Geth rakudo/udp_receive_hostname_port: 3edec176c0 | (Timo Paulssen)++ | src/core/IO/Socket/Async.pm
first draft of "datagram"-oriented API for udp sockets

ask the .Supply to give you :datagrams and you'll receive instances of IO::Socket::Async::Datagram with
  .data being the string or buffer of data and
  .hostname and .port being the sender's address.		 00:57
timotimo ^- i'd like feedback on this API. i'll likely rename the ":datagrams" parameter to ":datagram" 00:58
(requires a moarvm branch of the same name)
Geth rakudo/post-release: 0d7c4fe835 | (Zoffix Znet)++ | src/core/Hash.pm
Ensure Hash.clone does not lose descriptor
 01:37
rakudo/post-release: 8c4c979e35 | (Zoffix Znet)++ | src/core/Rakudo/Internals.pm
Fix `let`/`temp` on hashes losing Nils

Partially phixes github.com/rakudo/rakudo/issues/1432		 01:49
roast/post-release: 79c8fb7204 | (Zoffix Znet)++ | 2 files
Test `let`/`temp` do not lose Nils

  github.com/rakudo/rakudo/issues/1432
rakudo/post-release: 45560ac9ac | (Zoffix Znet)++ | src/core/Array.pm
Fix Array.clone losing descriptor

Rip out code from Array.from-iterator original code indirectly uses and make it use cloned descriptor when creating the clone.
Phixes: github.com/rakudo/rakudo/issues/1435		 02:22
roast/post-release: 29c95d23a2 | (Zoffix Znet)++ | S12-attributes/clone.t
Remove trailing whitespace
 02:23
roast/post-release: 286570e3d9 | (Zoffix Znet)++ | S12-attributes/clone.t
Test Hash/Array.clone do not lose descriptors

Closes github.com/rakudo/rakudo/issues/1435 Rakudo fixes: github.com/rakudo/rakudo/commit/0d7c4fe835
   github.com/rakudo/rakudo/commit/45560ac9ac		 02:24
roast/post-release: 311ef616b1 | (Zoffix Znet)++ | 2 files
Unfudge now passing let/temp tests

Closes github.com/rakudo/rakudo/issues/1432
nqp/post-release: 12 commits pushed by (Tom Browder)++, (Elizabeth Mattijsen)++, pmurias++, (Aleks-Daniel Jakimenko-Aleksejev)++, (Zoffix Znet)++
review: github.com/perl6/nqp/compare/74177...3ce69943cf		 02:25
rakudo/post-release: 32 commits pushed by (Tom Browder)++, (Aleks-Daniel Jakimenko-Aleksejev)++, (Zoffix Znet)++, (Alex Chen)++, (Elizabeth Mattijsen)++
review: github.com/rakudo/rakudo/compare/4...6f594ceac3
roast/post-release: 12 commits pushed by (Tom Browder)++, (Aleks-Daniel Jakimenko-Aleksejev)++, (Jan-Olof Hendig)++, (Zoffix Znet)++
review: github.com/perl6/roast/compare/311...b037ae173a		 02:26
AlexDaniel drops to bed 02:32
Geth nqp: 5167ad24dd | (Jeremy Studer)++ | 2 files
Don't die if parent doesn't have dump method

