[00:06] *** reportable6 left [00:10] *** linkable6 joined [03:13] *** frost joined [04:08] *** reportable6 joined [04:24] *** [Coke] left [04:24] *** [Coke] joined [06:08] *** reportable6 left [06:10] *** reportable6 joined [09:55] *** evalable6 left [09:55] *** linkable6 left [09:55] *** evalable6 joined [09:56] *** linkable6 joined [10:05] Files=1351, Tests=117094, 325 wallclock secs (36.65 usr 9.96 sys + 4536.90 cusr 359.92 csys = 4943.43 CPU) [10:05] vrurg: so yeah, the 346 wallclock secs was a fluke [10:06] although we seem to have lost about 800 tests since the last time? [10:56] *** evalable6 left [10:56] *** linkable6 left [10:56] *** linkable6 joined [10:58] *** evalable6 joined [11:30] *** Altai-man joined [11:41] *** Altai-man left [11:42] *** Altai-man joined [12:07] *** reportable6 left [13:10] *** reportable6 joined [14:04] ¦ nqp: 2ea01db9dd | (Stefan Seifert)++ | 2 files [14:04] ¦ nqp: Map new ops for unsigned handling [14:04] ¦ nqp: [14:04] ¦ nqp: New ops: bindpos_u, atpos_u, atposref_u, iscont_u, assign_u and captureposarg_u [14:04] ¦ nqp: review: https://github.com/Raku/nqp/commit/2ea01db9dd [14:24] lizmat: ot [14:24] lizmat: sorry... No, it's a fake. I just wrapped some into subtests within a test-assertion. [14:27] https://github.com/Raku/roast/commit/250a246ded [14:29] *** frost left [15:44] I see :-) [15:53] *** shinto joined [16:34] *** carlmasak joined [16:59] ¦ nqp: 7c20873fa4 | (Stefan Seifert)++ | 14 files [16:59] ¦ nqp: Replace flatnamed arg flag by flat | named [16:59] ¦ nqp: [16:59] ¦ nqp: No need to have a separate flag when the combination of the existing flags [16:59] ¦ nqp: means the same thing. Will free up the flag for other use in MoarVM. [16:59] ¦ nqp: review: https://github.com/Raku/nqp/commit/7c20873fa4 [17:08] *** Core5549 joined [17:11] *** shinto left [17:16] *** shinto joined [17:16] *** Core5549 left [17:18] *** shinto left [17:25] *** shinto joined [17:52] *** carlmasak100 joined [17:54] *** carlmasak left [18:01] *** carlmasak100 is now known as carlmasak [18:01] *** carlmasak left [18:08] *** reportable6 left [18:17] *** nine_ is now known as nine [19:09] *** reportable6 joined [19:11] *** Core6356 joined [19:11] *** shinto left [19:12] *** shinto joined [19:14] *** Core2487 joined [19:15] *** Core4335 joined [19:15] *** Core6356 left [19:16] *** Core9796 joined [19:17] *** shinto left [19:18] *** Core2487 left [19:19] *** Core4335 left [19:29] *** Altai-man left [20:29] *** evalable6 left [20:29] *** linkable6 left [20:31] *** evalable6 joined [20:45] *** shinto joined [20:46] *** Core3143 joined [20:49] *** Core9796 left [20:50] *** shinto left [21:54] m: sub s(\l where *.^can('rotor'|'iterator')) {}; s([]); [21:54] rakudo-moar f77fda7d1: OUTPUT: «This type cannot unbox to a native string: P6opaque, Junction␤ in sub s at line 1␤ in block at line 1␤␤» [21:54] Am I asking for to much? [21:58] yes [21:59] that basically winds up in NQP land, and NQP land doesn't know about Junctions really [22:19] ¦ problem-solving: vrurg++ created pull request #310: Add resolution document for #135 [22:19] ¦ problem-solving: review: https://github.com/Raku/problem-solving/pull/310 [22:23] greppable6: ValueList [22:23] vrurg, 35 lines, 3 modules: https://gist.github.com/23ab16ee883767983c75f0fc82b437dc [22:24] greppable6: ValueMap [22:24] vrurg, Found nothing! [22:26] vrurg: Shouldn't that be *Scalar* decontainerization? Or were you implying converting sub-Arrays to sub-ValueLists recursively? [22:29] japhb: I was keeping in mind "Scalar", as it was always implied by the discussion surrounding Map conversion to non-value type. But you made me confused now... [22:30] Oh dear, I'm now an agent of chaos. [22:30] You are my dear, you are... :D [22:30] :-D [22:31] Though I tend to make it scalar only. Because if go as deep as deconting arrays and hashes, then why stopping there? And where stop then? [22:33] At deeper level we may consider introducing a protocol of converting an object type into a value type. [22:35] But this would require some additional theoretical work. In the mean time it is sufficient for Value* structures to know that they contain the same array, no matter what is Array's content. [22:35] Actually, deep conversion of data structures in both directions would be a useful thing, but yeah, needs thought. [22:36] Right. So maybe specify Scalar decontainerization in this doc, and leave other sub-container types open for future expansion? [22:36] Oh, I'm recalling that codesections started working on immutable data types. That might eliminate the need for deep value-fication. [22:37] Yep, thanks for reminding me. I lost track and forgot to update the resolution. :) [22:41] Heh [22:45] Done. [23:03] Updated with parameterization section and that's probably it for today. [23:30] *** linkable6 joined