke++ | / (10 files):
15:15 dalek perl6-roast-data: today (automated commit)
15:15 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/f357b2c59e
15:16 colomon m: reduce( -> $a, $b { $a.{$b}; }, {a => {b => {c => 42}}}, <a b c>).perl.say
15:16 GLRelia rakudo-moar fb323f: OUTPUT«42␤»
15:16 camelia rakudo-moar 47ddca: OUTPUT«42␤»
15:16 colomon m: reduce( -> $a, $b { $a.WHAT.say; $a.{$b}; }, {a => {b => {c => 42}}}, <a b c>).perl.say
15:16 GLRelia rakudo-moar c563ec: OUTPUT«␤␤␤42␤»
15:16 camelia rakudo-moar 47ddca: OUTPUT«(Hash)␤(Hash)␤(Hash)␤42␤»
15:17 colomon m: (a => 42).WHAT.say
15:17 camelia rakudo-moar 47ddca: OUTPUT«(Pair)␤»
15:17 GLRelia rakudo-moar c563ec: OUTPUT«␤»
15:17 colomon umm....
15:17 colomon m: (a => 42)[0].WHAT.say
15:17 GLRelia rakudo-moar c563ec: OUTPUT«␤»
15:17 camelia rakudo-moar 47ddca: OUTPUT«(Pair)␤»
15:17 jnthn That looks wrong
15:18 colomon it’s not a Pair, it’s something without a name containing a Pair?
15:19 nine m: (a => 2)[0].^name.say;
15:19 camelia rakudo-moar 47ddca: OUTPUT«Pair␤»
15:19 GLRelia rakudo-moar c563ec: OUTPUT«Pair␤»
15:19 nine It seems to be just WHAT that's strange
15:19 colomon oh, never mind on that, got confused by the camelias switching places
15:19 colomon m: say Pair.WHAT
15:19 camelia rakudo-moar 47ddca: OUTPUT«(Pair)␤»
15:19 GLRelia rakudo-moar c563ec: OUTPUT«␤»
15:19 nine m: (a => 2)[0].WHAT.^name.say
15:19 camelia rakudo-moar 47ddca: OUTPUT«Pair␤»
15:19 GLRelia rakudo-moar c563ec: OUTPUT«Pair␤»
15:20 jnthn Normally it's 'cus there's a .gist that isn't constrained by :D
15:20 colomon m: say %h; say %h.WHAT
15:20 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/2N_f_st9RL␤Variable '%h' is not declared␤at /tmp/2N_f_st9RL:1␤------> 3say 7⏏5%h; say %h.WHAT␤»
15:20 GLRelia rakudo-moar c563ec: OUTPUT«5===SORRY!5=== Error while compiling /tmp/TFKpWDbhdh␤Variable '%h' is not declared␤at /tmp/TFKpWDbhdh:1␤------> 3say 7⏏5%h; say %h.WHAT␤»
15:20 colomon m: my %h; say %h.WHAT
15:20 GLRelia rakudo-moar c563ec: OUTPUT«␤»
15:20 camelia rakudo-moar 47ddca: OUTPUT«(Hash)␤»
15:20 dalek perl6-examples: d8073d1 | (Steve Mynott)++ | t/categories/cookbook/09directories.t:
15:20 dalek perl6-examples: work around changed behaviour of File::Temp
15:20 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/d8073d1df5
15:20 dalek perl6-examples: a39a3a6 | (Steve Mynott)++ | t/000-check-dependencies.t:
15:20 dalek perl6-examples: fix number of tests by counting array
15:20 dalek perl6-examples: review: https://github.com/perl6/perl​6-examples/commit/a39a3a69bc
15:21 brrt joined #perl6
15:21 pmurias joined #perl6
15:22 [Coke] m:  my $foo="+";&infix:($foo)(3,4).say
15:22 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ZvKoN5SVAo␤You can't adverb ␤at /tmp/ZvKoN5SVAo:1␤------> 3my $foo="+";&infix:($foo)7⏏5(3,4).say␤»
15:22 GLRelia rakudo-moar c563ec: OUTPUT«5===SORRY!5=== Error while compiling /tmp/H59FHLSUEQ␤You can't adverb ␤at /tmp/H59FHLSUEQ:1␤------> 3my $foo="+";&infix:($foo)7⏏5(3,4).say␤»
15:22 pmurias_ joined #perl6
15:22 domidumont joined #perl6
15:23 FROGGS m: my $a = <stringify args>; note $a
15:23 GLRelia rakudo-moar c563ec: OUTPUT«stringifyargs␤»
15:23 camelia rakudo-moar 47ddca: OUTPUT«stringify args␤»
15:25 nine Fix for that coming up
15:25 nine for 1.WHAT.say that is
15:25 nine m: 1.WHAT.say
15:25 camelia rakudo-moar 47ddca: OUTPUT«(Int)␤»
15:25 GLRelia rakudo-moar c563ec: OUTPUT«␤»
15:26 dalek doc: cdd3c1d | paultcochrane++ | lib/Language/modules.pod:
15:26 dalek doc: Add information about basic module structure and file extensions
15:26 dalek doc: review: https://github.com/perl6/doc/commit/cdd3c1ddef
15:26 dalek rakudo/glr: 814d421 | jnthn++ | src/core/array_slice.pm:
15:26 dalek rakudo/glr: Implement *-1 style things in multi-dim.
15:26 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/814d4214d7
15:26 colomon nine: yeah, that seems wrongish.  :)
15:26 smls Oh, I see jnthn is back to multidim work?
15:27 jnthn smls: Well, re-instating @a[1;1] style things
15:27 nine smls: yep, he just didn't find something else to do ;)
15:27 JimmyZ seq.t would be good one
15:28 dalek rakudo/glr: 3deda75 | (Stefan Seifert)++ | src/core/io_operators.pm:
15:28 dalek rakudo/glr: Fix 1.WHAT.say not printing anything
15:28 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/3deda7574f
15:29 nine m: my ($x, @b)  = ('xxx', (1,2,3).map(*.Int)); say @b.perl;
15:29 camelia rakudo-moar 47ddca: OUTPUT«[1, 2, 3]<>␤»
15:29 GLRelia rakudo-moar c563ec: OUTPUT«[(1, 2, 3).Seq,]␤»
15:29 nine jnthn: any hint on where to look for that ^^^ would be appreciated
15:30 dalek roast/glr: bc14686 | (Stefan Seifert)++ | S16-io/say-and-ref.t:
15:30 dalek roast/glr: Fix fallout from GLRification of say
15:30 dalek roast/glr: review: https://github.com/perl6/roast/commit/bc14686baf
15:31 khw joined #perl6
15:31 FROGGS m: my @a = any(3, 4); my $ok = 0; my $iterations = 0; for @a -> $x { $ok = 1 if $x ~~ Junction; $iterations++; }; say $ok; say $iterations
15:31 GLRelia rakudo-moar 814d42: OUTPUT«0␤1␤»
15:31 camelia rakudo-moar 47ddca: OUTPUT«1␤1␤»
15:32 colomon nine: I don’t think that was actually the problem I’m seeing?
15:32 nine unofficial count: 74
15:32 dalek doc: f3935b3 | paultcochrane++ | lib/Type/X/Syntax/AddCatego​rial/MissingSeparator.pod:
15:32 dalek doc: Add missing title to X::Syntax::AddCategorial::MissingSeparator
15:32 dalek doc: review: https://github.com/perl6/doc/commit/f3935b3227
15:32 jnthn nine: LIST.STORE
15:32 rurban joined #perl6
15:32 jnthn nine: Uh, List.STORE
15:33 timotimo STOPE IT, LIST
15:33 JimmyZ m: my @b  = (1,2,3).map(*.Int); say @b.perl;
15:33 camelia rakudo-moar 47ddca: OUTPUT«[1, 2, 3]<>␤»
15:33 GLRelia rakudo-moar 814d42: OUTPUT«[1, 2, 3]␤»
15:33 colomon verb [ no obj. ] (usu. as noun stoping)
15:33 colomon (in mining) excavate a series of steps or layers in (the ground or rock).
15:33 colomon • (as noun stoping) Geology the process by which country rock is broken up and removed by the upward movement of magma.
15:33 colomon (sorry, that’s stope).  I had no idea.
15:33 masak jnthn: "Unhandled exception: ctxlexpad needs an MVMContext" -- what does this mean? :)
15:33 FROGGS m: my @a = any(3, 4); my $ok = 0; for @a -> $x { say 'loop'; say $x }
15:33 GLRelia rakudo-moar 814d42: OUTPUT«loop␤4␤»
15:33 camelia rakudo-moar 47ddca: OUTPUT«loop␤any(3, 4)␤»
15:34 Ven .oO( it means you're masak`grr now )
15:34 FROGGS m: my @a = any(3, 4); say @a
15:34 GLRelia rakudo-moar 814d42: OUTPUT«4␤»
15:34 camelia rakudo-moar 47ddca: OUTPUT«any(3, 4)␤»
15:34 jnthn masak: You used nqp::ctxlexpad on something that was't what it wanted
15:35 masak ok.
15:35 FROGGS masak: pass it nqp::ctx for example
15:35 timotimo it's possible that masak didn't use that himself :)
15:35 FROGGS m: nqp::ctxlexpad(nqp::ctx)
15:35 GLRelia rakudo-moar 814d42: OUTPUT«======================================​=========================================␤The use of nqp::operations has been deprecated for non-CORE code.  Please␤change your code to not use these non-portable functions.  If you really want␤to keep using nqp:…»
15:35 camelia rakudo-moar 47ddca: OUTPUT«======================================​=========================================␤The use of nqp::operations has been deprecated for non-CORE code.  Please␤change your code to not use these non-portable functions.  If you really want␤to keep using nqp:…»
15:35 masak actually all I did was run perl6-debug-m on my code :)
15:35 itz try turning it off and on again?
15:35 FROGGS m: use nqp; nqp::ctxlexpad(nqp::ctx)
15:35 camelia rakudo-moar 47ddca: ( no output )
15:35 GLRelia rakudo-moar 814d42: ( no output )
15:35 * Ven invokes ./masak-debug-m
15:35 colomon m: my @a; @a.push(Hash); say @a.perl
15:35 GLRelia rakudo-moar 814d42: OUTPUT«[]␤»
15:35 camelia rakudo-moar 47ddca: OUTPUT«[Hash]<>␤»
15:35 masak Ven: ctxlexpad! blblblbl!
15:35 colomon THAT’S what’s causing the problem I’m looking at.
15:36 masak ok, Grammar::Debugger also dies with a random unrelated error... :)
15:36 masak conclusion: my problem is too difficult for debugging tools!
15:36 masak "figure it out yourself, masak!" :P
15:36 Ven .oO( who debugs the debugger? )
15:36 araujo joined #perl6
15:36 * masak .oO( qui debuggiet ipsos debuggerem? )
15:37 araujo joined #perl6
15:38 [Coke] Ven: Pretty sure that's jnthn.
15:39 colomon I think I’d doing someone wrong.  People keep on going out and coming back with alcohol.  I went out and bought water….
15:39 dalek rakudo/glr: 810daba | lizmat++ | src/core/Any.pm:
15:39 dalek rakudo/glr: Add an .nl method so that we can concatenate .say
15:39 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/810dabaaf2
15:39 dalek rakudo/glr: 38d336d | lizmat++ | src/core/io_operators.pm:
15:39 dalek rakudo/glr: Concatenate .say/.note/.print
15:39 dalek rakudo/glr:
15:39 dalek rakudo/glr: The idea being that nqp::concat on MoarVM doesn't actually concatenate
15:39 dalek rakudo/glr: strings, but rather creates a linked list of the original strings in
15:39 dalek rakudo/glr: memory.  So, concatenating strings this way should not create any memory
15:39 dalek rakudo/glr: pressure.  It should however increase the change that different threads
15:39 dalek rakudo/glr: saying stuff, will actually not interleave their messages.
15:39 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/38d336db77
15:40 AlexDaniel joined #perl6
15:40 dalek rakudo/glr: 15e124a | jnthn++ | src/core/array_slice.pm:
15:40 dalek rakudo/glr: Fix .[*-2, *-1; 0].
15:40 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/15e124a408
15:41 timotimo the circularity saw ... it bites :(
15:41 smls m: say (1, 2).WHAT
15:41 camelia rakudo-moar 47ddca: OUTPUT«(Parcel)␤»
15:41 GLRelia rakudo-moar 3deda7: OUTPUT«(List)␤»
15:42 smls hm, locally the GLR branch prints "Invocant requires an instance of type List, but a type object was passed." for that
15:42 masak timotimo: the fact that the circularity saw is *sharp* is probably the least fascinating fact about it :P
15:43 dalek roast/glr: 771f831 | jnthn++ | S09-subscript/multidim-assignment.t:
15:43 dalek roast/glr: Update test for GLR.
15:43 dalek roast/glr: review: https://github.com/perl6/roast/commit/771f83120c
15:43 timotimo i'm not so successful with my endeavour
15:43 FROGGS m: my @a = 1 .. 5; @a[0,1] = @a[1,0]; say @a[0,1].WHAT
15:43 camelia rakudo-moar 47ddca: OUTPUT«(Parcel)␤»
15:43 GLRelia rakudo-moar 3deda7: OUTPUT«(List)␤»
15:44 colomon m: my @a; @a.push(Hash); say @a.perl
15:44 GLRelia rakudo-moar 3deda7: OUTPUT«[]␤»
15:44 camelia rakudo-moar 47ddca: OUTPUT«[Hash]<>␤»
15:44 colomon ^^ is GLR correct here, or is this another push issue?
15:44 FROGGS m: my @a = 1 .. 5; @a[0,1] = @a[1,0]; say @a[0,1] # I think the List.STORE method tries to be too cheat here too much
15:44 GLRelia rakudo-moar 15e124: OUTPUT«22␤»
15:44 camelia rakudo-moar 47ddca: OUTPUT«2 1␤»
15:45 FROGGS err, proper english difficult is
15:45 FROGGS m: my @a; @a.push(Hash); say @a.elems
15:45 GLRelia rakudo-moar 15e124: OUTPUT«0␤»
15:45 camelia rakudo-moar 47ddca: OUTPUT«1␤»
15:45 FROGGS colomon: that is the same issue as Int.elems being zero
15:46 colomon FROGGS: … okay, but is that a bug?
