[01:30] *** Xliff_ left [02:41] *** ggoebel_ left [03:48] *** evalable6 left [03:48] *** linkable6 left [03:48] *** evalable6 joined [03:49] *** linkable6 joined [04:05] *** melezhik joined [04:19] *** melezhik left [05:55] *** travis-ci joined [05:55] Rakudo build passed. Elizabeth Mattijsen 'Remove postcircumfix[int] candidates [05:55] https://travis-ci.org/rakudo/rakudo/builds/747458945 https://github.com/rakudo/rakudo/compare/0f4970da2fb8...7c0956b81b4b [05:55] *** travis-ci left [06:18] *** domidumont joined [06:31] *** frost-lab joined [07:50] <[Tux]> timotimo, which one is twice as fast? [07:58] *** sena_kun joined [09:06] *** Altai-man joined [09:09] *** sena_kun left [09:53] *** patrickb joined [10:39] *** Xliff joined [10:40] lizmat: I think I got packages in Gentoo working [10:41] *** ggoebel_ joined [10:42] *** moray joined [10:42] *** moray is now known as morayj [10:47] as i understand it, if you've got a shaped native 1dimensional array like my int @blerp[99] then "most accesses" will be 2x as fast as when you leave out the [99] part; i imagine 2x as fast means instead of 10000 accesses per second you get 20000 (completely fictitious numbers) [10:49] m: my @a = gather for ^3 { take (0, 1, 2) }; @a.gist.say [10:49] rakudo-moar 2c5b54576: OUTPUT: «[(0 1 2) (0 1 2) (0 1 2)]␤» [10:49] m: my @a = gather for ^3 { take |(0, 1, 2) }; @a.gist.say [10:49] rakudo-moar 2c5b54576: OUTPUT: «[(0 1 2) (0 1 2) (0 1 2)]␤» [10:49] m: my @a = gather for ^3 { my @a = (0, 1, 2); take |@a }; @a.gist.say [10:49] rakudo-moar 2c5b54576: OUTPUT: «[(0 1 2) (0 1 2) (0 1 2)]␤» [10:49] m: my @a = gather for ^3 { my @a = (0, 1, 2); take |@a.flat }; @a.gist.say [10:50] rakudo-moar 2c5b54576: OUTPUT: «[(0 1 2) (0 1 2) (0 1 2)]␤» [10:50] Hum. [11:02] *** patrickb left [11:14] *** morayj left [11:31] *** patrickb joined [12:01] *** Xliff left [12:35] my @a = flat gather for ^3 { my @a = (0, 1, 2); take @a }; say @a [12:35] moritz, rakudo-moar 2c5b54576: OUTPUT: «[0 1 2 0 1 2 0 1 2]␤» [12:52] ¦ rakudo: 3def3ced38 | (Elizabeth Mattijsen)++ | 2 files [12:52] ¦ rakudo: Properly mark native shaped1 candidates [12:52] ¦ rakudo: [12:52] ¦ rakudo: Some were missing "is raw", all were missing "is default" to resolve [12:52] ¦ rakudo: dispatch problems with upcoming unshaped native candidates. [12:52] ¦ rakudo: review: https://github.com/rakudo/rakudo/commit/3def3ced38 [13:05] *** lucasb joined [13:08] *** sena_kun joined [13:09] *** Altai-man left [14:04] *** frost-lab left [14:54] *** patrickb left [14:55] *** patrickb joined [15:14] *** melezhik joined [15:15] hi! do we have a list of GH issue are going to resolved in the next release? [15:15] ¦ rakudo: 705e6e62c5 | (Elizabeth Mattijsen)++ | 4 files [15:15] ¦ rakudo: Give native arrays their own postcircumfix candidates [15:15] ¦ rakudo: [15:15] ¦ rakudo: This makes native array access about 40% faster, making it now 25% [15:15] ¦ rakudo: faster again than shaped native array access. [15:15] ¦ rakudo: [15:15] ¦ rakudo: Remove some candidates from the shaped native array circumfixes [15:15] ¦ rakudo: that can be handled by the non-shaped native array access ones. [15:15] ¦ rakudo: [15:15] ¦ rakudo: Adapted error message test to check for the mention of type of [15:15] ¦ rakudo: native array. [15:15] to resolve -> to be resolved [15:15] ¦ rakudo: review: https://github.com/rakudo/rakudo/commit/705e6e62c5 [15:15] the ones marked "blocker: I guess ? [15:18] https://github.com/rakudo/rakudo/issues?q=is%3Aissue+label%3ABLOCKER+is%3Aopen only 2 issues here? [15:18] melezhik: yeah, looks like [15:20] I am looking to the closed ones - https://github.com/rakudo/rakudo/issues?q=is%3Aissue+is%3Aclosed , which one from this group are going to be included in the next release? [15:22] basically all that got closed since the last release ? [15:23] ok, let me put it this way. I am thinking