The dump_extra_node_info method of the QAST::SpecialArg role would look up the method of the same name on the immediate parent and invoke it, dying if that method didn't exist and could not be invoked (on QAST::Want, for instance). Modifying the behavior so that it checks whether the there is an invokable method, and if not continue along the parent chain until one is found or all are exhausted.		 03:55
travis-ci NQP build failed. Jeremy Studer 'Don't die if parent doesn't have dump method 04:11
travis-ci.org/perl6/nqp/builds/332129274 github.com/perl6/nqp/compare/92221...67ad24ddc8
nine lizmat: your Perl 6 in production section motivated me to write a blog post :) niner.name/blog/production_on_perl_6/index.html 08:28
lizmat nine++ 08:42
nine More production stories yet to come. I didn't want to cram everything into a single post 08:44
lizmat cool!
keep 'm coming!
m: sub undef(\a) { }; undef my $x # aaaw
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of undef as a value; in Perl 6 please use something more specific:
an undefined type object such as Any or Int,
:!defined as a matcher,
Any:U as a type constraint,
Nil as theā€¦
lizmat shouldn't it look there is a undef sub ? 08:45
Geth rakudo: ee901cf3fa | (Zoffix Znet)++ (committed using GitHub Web editor) | docs/ops.markdown
Fix typo
 09:06
lizmat Files=1233, Tests=76085, 339 wallclock secs (15.19 usr 5.23 sys + 2306.31 cusr 237.87 csys = 2564.60 CPU) 09:16
Zoffix What does nqp::p6capturelex do? I can surmise it's capturing the current lexical symbols, but I don't get what exactly the `p6capturelex(Block.clone)` thing does 09:21
m: gist.github.com/zoffixznet/beeafcd...2a894fc400
camelia 4
42
1
42
Zoffix ^ would've though that to be `4 100 1 42`, with how I imagine p6capturelex working
(and even then, I don't get what the clone is for)
lizmat jnthn would know 09:22
Zoffix void MVM_frame_capturelex /* Given the specified code object, sets its outer to the current scope. */ k, now I get why clone is needed, but don't get why .outer doesn't get changed in this code: 09:37
m: gist.github.com/zoffixznet/97f3f29...af21a1f40f
camelia -> { #`(Block|61940520) ... }
-> { #`(Block|61940520) ... }
42
42
El_Che (just keeping an eye on the release) 09:52
Zoffix m: use nqp; nqp::p6capturelexwhere 10:02
camelia ===SORRY!===
No registered operation handler for 'p6capturelexwhere'
Zoffix and yet there's a bunch of these in BOOTSTRAP; :/
jnthn Zoffix: p6capturelex is meaningless if you don't call it in the immediate outer scope of the block in question 10:04
(It'll silently do nothing at the moment, we might in the future make it blow up loudly)
Also, I suspect p6capturelex will go away at some point, in favor of a getattr and using the MoarVM capturelex op 10:07
(As part of the long-term plan to be rid of extops)
Zoffix jnthn: I don't get what it's for then. Does it just set the outer when none are set yet? 10:08
And then what's the clone for then?
jnthn It just sets the outer
The two together result in a closure 10:09
That's all a closure is: an instance of a Code (or some subclass of it) that refers to an outer scope ("outer frame" in MoarVM terms) 10:10
In something like for ^5 -> $i { push @a, { $i } } we need to clone that inner block each loop iteration, so it closes over the correct $i, for example 10:11
Zoffix m: 1 andthen <a b c>[$_.self xx 2].say
camelia Use of Nil in string context
in block at <tmp> line 1
Unable to call postcircumfix [ (Any) ] with a type object
Indexing requires a defined object
in block <unit> at <tmp> line 1
Zoffix Was trying to debug this one.
m: $_ = 2; <a b c>[$_ xx 2].say 10:12
camelia (c c)
Zoffix m: $_ = 2; 1 andthen <a b c>[$_.self xx 2].say
camelia (c c)
jnthn Those sorts of things are often due to mis-nesting of blocks in the produced QAST 10:13
Zoffix Ah. OK. I'll try to explode that path. 10:14
*explore 10:16
I think that is it, with what you said about "p6capturelex is meaningless if you don't call it in the immediate outer scope"; cause the QAST setup for the buggy version is more like my gist 10:18
jnthn++
Geth rakudo: 8bd9fd3f82 | (Zoffix Znet)++ (committed using GitHub Web editor) | docs/ops.markdown
Document p6capturelex

Per irclog.perlgeek.de/perl6-dev/2018-...i_15725556		 10:24
pmurias I have a dilemma with regards to changing the return values of eqat* ops 10:25
on one hand if they return 0 when the string doesn't match then they don't work for a "" needle correctly 10:26
on the other if they return -1, nqp would need to be rebootstrapped as matching literals uses nqp::eqat 10:27
Geth nqp: 735865df3e | (Tom Browder)++ (committed using GitHub Web editor) | src/HLL/World.nqp
fix spelling
 11:26
nqp: 1ea46092d1 | (Tom Browder)++ (committed using GitHub Web editor) | src/HLL/Grammar.nqp
align code
 11:33
travis-ci NQP build passed. Tom Browder 'fix spelling' 11:42
travis-ci.org/perl6/nqp/builds/332252928 github.com/perl6/nqp/compare/5167a...5865df3ecd
NQP build passed. Tom Browder 'align code' 11:50
travis-ci.org/perl6/nqp/builds/332254819 github.com/perl6/nqp/compare/73586...a46092d11f
Geth nqp: d8f194b28e | (Tom Browder)++ (committed using GitHub Web editor) | src/NQP/Grammar.nqp
align code
nqp: bbaa5a5eb9 | (Tom Browder)++ (committed using GitHub Web editor) | src/NQP/World.nqp
align code
 11:54
travis-ci NQP build failed. Tom Browder 'align code' 12:06
travis-ci.org/perl6/nqp/builds/332260871 github.com/perl6/nqp/compare/1ea46...f194b28ea6
Geth nqp: 86e062e0fb | (Tom Browder)++ (committed using GitHub Web editor) | docs/using-pod-in-nqp.md
document pod use in nqp
 12:22
tbrowder looks like tabs in the source... 12:24
pmurias how do I rebootstrap NQP? 12:44
timotimo "make m-bootstrap-files" 13:05
pmurias: ^
Geth rakudo: 7361fe2b6d | (Elizabeth Mattijsen)++ | src/core/Baggy.pm
Fix "say (bag 'a').pick(1)"