15:46 FROGGS I very much think so
15:47 colomon it’s intentional:
15:47 rindolf joined #perl6
15:47 colomon proto method elems(|) is nodal { * }
15:47 colomon multi method elems(Any:U:) { 0 }
15:47 colomon multi method elems(Any:D:) { self.list.elems }
15:47 FROGGS colomon: we discussed that earlier... that zero there is a thinko
15:48 colomon FROGGS: … okay… it should be a 1?
15:48 FROGGS yes
15:48 colomon seems like a very easy fix?
15:48 FROGGS but that will cause other problems
15:48 colomon oh?
15:48 FROGGS m: my $a; say $a.elems
15:48 camelia rakudo-moar 47ddca: OUTPUT«0␤»
15:48 GLRelia rakudo-moar 15e124: OUTPUT«0␤»
15:48 FROGGS this will be 1 then
15:49 FROGGS which might be okay, but I can't judge on that
15:49 FROGGS m: my $a; for $a { say 42 }
15:49 GLRelia rakudo-moar 15e124: ( no output )
15:49 camelia rakudo-moar 47ddca: OUTPUT«42␤»
15:49 FROGGS yeah, it potentially should output 1 in the script before the last
15:49 dalek perl6-roast-data: 4dd10c4 | coke++ | / (2 files):
15:49 dalek perl6-roast-data: another glr run for 'today'
15:49 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/4dd10c44aa
15:50 * [Coke] can now enter his glr-rebuild-retest loop for the day, whee.
15:51 FROGGS colomon: also change method end
15:52 colomon FROGGS: to 0?
15:52 FROGGS aye
15:53 colomon those changes are old?  lizmat++ made them in Jan?
15:53 FROGGS aye
15:53 colomon FROGGS: trying the changes to see what happens
15:53 FROGGS but pre-glr still iterated over things that had zero elems
15:53 lizmat what did I change in Jan ?
15:53 colomon lizmat: https://github.com/rakudo/rakudo/commit/​6a3fc316b185057eee161d210493e8a442ec0a22
15:53 dalek rakudo/glr: 3459537 | jnthn++ | src/core/ (2 files):
15:54 dalek rakudo/glr: Implement {;} style indexing.
15:54 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/3459537c6b
15:54 colomon we’re pondering changing the values returned by the :U version of both of these
15:54 rindolf joined #perl6
15:54 lizmat to  Nil or so?
15:54 [Coke] ... me kills and restarts to get jnthn++'s commit. :P
15:54 colomon m: my @a; say @a.elems
15:54 GLRelia rakudo-moar 15e124: OUTPUT«0␤»
15:54 camelia rakudo-moar 47ddca: OUTPUT«0␤»
15:54 colomon m: my @a; say @a.defined
15:54 camelia rakudo-moar 47ddca: OUTPUT«True␤»
15:54 GLRelia rakudo-moar 15e124: OUTPUT«True␤»
15:54 FROGGS lizmat: https://github.com/rakudo/rakudo/commit/​6a3fc316b185057eee161d210493e8a442ec0a22
15:55 colomon hurm.
15:55 colomon that didn’t fix the issue I’ve been looking at.
15:55 FROGGS :/
15:56 nine colomon: @a being empty after @a.push(Hash) is actually constitent with GLR semantics
15:56 colomon nine: you think?
15:56 nine FROGGS: ^^^
15:56 FROGGS nine: how do you push a typo object then?
15:56 skids Hash =:= {} though.
15:56 nine colomon: yes, because of the single-arg rule again. Hash is iterable, so it is flattened when being passed to push
15:56 skids erm !
15:57 colomon oh, maybe that’s why it didn’t fix my issue.
15:57 colomon I think I’m just going to change the test…. ;0
15:57 nine Or...maybe it shouldn't because it's not actually a Hash but the type object for a Hash
15:57 skids Type objects probably should not be considered themselves iterable.
15:57 FROGGS nine: that's what I am thinking
15:58 [Coke] jnthn: https://github.com/coke/perl6-roast-data/b​lob/master/log/rakudo.moar-glr_summary.out
15:58 [Coke] Working on refreshing now.
15:59 timotimo m: say "Perl 6".ords
15:59 GLRelia rakudo-moar 345953: OUTPUT«80101114108823954␤»
15:59 camelia rakudo-moar 47ddca: OUTPUT«80 101 114 108 8239 54␤»
15:59 skids .oO(unless they have an identity value that is iterable)
16:00 lizmat m: say "Perl 6".ords.item
16:00 GLRelia rakudo-moar 345953: OUTPUT«80101114108823954␤»
16:00 camelia rakudo-moar 47ddca: OUTPUT«80 101 114 108 8239 54␤»
16:01 dalek roast/glr: 1a0223c | colomon++ | S32-list/reduce.t:
16:01 dalek roast/glr: Rework test to not push type objects.
16:01 dalek roast/glr:
16:01 dalek roast/glr: Previously it pushed the $hash.WHAT, then checked whether or not it was
16:01 dalek roast/glr: Hash.  Now it pushes the result of $hash ~~ Hash, and checks that it was
16:01 dalek roast/glr: true.
16:01 dalek roast/glr: review: https://github.com/perl6/roast/commit/1a0223c26d
16:01 colomon okay, that gets S32-list/reduce.t passing
16:02 [Coke] m: my @a := gather { for 1..3 { take $_; say @a.perl } }; say @a.perl
16:02 camelia rakudo-moar 47ddca: OUTPUT«()␤()␤()␤This type does not support positional operations␤  in block <unit> at /tmp/dKLzM_Yc_j:1␤␤»
16:02 GLRelia rakudo-moar 345953: OUTPUT«Type check failed in binding; expected 'Positional' but got 'Seq'␤  in block <unit> at /tmp/1dOVHEFQDc:1␤␤»
16:02 [Coke] ^^ RT #66820
16:02 synbot6 Link:  https://rt.perl.org/rt3/Publ​ic/Bug/Display.html?id=66820
16:04 rindolf joined #perl6
16:05 lizmat m: sub a($a) { say $a.perl }; my %h = a => 42, b => 666; a %h
16:05 camelia rakudo-moar 47ddca: OUTPUT«{:a(42), :b(666)}␤»
16:05 GLRelia rakudo-moar 345953: OUTPUT«${:a(42), :b(666)}␤»
16:06 jnthn m: my %h = a => 1, b => 2; my @a = %h; say @a.perl
16:06 camelia rakudo-moar 47ddca: OUTPUT«[:a(1), :b(2)]<>␤»
16:06 GLRelia rakudo-moar 345953: OUTPUT«[:a(1), :b(2)]␤»
16:06 [Coke] m: say (^10).map: { $^n * 2 + 1 }.perl
16:06 camelia rakudo-moar 47ddca: OUTPUT«Cannot call map(Int: Int, Int, Int, Int, Int, Int, Int, Int, Int, Str); none of these signatures match:␤    ($: Whatever, *%_)␤    ($: &block, :$label, *%_)␤  in block <unit> at /tmp/W9wykctx3Y:1␤␤»
16:06 GLRelia rakudo-moar 345953: OUTPUT«Cannot call map(Range: Str); none of these signatures match:␤    ($: &block, :$label, *%_)␤    (HyperIterable:D $: &block, :$label, *%_)␤  in block <unit> at /tmp/ZWpV0OZm_I:1␤␤»
16:06 [ptc] Ven: http://stackoverflow.com/questions/​1273616/how-do-i-compare-two-hashes​-in-perl-without-using-datacompare
16:06 jnthn m: my %h = a => 1, b => 2; my @a = %h; say @a.map(*.WHAT).say
16:06 GLRelia rakudo-moar 345953: OUTPUT«Cannot call map(Array: Whatever); none of these signatures match:␤    ($: &block, :$label, *%_)␤    (HyperIterable:D $: &block, :$label, *%_)␤  in block <unit> at /tmp/OCTDnu6EC_:1␤␤»
16:06 camelia rakudo-moar 47ddca: OUTPUT«a => 1 b => 2␤True␤»
16:06 FROGGS m: my @a = gather { for 1..3 { take $_; say @a.perl } }; say @a.perl
16:06 GLRelia rakudo-moar 345953: OUTPUT«[]␤[]␤[]␤[1, 2, 3]␤»
16:06 camelia rakudo-moar 47ddca: OUTPUT«[]<>␤[]<>␤[]<>␤[1, 2, 3]<>␤»
16:06 lizmat m: sub a(\a) { say a.perl }; my %h = a => 42, b => 666; a %h
16:06 camelia rakudo-moar 47ddca: OUTPUT«{:a(42), :b(666)}<>␤»
16:06 GLRelia rakudo-moar 345953: OUTPUT«{:a(42), :b(666)}␤»
16:06 [ptc] Ven: as well as http://learn.perl.org/faq/perlfaq4.html#How-do​-I-test-whether-two-arrays-or-hashes-are-equal
16:06 Ven [ptc]: thanks
16:07 lizmat m: sub a(**@a) { say @a.perl }; my %h = a => 42, b => 666; a %h
16:07 GLRelia rakudo-moar 345953: OUTPUT«[{:a(42), :b(666)},]␤»
16:07 camelia rakudo-moar 47ddca: OUTPUT«(:a(42), :b(666))␤»
16:07 lizmat m: sub a(*@a) { say @a.perl }; my %h = a => 42, b => 666; a %h
16:07 GLRelia rakudo-moar 345953: OUTPUT«[:a(42), :b(666)]␤»
16:07 camelia rakudo-moar 47ddca: OUTPUT«[:a(42), :b(666)]<>␤»
16:08 araujo joined #perl6
16:08 araujo joined #perl6
16:08 lizmat m: sub a(*@a) { say @a.perl }; a { a => 42, b => 666 }
16:08 camelia rakudo-moar 47ddca: OUTPUT«[{:a(42), :b(666)}]<>␤»
16:08 GLRelia rakudo-moar 345953: OUTPUT«[:a(42), :b(666)]␤»
16:08 lizmat m: sub a(*@a) { say @a.perl }; a ${ a => 42, b => 666 }
16:08 GLRelia rakudo-moar 345953: OUTPUT«[{:a(42), :b(666)},]␤»
16:08 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/S_7n2s6Iq9␤Unsupported use of ${ a => 42, b => 666 }; in Perl 6 please use $( a => 42, b => 666 )␤at /tmp/S_7n2s6Iq9:1␤------> 3 say @a.perl }; a ${ a => 42, b => 666 }7⏏5<EOL>␤»
16:10 dalek rakudo/glr: 9ba8e43 | TimToady++ | src/core/ (3 files):
16:10 dalek rakudo/glr: 1st whack at coping with self-referential data
16:10 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/9ba8e43d1f
16:10 [Coke] m: sub foo(@a) { @a[1] = "NOES" }; my @a = <OH HAI>; foo(@a); say ~@a
16:10 camelia rakudo-moar 47ddca: OUTPUT«OH NOES␤»
16:10 GLRelia rakudo-moar 345953: OUTPUT«OH NOES␤»
16:10 [Coke] m: sub foo(@a) { @a[0, 1] = <OH NOES> }; my @a = 0, 1; foo(@a)
16:11 camelia rakudo-moar 47ddca: ( no output )
16:11 GLRelia rakudo-moar 345953: ( no output )
16:11 [Coke] m: sub foo(@a) { @a[0, 1] = <OH NOES> }; my @a = 0, 1; foo(@a); say ~@a;
16:11 camelia rakudo-moar 47ddca: OUTPUT«OH NOES␤»
16:11 GLRelia rakudo-moar 345953: OUTPUT«OH NOES␤»
16:11 FROGGS nine: what are you currently working on?
16:12 [Coke] masak: can you please review RT #74902?
16:12 synbot6 Link:  https://rt.perl.org/rt3/Publ​ic/Bug/Display.html?id=74902
16:12 * masak does
16:12 dalek roast/glr: 3fa8b4d | lizmat++ | S16-io/ (3 files):
16:12 dalek roast/glr: note/print/say stupid test clean up
16:12 dalek roast/glr: review: https://github.com/perl6/roast/commit/3fa8b4d1cd
16:13 colomon m: rindex( 'xxy','y', -1e34 )
16:13 camelia rakudo-moar 47ddca: OUTPUT«WARNINGS:␤Useless use of "rindex( 'xxy','y', -1e34 )" in expression "rindex( 'xxy','y', -1e34 )" in sink context (line 1)␤»
16:13 GLRelia rakudo-moar 345953: OUTPUT«WARNINGS:␤Useless use of "rindex( 'xxy','y', -1e34 )" in expression "rindex( 'xxy','y', -1e34 )" in sink context (line 1)␤»
16:13 colomon m: say rindex( 'xxy','y', -1e34 )
16:13 camelia rakudo-moar 47ddca: OUTPUT«Nil␤»
16:13 GLRelia rakudo-moar 345953: OUTPUT«Use of Nil in string context  in block <unit> at /tmp/FgRhUOy8do:1␤␤»
16:13 masak [Coke]: no.
16:13 lizmat m: say rindex( 'xxy','y', -1 )
16:13 GLRelia rakudo-moar 345953: OUTPUT«2␤»
16:13 camelia rakudo-moar 47ddca: OUTPUT«Nil␤»
16:14 masak [Coke]: I think you misunderstood the goal of the ticket.
16:14 nine FROGGS: I'm now changing everything, so itemization prevents from flattening under the single-arg-rule
16:14 [Coke] one of the goals was "these should be consistent"
16:14 [Coke] but that wasn't the original request.
16:15 lizmat m: say $*FOO
16:15 camelia rakudo-moar 47ddca: OUTPUT«Dynamic variable $*FOO not found␤  in block <unit> at /tmp/xh0SL6RIGZ:1␤␤Actually thrown at:␤  in block <unit> at /tmp/xh0SL6RIGZ:1␤␤»
16:15 GLRelia rakudo-moar 345953: OUTPUT«Dynamic variable $*FOO not found␤  in block <unit> at /tmp/OV47SxHDTI:1␤␤Actually thrown at:␤  in block <unit> at /tmp/OV47SxHDTI:1␤␤»
16:15 [Coke] "so something is inconsistent, at least."
