[00:02] *** reportable6 left [00:02] *** reportable6 joined [00:07] *** kjp joined [00:31] Got this building Rakudo HEAD just now: [00:31] t/04-nativecall/02-simple-args.t ................................ Dubious, test returned 1 (wstat 256, 0x100) [00:31] No subtests run [00:31] nine: ^^ [00:38] Rebuilt again and it passed this time. Flaky test? [00:50] *** kjp left [01:04] apparently that's a bit of a problem child. just at a quick glance there are issues open for failures in that file on ARM, M1 macs, WSL, and clang builds of rakudo on linux [01:21] Ooof [01:21] *** kjp joined [02:55] *** Kaipi left [02:56] *** Kaipi joined [06:03] *** reportable6 left [06:03] *** reportable6 joined [06:48] *** patrickb joined [06:57] o/ [06:59] re yesterdays discussion wrt what to do with the Star Bundle. I'm pretty sure the module list wasn't properly maintained. So yes giving the list some love is in order. [07:02] In my experience the module search has been sufficient to rather quickly find all possible modules to solve a given problem. [07:03] *** vrurg_ joined [07:04] To more quickly judge which module to choose it's all a question of which metrics we provide the user. Tests passing is definitely a good thing to have. IIRC we even once had something similar to smoke (automatic test reporting of users test runs) in zef, but it was turned off soon and I thing the corresponding webservice doesn't exist anymore. [07:05] *** vrurg left [07:07] Metrics I think would help me to decide which module to use: When was the module initially released? When was the last release? How many releases have there been in total? Upstream and downstream count. [07:08] But that's more of a topic for raku.land. [07:44] *** Geth left [07:44] *** Geth joined [07:59] *** Xliff joined [07:59] o/ [08:00] \o [08:17] Totally agree on the above metrics, I'm tempted to steal the new bus factor from metacpan too [08:51] JRaspass: Thanks for your work on raku.land! [08:53] so what is our feeling about this: [08:53] m: for ^10 { LAST last } [08:53] rakudo-moar 1bd0411c0: OUTPUT: «===SORRY!===␤last without loop construct␤» [08:53] aka, using control exception constructs inside phasers [08:54] m: for ^10 { NEXT last } # there's actually a spectest for this one [08:54] rakudo-moar 1bd0411c0: ( no output ) [08:54] JRaspass: If I may wish for one more thing: Displaying the ecosystem of a module in the module page and the search list. Workaround is looking in the address bar, so not high prio,but I think it's a good thing to give the ecos some visibility. [09:00] Oh that's a great idea, I shall add it to the issues [09:00] done, https://gitlab.com/raku-land/raku-land/-/issues/6 [09:01] Yay! [09:22] hi, looking at raku.land i see dups on prolific author list [09:24] JRaspass: ++ for raku.land, esp for the alpha sorted list! [09:26] if you're looking for suggestion: add sort by date of publishing [09:28] correction, the dups are on the popular dist list [09:45] *** frost joined [10:59] *** Kaipi left [11:00] *** Kaipi joined [11:14] *** frost left [11:28] *** frost joined [12:02] *** reportable6 left [12:03] *** reportable6 joined [12:56] .tell JRaspass ^^^ [12:57] tbrowder, I'll pass your message to JRaspass [13:01] .ask lizmat is there an issue with my rakudo docs PR #4441? my intent was to get published pretty much as is and let you experts add to it--a living WIP [13:01] tbrowder, I'll pass your message to lizmat [14:51] *** melezhik joined [14:51] . [14:51] releasable: status [14:51] melezhik, Next release in ≈15 days and ≈4 hours. 1 blocker. Changelog for this release was not started yet [14:51] melezhik, Details: https://gist.github.com/b4816c38719559d3b14857165907813a [15:04] lizmat: `LAST last` That one is tricky. Because it's a NOP, and such could be implemented as a NOP, but is also most likely a mistake. In a dynamically optimised language any NOP should vanish. I would impelent it as a NOP and warn. [15:06] lizmat: The tricky one is `NEXT next` because that can have side effects and therefore must be executed. But again, it can also be a mistake. [15:09] *** melezhik left [15:28] ¦ rakudo/new-disp: cb029f1a51 | (Jonathan Worthington)++ | 4 files [15:28] ¦ rakudo/new-disp: Start adapting Raku to always use new-disp on Moar [15:28] ¦ rakudo/new-disp: [15:28] ¦ rakudo/new-disp: Switch all the places that were guarded by an environment variable to [15:28] ¦ rakudo/new-disp: instead be VM conditional. The code to make callmethod compile into a [15:28] ¦ rakudo/new-disp: dispatch op goes away, as NQP on MoarVM will now do that by default (and [15:28] ¦ rakudo/new-disp: also for call ops). [15:28] ¦ rakudo/new-disp: review: https://github.com/rakudo/rakudo/commit/cb029f1a51 [15:28] ¦ rakudo/new-disp: 9e66de3ac6 | (Jonathan Worthington)++ | src/Perl6/bootstrap.c/BOOTSTRAP.nqp [15:28] ¦ rakudo/new-disp: Use NQP dispatchers as Raku ones during bootstrap [15:28] ¦ rakudo/new-disp: [15:28] ¦ rakudo/new-disp: Just to get us through the dispatches we need to do during the process [15:28] ¦ rakudo/new-disp: of setting things up. Since we don't currently wrap any code objects up [15:28] ¦ rakudo/new-disp: as Raku ones, and the meta-objects are compatible so far as using the [15:28] ¦ rakudo/new-disp: `find_method` API goes, this is good enough. [15:28] ¦ rakudo/new-disp: review: https://github.com/rakudo/rakudo/commit/9e66de3ac6 [15:47] *** patrickb left [15:48] Oooh [16:11] *** smarton left [16:11] *** smarton joined [17:25] *** Kaipi left [17:26] *** Kaipi joined [17:26] *** cognominal joined [18:02] *** reportable6 left [18:03] *** reportable6 joined [18:45] *** Kaipi left [18:45] *** Kaipi joined [18:47] *** Kaipi left [18:47] *** Kaipi joined [18:50] *** Kaipi left [18:50] *** Kaipi joined [20:09] *** Xliff left [22:34] *** dogbert17 left [22:53] *** Kaiepi joined [22:53] *** Kaipi left