Not sure what was going on here, but the idea was that $!todo would not be a container, but always bound. Fixed now by binding the original todo value as well. This feels like magic, and may be indicative of a deeper problem.		 13:06
pmurias hmm, if I want to change the way eqat_s works to make it match the way eqatim will work what's a sane way to do it? 14:04
timotimo add a second op for the time being? :)
Geth roast: lefth++ created pull request #379:
Add a test for picking a Bag with an argument smaller than the Bag size.		 14:10
roast: d5c1b77096 | (Dan Zwell)++ (committed by Dan) | S02-types/bag.t
Add a test for picking a Bag with an argument smaller than the Bag size.

  github.com/rakudo/rakudo/issues/1438		 14:12
roast: 0047ed7eaf | (Zoffix Znet)++ (committed using GitHub Web editor) | S02-types/bag.t
Merge pull request #379 from lefth/test-for-1438

Add a test for picking a Bag with an argument smaller than the Bag size.
lizmat "obscure companies are using it for trivial tasks?" I guess it's up from "not being used" :-) 14:43
Geth nqp: tbrowder++ created pull request #396:
improve formatting		 14:53
jnthn Well, >99% of companies in the world are "obscure" (if that means "you haven't heard of them"), and >99% of code that gets written is solving "trivial tasks" (if that means "solving everyday problems"). 14:58
El_Che jnthn: I appreciate you don't get discourage by negative comments 15:07
[Coke] core devs I think are self-selected to be slightly immune. 15:08
El_Che [Coke]: Well, most of them are still people 15:09
(I have my doubts about a few) :)
lizmat when I first started my own company, my then boss warned me that most companies go bankrupt within a year
[Coke] Right, I'm just saying that if they're still here, they've already managed to get past some of that.
lizmat guess which company *did* go bankrupt 1.5 years after I left 15:10
[Coke] lizmat: for me, it was Enron. ;)
lizmat: and now I'm apparently working on GE.
[Coke] guesses it was your bosses' company. :)
lizmat yup 15:11
looks like the next indexing of modules.perl6.org will put it past 1000 :-) 15:30
timotimo nice. 15:35
you see all those pingbacks from random copy-paste blogs on the 'net, lizmat? some bots have picked up on that post getting lots of views from hackernews i guess 15:36
lizmat timotimo: no, didn't look yet and will not for a few hours 15:42
timotimo no worries, i'd consider it spam 15:43
lizmat afk for a few hours& 15:47
Geth nqp: e86e001435 | (Tom Browder)++ | docs/using-pod-in-nqp.md
improve formatting
 15:50
nqp: 9a89c769cc | (Tom Browder)++ | docs/using-pod-in-nqp.md
clarify what 'word' is being discussed
nqp: 91e4a3e32b | (Tom Browder)++ (committed using GitHub Web editor) | docs/using-pod-in-nqp.md
Merge pull request #396 from tbrowder/master