16:15 masak [Coke]: the spec said (and still *says*, I think) that the *contents* of the array parameter are readonly. that's what the "one level down" thing refers to.
16:15 masak [Coke]: I might try to find the relevant bit of spec for you.
16:16 dalek rakudo/glr: 7467173 | TimToady++ | src/core/Array.pm:
16:16 dalek rakudo/glr: remove useless dynvar declarations
16:16 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/7467173fe3
16:16 [Coke] please add it to the ticket if so, yes.
16:17 rangerprice joined #perl6
16:18 TimToady m: my $a = 1; my $b = 2; $a = ($a, $b); say $a[0]
16:18 GLRelia rakudo-moar 9ba8e4: OUTPUT«(\List_140492816507288 = (List_140492816507288 2))2␤»
16:18 camelia rakudo-moar 47ddca: OUTPUT«Memory allocation failed; could not allocate 29952 bytes␤»
16:18 dalek roast/glr: b057321 | FROGGS++ | S03-operators/repeat.t:
16:18 dalek roast/glr: fix plan, was busted by 06b5ba0
16:18 dalek roast/glr: review: https://github.com/perl6/roast/commit/b057321a2d
16:19 masak [Coke]: However, the default forces any container argument to also
16:19 masak be treated as an immutable value.  This extends down only one level;
16:19 masak an immutable container may always return an element that is mutable if
16:19 masak it so chooses.
16:19 TimToady m: my $a = 1; my $b = 2; $a = (a => 2,:b($a, $b)); say $a.perl
16:19 GLRelia rakudo-moar 9ba8e4: OUTPUT«(my \List_140420932479392 = (:a(2), :b($(List_140420932479392, 2))))␤»
16:19 masak S06:718
16:19 synbot6 Link: http://design.perl6.org/S06.html#line_718
16:19 FROGGS m: my $a = 1; my $b = 2; $a = (a => 2,:b($a, $b)); say $a.perl.EVAL
16:19 GLRelia rakudo-moar 9ba8e4: OUTPUT«5===SORRY!5=== Error while compiling EVAL_0␤Variable '&List_140203337335216' is not declared␤at EVAL_0:1␤------> 3my \List_140203337335216 = (:a(2), :b($(7⏏5List_140203337335216, 2))))␤»
16:19 TimToady m: my $a = 1; my $b = 2; $a = {a => 2,:b($a, $b)}; say $a.perl
16:19 GLRelia rakudo-moar 9ba8e4: OUTPUT«(my \Hash_139742432439992 = {:a(2), :b($(Hash_139742432439992, 2))})␤»
16:20 * masak adds this to the ticket
16:20 camelia rakudo-moar 47ddca: OUTPUT«(timeout)»
16:20 camelia rakudo-moar 47ddca: OUTPUT«(timeout)»
16:20 camelia rakudo-moar 47ddca: OUTPUT«(timeout)»
16:20 FROGGS m: my $a = 1; my $b = 2; $a = {a => 2,:b($a, $b)}; say $a.perl.EVAL
16:20 GLRelia rakudo-moar 9ba8e4: OUTPUT«5===SORRY!5===␤Variable '&Hash_140342462502600' is not declared␤at EVAL_0:1␤------> 3my \Hash_140342462502600 = {:a(2), :b($(7⏏5Hash_140342462502600, 2))})␤Variable '&Hash_140342462502600' is not declared␤at EVAL_0:1␤------> 3my \Has…»
16:20 FROGGS k, same
16:21 TimToady looks like my \foo isn't declaring the symbol soon enough
16:21 camelia rakudo-moar 47ddca: OUTPUT«(timeout)»
16:21 dalek nqp/nfa_to_statelist: e4f5004 | timotimo++ | src/QRegex/ (2 files):
16:21 dalek nqp/nfa_to_statelist: try to nuke statelist and keep fates
16:21 dalek nqp/nfa_to_statelist:
16:21 dalek nqp/nfa_to_statelist: but we're not getting far at all yet.
16:21 dalek nqp/nfa_to_statelist: review: https://github.com/perl6/nqp/commit/e4f5004ab9
16:22 TimToady std: (my \Hash_139742432439992 = {:a(2), :b($(Hash_139742432439992, 2))})
16:22 camelia std 28329a7: OUTPUT«ok 00:00 142m␤»
16:22 FROGGS TimToady: it does so potentially after the initializer?
16:22 TimToady std declares it soon enough :)
16:23 TimToady m: (my \Hash_139742432439992 = {:a(2), :b($(Hash_139742432439992, 2))})
16:23 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5===␤Variable '&Hash_139742432439992' is not declared␤at /tmp/5tJjoJTYeL:1␤------> 3my \Hash_139742432439992 = {:a(2), :b($(7⏏5Hash_139742432439992, 2))})␤Variable '&Hash_139742432439992' is not declared␤at /tmp/5tJjoJTYeL:1…»
16:23 colomon how do we rebuild GLRelia?
16:23 GLRelia joined #perl6
16:23 GLRelia std : OUTPUT«Can't chdir to '/home/steve/std/snap': No such file or directory at lib/EvalbotExecuter.pm line 171.␤      EvalbotExecuter::_auto_execute(HASH(0x206fe38), "(my \\Hash_139742432439992 = {:a(2), :b(\$(Hash_139742432439992"..., GLOB(0x2417bc8), "/tmp/AKTCivk9Ni", "std") calle…»
16:24 colomon or is it only one commit behind?
16:24 masak ok, 007 now has custom operators \o/
16:24 colomon m: say rindex( 'xxy','y', -1e34 )
16:24 camelia rakudo-moar 47ddca: OUTPUT«Nil␤»
16:24 GLRelia rakudo-moar 746717: OUTPUT«Use of Nil in string context  in block <unit> at /tmp/vdHBDNEDhb:1␤␤»
16:25 MilkmanDan joined #perl6
16:25 * colomon is really confused about the difference he is getting between evalbot and his local p6
16:26 colomon > say rindex( 'xxy','y', -1e34 )
16:26 colomon index start offset out of range
16:26 masak still have custom macro ops left to do...
16:29 [Coke] what's your local p6 version?
16:29 colomon [Coke]: 7467173fe344366f0c94f431dc42657c69d0f13c
16:29 masak that's GLR, it seems.
16:29 colomon yes
16:29 colomon it’s the latest, so far as I know
16:29 colomon I have been pulling regularly all afternoon
16:30 colomon and GLRelia has been updated to it
16:30 * moritz will shortly go off to celebrate his 5th wedding aniversary
16:30 brrt moritz++
16:30 colomon moritz++
16:31 FROGGS moritz: have fun! :o)
16:31 [Coke] moritz: congrats!
16:32 [Coke] m: say (2015-1993)/(2015-1971)
16:32 camelia rakudo-moar 47ddca: OUTPUT«0.5␤»
16:32 GLRelia rakudo-moar 746717: OUTPUT«0.5␤»
16:32 moritz thanks everybody
16:32 hoelzro moritz: congrats!
16:32 timotimo moritz o/
16:32 jdv79 is there a terse way to coerce in a case like this?: https://gist.github.com/jdv/b92fa2a926d801804f83
16:32 FROGGS TimToady: the Hash_* is declared it the actions here: rakudo/src/Perl6/Actions.nqp:2435:        elsif $<defterm> {
16:33 moritz m: try { die 42 }; my $x = $!.WHAT; say $x;
16:33 camelia rakudo-moar 47ddca: OUTPUT«(X::AdHoc)␤»
16:33 GLRelia rakudo-moar 746717: OUTPUT«Use of uninitialized value $x of type X::AdHoc in string context  in block <unit> at /tmp/CCpdXAmFhs:1␤␤»
16:33 carlin joined #perl6
16:33 [Coke] ^^ as of tomorrow, I'll have been alive only 2x as long as I've been married. Ish.
16:33 masak moritz: congrats
16:33 moritz [Coke]: wow.
16:33 [Coke] And I'm spending it with you guys!
16:33 masak aww :)
16:34 moritz m: try { die 42 }; my $x = $!.WHAT; say $x.gist;
16:34 camelia rakudo-moar 47ddca: OUTPUT«(X::AdHoc)␤»
16:34 GLRelia rakudo-moar 746717: OUTPUT«(X::AdHoc)␤»
16:34 tadzik moritz: congratulations to you both :)
16:34 tadzik moritz: have a lot of fun
16:34 moritz tadzik, masak, hoelzro: thanks
16:35 dalek perl6-roast-data: c74ed93 | coke++ | / (2 files):
16:35 dalek perl6-roast-data: another glr run for 'today'
16:35 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/c74ed93024
16:36 [Coke] only 9 more passing tests. hurry up, people! ;)
16:37 carlin has the panda precomp change caused anything submitted to testers.perl6 to have a build stage grade of "NA"?
16:38 [Coke] *furious investigation*
16:38 tadzik ...could have
16:38 pmurias_ timotimo: the nfa handling is getting changed again?
16:39 tadzik panda's been pretty aggresively mangled by me today
16:39 dalek rakudo/glr: 355b999 | TimToady++ | src/core/Enum.pm:
16:39 dalek rakudo/glr: handle self-referential pair as well in gist/perl
16:39 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/355b9999b4
16:39 timotimo pmurias_: just an optimization
16:39 timotimo but i'm not doing well there at all
16:40 dalek roast/glr: 8256c7f | lizmat++ | S32-io/note.t:
16:40 dalek roast/glr: Fix another note concat issue
16:40 dalek roast/glr: review: https://github.com/perl6/roast/commit/8256c7fe45
16:41 dalek rakudo/glr: 8d17aab | (Stefan Seifert)++ | src/core/Array.pm:
16:41 dalek rakudo/glr: GLRR: itemization should prevent flattening single args
16:41 dalek rakudo/glr:
16:41 dalek rakudo/glr: First part:
16:41 dalek rakudo/glr: my @arr = 1, 2;
16:41 dalek rakudo/glr: Array.new(@arr) and [@arr] return: [1, 2]
16:41 dalek rakudo/glr: Array.new($@arr) and [$@arr] return: [[1, 2]]
16:41 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/8d17aaba32
16:42 FROGGS m: say (1, 2, 3).perl
16:42 camelia rakudo-moar 47ddca: OUTPUT«(1, 2, 3)␤»
16:42 GLRelia rakudo-moar 746717: OUTPUT«(1, 2, 3)␤»
16:42 colomon m: permutations(3).list.perl
16:42 GLRelia rakudo-moar 746717: ( no output )
16:42 camelia rakudo-moar 47ddca: ( no output )
16:43 colomon m: permutations(3).list.perl.say
16:43 camelia rakudo-moar 47ddca: OUTPUT«((0, 1, 2), (0, 2, 1), (1, 0, 2), (1, 2, 0), (2, 0, 1), (2, 1, 0))␤»
16:43 GLRelia rakudo-moar 746717: OUTPUT«((0, 1, 2), List_139685018653472, List_139685018725080, List_139685018769784, List_139685018841176, List_139685018885816)␤»
16:43 nine m: ((1,2,3),(2,3,4),(3,4,5)).perl.say
16:43 camelia rakudo-moar 47ddca: OUTPUT«((1, 2, 3), (2, 3, 4), (3, 4, 5))␤»
16:43 GLRelia rakudo-moar 746717: OUTPUT«((1, 2, 3), (2, 3, 4), (3, 4, 5))␤»
16:44 nine TimToady: ^^^ you got a bug in .perl. Also the reason for t/spec/S32-list/combinations.t failing
16:45 colomon oh, it’s .perl that is the issue?
16:45 colomon m: permutations(3).list.say
16:45 camelia rakudo-moar 47ddca: OUTPUT«0 1 2 0 2 1 1 0 2 1 2 0 2 0 1 2 1 0␤»
16:45 GLRelia rakudo-moar 8d17aa: OUTPUT«(0 1 2)(0 2 1)(1 0 2)(1 2 0)(2 0 1)(2 1 0)␤»
16:47 pmurias_ timotimo: ok, I'm porting over the bitpacking from the CODEPOINT_I_LL and it's a bit annoying
16:48 timotimo ah, yeah
16:48 timotimo i can see that
16:50 dalek ecosystem: 483d972 | (John Scoles)++ | META.list:
16:50 dalek ecosystem: Acme::Flutterby - An object-oriented interface to a butterfly. In what else but Perl 6.
16:50 dalek ecosystem:
16:50 dalek ecosystem: This module provides a simplistic, but powerful, interface to a Butterfly.
16:50 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/483d972967
16:50 dalek ecosystem: 1ab273a | FROGGS++ | META.list:
16:50 dalek ecosystem: Merge pull request #54 from byterock/patch-1
16:50 dalek ecosystem:
16:50 dalek ecosystem: Acme::Flutterby - An object-oriented interface to a butterfly. In wha…
16:50 dalek ecosystem: review: https://github.com/perl6/e​cosystem/commit/1ab273ae85
16:50 tadzik hah! Along with Acme::Meow
16:52 [Coke] nqp::shell complains it requires 7 operands. Not in the docs. :P
16:53 [Coke] (trying to write a test that will fail on the jvm)
16:54 dalek rakudo/glr: 4050d34 | jnthn++ | src/core/Any.pm:
16:54 dalek rakudo/glr: Bring back .tree.
