[01:04] *** kylese left [01:04] *** hulk joined [01:19] *** sibl joined [01:25] *** _________ left [01:26] *** _________ joined [01:33] *** MasterDuke joined [02:02] *** kylese joined [02:02] *** hulk left [02:10] *** MasterDuke left [02:15] *** kylese left [02:15] *** kylese joined [03:02] *** lichtkind_ joined [03:05] *** lichtkind left [03:39] *** sibl left [03:45] *** sibl joined [03:55] *** sibl left [04:00] *** lichtkind__ joined [04:03] *** lichtkind_ left [05:56] *** Inline left [06:00] *** sibl joined [06:03] *** Inline joined [06:06] *** sibl left [06:10] *** sibl joined [06:12] *** kylese left [06:13] *** kylese joined [07:00] *** sibl left [07:09] *** Sgeo left [07:24] *** sibl joined [07:38] *** sibl left [07:51] *** dakkar joined [08:12] *** dmvrtx left [08:13] *** dmvrtx joined [08:30] japhb: The readme on the Selkie repo (https://github.com/m-doughty/Selkie) and API docs (https://github.com/m-doughty/Selkie/blob/master/docs/api/index.md) are pretty comprehensive imo. [08:34] Notcurses::Native 0.4.0 is released, fixes some potential memory leaks by annotating which objects are user-freed and which are system-freed, pulls notcurses from my fork instead of vendoring it (makes submitting upstream fixes easier), fixes a memory leak which was silently allocating Str and not freeing underlying char* buffer. [08:48] I am about to push Selkie 0.8.0 which has a couple of default flips (bc breaks). Selkie subscriptions now default to deep check equality, you can remove :deep-equality-check at all callsites and if you want the old default behavior, you now need to pass :identity-check-only. Effect payloads are now *required* to be associative (most probably already are) but if you have any that aren't, they now need [08:48] wrapping. An empty hash will have no payload effect. It also uses Notcurses::Native 0.4.0 which fixes some memory leaks. [08:49] These breaks are documented in Changes [08:49] *** sibl joined [08:49] SmokeMachine: Relevant for Selkie::UI [08:50] BTW I would recommend Selkie::UI pins exact version of Selkie if it doesn't already until 1.0.0, at 1.0.0 there will be no bc breaks until 2.0.0 [08:50] Pre-1.0.0 minors are allowed a bc break, no bc breaks in patch versions [10:06] *** Pixi` joined [10:08] *** Pixi left [10:10] *** sibl left [10:24] *** donaldh joined [10:36] *** librasteve_ left [10:42] *** kurahaupo left [11:12] *** kurahaupo joined [11:31] *** kurahaupo left [11:31] *** donaldh left [11:31] *** dmvrtx left [11:31] *** donaldh joined [11:32] *** dmvrtx joined [11:36] *** sibl joined [11:39] *** kurahaupo joined [11:43] apogee_ntv: it is currently setting the minimum version… I’ll update it to a specific version [11:44] Thanks [11:45] I dont know if there's a way to pin a minor or major in zef [11:45] If you can pin the minor I will not break bc on a minor. [11:46] Once it gets to 1.0 only a major will break bc unless you rely on undocumented behavior (bugs) [11:52] m: say v1.0.3 ~~ v1.*; say v2.0.0 ~~ v1.* # I think it is possible [11:52] SmokeMachine, rakudo-moar 6f0f1f5a8: OUTPUT: «True␤False␤» [11:54] apogee_ntv: I was reading about your terminal and I have 1 question: have you thought on using Raku itself for configuration? Maybe with something like this? https://raku.land/zef:FCO/Configuration [11:58] apogee_ntv: about breaking changes, on Red, when I do breaking changes (happened once) I change the api… (I mean the api version) so, now we are on Red:api<2> [11:59] SmokeMachine: We would be on api ~18 by now :D [11:59] apogee_ntv: I don’t see that as a problem… [12:01] Maybe not, tbh once we're on 1.0 the API will be locked. It will only be added to with new modules & new kwargs, defaults will not change. [12:01] 👍 [12:02] What about the configuration in Raku for your terminal? [12:02] For my