improve formatting
[Tux] Rakudo version 2017.12-287-g7361fe2b6 - MoarVM version 2017.12.1-62-ga04d1099b
csv-ip5xs1.141 - 1.156
csv-ip5xs-2012.811 - 13.803
csv-parser12.098 - 12.105
csv-test-xs-200.444 - 0.464
test11.250 - 11.879
test-t2.840 - 2.978
test-t --race1.163 - 1.172
test-t-2049.810 - 51.634
test-t-20 --race17.387 - 17.729
 16:14
[Coke] win 10 16:18
buggable [Coke], Thank you for entering Accidental /win Lottery! The next draw will happen in 1 week, 1 day, 7 hours, 41 minutes, and 20 seconds
jnthn m: my $(42) 17:23
camelia 5===SORRY!5=== Error while compiling <tmp>
The () shape syntax in variable declarations is reserved
at <tmp>:1
------> 3my $(427ā5)
expecting any of:
constraint
formal parameter
tbrowder AlexDaniel: i plan to add a pod slang to nqp as a warm up and learning exercise for doing the same for p6. if there is any more documentation on doing so than i have found around repos nqp and rakudo i would love to get links. i have looked at jnthnā€™s slide archives briefly, and i found his presentation in japan given in 2013, but i havenā€™t had a chance to watch but a few minutes so far. as a starter to the 17:42
nqp pod slang, i would like to add a new repo doc file called something like nqp/docs/adding-a-pod-slang.md and document the steps i take. the new doc i added today is a precursor to that. i wonā€™t move farther along unless you think the idea is worthwhile.
AlexDaniel tbrowder: me? But I'm not a specialist in that area at all 17:56
:)
have you had any luck getting supernova to run? For example, does it pass its tests? 17:57
Geth nqp: tbrowder++ created pull request #397:
convert tabs to 4 spaces, correct some obvious code misalignment		 17:58
tbrowder supernova runs, but some tests fail (i added all the roast pod tests), i haven't looked into the failures deeply yet. 18:01
AlexDaniel cool 18:02
tbrowder you don't need to be a specialist, you're the current release manager and can try to delegate (or cajole or sweet-talk). i just want my plan to be approved OR shot down by core devs. if the plan is approved, i want my work visible for critique and helpful suggestions . 18:04
got to leave for a while... 18:06
AlexDaniel tbrowder: well, at that point whatever gets you closer to the target should work 18:17
tbrowder: if experimenting with nqp can potentially help you then sure
tbrowder: the long-term plan is better pod support in the core, and that is definitely approved 18:21
samcv after Travis finishes testing should be ready to merge my MVM PR that fixes White_Space and as a bonus fixes 'space' 18:28
plus also makes it so that you can use not just 'Foo_Bar' and 'FooBar' and 'foobar' but also 'foo_bar' now will also work for all of the property names 18:29
and then probably may fix one or two properties we may not have tests yet for or just make sure we don't have duplicate items in the properties c hash 18:30
/win totally an accident 18:31
/win 99
buggable samcv, Thank you for entering Accidental /win Lottery! The next draw will happen in 1 week, 1 day, 5 hours, 28 minutes, and 45 seconds
samcv /win 99 an accident
/win 99
buggable samcv, Thank you for entering Accidental /win Lottery! The next draw will happen in 1 week, 1 day, 5 hours, 28 minutes, and 29 seconds
nine samcv: cat on the keyboard?
El_Che maybe the cat has good utf8 symbols shortcuts 18:32
samcv no. trying to test the win lottery
AlexDaniel huggable: star 19:15
huggable AlexDaniel, Estimated Rakudo Star releases for 2017: .01, .04, .07 & .10
lizmat ok, so I'm working on getting scalar / wantarray working some on Perl 6: github.com/lizmat/P5wantarray/blob...tarray.pm6 22:06
I cargo-culted some from DrForr's Roman slang
and from Perl6/Grammar and Perl6/Actions 22:07
the idea being that "scalar" would take a blorst
and first set a dynamic variable before calling the blorst
so that wantarray could return True/False on the value of that dynamic variable 22:08
the thing is that it dies deeply at: github.com/lizmat/P5wantarray/blob...ay.pm6#L29
Cannot invoke this object (REPR: Null; VMNull) 22:09
at gen/moar/stage2/QRegex.nqp:1301 (/Users/liz/Github/rakudo.moar/install/share/nqp/lib/QRegex.moarvm:check_LANG_oopsies) 22:10
the actual role being mixed in is only a single token: 22:12
token statement_prefix:sym<scalar> { <sym><.kok> <blorst> }
DrForr Aroo? 22:14
lizmat que? 22:16
DrForr Heard my nick being taken in vain, but I must off to bed, unfortunately.
lizmat good night, DrForr! 22:17
DrForr Glad that module could help someone...
lizmat good night, #perl6-dev! 22:57
El_Che lizmat: isn't wantarray the reason got bad security press? 23:18
timotimo El_Che: that and automagic flattening of arguments that can also become named arguments overwriting earlier named arguments 23:20

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!