16:54 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/4050d341b3
16:54 * FROGGS .oO( bring back, bring back, bring back the tree to me, to meeee! )
16:56 maettu dinner @7:30 (outside)
16:56 tadzik \o/
16:56 * maettu can already smell it
16:57 colomon m: permutations(3).list[0].WHAT.say
16:57 camelia rakudo-moar 47ddca: OUTPUT«(Parcel)␤»
16:57 GLRelia rakudo-moar 8d17aa: OUTPUT«Use of uninitialized value of type List in string context  in block <unit> at /tmp/6U6qj8LGdf:1␤␤»
16:57 colomon m: permutations(3)[0].WHAT.say
16:57 camelia rakudo-moar 47ddca: OUTPUT«(Parcel)␤»
16:57 GLRelia rakudo-moar 8d17aa: OUTPUT«Use of uninitialized value of type List in string context  in block <unit> at /tmp/5WEs8A9jds:1␤␤»
16:58 [Coke] maettu++
16:59 masak m: sub x($/) { given "foo" { when /foo/ { say "yay" } } }; x(42)
16:59 camelia rakudo-moar 47ddca: OUTPUT«Cannot assign to a readonly variable or a value␤  in sub x at /tmp/3G_glWnQvA:1␤  in block <unit> at /tmp/3G_glWnQvA:1␤␤»
16:59 GLRelia rakudo-moar 4050d3: OUTPUT«Cannot assign to a readonly variable or a value␤  in sub x at /tmp/skf5wIwFrJ:1␤  in block <unit> at /tmp/skf5wIwFrJ:1␤␤»
16:59 masak we could detect this statically.
16:59 dalek rakudo/glr: 066bec0 | TimToady++ | src/core/ (4 files):
16:59 dalek rakudo/glr: allow reuse of objects in self-ref constructs
16:59 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/066bec04fc
16:59 masak if I ask "should we?", will someone pop in and say "no, that's too much complexity for too little gain" ? :)
17:00 TimToady colomon: that should fix your permutations vs perl bug
17:00 colomon TimToady: building now...
17:01 colomon done
17:01 colomon indeed!
17:01 colomon TimToady++
17:01 masak background is this: in sufficiently advanced action methods, I tend to do some smartmatching.
17:01 colomon combinations still failing something
17:01 masak but I have to do it inside of a lexical sub which I call, completely for that purpose
17:03 [Coke] FROGGS: how do I run a command and get at the output?
17:03 tadzik jvm: nqp::ordat('', 0)
17:04 tadzik m: 1
17:04 camelia rakudo-moar 47ddca: OUTPUT«WARNINGS:␤Useless use of constant integer 1 in sink context (line 1)␤»
17:04 GLRelia rakudo-moar 066bec: OUTPUT«WARNINGS:␤Useless use of constant integer 1 in sink context (line 1)␤»
17:05 tadzik j: nqp::ordat('', 0)
17:05 GLRelia rakudo-jvm 066bec: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
17:05 camelia rakudo-jvm 47ddca: OUTPUT«===SORRY!===␤java.l​ang.NullPointerException␤»
17:05 FROGGS [Coke]: in nqp?
17:07 dalek roast/glr: af9c772 | jnthn++ | S04-statements/for-scope.t:
17:07 dalek roast/glr: Update test for [...] semantics.
17:07 dalek roast/glr: review: https://github.com/perl6/roast/commit/af9c772c8f
17:07 dalek roast/glr: c1677a5 | jnthn++ | S06-signature/passing-arrays.t:
17:07 dalek roast/glr: for $a { } with Iterable now iterates.
17:07 dalek roast/glr: review: https://github.com/perl6/roast/commit/c1677a5499
17:07 [Coke] updated question: how to do .perl in nqp?
17:08 FROGGS [Coke]: there is nothing like that IMO
17:08 dalek perl6-roast-data: 67f9a94 | coke++ | / (2 files):
17:08 dalek perl6-roast-data: another glr run for 'today'
17:08 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/67f9a94e01
17:08 [Coke] trying to diagnose why nqp::shell fails in the jvm; hard when there's no tests. :|
17:09 ribasushi joined #perl6
17:10 dalek roast/glr: 6401b9a | jnthn++ | S06-signature/sigilless.t:
17:10 dalek roast/glr: This isn't a good way to test itemness any more.
17:10 dalek roast/glr: review: https://github.com/perl6/roast/commit/6401b9a9e1
17:11 FROGGS TimToady: I tried to make @recursive.perl.EVAL work: https://gist.github.com/FR​OGGS/da2607dd8fc9d9774d6d
17:11 FROGGS though, only partially successful
17:12 tadzik jvm broken!
17:13 tadzik j: nqp::null()
17:13 GLRelia rakudo-jvm 066bec: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
17:13 camelia rakudo-jvm 47ddca: OUTPUT«===SORRY!===␤java.l​ang.NullPointerException␤»
17:15 FROGGS colomon: how far did you get with the Any.elems patch?
17:15 colomon FROGGS: I aborted it in confusion.  do you still reckon it’s the right thing to do?  or should I just test it fully?
17:15 colomon FROGGS: I’m multi-tasking a bit here… also trying to sort out smoker issues.  :)
17:16 * colomon tries building it again.
17:16 FROGGS colomon: I think it is the right thing todo, but I can pick up that task again
17:17 FROGGS colomon: I'll do it
17:18 TimToady FROGGS: I was considering adding a <deftermnow> that reduces before parsing the initializer, then we could put it into Actions
17:19 domidumont joined #perl6
17:19 daxim joined #perl6
17:19 FROGGS TimToady: sounds like a plan
17:20 lizmat m: my %h{Any}; given a => 42 { say $_.key.WHICH; (%h.AT-KEY($_.key.WHICH) //= ($_.key => 0)).value += $_.value }; say %h.pairs   # broken on GLR
17:20 camelia rakudo-moar 47ddca: OUTPUT«Str|a␤Str|a => a => 42␤»
17:20 GLRelia rakudo-moar 066bec: OUTPUT«Str|a␤Use of uninitialized value of type Any in numeric context  in block <unit> at /tmp/Pvt0qV6JdZ:1␤Use of uninitialized value of type Any in numeric context  in block <unit> at /tmp/Pvt0qV6JdZ:1␤ObjAt.new => :a(42)␤»
17:20 domidumont joined #perl6
17:22 jnthn m: my %h{Any}; given a => 42 { say $_.key.WHICH; (%h.AT-KEY($_.key.WHICH) //= ($_.key => 0)).perl.say };
17:22 camelia rakudo-moar 47ddca: OUTPUT«Str|a␤:a(0)␤»
17:22 GLRelia rakudo-moar 066bec: OUTPUT«Str|a␤Use of uninitialized value of type Any in numeric context  in block  at /tmp/EYJQmrrKyq:1␤:a(0)␤»
17:23 jnthn m: my %h{Any}; given a => 42 { (($_.key => 0)).perl.say };
17:23 camelia rakudo-moar 47ddca: OUTPUT«:a(0)␤»
17:23 GLRelia rakudo-moar 066bec: OUTPUT«Use of uninitialized value of type Any in numeric context  in block  at /tmp/mh9G0L58YP:1␤:a(0)␤»
17:23 jnthn m: given a => 42 { (($_.key => 0)).perl.say };
17:23 camelia rakudo-moar 47ddca: OUTPUT«:a(0)␤»
17:23 GLRelia rakudo-moar 066bec: OUTPUT«Use of uninitialized value of type Any in numeric context  in block  at /tmp/nLbZHXs7xa:1␤:a(0)␤»
17:23 jnthn m: given a => 42 { (($_.key => 0)).say };
17:23 camelia rakudo-moar 47ddca: OUTPUT«a => 0␤»
17:23 GLRelia rakudo-moar 066bec: OUTPUT«Use of uninitialized value of type Any in numeric context  in block  at /tmp/WgI9wOuUZF:1␤:a(0)␤»
17:23 jnthn m: given a => 42 { (($_.key => 0)).WHAT.say };
17:23 camelia rakudo-moar 47ddca: OUTPUT«(Pair)␤»
17:23 GLRelia rakudo-moar 066bec: OUTPUT«Use of uninitialized value of type Pair in string context  in block  at /tmp/4EtK1PD6fL:1␤␤»
17:24 rindolf joined #perl6
17:24 jnthn m: given a => 42 { (($_.key => 0)).DEFINITE };
17:24 camelia rakudo-moar 47ddca: ( no output )
17:24 GLRelia rakudo-moar 066bec: ( no output )
17:24 jnthn m: given a => 42 { (($_.key => 0)).DEFINITE.say };
17:24 camelia rakudo-moar 47ddca: OUTPUT«True␤»
17:24 GLRelia rakudo-moar 066bec: OUTPUT«True␤»
17:24 daxim hahainternet, what did you find re devanagari codepoints?
17:24 jnthn m: given a => 42 { (($_.key => 0)).^name.say };
17:24 camelia rakudo-moar 47ddca: OUTPUT«Pair␤»
17:24 GLRelia rakudo-moar 066bec: OUTPUT«Pair␤»
17:25 jnthn m: given a => 42 { .key.say };
17:25 camelia rakudo-moar 47ddca: OUTPUT«a␤»
17:25 GLRelia rakudo-moar 066bec: OUTPUT«a␤»
17:25 jnthn m: given a => 42 { (.key => 'x').say };
17:25 camelia rakudo-moar 47ddca: OUTPUT«a => x␤»
17:25 GLRelia rakudo-moar 066bec: OUTPUT«Use of uninitialized value of type Any in numeric context  in block  at /tmp/tAPfJOGESb:1␤:a("x")␤»
17:26 araujo joined #perl6
17:26 araujo joined #perl6
17:26 TimToady m: given a => 42 { (($_.key => 0)).WHICH.say }
17:26 camelia rakudo-moar 47ddca: OUTPUT«Pair|67325440␤»
17:26 GLRelia rakudo-moar 066bec: OUTPUT«Pair|52411280␤»
17:27 hahainternet daxim: that whatever is in rakudo as it stands doesn't work as the spec dictates
17:27 hahainternet so i'll just wait
17:27 TimToady m: given a => 42 { (($_.key => 0)).WHAT }
17:27 camelia rakudo-moar 47ddca: ( no output )
17:27 GLRelia rakudo-moar 066bec: ( no output )
17:27 hahainternet i'm not fussy, i appreciate there's a lot of work to be done
17:28 daxim I see
17:28 TimToady m: given a => 42 { (($_.key => 0)).WHAT.say }
17:28 camelia rakudo-moar 47ddca: OUTPUT«(Pair)␤»
17:28 GLRelia rakudo-moar 066bec: OUTPUT«Use of uninitialized value of type Pair in string context  in block  at /tmp/tWRmZ5JHer:1␤␤»
17:28 hahainternet i didn't bother anyone else here with it yet
17:28 hahainternet m: "नि".chars
17:28 camelia rakudo-moar 47ddca: ( no output )
17:28 GLRelia rakudo-moar 066bec: ( no output )
17:28 hahainternet m: say "नि".chars
17:28 camelia rakudo-moar 47ddca: OUTPUT«2␤»
17:28 GLRelia rakudo-moar 066bec: OUTPUT«2␤»
17:28 hahainternet so ya, if that doesn't work, nothing else will
17:29 hahainternet but i know it's been on the list for ages so either spec or similar will be updated eventually
17:29 hahainternet glr probably lot more important :D
17:29 lizmat m: my %b := BagHash.new(<a b o>,< p a p o o>); my %h = %b; say %b.perl
17:29 camelia rakudo-moar 47ddca: OUTPUT«Odd number of elements found where hash initializer expected␤  in block <unit> at /tmp/_McWAB6VDB:1␤␤»
17:29 GLRelia rakudo-moar 066bec: OUTPUT«("a"=>2,"p"=>2,"b"=>1,"o"=>3).BagHash␤»
17:30 jnthn m: say "नि".NFD.list.say
17:30 GLRelia rakudo-moar 066bec: OUTPUT«23442367␤True␤»
17:30 camelia rakudo-moar 47ddca: OUTPUT«2344 2367␤True␤»
17:30 jnthn m: say "नि".NFC.list.say
17:30 camelia rakudo-moar 47ddca: OUTPUT«2344 2367␤True␤»
17:30 GLRelia rakudo-moar 066bec: OUTPUT«23442367␤True␤»
17:30 jnthn m: say uniprop('Canonical_Combining_Class', 2367)
17:30 camelia rakudo-moar 47ddca: OUTPUT«Cannot call uniprop(Int, Int); none of these signatures match:␤    (Str $str, |c)␤    (Int $code, Stringy $propname = { ... })␤  in block <unit> at /tmp/muOYuduNGf:1␤␤»
17:30 GLRelia rakudo-moar 066bec: OUTPUT«Cannot call uniprop(Int, Int); none of these signatures match:␤    (Str $str, |c)␤    (Int $code, Stringy $propname = { ... })␤  in block <unit> at /tmp/InTi71kC3O:1␤␤»
17:30 jnthn m: say uniprop(2367, 'Canonical_Combining_Class')
17:30 camelia rakudo-moar 47ddca: OUTPUT«0␤»
17:30 GLRelia rakudo-moar 066bec: OUTPUT«0␤»
17:31 hahainternet jnthn: don't spend any time on it, it was just a trivial post on HN i noticed
17:31 hahainternet so please don't divert yourself on my account :)
17:31 jnthn hahainternet: Yeah, but anyway that explains everything; things that are combiners are identified in the NFG algo by having a CCC > 0
17:32 hahainternet i gotta say i don't quite understand that, but fwiw, that character is literally taken from https://raw.githubusercontent.com/​perl6/specs/master/S15-unicode.pod
17:32 hahainternet which indicates that .chars would return 1
17:34 jnthn Well, the doc pre-dates any actual NFG implementation work
17:34 jnthn So may need bringing in line with what actually got done.
17:34 hahainternet perhaps, there still seems to be some disagreement there but it's not with me, here is the HN comment relating to it: https://news.ycombinator.com/item?id=10130002
17:35 hahainternet so one assumes there should be a way to get a count that = 1 grapheme, but buggered if i know where it'd be, thanks for elucidating jnthn
17:36 araujo joined #perl6
17:36 araujo joined #perl6
17:37 araujo_ joined #perl6
17:38 FROGGS m: say "".split("").perl
17:38 GLRelia rakudo-moar 066bec: OUTPUT«Nil␤»
17:38 camelia rakudo-moar 47ddca: OUTPUT«Nil␤»
17:38 tony-o lizmat: Q about the recommendation manager in S22. is the recommendation manager referring to .candidates() in a CUR, or is the recommendation manager something further up the food chain that is expected query all CURs in the scope and combine the results of their .candidates
17:38 FROGGS m: say "".split("").elems
17:38 camelia rakudo-moar 47ddca: OUTPUT«0␤»
17:38 GLRelia rakudo-moar 066bec: OUTPUT«0␤»
17:39 FROGGS m: say "".split("").elems; say Nil.elems
17:39 camelia rakudo-moar 47ddca: OUTPUT«0␤0␤»
17:39 GLRelia rakudo-moar 066bec: OUTPUT«0␤0␤»
17:39 nine jnthn: what's the difference between [,] and my &infix:<op> := &[,]; [op]?