about having all to be shipped as a next release resolved issues in RICH [15:24] so that devs could take a look at the related RP tests and verify that issues are really gone against HEAD commit [15:25] that would be very nice... but some issues get closed without any code changes... [15:25] so far it's just a set of some (recent) issue I greped from GH issues pages - https://rakudist.raku.org/play/rakudo-issues [15:25] so I don't think you can automate that so easily [15:25] yeah, I see. I am only talking about the type of issues where bug fixes and one can verify it through a RP test [15:26] and this is main point in here - https://sparrowdo.wordpress.com/2020/12/03/rakudo-issues-player/ [15:26] I guess we could think about some standard format in issues [15:26] to make this easier [15:27] yeah ... [15:27] "Let's make it easier for Richie" :-) [15:27] let's make it easier for everyone :] [15:28] so extracting code samples from GH could be a final goal, so far it's just a matter of creating some (unit Test ?) via RP and named it as "issue-$issue_numer" [15:29] number [15:29] after that Sparky cron will start replaying the scenario for all other new commits and have summary at the RICH page [15:33] cool! [15:34] http://rakudist.raku.org/sparky/report/RakuPlay-1/2740 just an example. issue-4085 in the build title matters, this is how RICH cron knows it needs to rerun the same build for all new rakudo commits [16:24] *** melezhik left [16:30] *** cog joined [16:57] *** Xliff joined [17:01] *** domidumont left [17:07] *** Altai-man joined [17:09] *** sena_kun left [17:11] *** moray joined [17:11] *** moray is now known as morayj [17:14] <[Tux]> Rakudo v2020.11-93-g705e6e62c (v6.d) on MoarVM 2020.11-54-ga595d9ddc [17:14] <[Tux]> csv-test-xs-20 0.388 - 0.398 [17:14] <[Tux]> csv-ip5xs 0.832 - 0.839 [17:14] <[Tux]> test-t --race 0.844 - 0.926 [17:14] <[Tux]> test-t 1.884 - 1.893 [17:14] <[Tux]> test 7.571 - 7.860 [17:14] <[Tux]> csv-ip5xs-20 7.880 - 8.673 [17:14] <[Tux]> test-t-20 --race 9.364 - 9.747 [17:14] <[Tux]> csv-parser 24.635 - 25.020 [17:14] <[Tux]> test-t-20 31.972 - 32.423 [17:15] ¦ nqp: e7b2b7e0c8 | (Elizabeth Mattijsen)++ | docs/ops.markdown [17:15] ¦ nqp: Some ops documentation reshuffling [17:15] ¦ nqp: [17:15] ¦ nqp: - moved some lemma's to a proper chapter [17:15] ¦ nqp: - created new "Profiling" and "Introspection" section [17:15] ¦ nqp: review: https://github.com/Raku/nqp/commit/e7b2b7e0c8 [17:15] ¦ nqp: 50862433a3 | (Elizabeth Mattijsen)++ | docs/ops.markdown [17:15] ¦ nqp: Restore Miscellaneous section [17:15] ¦ nqp: [17:16] ¦ nqp: It got clobbered in the last commit [17:16] ¦ nqp: review: https://github.com/Raku/nqp/commit/50862433a3 [17:20] ¦ nqp: 04a2ddb79f | (Elizabeth Mattijsen)++ | docs/ops.markdown [17:20] ¦ nqp: Some more moving lemma's around [17:20] ¦ nqp: [17:20] ¦ nqp: Remove the Variable section, added a Time section [17:20] ¦ nqp: review: https://github.com/Raku/nqp/commit/04a2ddb79f [18:17] *** morayj left [18:21] ¦ rakudo: 4304e25059 | (Elizabeth Mattijsen)++ | 3 files [18:21] ¦ rakudo: Fixes to native @a[Iterable] [18:21] ¦ rakudo: [18:21] ¦ rakudo: - Return a list of the same native type, instead of an Opaque one [18:21] ¦ rakudo: - Handle Iterables in containers correctly [18:21] ¦ rakudo: review: https://github.com/rakudo/rakudo/commit/4304e25059 [18:21] *** domidumont joined [18:24] *** domidumont left [19:41] *** MasterDuke joined [20:13] ¦ rakudo: 6c7044ec31 | (Elizabeth Mattijsen)++ | 3 files [20:13] ¦ rakudo: Tighten up native @a[] invocant signatures [20:13] ¦ rakudo: [20:13] ¦ rakudo: Candidates can't handle type objects as invocant, so don't allow that. [20:13] ¦ rakudo: review: https://github.com/rakudo/rakudo/commit/6c7044ec31 [21:05] *** Altai-man left [21:22] *** patrickb left [21:23] *** patrickb joined [23:10] *** leont joined [23:13] *** patrickb left