Time Nick Message 16:59 [MTMatrix] Whats the file in minetest repo that is able the read the server(or the world) that makes the game playable 17:06 celeron55 "that makes the game playable" doesn't make a lot of sense. i'll ask the classic tech support counter question: what are you trying to do? 17:06 Krock @Primo Kidd I'd say that 50% of the entire code is responsible for that. The other is for Lua API bindings and main menu 17:07 Krock I suppose you're trying to fix a printer 18:00 MTDiscord Woah woah, the Lua mapgen thing is a HUGE feature for modders. Actually useable Lua mapgen is a game changer for me. Kinda hard to run a server of any kind when your main thread locks up Everytime you load a new mapblock 18:01 MTDiscord And yeah, at least 5.10 will have some* cool new stuff. 18:07 MTDiscord I'm actually excited for this year's game jam;) 18:12 [MTMatrix] why? There is not much new you can do compared to previous versions if 5.10 isn't released before the jam 18:44 Krock it might be time to go through the 5.9.0 milestone the next few days to see what we'd like to fix before release 18:45 Krock so far it does not appear that there's any blocker 19:12 [MTMatrix] Thanks for ignoring my 5.10 messages from yesterday..? I'm sorry but this is not how things should be handled - with silence. I guess that I should stomp my feet (yet again) as heck did a few days ago to be considered? Have some respect for the time I spent motivating why I think it's a bad idea 19:15 Krock checking logs ... 19:20 Krock @Zughy What exactly were you asking for? 19:21 Krock as for #13987 - that's too late considering that we're already in feature freeze 19:21 ShadowBot https://github.com/minetest/minetest/issues/13987 -- [pls do not squash :3] Allow limiting object observers by appgurueu 19:22 [MTMatrix] in short: 19:22 [MTMatrix] 1. there's no way we can put all those big features in 5.10 -> some people will be let down, whether it's modders or contributors 19:22 [MTMatrix] 2. modders won't have much to play with in 5.9. 13987 would have been a great tool to make people happy. It was already there when the feature freeze started and nobody opposed the fact there was a feature still hanging 19:23 Krock 1. don't worry. I think that not all of them will make it into 5.10, either because drowning in rebase hell or absent authors 19:23 Krock 2. It can't take forever releasing a new version, I think we should definitely start releasing more often after 5.9 19:23 sfan5 I thought 13987 was planned to be merged for 5.9? 19:24 Krock originally yes, but a few weeks into feature freeze, I noticed that this PR was still open, a feature, thus would not fit to the rules 19:24 Krock I mean merging it wouldn't be the problem - it's fixing the issues that yet aren't known. 19:26 sfan5 we can make exceptions of course but it indeed would be better to have it earlier than later due to testing 19:26 [MTMatrix] The (probably) most popular server has been using it for more than a month, I think it's pretty safe 19:26 sfan5 a possible way out of this is to merge it now anyway and release an RC for more testing exposure 19:26 [MTMatrix] (also worst case scenarios we already have broken visibility, see #12403) 19:26 ShadowBot https://github.com/minetest/minetest/issues/12403 -- `is_visible = false` doesn't work on players 19:27 Krock how extensive have they used the API? Granted, long-term tests are great but one or two mods will not test all edge-cases 19:28 [MTMatrix] In the past we merged things that turned to be quite broken, like the custom celestial vault. It took two years to notice because apparently I was the only one who really started playing with such feature *and* bothered to open a few issues/PRs 19:30 [MTMatrix] also see fog in 5.8. The design is debatable, sfan5 fixed a few things in 5.9-dev 19:30 Krock fair point. dev versions aren't tested well either in some cases, thus the RC build would make sense to me to speed things up a little bit 19:34 Krock currently reviewing that PR. let's see how it goes. 19:34 [MTMatrix] thank you Krock for understanding 19:35 [MTMatrix] Since we're at it, can we write down somewhere the idea of smaller dev cycles (e.g. 3 or 4 months?) 19:36 Krock it's been noted down i 2-3 meetings already 19:36 Krock that's over the course of years. it's always been a target but we always ended up with this rythm 19:37 sfan5 speaking of #14342 I think we need to revert what led to this. we can't pull in such a far-reaching change close to release 19:37 ShadowBot https://github.com/minetest/minetest/issues/14342 -- Send sound to players which comes to hear distance. by sfence 19:38 Krock I thought the same thing. Yes. 19:39 Krock it's labelled as Bugfix but the logic that it implements seem more like a feature ... 19:40 MTDiscord Coming soon: Fix broken double wielding support 19:41 MTDiscord Also: Fix codebase accidentally being written in C++ instead of Rust 19:41 sfan5 merging #14841, #14849, #14648 in 10 min 19:41 ShadowBot https://github.com/minetest/minetest/issues/14841 -- Allow main menu gamebar scroll to loop (#14835) by Kazooo100 19:41 ShadowBot https://github.com/minetest/minetest/issues/14849 -- Un-deprecate redis database backend by sfan5 19:41 ShadowBot https://github.com/minetest/minetest/issues/14648 -- Better AppStream metadata by AFCMS 19:41 sfan5 the mainmenu PR is arguably a UI bugfix 19:44 Krock fine my be 19:53 Desour regarding milestone issues, I've recently noticed that #13146 was merged before 5.8.0 was released. so, the regressions caused by it are not 5.9-dev regressions. by updating from 5.8 to 5.9, users won't get a downgrade due to them. hence, I think we can remove those issues from the milestone, no? 19:54 ShadowBot https://github.com/minetest/minetest/issues/13146 -- Inventory mouse shortcut improvements by OgelGames 20:13 Krock Desour: you're right. the milestone'd issues and PRs aren't marked as blocker either, thus could be removed from there 20:14 Krock or moved to 5.10 so that they're not forgotten 8) 20:14 MTDiscord off the cuff idea about 5.10, no idea if its any good. maybe we finally get 5.9 out the door (at some point), and then merge a predefined list of prs (for example lua mapgen, object visibility per player, async ipc, etc) that are pretty much ready, and go straight to feature freeze for 5.10? 20:16 Desour ah, I didn't know the blocker label is used like this 20:16 Krock heh that's an ambitious idea. it's indeed that feature PRs are merged very often early on, giving them time for testing. however, if we'd have the default 2 week testing, it would not be enough. 20:17 MTDiscord sure, but at least for the three listed prs, they have been tested by at least a few people for a decent bit of time 20:17 MTDiscord and lets by honest, feature freeze is always longer than 2 weeks lol 20:17 MTDiscord *be 20:17 Krock never said that I wouldn't agree with you :) 20:17 Desour wsor4035: we already have lua mapgen. it would be funny though, if our feature freezes get so long, that 90% of release period is freeze 20:18 Krock we've already had 2 weeks, sometimes 3. this time it's pretty long, though. 20:18 MTDiscord desour: async lua mapgen 20:19 Krock isn't the Linux kernel moving at a similar pace? merge all features at once, then enter feature freeze for fixes shortly after? 20:20 MTDiscord seems async lua mapgen was merged for 5.9, anyways, other two examples still stand 20:20 Desour wsor4035: yes, the mapgen thread PR. it was merged 20:22 MTDiscord krock: im not sure. kernel does have a lot more steps before stuff gets to linus for merging