17:40 pink_mist precedence?
17:40 TimToady &[,] is always treated as binary
17:40 TimToady [,] will be a list associative operator
17:41 nine so [,] will hit a different meta op implementation than [op]?
17:41 TimToady yup
17:42 nine That would explain:
17:42 nine m: my &infix:<op> := &[,]; say ([op] [1,2,3]).perl; say ([,] [1,2,3]).perl;
17:42 camelia rakudo-moar 47ddca: OUTPUT«([1, 2, 3],)␤([1, 2, 3],)␤»
17:42 GLRelia rakudo-moar 066bec: OUTPUT«((1, 2), 3)␤(1, 2, 3)␤»
17:42 TimToady m: my &infix:<op> := &infix:<,>; say ([op] [1,2,3]).perl; say ([,] [1,2,3]).perl;
17:42 masak TimToady: is there a way to refer to the list associative operator as a noun?
17:42 camelia rakudo-moar 47ddca: OUTPUT«([1, 2, 3],)␤([1, 2, 3],)␤»
17:42 GLRelia rakudo-moar 066bec: OUTPUT«((1, 2), 3)␤(1, 2, 3)␤»
17:43 lizmat dinner&
17:43 TimToady hmm, looks like the long form is also misbehaving
17:43 Peter_R joined #perl6
17:45 TimToady m: say &infix:<,>.prec
17:45 camelia rakudo-moar 47ddca: OUTPUT«assoc => list, prec => g=␤»
17:45 GLRelia rakudo-moar 066bec: OUTPUT«Use of uninitialized value of type Any in numeric context  in block <unit> at /tmp/DH2Qt_UjM0:1␤Use of uninitialized value of type Any in numeric context  in block <unit> at /tmp/DH2Qt_UjM0:1␤:prec("g="):assoc("list")␤»
17:45 TimToady well, it says it's list associative
17:49 muraiki does anyone have any advice for daemonizing a perl6 script?
17:50 hahainternet what platform?
17:50 muraiki ubuntu 14.04
17:50 hahainternet a simple systemd unit file would do it for you in the 'correct' ubuntu-y way
17:51 muraiki 14.04 is still on upstart, which I was giong to use, but I didn't know if there was perhaps a better way
17:51 araujo joined #perl6
17:51 araujo joined #perl6
17:51 hahainternet i was /just/ about to ask that :)
17:51 muraiki hehe :)
17:51 hahainternet i realised 14.04 was a bit older than i thought
17:51 muraiki a lot happens in a 1.5 years!
17:51 hahainternet but yeah, there are lots of options but i prefer the 'simple' systemd approach
18:05 kernel joined #perl6
18:05 MilkmanDan joined #perl6
18:13 araujo joined #perl6
18:15 Sqirrel joined #perl6
18:16 arafatkamaal joined #perl6
18:17 arafatkamaal Hi
18:17 arafatkamaal a noob question
18:17 arafatkamaal $ perl6 > my @a = 1, 2; 1 2 > say @a.elems 2 > say @a[10]   (Any) >
18:17 arafatkamaal shouldn't this be an exception?
18:20 laben no, because it simply is not set. btw, if you assign, it automatically creates the memory slot. check autovivification
18:21 arafatkamaal i get it, but I was trying to access something that didn't exist
18:21 lucasb joined #perl6
18:21 ugexe m:  my @a = 1, 2; 1 2 > say @a.elems 2 > say @a[10]
18:21 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/tMIwJwyIxj␤Two terms in a row␤at /tmp/tMIwJwyIxj:1␤------> 3my @a = 1, 2; 17⏏5 2 > say @a.elems 2 > say @a[10]␤    expecting any of:␤        infix␤        infix stopper␤        statement end…»
18:21 GLRelia rakudo-moar 066bec: OUTPUT«5===SORRY!5=== Error while compiling /tmp/rmyErt_y2J␤Two terms in a row␤at /tmp/rmyErt_y2J:1␤------> 3my @a = 1, 2; 17⏏5 2 > say @a.elems 2 > say @a[10]␤    expecting any of:␤        infix␤        infix stopper␤        statement end…»
18:21 ugexe or do you mean the (Any) > should throw an exception
18:21 laben m:  my @a = 1, 2; 1 2; say @a.elems; say @a[10]
18:21 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/g5awu6DryG␤Two terms in a row␤at /tmp/g5awu6DryG:1␤------> 3my @a = 1, 2; 17⏏5 2; say @a.elems; say @a[10]␤    expecting any of:␤        infix␤        infix stopper␤        statement end␤ …»
18:21 GLRelia rakudo-moar 066bec: OUTPUT«5===SORRY!5=== Error while compiling /tmp/KFd66JBuWU␤Two terms in a row␤at /tmp/KFd66JBuWU:1␤------> 3my @a = 1, 2; 17⏏5 2; say @a.elems; say @a[10]␤    expecting any of:␤        infix␤        infix stopper␤        statement end␤ …»
18:21 laben m:  my @a = 1, 2 say @a.elems; say @a[10]
18:21 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ZZFn7SkwuF␤Two terms in a row␤at /tmp/ZZFn7SkwuF:1␤------> 3my @a = 1, 27⏏5 say @a.elems; say @a[10]␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        state…»
18:21 GLRelia rakudo-moar 066bec: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Ef0b3Czhbs␤Two terms in a row␤at /tmp/Ef0b3Czhbs:1␤------> 3my @a = 1, 27⏏5 say @a.elems; say @a[10]␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        state…»
18:21 laben aargh
18:21 laben m:  my @a = 1, 2; say @a.elems; say @a[10]
18:21 camelia rakudo-moar 47ddca: OUTPUT«2␤(Any)␤»
18:21 GLRelia rakudo-moar 066bec: OUTPUT«2␤Use of uninitialized value @a of type Any in string context  in block <unit> at /tmp/LaW8Ks804c:1␤␤»
18:21 arafatkamaal That was copy pasted from the repl
18:22 cgfbee joined #perl6
18:22 geekosaur hm, LTA error from glr there
18:22 laben i think he means that @a[10] should throw exception since it doesnt exists
18:22 arafatkamaal yes
18:23 [Coke] last batch run is a mixed bag, some improvements, some new failures.
18:23 geekosaur perl6 is still perl; @a does not have a defined size so uninitialized values are undef ((Any) since it is also not typed)
18:23 arafatkamaal ok
18:23 laben arafatkamaal: it doesnt because of autovivification. since you did not declare a size for it (not a common case), it automatically expands
18:24 arafatkamaal so any list is basically of an unlimited size?
18:24 ugexe my @a[10] (shaped arrays) will do what you want but its not implemented yet i dont think
18:24 arafatkamaal ok
18:24 laben @a is an Array. it is not of unlimited size, but can be prolonged how much you need/want
18:24 pink_mist my @a = 1..Inf; #yes they are
18:25 [Coke] m: my @a=1..Inf; say @a.WHAT # pedant.
18:25 camelia rakudo-moar 47ddca: OUTPUT«(Array)␤»
18:25 GLRelia rakudo-moar 066bec: OUTPUT«Cannot look up attributes in a type object␤  in block <unit> at /tmp/TtrkN57RCw:1␤␤»
18:25 laben m: my @a = 1..5; say @a.elems; @a[20]; say @a.elems; @a[20] = 20; say @a.elems;
18:25 camelia rakudo-moar 47ddca: OUTPUT«5␤5␤21␤»
18:25 GLRelia rakudo-moar 066bec: OUTPUT«5␤5␤21␤»
18:25 laben see ^^^
18:25 arafatkamaal ok
18:25 [Coke] pedant check passed, thoughI don't think GLR is right there.
18:26 laben btw, why did panda switch to JSON::Fast? Now i need to actually fix that up for glr...
18:27 [Coke] 516 failing tests remain on glr.
18:28 pink_mist [Coke]: my "yes they are" was to arafatkamaal anyway, and he mentioned lists not arrays :P
18:30 arafatkamaal my @a = 1 .. Inf;
18:30 arafatkamaal say @a.elems
18:30 arafatkamaal should give Inf?
18:30 arafatkamaal or just run forever
18:31 lucasb m: say class { method gist { 'oh' }; method Str { 'hai' } }
18:31 camelia rakudo-moar 47ddca: OUTPUT«oh␤»
18:31 GLRelia rakudo-moar 066bec: OUTPUT«hai␤»
18:32 rindolf pink_mist: hi, sup?
18:33 pink_mist evening rindolf .. been darn tired all day
18:33 arnsholt joined #perl6
18:34 rindolf pink_mist: ah, that sucks.
18:34 rindolf pink_mist: do you know why?
18:34 pink_mist no :/
18:34 rindolf pink_mist: ah.
18:34 rindolf pink_mist: :-(
18:34 rindolf pink_mist: I'm going to hit the shower now.
18:34 rindolf Bye.
18:35 domidumont1 joined #perl6
18:35 TimToady m: my &infix:<op> := &infix:<,>; say &infix:<op>.prec
18:35 camelia rakudo-moar 47ddca: OUTPUT«assoc => list, prec => g=␤»
18:35 GLRelia rakudo-moar 066bec: OUTPUT«Use of uninitialized value of type Any in numeric context  in block <unit> at /tmp/Vy0P9xFYSP:1␤Use of uninitialized value of type Any in numeric context  in block <unit> at /tmp/Vy0P9xFYSP:1␤:prec("g="):assoc("list")␤»
18:39 [Coke] m: my @a=1..Inf; say @a.elems; # arafatkamaal
18:39 camelia rakudo-moar 47ddca: OUTPUT«Inf␤»
18:39 GLRelia rakudo-moar 066bec: OUTPUT«Cannot .elems a lazy list␤  in block <unit> at /tmp/3hE79y_dX3:1␤␤Actually thrown at:␤  in block <unit> at /tmp/3hE79y_dX3:1␤␤»
18:39 [Coke] GLR is the new way.
18:40 arafatkamaal on my rakudo installation it just runs forever, will install from the glr branch
18:40 arafatkamaal what should happen if @b = Inf .. 1;
18:45 MilkmanDan joined #perl6
18:46 dalek rakudo/glr: c5086d7 | TimToady++ | src/core/Enum.pm:
18:46 dalek rakudo/glr: fix Pair.perl uninit bug
18:46 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/c5086d7a05
18:46 [Coke] you can't sequence backwards.
18:46 [Coke] m: my @a = 3..1; say @a.perl;
18:46 GLRelia rakudo-moar 066bec: OUTPUT«[]␤»
18:46 camelia rakudo-moar 47ddca: OUTPUT«[]<>␤»
18:47 yqt joined #perl6
18:47 FROGGS m: use Test; my $should_ret_empty_list1 = sub { return; 1 }; is $should_ret_empty_list1().elems, 0, "our sub returned an empty list (1)";
18:47 GLRelia rakudo-moar 066bec: OUTPUT«ok 1 - our sub returned an empty list (1)␤»
18:47 camelia rakudo-moar 47ddca: OUTPUT«ok 1 - our sub returned an empty list (1)␤»
18:47 FROGGS I think this test is bogus
18:47 Peter_R joined #perl6
18:48 lucasb m: sub f(|) { 42 }; say 2 &f 3
18:48 camelia rakudo-moar 47ddca: OUTPUT«all(2, 42)␤»
18:49 GLRelia rakudo-moar 066bec: OUTPUT«(timeout)»
18:49 lucasb ^^ I think this should be an error, right? It should not call f, and it has nothing to do with junctions
18:49 [Coke] lucasb: it's parsing the & as a junction combiner.
18:49 colomon [Coke]: you can’t range backwards.  you can sequence however you want, if you ask nicely.
18:50 [Coke] whoops, you'd think I'd know the difference after 10 years.
18:50 TimToady m: use Test; my $should_ret_empty_list1 = sub { return slip; 1 }; is $should_ret_empty_list1().elems, 0, "our sub returned an empty list (1)";
18:50 GLRelia rakudo-moar 066bec: OUTPUT«ok 1 - our sub returned an empty list (1)␤»
18:50 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/CbcfF8mSbM␤Undeclared routine:␤    slip used at line 1. Did you mean 'flip', 'skip'?␤␤»
18:53 lucasb [Coke]: oh, understood. It is parsing as '2 & f(3)'
18:54 lucasb I was seeing '&f' as a single token and as expecting a TTIAR error
18:54 lucasb *was
19:06 Peter_R joined #perl6
19:06 FROGGS hmpf
19:07 FROGGS and now I don't know if return() should return Nil, Empty or Slip
19:07 FROGGS $ perl6 -e 'say Empty.elems, Nil.elems, Slip.elems'
19:07 FROGGS 011 # locally
19:07 FROGGS m: say Empty.elems, Nil.elems, Slip.elems
19:07 GLRelia rakudo-moar c5086d: OUTPUT«000␤»
19:07 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/MqfB4kAuRk␤Undeclared name:␤    Slip used at line 1␤␤»
19:07 FROGGS m: say Empty.elems, Nil.elems
19:07 camelia rakudo-moar 47ddca: OUTPUT«00␤»
19:07 GLRelia rakudo-moar c5086d: OUTPUT«00␤»
19:08 lucasb m: sub f {}; say @ = f
19:08 camelia rakudo-moar 47ddca: OUTPUT«(Any)␤»
19:08 GLRelia rakudo-moar c5086d: OUTPUT«(Any)␤»
19:08 FROGGS $ perl6 -e 'say Empty.elems, Nil.elems, Slip.new.elems'
19:08 FROGGS 010
19:08 lucasb oops, sorry
19:10 dalek rakudo/glr: f3d060e | TimToady++ | src/core/Array.pm:
19:10 dalek rakudo/glr: only add comma if single array elem is iterable
19:10 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/f3d060e997
19:10 dalek roast/glr: 485af56 | FROGGS++ | S04-statements/return.t:
19:10 dalek roast/glr: use non-deprecated isa-ok
19:10 dalek roast/glr: review: https://github.com/perl6/roast/commit/485af568ce
19:13 rindolf FROGGS: hi! How was dinner yesterday?
