| [Tux] | This is Rakudo version 2016.05-30-g4a7eaa0 built on MoarVM version 2016.05 | 07:06 | |
| test 20.323 | |||
| test-t 13.180 | |||
| csv-parser 34.510 | |||
| dalek | kudo/js: a80a5ae | (Pawel Murias)++ | src/vm/js/perl6-runtime/runtime.js: [js] Fetching stuff out of variables. |
10:30 | |
| lizmat | jnthn: is there a reason why Seq.new only accepts instantiated Iterators ? | 12:19 | |
| jnthn: in the case of the EmptyIterator, there's no actual need to instantiate | |||
| jnthn: but I guess I can make it a highlander :-) | 12:20 | ||
| jnthn | lizmat: 'cus it's almsot always not the right thing to have a type object there :) | 12:26 | |
| lizmat | yeah, I guess :-) | 12:28 | |
| dalek | kudo/js: 4e47144 | (Pawel Murias)++ | src/vm/js/ (2 files): [js] Implement nqp::p6box_n, nqp::p6box_i, nqp::p6box_s. |
12:32 | |
| p: 93b7c09 | (Pawel Murias)++ | src/vm/js/nqp-runtime/hll.js: [js] Implement hllizing nqp primitive types. |
12:36 | ||
| p: 7a6cba1 | (Pawel Murias)++ | t/nqp/097-hll.t: Test foreign_type_int, foreign_type_num, foreign_type_string. |
|||
| p: 2a22a16 | (Pawel Murias)++ | src/vm/js/nqp-runtime/runtime.js: [js] Throw an NQPException instead of a untyped on so it can be caught by try in rakudo-js. |
|||
| p: 7706bab | (Pawel Murias)++ | src/vm/js/nqp-runtime/runtime.js: [js] Allow looking up of an undeclared lexical, rakudo depends on that. |
|||
| tomboy64 | wow | 12:41 | |
| i should really start filing bugs against perl6 :D | |||
| brrt | :'-( | 12:48 | |
| and we have to fix them, surely | |||
| :-P | |||
| [Coke] | RT: 1307; CONC: 7; GLR: 6; JVM: 68; LHF: 1; LTA: 75; NYI: 28; OSX: 6; PERF: 15; POD: 3; PRECOMP: 3; RFC: 23; SEGV: 19; STAR: 1; TESTNEEDED: 29; TODO: 9; UNI: 5; WEIRD: 3 | 12:49 | |
| dalek | p: 4cd54b4 | (Pawel Murias)++ | src/vm/js/ (2 files): [js] Implement nqp::sayfh. |
13:35 | |
| tomboy64 | i have libffi and dyncall locally installed | 16:10 | |
| but even when specifying --has-libffi, the logic in Configure.pl 1. tests for libff, 2. tests for dyncall, 3. tests for nativecall, all three exclusively. and during the build itself, 3rdparty/dyncall is accessed nevertheless. | 16:12 | ||
| considering it, this is for #moarvm, please disregard | |||
| ZoffixW | 🎺🎺🎺 I wrote down the details for the work I plan to do on `is-approx` and I'll start in a day or two. If anyone has any objections or suggestions just say in channel (I'll read the backlog even if I'm not here): github.com/zoffixznet/debug/blob/m...-rework.md | 18:51 | |
| There are some non-backwards compat changes, but they affect non-specced and non-documented areas and the old behaviour is broken in some places, so I think even if someone is using those portions, they expect it to behave as specced/docced not as the sub currently does things. | 18:52 | ||
| 👋 | |||
| perlpilot | ZoffixW++ | 19:05 | |
| dalek | ast: df98e1d | (Tom Browder)++ | S26-documentation/07-tables.t: add critical test for known critical table pod bug (#128221) in latest rakudo release (2016.05) |
19:27 | |
| ast: 1a0127d | (Tom Browder)++ | S26-documentation/07-tables.t: correct num of elems (1 row) |
|||
| ast: 5d31d6e | (Tom Browder)++ | S26-documentation/07-tables.t: Merge pull request #121 from tbrowder/bad-pod-tables add critical test for known critical table pod bug (#128221) in lates… |
|||
| lizmat | hmmmm | 20:33 | |
| $ perl6 t/spec/S26-documentation/07-tables.t | |||
| ===SORRY!=== | |||
| Cannot iterate object with P6opaque representation | |||
| Am I missing something ?? | |||
| dalek | kudo/nom: b7e224f | lizmat++ | src/core/ (2 files): Make Rakudo::Internals::EmptyIterator a highlander Because Seq.new requires a Iterator:D, even though technically the EmptyIterator class didn't need instantiation to work. But it was felt that keeping the :D requirement for Seq.new was a smart thing. |
20:40 | |
| travis-ci | Rakudo build failed. Elizabeth Mattijsen 'Make Rakudo::Internals::EmptyIterator a highlander | 21:58 | |
| travis-ci.org/rakudo/rakudo/builds/134281984 github.com/rakudo/rakudo/compare/4...e224fadbc9 | |||
| dalek | kudo/nom: 33ab868 | lizmat++ | src/core/List.pm: Make iterating over a list about 10% faster again Basically by not checking for the index being in range. Since Lists are immutable and cannot have holes (unlike arrays), just checking for null is an equally good way to know you're done. This of course assumes that the VM *does* check for index being in range, but in a much faster way. |
22:11 | |
| lizmat | m: use nqp; sub a(@a) { say nqp::getattr(@a,List,q/$!todo/) }; a((1,2,3)) #@a is completely reified | 22:16 | |
| camelia | rakudo-moar b7e224: OUTPUT«(Mu)» | ||
| lizmat | m: use nqp; sub a(*@a) { say nqp::getattr(@a,List,q/$!todo/) }; a(1,2,3) # @a is completely lazy | ||
| camelia | rakudo-moar b7e224: OUTPUT«List::Reifier.new» | ||
| lizmat | m: sub a(*@a) { @a.push(4); dd @a }; a(1,2,3) # I assume this is correct behaviour? Or not ? | 22:17 | |
| camelia | rakudo-moar b7e224: OUTPUT«[1, 2, 3, 4]» | ||
| lizmat | jnthn: I could argue that a slurpy array is actually a List inside the block, and thus immutable | 22:18 | |
| jnthn: if that is too difficult, I would argue that we should probably completely reify slurpies when we can, possibly at compile time ? | 22:19 | ||
| travis-ci | Rakudo build failed. Elizabeth Mattijsen 'Make iterating over a list about 10% faster again | 23:14 | |
| travis-ci.org/rakudo/rakudo/builds/134302950 github.com/rakudo/rakudo/compare/b...ab86837a2e |