terminal? [12:02] I use Kitty [12:05] I dont maintain a terminal [12:07] Sorry, it seems I misread the author’s name… [12:08] Sorry [12:08] avuserow: have you thought about using Raku for your terminal’s configuration? Maybe something like https://raku.land/zef:FCO/Configuration ? [12:08] I do contribute to notcurses very occasionally. [12:09] The C lib that Selkie sits on top of. [12:09] I mean RelayTerm [12:09] Ah [12:10] Sorry for the confusion [12:10] No worries [12:30] *** johnjay left [13:09] *** sibl left [13:10] *** hurufu joined [13:49] o/ [13:51] Is any way to speed up Raku regexp ? Applying relatively simple regexp on array of few thousands short strings takes 10 minutes [13:51] Very slow [13:51] Maybe Rakuast will improve that ? [13:51] if you're not interested in the actual match, just *if* there is something, you're better of with .contains(/foo/) [13:52] as that will not create the expensive Match object [13:52] RakuAST in itself will not improve that [13:52] rakuast doesn't return match objects [13:52] ? [13:52] i dont know if that is intentional though [13:53] https://github.com/melezhik/Sparrow6/blob/master/lib/Sparrow6/Task/Check.rakumod#L223 [13:53] Yeah I am taking about regexp , not about contain … [13:53] ugexe: re rakuast not returning Match objects? what to do mean? [13:53] let me look what i actually mean. there is some variation of smartmatch that does not return the match object but the naive example i tried does return it so maybe im wrong [13:54] For my day job project it maybe a bottle neck [13:54] melezhik: [13:55] my $matched = $data.comb(/<$pattern>/,:match) doesn't do what you want ? [13:56] Is it any different from the current implementation? Could it be faster ? [13:57] my $m = 1 ~~ m/\w/; say "house" ~~ $m [13:57] ugexe, rakudo-moar 6f0f1f5a8: OUTPUT: «｢1｣␤» [13:57] Yeah I mean I can remove for sure and rewrite it your way [13:57] well, it would be doing less [13:57] that returns True on rakuast [13:57] which is *always* an advantage [13:58] Well the question how much faster it’ll be )) [13:58] But I can try ) [13:58] no idea [13:58] Yeah … [14:02] m: say run($*EXECUTABLE, q|-e|, q|my $m = 1 ~~ m/\w/; say "house" ~~ $m|, :merge).out.slurp(:close); BEGIN %*ENV = 1 [14:02] ugexe, rakudo-moar 6f0f1f5a8: OUTPUT: «True␤␤» [14:02] m: say run($*EXECUTABLE, q|-e|, q|my $m = 1 ~~ m/\w/; say "house" ~~ $m|, :merge).out.slurp(:close); [14:02] ugexe, rakudo-moar 6f0f1f5a8: OUTPUT: «｢1｣␤␤» [14:08] hmmm... that's a weird difference [14:09] *** Inline left [14:32] *** annamalai left [14:32] *** annamalai joined [14:38] *** kjp left [14:38] *** kjp joined [14:39] *** abraxxa joined [14:39] *** hurufu left [15:46] *** Inline joined [15:56] *** Inline left [16:08] *** annamalai left [16:19] *** Pixi` is now known as Pixi [16:21] *** annamalai joined [16:33] *** human-blip left [16:35] *** human-blip joined [16:37] *** dakkar left [16:59] *** Aedil3 left [17:01] *** Aedil3 joined [17:13] *** dg left [17:18] *** dg joined [18:37] *** belluzj joined [18:45] *** belluzj left [18:47] *** belluzj joined [18:47] *** belluzj left [19:28] *** ShimmerFairy left [19:30] *** ShimmerFairy joined [19:53] *** ShimmerFairy left [19:53] *** dg left [19:53] *** human-blip left [19:53] *** kurahaupo left [19:53] *** Pixi left [19:53] *** lichtkind__ left [19:53] *** dg joined [19:53] *** lichtkind__ joined [19:54] *** human-blip joined [20:02] *** kurahaupo joined [20:03] *** Inline joined [20:11] *** nicole left [21:15] *** simcop2387 left [21:22] *** simcop2387 joined [21:42] *** undermine joined [22:02] *** vasko4535586 joined [22:06] *** perryprog left [22:07] *** perryprog joined [22:07] *** ShimmerFairy joined [22:12] *** Sgeo joined [22:16] *** vrurg_ joined [22:20] *** vrurg left [22:23] *** leifgunnar left [22:28] *** leifgunnar joined