19:14 FROGGS rindolf: I don't remember ó.ò
19:14 rindolf FROGGS: ah.
19:14 Ven joined #perl6
19:15 jdv79 rindolf: do you do any p6 stuff?
19:15 rindolf jdv79: a little - Project Euler/etc.
19:15 rindolf jdv79: at the perl6-examples repository.
19:15 timotimo no lol! no project euler!
19:16 laben i still am not sure of this one-arg rule, why does [{a => 1, b => 1}] transform into [a => 1, b=> 2] instead of {a => 1, b =>2} ?
19:16 rindolf timotimo: PE is love! PE is life! ;-)
19:16 timotimo ban he!
19:16 skids laben: {...} is the one-arg for [...]
19:17 FROGGS okay... now I understand that Empty is not much more than Slip.new
19:17 laben m: sub foo(\a) { dd a }; foo([{a => 1, b => 2}])
19:17 camelia rakudo-moar 47ddca: OUTPUT«$ = [{:a(1), :b(2)}]␤»
19:17 GLRelia rakudo-moar f3d060: OUTPUT«[:a(1), :b(2)]␤»
19:18 skids Is Empty even staying?
19:18 timotimo what's our current stance on how that's supposed to work? what laben just wrote?
19:18 timotimo with a dollar sign?
19:18 FROGGS skids: perhaps as Sugar...
19:19 skids I'm surmising from the "GLRR" commit that it's been decided you need to itemize.
19:19 FROGGS m: sub foo(\a) { dd a }; foo([${a => 1, b => 2}])
19:19 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/7gsHX1U9ZJ␤Unsupported use of ${a => 1, b => 2}; in Perl 6 please use $(a => 1, b => 2)␤at /tmp/7gsHX1U9ZJ:1␤------> 3foo(\a) { dd a }; foo([${a => 1, b => 2}7⏏5])␤»
19:19 GLRelia rakudo-moar f3d060: OUTPUT«[{:a(1), :b(2)},]␤»
19:19 skids m: sub foo(\a) { dd a }; foo($[${a => 1, b => 2}])
19:19 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/T6hLavoSNh␤Unsupported use of ${a => 1, b => 2}; in Perl 6 please use $(a => 1, b => 2)␤at /tmp/T6hLavoSNh:1␤------> 3oo(\a) { dd a }; foo($[${a => 1, b => 2}7⏏5])␤»
19:19 GLRelia rakudo-moar f3d060: OUTPUT«[{:a(1), :b(2)},]␤»
19:20 skids m: sub foo(\a) { dd a }; foo($[{a => 1, b => 2}])
19:20 camelia rakudo-moar 47ddca: OUTPUT«$v = Any␤»
19:20 GLRelia rakudo-moar f3d060: OUTPUT«[:a(1), :b(2)]␤»
19:20 rangerprice i thing that Rust will be my preferred language after Perl
19:21 bin_005 joined #perl6
19:22 skids m: sub foo(\a) { dd a }; foo($[$[a => 1, b => 2]])
19:22 camelia rakudo-moar 47ddca: OUTPUT«Method 'Int' not found for invocant of class 'Pair'␤  in block <unit> at /tmp/QezSscG3tw:1␤␤»
19:22 GLRelia rakudo-moar f3d060: OUTPUT«[[:a(1), :b(2)],]␤»
19:23 timotimo ich habe mir "birdemic" 1 & 2 empfehlen lassen
19:23 timotimo ww
19:25 laben m: sub foo(\a) { dd a }; foo([%{a => 1, b => 2}])
19:25 GLRelia rakudo-moar f3d060: OUTPUT«[:a(1), :b(2)]␤»
19:25 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/WlWIJijYdI␤Unsupported use of %{a => 1, b => 2}; in Perl 6 please use %(a => 1, b => 2)␤at /tmp/WlWIJijYdI:1␤------> 3foo(\a) { dd a }; foo([%{a => 1, b => 2}7⏏5])␤»
19:25 laben m: sub foo(\a) { dd a }; foo([{a => 1, b => 2},])
19:25 camelia rakudo-moar 47ddca: OUTPUT«$ = [{:a(1), :b(2)}]␤»
19:25 GLRelia rakudo-moar f3d060: OUTPUT«[{:a(1), :b(2)},]␤»
19:25 tadzik soo, do we have a list of user-facing changes in glr?
19:25 tadzik like a porting guideline?
19:26 laben m: sub foo(\a) { dd a }; foo([{a => 1, b => 2},].elems)
19:26 camelia rakudo-moar 47ddca: OUTPUT«1␤»
19:26 GLRelia rakudo-moar f3d060: OUTPUT«1␤»
19:27 timotimo like a "Great Refactoring List"?
19:28 tadzik yes!
19:28 tadzik from another channel: [Perl is] a quantum superposition of everyone's favourite language
19:30 colomon jnthn:                 take @result.map(-> $x { $x }).list;
19:30 colomon did NOT work.  Same fail as before
19:30 laben timotimo: here is it a fix for JSON::Fast tests https://gist.github.com/6d4b4539b45368d70058
19:31 nine m: my @a = <a b c d e f>; say @a[[3, 4], 0,].perl;
19:31 camelia rakudo-moar 47ddca: OUTPUT«("c", "a")␤»
19:31 GLRelia rakudo-moar f3d060: OUTPUT«(("d", "e"), "a")␤»
19:32 nine m: my @a = <a b c d e f>; say @a[[3, 4]].perl;
19:32 GLRelia rakudo-moar f3d060: OUTPUT«("d", "e")␤»
19:32 camelia rakudo-moar 47ddca: OUTPUT«"c"␤»
19:33 nine m: my @a = <a b c d e f>; say @a[$[3, 4]].perl;
19:33 camelia rakudo-moar 47ddca: OUTPUT«Indexing requires an instance, tried to do: @a[ (Any) ]␤  in block <unit> at /tmp/bkatXDUy3b:1␤␤»
19:33 GLRelia rakudo-moar f3d060: OUTPUT«"c"␤»
19:34 laben m: my @a = <a b c d e f>; my @b = [3, 4]; say @a[@b].perl;
19:34 camelia rakudo-moar 47ddca: OUTPUT«("c",)␤»
19:34 GLRelia rakudo-moar f3d060: OUTPUT«("d", "e")␤»
19:34 laben m: my @a = <a b c d e f>; my @b = $[3, 4]; say @a[@b].perl;
19:34 camelia rakudo-moar 47ddca: OUTPUT«Indexing requires an instance, tried to do: @a[ (Any) ]␤  in block <unit> at /tmp/4CT76ieaNN:1␤␤»
19:34 GLRelia rakudo-moar f3d060: OUTPUT«("c")␤»
19:37 laben tadzik: panda/master is now the most updated branch, right? did you merge glr into master?
19:37 tadzik laben: I did not
19:37 tadzik was glr merged into rakudo master?
19:37 tadzik I mean, nom :)
19:37 tadzik laben: I was developing on nom all day
19:37 laben ok, so i will merge it in my repo manually
19:37 tadzik oh, thank you, that'd be sweet
19:38 laben i meant locally, or do you want me to push upstream?
19:38 dalek rakudo/glr: ba5494a | TimToady++ | src/Perl6/ (2 files):
19:38 dalek rakudo/glr: introduce my \term before initializer is parsed
19:38 dalek rakudo/glr:
19:38 dalek rakudo/glr: (Still doesn't work, since we haven't really closed the definitional loop yet.)
19:38 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/ba5494a345
19:39 laben omfg a ton of conflicts, :(
19:39 tadzik :(
19:39 tadzik sorry
19:39 tadzik I may have changed whitespaces in places :(
19:40 tadzik totally didn't think of that
19:40 timotimo laben: is/isn't that going to be [${...}] in the future?
19:41 dalek roast/glr: 9081d01 | (Stefan Seifert)++ | S (2 files):
19:41 dalek roast/glr: Adjust flatteing.t to new GLRR semantics
19:41 dalek roast/glr:
19:41 dalek roast/glr: In a nutshell:
19:41 dalek roast/glr: * there is no implicit flattening
19:41 dalek roast/glr: * unless it's a container passed as single argument to a method or function
19:41 dalek roast/glr: * however, itemization prevents flattening even there
19:41 dalek roast/glr:
19:41 dalek roast/glr: foo(@flattens)
19:41 dalek roast/glr: foo($@doesnt-flatten)
19:41 dalek roast/glr: foo(@we, @do, @no, @flatten)
19:41 dalek roast/glr: review: https://github.com/perl6/roast/commit/9081d01282
19:41 timotimo flat teing?
19:41 laben timotimo: ah is it? for now that does not work, even throws
19:41 timotimo mhh, i lost track :S
19:42 laben m: sub foo(\a) { dd a }; foo([$%{a => 1, b => 2}])
19:42 GLRelia rakudo-moar f3d060: OUTPUT«[{:a(1), :b(2)},]␤»
19:42 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/9l0q3FLfS8␤Unsupported use of %{a => 1, b => 2}; in Perl 6 please use %(a => 1, b => 2)␤at /tmp/9l0q3FLfS8:1␤------> 3oo(\a) { dd a }; foo([$%{a => 1, b => 2}7⏏5])␤»
19:43 laben m: sub foo(\a) { dd a }; foo([${a => 1, b => 2}])
19:43 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/KcB_WqMdsZ␤Unsupported use of ${a => 1, b => 2}; in Perl 6 please use $(a => 1, b => 2)␤at /tmp/KcB_WqMdsZ:1␤------> 3foo(\a) { dd a }; foo([${a => 1, b => 2}7⏏5])␤»
19:43 GLRelia rakudo-moar f3d060: OUTPUT«[{:a(1), :b(2)},]␤»
19:43 laben oh now it works
19:43 laben but cannot do, as it breaks nom-living perl6ers
19:44 virtualsue joined #perl6
19:46 TimToady m: my $a = 1; my $b = 2; $a = [1,2,($a, $b)]; say $a.perl.EVAL
19:46 camelia rakudo-moar 47ddca: OUTPUT«1 2 1 2␤»
19:46 GLRelia rakudo-moar ba5494: OUTPUT«12((Mu) 2)␤»
19:47 timotimo hm, so it displays it with a trailing comma ... maybe that's how it's going to stay?
19:47 TimToady progress, doesn't blow up, but we need a way of delaying evaluation of terms that are being defined until the binding
19:48 jdv79 masak: were you working on fixing eqv?  i vaguely remember you were vocal about it.
19:48 TimToady timotimo: it's hard to do it otherwise
19:48 TimToady you end up with things like [$1..10]
19:48 skids GLRR++ now I know what the heck I'm supposed to be doing :-)
19:48 timotimo OK, if that's how it's going to stay, i'll merge laben's pull request for JSON::Fast
19:49 TimToady trailing comma will continue to work, anyway, even if it's not the preferred form for arrays
19:50 timotimo m: OK
19:50 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/A05g_4Id2b␤Undeclared name:␤    OK used at line 1␤␤»
19:50 GLRelia rakudo-moar ba5494: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ocaa_B24sY␤Undeclared name:␤    OK used at line 1␤␤»
19:50 colomon jnthn: https://gist.github.com/co​lomon/14fab50530ef7011b316
19:50 dalek rakudo/glr: e26a731 | colomon++ | src/core/List.pm:
19:50 dalek rakudo/glr: Fix for wonky combinations bug.
19:50 dalek rakudo/glr:
19:50 dalek rakudo/glr: You've got to copy that array before returning it.  Um, for some
19:50 dalek rakudo/glr: reason.
19:50 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/e26a731173
19:50 dalek rakudo/glr: b90cac5 | colomon++ | src/core/List.pm:
19:50 dalek rakudo/glr: Check in golfed demo of the old combinations bug.
19:50 dalek rakudo/glr:
19:50 dalek rakudo/glr: This is not intended to be a long term part of Rakudo.  This
19:50 dalek rakudo/glr: is just for the purpose of debugging the insanity.
19:50 timotimo oops, the m: stayed
19:50 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/b90cac523e
19:51 nine 68 failing now, official count!
19:51 colomon nine: was that counting my patch above?  that fixed another file.
19:51 skids \o/
19:51 tadzik \o/
19:52 jdv79 nine++
19:52 nine colomon: ah, I'm lagging :) Rerunning...
19:52 colomon nine++
19:52 laben what is this GLRR you all speak of?
19:52 timotimo moving the goalpost again :)
19:52 telex joined #perl6
19:52 dalek doc: 33dc2ec | paultcochrane++ | lib/Type/Exception.pod:
19:52 dalek doc: Correct headings to allow die and warn to show up in search
19:52 dalek doc: review: https://github.com/perl6/doc/commit/33dc2ec0d4
19:52 timotimo obviously it's not that "Great" if it only breaks so few tests
19:52 nine Great List Refactor Refinement :)
19:52 jdv79 what happens when the fail count hits 0j?
19:52 jdv79 *0
19:53 colomon jdv79: ship it!
19:53 skids It prevented the slaughter of hundreds of poor tests :-)
19:53 * TimToady --> horizontality
19:53 Peter_R joined #perl6
19:54 laben GLR Refinement? is that a soviet for Perl6 O.o?
19:54 colomon At some point someone was claiming that one of the modules completely messed up panda when you installed it.  Anyone remember what module it was?
19:55 masak jdv79: yes, I was involved. I posted a gist about it. it's in today's backlog.
19:55 jdv79 oh...
19:55 timotimo laben: applied the patch and at the same time pushed some error reporting improvements i had uncommitted locally
19:56 masak jdv79: here it is: https://gist.github.com/masak/b694a4050f4e91bd254c
19:56 masak jdv79: note that this gist documents things the way I'd like them, not the way things are now
19:56 jdv79 oh, before bedtime reading - thanks
19:57 masak I think I saw in the backlog today that TimToady++ was doing something with .perl and cycles
19:58 skids TimToady++ tackling bears.
19:59 arafatkamaal joined #perl6
20:00 timotimo Your branch is behind 'origin/glr' by 165 commits, and can be fast-forwarded.
20:00 tadzik hah!
20:00 timotimo big ups
20:01 arafatkamaal https://gist.github.com/ano​nymous/3c294b28eda4e10aa7f8
20:01 nine 67 failing indeed!
20:01 arafatkamaal Boolean values like 'True' and 'False' can be variable names
20:03 timotimo arafatkamaal: yup, we don't reserve these words
20:03 tadzik so, on GLR my $c = self.bless(:$height, :$width, :$p, :$f); results with Method 'gimme' not found for invocant of class 'Seq'
20:03 tadzik does that ring a bell, or am I missing something obvious?
20:03 nine tadzik: there's no gimme anymore in rakudo's source code
20:04 moritz well, Mu has a gimme method that throws an error
20:04 moritz so you should never get a "Method 'gimme' not found" error
20:04 timotimo for some reason my stage parse has gone above 60 seconds :|
20:05 tadzik so apparently I'm not actually on glr
20:05 moritz tadzik: but nom doesn't have a Seq type
20:05 moritz tadzik: so something with your rakudo is *very* wonky
20:05 tadzik (╯°□°)╯︵ ┻━┻
20:06 mr-foobar joined #perl6
20:07 timotimo congratulations!
20:07 dalek perl6-roast-data: b23c606 | coke++ | / (2 files):
20:07 dalek perl6-roast-data: another glr run for 'today'
20:07 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/b23c6066d4
20:08 dalek perl6-roast-data: fcf7109 | coke++ | / (2 files):
20:08 dalek perl6-roast-data: another glr run for 'today'
20:08 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/fcf710965b
20:08 timotimo ah!
20:08 timotimo i have a moarvm --optimize=0
20:09 nine skids: thanks for your work on push and unshift. It helped showing what we really need to do.
20:10 dalek roast/glr: 0fd48d2 | (Stefan Seifert)++ | S02-types/array.t:
20:10 dalek roast/glr: Need to flatten explicitly in S02-types/array.t
20:10 dalek roast/glr: review: https://github.com/perl6/roast/commit/0fd48d2881
20:14 brrt joined #perl6
20:15 skids nine: np glad it worked out.
20:15 nebuchadnezzar joined #perl6
20:16 dalek rakudo/glr: 268ca08 | lizmat++ | / (2 files):
20:16 dalek rakudo/glr: Make "done" a full supply citizen
20:16 dalek rakudo/glr:
20:16 dalek rakudo/glr: Tomorrow, Jonathan is going to give yet another installment of his latest
20:16 dalek rakudo/glr: async presentation.  We want people to use the new goodies as soon as
20:16 dalek rakudo/glr: possible and not have to change their async code again.  So I thought it
20:16 dalek rakudo/glr: would be a good idea to rename supply-emit and supply-done to just "emit"
20:16 dalek rakudo/glr: and "done".  This then interferes with the "done" from Test.pm (which I
20:16 dalek rakudo/glr: recently deprecated).  It seemed like a good time, during this upheaval,
20:16 dalek rakudo/glr: to at least make this work in the GLR branch.  So "done" is eradicated from
20:16 dalek rakudo/glr: Test.pm (please use "done-testing" instead if you must: you should probably
20:16 dalek rakudo/glr: use "plan" anyway).
20:16 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/268ca0802e
20:17 tadzik from done_testing to done to done-testing :)
20:17 tadzik another 200 PRs to all my modules :P
20:17 dalek roast/glr: d542bef | lizmat++ | S17-supply/syntax.t:
20:17 dalek roast/glr: Make the changes of 268ca0802e6106de2880ddf in glr
20:17 dalek roast/glr: review: https://github.com/perl6/roast/commit/d542bef70b
20:20 llfourn joined #perl6
20:21 jdv79 How about the idea of not polluting namespaces with exports?  Test::done is pretty short and easy.
20:21 * skids spares tadzik from PRs on Grammar::BNF.
20:21 nine jdv79: Test exports so many functions already. The one won't make a difference...
20:21 jdv79 or .done
20:22 tadzik skids: <3
20:22 colomon lizmat: that was a harmless bug!
20:22 colomon ;)
20:23 jdv79 i guess in context.  but in general seems kinda lame.
20:23 lizmat I *squashed* it!
20:23 tadzik skids: ooh
20:23 lizmat well, I guess I could leave sub done in there, just not export it
20:23 tadzik but now it doesn't work on nom anymore
20:23 lizmat but then people would need to change their test anyway
20:23 tadzik maybe we should just make them use plan()
20:23 skids Oh, nom doesn't have done-testing yet?
20:24 skids I'll do that then.
20:24 lizmat and changing it to done-testing sort of matches the P5 experience anyway (I'm told)
20:24 tadzik nope, still on old-done() :)
20:24 jdv79 really?  we've already had the plan vs done debate in the p5 world.
20:24 tadzik we used to have done_testing in p6 world
20:24 tadzik then we changed that to done() because done_testing was RONG
20:24 lizmat well, that was before my time...
20:25 tadzik I don't remember the actual reasonong
20:25 tadzik or ning
20:25 tadzik there's always Test::done() :)
20:26 colomon Test:I-really-think-I'm-done-now
20:27 ugexe 1;
20:27 moritz Test::ichhabefertig
20:27 skids Test::mike-drop
20:27 tadzik or let's use ]
20:27 tadzik like bash
20:27 lizmat tadzik: done was implemented by moritz in 6a9a38f3b4d322a3ea93b
20:27 colomon skids: +1
20:27 lizmat on the last day of 2010
20:27 tadzik hm
20:27 tadzik I'm that old? :o
20:30 tadzik http://blogs.perl.org/users/byterock/201​5/08/my-first-perl6-part-the-first.html byterock++'s notes fromthe hackathon
20:30 lizmat yeah, byterock++
20:31 larion joined #perl6
20:31 dalek specs: f3822a7 | lizmat++ | S99-glossary.pod:
20:31 dalek specs: Add GLRR lemma
20:31 dalek specs: review: https://github.com/perl6/specs/commit/f3822a7f7d
20:32 CQ joined #perl6
20:32 tadzik I guess that looking at roast commits would be a good way to write down GRL?
20:33 lizmat guess so
20:33 TEttinger joined #perl6
20:36 ugexe this 2 hour time limit on macincloud is brutal
20:36 timotimo .o( is that a video game? )
20:37 ugexe its a OSX VM company
20:37 timotimo mhm
20:37 timotimo i wanted to build a website called "macontisch" a long time ago
20:37 timotimo which is a pun on "mac on table" in german
20:40 ugexe where im from it we had the phrase `crackentosh` which referred to the phenomenon of crack addicted people to always happen to have an apple laptop for sale (obviously stolen)
20:40 nine m: my @arr = 1, 2, 3; my ($x, @arr) = @arr; say @arr.perl;
20:40 camelia rakudo-moar 47ddca: OUTPUT«5===SORRY!5=== Error while compiling /tmp/lxhjImXpRp␤Redeclaration of symbol @arr␤at /tmp/lxhjImXpRp:1␤------> 3my @arr = 1, 2, 3; my ($x, @arr7⏏5) = @arr; say @arr.perl;␤    expecting any of:␤        shape declaration␤»
20:40 GLRelia rakudo-moar 268ca0: OUTPUT«5===SORRY!5=== Error while compiling /tmp/eDigtLlrG2␤Redeclaration of symbol @arr␤at /tmp/eDigtLlrG2:1␤------> 3my @arr = 1, 2, 3; my ($x, @arr7⏏5) = @arr; say @arr.perl;␤    expecting any of:␤        shape declaration␤»
20:40 nine m: my @arr = 1, 2, 3; my $x; ($x, @arr) = @arr; say @arr.perl;
20:40 camelia rakudo-moar 47ddca: OUTPUT«[2, 3]<>␤»
20:40 GLRelia rakudo-moar 268ca0: OUTPUT«[2, 3]␤»
20:41 rurban joined #perl6
20:42 itz joined #perl6
20:42 dalek rakudo/glr: c944e43 | (Stefan Seifert)++ | src/core/List.pm:
20:42 dalek rakudo/glr: Fix list assignment evaluating the RHS too late
20:42 dalek rakudo/glr:
20:42 dalek rakudo/glr: Fixes ($a, $b) = ($b, $a)
20:42 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/c944e43b21
20:44 laben tadzik: i have resolved the merge conflicts and pushed to my fork of panda, what do now? want a PR?
20:45 lizmat m: say Hash[Any].new( <a b c>, 42 ).perl
20:45 camelia rakudo-moar 47ddca: OUTPUT«Hash[Any].new("a" => "b", "c" => 42)␤»
20:45 GLRelia rakudo-moar 268ca0: OUTPUT«Hash[Any].new(:a("b"), :c(42))␤»
20:46 lizmat I think in the GLR, the latter is wrong
20:46 tadzik laben: yeah, sure :)
20:46 lizmat m: say Hash[Any].new( <a b c> => 42 ).perl
20:46 camelia rakudo-moar 47ddca: OUTPUT«Type check failed in binding key; expected 'Str' but got 'Parcel'␤  in block <unit> at /tmp/nCL8U3nmev:1␤␤»
20:46 GLRelia rakudo-moar 268ca0: OUTPUT«Type check failed in binding key; expected 'Str' but got 'List'␤  in block <unit> at /tmp/4oK1OX1jbq:1␤␤»
20:46 laben tadzik: https://github.com/tadzik/panda/pull/208
20:47 tadzik awesome, thanks!
20:47 laben i hope nothing broke panda/glr with all the commits
20:47 tadzik that's not all that much :)
20:48 tadzik I like it
20:48 tadzik I thought it'll be very mysterious and strange
20:48 laben yea it turned out i had to delete a ton of lines and that's basically it
20:48 tadzik awesome
20:48 tadzik well done :)
20:48 tadzik do you know if it's backwards compatible with nom?
20:48 laben our work up till now it sure was
20:49 laben our = mine + nine++
20:49 lizmat m: BagHash.new(1, [ foo => 10, bar => 17, baz => 42, santa => 0 ]).perl.say
20:49 camelia rakudo-moar 47ddca: OUTPUT«([:foo(10), :bar(17), :baz(42), :santa(0)]=>1,1=>1).BagHash␤»
20:49 GLRelia rakudo-moar 268ca0: OUTPUT«(:foo(10)=>1,1=>1,:santa(0)=>​1,:bar(17)=>1,:baz(42)=>1).BagHash␤»
20:50 skids lizmat: probably a flat or *@ where a **@ wants to be.
20:50 lizmat skids: spot on
20:50 tadzik laben: perfect :) I'll do some sanity checks and merge it tomorrow morning
20:50 laben i did not test the new version of panda/nom before merging nor did i test on the newest glr
20:50 tadzik alright
20:50 hudo joined #perl6
20:51 laben you made a ton of commits
20:51 laben O.o newest rakudo/glr got this: t/01-sanity/99-test-basic.t .......... All 71 subtests passed         (less 4 skipped subtests: 67 okay)
20:51 laben smells like trouble
20:53 geekosaur skipped presumably means either missing prerequisite or known issue stubbed out
20:54 laben but it was working up till some hours ago
20:55 masak wohoo! it seems I accidentally made my first talk deliverable! :)
20:55 masak (in that it now contains 0 TODOs)
20:56 dalek joined #perl6
20:57 colomon joined #perl6
20:57 skids 99-test-basic.t tests Test, intentionally todoing and skipping things.  So there would have had to have been some finagling to not report skips.
20:58 hudo it seems that $suchbegriff is not initialized
20:58 hudo although it contains a string...
21:02 dalek roast/glr: 53c00f1 | (Stefan Seifert)++ | S02-types/nested_arrays.t:
21:02 dalek roast/glr: We don't auto flatten nested Lists anymore
21:02 dalek roast/glr: review: https://github.com/perl6/roast/commit/53c00f1c52
21:03 dwarring joined #perl6
21:04 bin_005 joined #perl6
21:10 colomon joined #perl6
21:12 FROGGS joined #perl6
21:12 FROGGS m: my @a = 1 .. 5; @a[0,1] = @a[1,0]; say @a
21:12 camelia rakudo-moar 47ddca: OUTPUT«2 1 3 4 5␤»
21:12 GLRelia rakudo-moar c944e4: OUTPUT«21345␤»
21:12 laben tadzik: found out that on newest glr, panda need another slight fix, want me to add it to the merge PR or make another?
21:13 FROGGS nine++ # I was about fixing that earlier, but would not have found the patch me thinks...
21:13 tadzik laben: same one is fine
21:17 laben tadzik: and pushed, now to try bootstrap...
21:18 FROGGS m: sub bare_return { return }; my $s = "foo"; $s = bare_return(); say $s.defined
21:18 GLRelia rakudo-moar c944e4: OUTPUT«False␤»
21:18 camelia rakudo-moar 47ddca: OUTPUT«False␤»
21:18 FROGGS m: sub bare_return { return }; my $s = "foo"; $s = bare_return(); say $s.WHAT
21:18 camelia rakudo-moar 47ddca: OUTPUT«(Any)␤»
21:18 GLRelia rakudo-moar c944e4: OUTPUT«Use of uninitialized value of type Any in string context  in block <unit> at /tmp/iKLXU0QTaa:1␤␤»
21:19 FROGGS hmmm
21:19 nine Official count: 64 spec test files failing
21:19 FROGGS when I make 'return;' returning an Empty, it will be a defined Slip when assigning it to a scalar
21:19 llfourn joined #perl6
21:20 FROGGS wow
21:20 FROGGS 64 is a nice number
21:20 timotimo it won't be as nice again until we hit 32 :(
21:20 FROGGS 42 is awesome too
21:21 timotimo mhh, true
21:21 FROGGS true indeed
21:22 timotimo mhh, defined
21:23 virtualsue joined #perl6
21:27 laben tadzik: why does bootstrap install the modules in <site> but panda lib and bin in $prefix?
21:28 lizmat hotel&
21:29 trnh joined #perl6
21:30 laben what's more, why does it not install the .moarvm files anymore for modules but does so for panda?
21:32 timotimo because precompilation is gone from panda
21:32 timotimo demon be gone!
21:32 timotimo foul, foul precompilation!
21:32 jdv79 interesting question - could the various vm's gen the precomp files themselves?
21:33 jdv79 either if missing or on dep match fail?  that seems like a better idea.
21:33 FROGGS jdv79: rakudo potentially should do that (automagically)
21:33 jdv79 *vms
21:33 jdv79 i have no idea why i inserted a '
21:33 laben seems it's not gone away completely as panda does precomp itself, as i just said
21:33 jdv79 so just a NYI?
21:33 timotimo yes, NYI
21:33 jdv79 neat
21:34 timotimo precompiled files were supposed to only be a cache
21:34 laben FROGGS: and where would rakudo put the precomp'd file then?
21:34 FROGGS laben: into a CUR
21:34 timotimo but nowadays the presence of a "wrong" precompiled file causes execution to fail/bail out
21:34 timotimo that's obviously wrong
21:34 timotimo if that's encountered, the vm should just delete the precompiled file that caused the problem and compile anew
21:34 * timotimo disappears into bed
21:35 laben CUR is the CompUnitRepo, right? where would it be located physically?
21:35 laben i mean computer-physically
21:35 timotimo it's the CompUnitRepo, but it's only a role
21:35 timotimo you're interested in CompUnitRepo::Installation i think
21:35 ugexe can you precompile just 1 file without having to precompile everything that is connected to it?
21:35 timotimo that's the one with the serialized database for installed things
21:35 laben this whole thing smells like incredible trouble for packaging
21:36 timotimo ugexe: depends on what you mean with "connected to it"
21:36 ugexe anything up stream might have been updated or deleted, and downstream would depend on it
21:36 FROGGS laben: the CUR decides how that happens, there is a database like CURLI and a P5/panda like CURLF
21:36 timotimo things a file depends on have to be compiled first, of course. and then they ought to be precompiled as well, i think
21:37 FROGGS m: say Empty =:= Empty
21:37 GLRelia rakudo-moar c944e4: OUTPUT«True␤»
21:37 camelia rakudo-moar 47ddca: OUTPUT«True␤»
21:37 ugexe right, but just having the vm automatically do it could then lead to unknown wait time
21:37 jdv79 no reason something like panda couldn't do it as an optimization, right ?
21:38 dalek nqp: 0a45369 | hoelzro++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
21:38 dalek nqp: Make sure STAT_IS* returns false for non-existent files on JVM
21:38 dalek nqp:
21:38 dalek nqp: Previously, non-existent files would return -1, which is still
21:38 dalek nqp: a truthy value.
21:38 dalek nqp: review: https://github.com/perl6/nqp/commit/0a453697e6
21:38 ugexe CURLI should be able to do it itself
21:38 laben what i'm thinking of right now is that: if you put the files/db in the user home, you risk wipes as (s)he could remove files; if you put it in a privileged position, you need privileges to rebuild them.
21:39 jdv79 yeah.   might require a bit of sophistication .
21:39 ugexe in the CURLI MANIFEST you just have to store the dependency build order (and provides build order), then it can rebuild itself
21:40 jdv79 where is this manifest you speak of?
21:40 ugexe also involves saving the hooks, %*RESOURCES, but i can see where this is going
21:40 ugexe %*CUSTOM_LIB<site>.child('MANIFEST')
21:41 ugexe *if* you have anythign installed to CURLI
21:41 jdv79 should i have one now?
21:41 jdv79 i installed someting via panda
21:41 ugexe i dont know if panda installs with CURLI
21:42 jdv79 ah
21:42 jdv79 i saw a statefile thingee
21:42 laben i guess you could install a cache in a standardized xdg directory, but do you imagine having to explain why everytime you use a module, it has to check the fs to see if it has been updated?
21:42 ugexe zef installs to a CURLI. panda to a CURLF.
21:42 FROGGS CURLI is not the default atm
21:42 FROGGS aye
21:43 jdv79 ah
21:43 FROGGS CURLI though has currently bugs, and that's the reason it is not the default
21:43 jdv79 xdg is what now?
21:43 ugexe what bugs?
21:43 laben freedesktop.org standard
21:44 jdv79 what about windows?
21:44 FROGGS it is quite veeeeery slow when you install a lot of modules
21:44 pink_mist jdv79: standard for where config/cache/data files used by programs should be
21:44 ugexe I was working on that problem today. its because to-json is so slow. i was trying to sneak an outside to-json into CURLI with .wrap but it didnt work
21:45 laben well, since it already checks if it's on windows, having a dedicated WIN folder won't hurt
21:45 FROGGS and the solution to the slowness sits in a branch (jsoff), but this also is not bugfree (it crashes)
21:45 jdv79 i don't think the vm would have to check all the time.  as long as the pkg mgr blows the precomp file away on a module update it should be fine.
21:45 japhb Wow, long backlog is long!  Y'all have been rather active over there.  :-)
21:46 ugexe this can output the projects.json in like 0.5 seconds vs default to-json of 2-3 minutes
21:46 * japhb just got rickrolled by his music service
21:46 laben jdv79: the pkg manager would have to rebuild every reverse dep every time the module changes
21:46 ugexe tony-o json::faster
21:47 japhb What actually makes json::fast{,er} ... well, faster?
21:47 ugexe dunno, probably because it doesnt try to be clever
21:47 jdv79 laben: what?  do you mean pre-comp?  it could.  but it could also just rm the ones directly related to the changing modules and let the vm fault it out
21:48 jdv79 right now panda likes to fully reinstall revdeps so there is much room for improvement:)
21:52 laben but where did the pkg manager put them in the first place? generally they use privileged access and unless you want to give it to rakudo too, it wont be able to rm/update them as it pleases
21:56 FROGGS ugexe: I'd be interested in your bare perl6 startup time when you installed hundred modules...
21:56 FROGGS 'perl6 -e 1' went to more than five seconds IIRC
21:56 ugexe oh for reading. i was talking about for *writing* the manifest after it gets huge
21:56 dalek perl6-roast-data: 2e63d46 | coke++ | / (2 files):
21:56 dalek perl6-roast-data: another glr run for 'today'
21:56 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/2e63d46f9a
21:57 laben FROGGS: doesn't rakudo avoid searching for deps until asked (at compile time)?
21:58 FROGGS laben: might be true nowadays
21:58 tadzik laben: ...it shouldn't
21:58 FROGGS but then you have to wait the extra seconds once on first module use
21:58 laben why it shouldn't? if i say "perl6 -e 1", it shouldnt loadn any modules
21:59 FROGGS true, and it wont load any CUR code IIRC
21:59 tony-o japhb: it's json::faster is significantly faster, not sure what makes it faster than, say, json::fast or the built in to-json because i didn't bother looking at them
21:59 FROGGS so it wont deserialize MANIFEST.json
22:00 tony-o it's just a recursive method to convert hash/array to json
22:00 ugexe with 100 files (not distributions) in site/manifest the startup is still instant
22:00 laben the best way to make something faster is to make it work less
22:01 tony-o laben: https://github.com/tony-o/perl6-j​son-faster#benchmark-vs-jsonfast
22:02 laben i was talking about loading modules and what not, but congrats :)
22:03 tony-o ah
22:04 FROGGS tony-o: you do not plan to implement a fast json reader?
22:04 FROGGS because that's what is the bottleneck for CURLF
22:05 FROGGS writing json does not happen that often
22:05 tony-o the reader seemed like it was fast enough but i'll take a look at trying to make it faster
22:05 FROGGS use a large json blob and you'll see what I mean...
22:05 FROGGS with that...
22:05 FROGGS gnight everyone
22:05 tony-o later FROGGS
22:07 [Coke] Curse you Netflix.
22:11 japhb tony-o: Can you add the builtin to-json performance to that comparison?
22:12 japhb [Coke]: What has Netflix done to you today?
22:12 tony-o japhb: i can add that, yes.  it's roughly the same as the json::fast and json::tiny
22:12 tony-o both of those are comparable to the built-in
22:12 japhb ISTR the built-in was just directly taken from JSON::Tiny's code
22:13 tony-o i wouldn't doubt it
22:15 khw joined #perl6
22:18 [Coke] japhb: SG:atlantis
22:22 kinslayer joined #perl6
22:23 braintwist joined #perl6
22:23 dalek rakudo/glr: 817767b | (Stefan Seifert)++ | src/core/List.pm:
22:23 dalek rakudo/glr: Fix list assignment involving sublists
22:23 dalek rakudo/glr:
22:23 dalek rakudo/glr: Fixes ($x, @a[0, 1], $y) = 1, 2, 3, 4;
22:23 dalek rakudo/glr: We didn't handle the case that a finite List was contained on the left hand
22:23 dalek rakudo/glr: side. This caused the list in the middle to slurp all the remaining
22:23 dalek rakudo/glr: values, so the following scalar would end up empty.
22:23 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/817767b70d
22:31 bbkr joined #perl6
22:32 huf joined #perl6
22:32 dalek roast/glr: efb26f8 | (David Warring david.warring@gmail.com)++ | integration/99problems-51-to-60.t:
22:32 dalek roast/glr: GLRify 99problems-51-to-60.t
22:32 dalek roast/glr: review: https://github.com/perl6/roast/commit/efb26f8ad0
22:39 skids joined #perl6
22:41 japhb [Coke]: *chuckle*  Watching it for the first time?
22:44 bbkr r: sub foo ($x) { say $x; samewith($x+1) if $x < 10 }; foo(1)
22:44 GLRelia rakudo-jvm 817767: OUTPUT«Can't open perl script "/home/steve/rakudo-inst/bin/eval-client.pl": No such file or directory␤»
22:44 GLRelia ..rakudo-moar 817767: OUTPUT«1␤Could not find dispatcher␤  in sub foo at /tmp/tmpfile:1␤  in block <unit> at /tmp/tmpfile:1␤␤»
22:44 camelia rakudo-{moar,jvm} 47ddca: OUTPUT«1␤Could not find dispatcher␤  in sub foo at /tmp/tmpfile:1␤  in block <unit> at /tmp/tmpfile:1␤␤»
22:44 bbkr what does this error mean?
22:46 dalek rakudo/glr: 087a6c5 | (Stefan Seifert)++ | src/core/List.pm:
22:46 dalek rakudo/glr: Fix Array handling in list assignment
22:46 dalek rakudo/glr:
22:46 dalek rakudo/glr: While Arrays are Lists, too, they do swallow up all values stuffed into
22:46 dalek rakudo/glr: them, so we may not treat them like slices (which are Lists).
22:46 dalek rakudo/glr:
22:46 dalek rakudo/glr: Fixes my (@a) = (1, 2, 3); broken by my implementation of sublist
22:46 dalek rakudo/glr: support. No other regressions appeared.
22:46 dalek rakudo/glr: review: https://github.com/rakudo/rakudo/commit/087a6c5fd7
22:46 nine Ok, that's it for today. Official count is now at 63 failing spec test files :)
22:46 nine Good night #perl6!
22:47 mattp_ jnthn: is a new 6guts blogpost coming soon? :)
22:52 [Coke] japhb: nah, but it was a favorite. pretty much just jumped to the top of my passive rewatch list.
22:54 japhb :-)
22:54 dalek perl6-roast-data: e3d5c20 | coke++ | / (2 files):
22:54 dalek perl6-roast-data: another glr run for 'today'
22:54 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/e3d5c20a84
22:55 bbkr is it possible to get method handle within method itself?  let's say I have recursive "sub foo { foo() }" and I don't want to repeat method name explicitly but to express "sub foo { call-the-method-i-am-in() }"
22:56 bbkr inner call may have different set of params
23:07 araujo joined #perl6
23:08 [Sno] joined #perl6
23:09 araujo joined #perl6
23:09 skids bbkr: $?ROUTINE maybe?
23:09 skids But if you want to do inheritance/MMD fallback look at callwith/nextwith.
23:10 skids Erm &?ROUTINE
23:10 bbkr no multis, no inheritance
23:10 araujo joined #perl6
23:11 bbkr perfect, &?ROUTINE works. thanks!
23:16 skids m: my @a = 1,2,3; my @b = <a b c>; @a = @a Z~ @b; @a.perl.say;
23:16 camelia rakudo-moar 47ddca: OUTPUT«["1a", "2b", "3c"]<>␤»
23:16 GLRelia rakudo-moar 087a6c: OUTPUT«["1a", "2b", "3c"]␤»
23:16 skids m: my @a = 1,2,3; my @b = <a b c>; @a = @a Z~ @b; @a[1,2,3].perl.say;
23:16 GLRelia rakudo-moar 087a6c: OUTPUT«("2b", "3c", Any)␤»
23:16 camelia rakudo-moar 47ddca: OUTPUT«("2b", "3c", Any)␤»
23:16 skids m: my @a = 1,2,3; my @b = <a b c>; @a[0,1,2] = @a Z~ @b; @a.perl.say;
23:16 camelia rakudo-moar 47ddca: OUTPUT«["1a", "2b", "3c"]<>␤»
23:16 GLRelia rakudo-moar 087a6c: OUTPUT«This Seq has already been iterated, and its values consumed␤  in block <unit> at /tmp/kLud5DYbFB:1␤␤»
23:20 llfourn joined #perl6
23:23 dalek perl6-roast-data: 43c8544 | coke++ | / (2 files):
23:23 dalek perl6-roast-data: another glr run for 'today'
23:23 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/43c8544915
23:26 dalek doc: 889e20a | (Tim Smith)++ | lib/Type/IO/Handle.pod:
23:26 dalek doc: Document .words and .ins
23:26 dalek doc: review: https://github.com/perl6/doc/commit/889e20a726
23:41 dalek doc: 15603c8 | coke++ | lib/Type/Str.pod:
23:41 dalek doc: comb's :$match doesn't exist on Str pattern
23:41 dalek doc: review: https://github.com/perl6/doc/commit/15603c8e13
23:45 dalek rakudo/nom: 30d6fe7 | (Geoffrey Broadwell)++ | docs/ChangeLog:
23:45 dalek rakudo/nom: Unsplit a ChangeLog entry
23:45 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/30d6fe7757

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

Perl 6 | Reference Documentation | Rakudo