Time Nick Message 07:35 paramat will merge trivial #7942 in 15 mins 07:35 ShadowBot https://github.com/minetest/minetest/issues/7942 -- Declare mapgens v5 and flat stable. Add missing carpathian by paramat 07:36 nerzhul +1 07:43 CoderForTheBette By the way, I made a PR to fix player rotations from this issue: https://github.com/minetest/minetest/issues/7938 and here is the PR: https://github.com/minetest/minetest/pull/7941 it is a one line change 07:45 paramat thanks 07:49 CoderForTheBette No problem. Goodnight. 08:18 paramat merging 7942 08:19 paramat done 08:27 nerzhul merging #7941 bugfix too 08:27 ShadowBot https://github.com/minetest/minetest/issues/7941 -- Fix player rotations by CoderForTheBetter 08:28 nerzhul trivial bugfix already discussed in the issue 08:28 nerzhul ty CoderForTheBetter 08:43 nerzhul a unittest on that part can be nice but the AO event model is bad 14:34 rubenwardy suggestion for generic syntax to theme things: 14:34 rubenwardy style[itemname;bgcolor;#333] 14:34 ShadowBot https://github.com/minetest/minetest/issues/333 -- Shaders - Darker bottom surfaces (to match brighter top surfaces) by MirceaKitsune 14:34 rubenwardy style_default[button;bgcolor;color] 14:35 rubenwardy kind of a pain to add support for this to most elements, due to irrlicht theming 14:37 rubenwardy the purpose of this is to avoid adding loads and loads of new parameters to formspec elements. It would be implemented by 2 maps, one for elename -> style, another for eletype -> style 14:38 rubenwardy honest to god, I hate the implementation of formspecs so much 14:42 twoelk yet no replacement seems in sight that the community would agree on 14:43 sfan5 I don't think that's the issue 14:43 sfan5 the issue that replacing formspecs is a big task 14:43 rubenwardy yeah 14:45 Unarelith could start with a Lua API extension `f = formspec.new() f:add_button(x, y, text)` included in MT 14:45 rubenwardy that doesn't help 14:45 Unarelith actually it does 14:45 twoelk uhm, so what good replacement was suggested that just needs the man-power and loads of work-hours? 14:45 rubenwardy it doesn't fix the layouting or the ability of anything 14:45 rubenwardy the format is the least of the issues 14:46 Unarelith not at the beginning 14:46 rubenwardy also, http://github.com/rubenwardy/smartfs 14:46 Unarelith i know, exactly this 14:47 Unarelith it's easier to work with this kind of API than with a string 14:47 sfan5 you can't throw backwards compatibility out of the window 14:48 Unarelith this API can start by using formspecs, but once most of the mods will use this API it will be easier to drop the formspec string 14:48 rubenwardy a replacement would have to be more of an alternative than a replacement 14:49 Unarelith include something like this in MT and nobody will use the formspec string 14:49 rubenwardy but yeah, the correct approach for this is to have the 2 systems side by side and deprecate formspecs 14:49 rubenwardy then remove formspecs in 6.0 for the lols 14:49 twoelk so why not fix the layouting issues of formspecs? or rather how should that be achieved as I havn't got a clue why it renders as irritating as it currently does 14:49 rubenwardy it can't be fixed without causing backwards compat issues 14:51 twoelk I don't know wether inserting yet another layer makes things less complicated 14:51 rubenwardy it makes things more complicated, especially if you have to translate between your API and formspec co-ordinates 14:51 rubenwardy so it shouldn't be done on top of formspecs 14:55 twoelk maybe we are looking from the wrong side all to mt centric. Maybe we should look what works well on for example Android and then try to link that to minetest instead of looking at the user through Irrlicht-eyes 14:56 Unarelith +1 14:57 sfan5 most pre-existing solutions are unsuitable as the gui control code is not running on the same device as the gui 14:59 rubenwardy ^ 14:59 Unarelith would be better to just write a small GUI lib then, it's not that hard if you keep it really basic (at MT doesn't need more anyway) 14:59 rubenwardy also, Android-style UIs would be inappropriate, as they're not game-focuses 14:59 Unarelith (as* 14:59 rubenwardy I imagine you mean the actual game itself 14:59 Unarelith ? 15:00 twoelk Android was just an example although at the moment it might be the biggest user group - so not really unimportant. 15:00 rubenwardy #6527 15:00 ShadowBot https://github.com/minetest/minetest/issues/6527 -- Formspec replacement 15:04 rubenwardy This is something that can wait until after 5.0, however 15:07 Unarelith yep 16:13 nerzhul i think it should, it seems we are in a feature less dev period these days 16:14 nerzhul Unarelith if you want to check i published second refactor part on #7939 16:14 ShadowBot https://github.com/minetest/minetest/issues/7939 -- Add an activeobject manager to hold active objects by nerzhul 16:15 nerzhul it's nearly complete, it needs a little bit more unit tests on the server part + clang-format run 16:19 Unarelith nerzhul, +1 for those changes, though the usage of `unique_ptr` for storage could be really good since the variable lifetime is known here 16:20 nerzhul on which part ? the manager itself ? 16:21 Unarelith nerzhul, https://github.com/minetest/minetest/pull/7939#pullrequestreview-181768033 16:22 nerzhul objects are not removed when outside of the map 16:22 nerzhul look at the clear lambda 16:22 nerzhul there are some cases where it's not the case if i'm correct 16:23 nerzhul and there is a condition to remove the object 16:23 nerzhul i don't changed the function part on the object lifetime it's a little bit tricky i prefer to keep the same functional at this moment 16:24 Unarelith ok I see np, though I'm interested to know which class handles object lifetime then 16:25 nerzhul it's this class but handled in a lambda with too many conditions :p 16:25 nerzhul if it was just remove from the map => bye 16:25 nerzhul but no 16:25 nerzhul also the pointer is stored in static object if i remember 16:25 nerzhul it's not easy, i hope refactoring and reducing the object sizes can help to have more proper lifetimes 16:27 * twoelk assumes it is one of the basic rights of an object to have a proper lifetime 16:28 Unarelith nerzhul, sure, I can assure you we'll work on that once 5.0.0 will be ready, until then I'm free for whatever thing you need to release this long awaited version 16:29 Unarelith but you're right, it's better to keep it like this for now 16:29 nerzhul yep 16:30 nerzhul it's mergeable in any window without problem :) 16:30 nerzhul if you want to help look at blocker tag or 5.0.0 milestone 16:30 nerzhul PR increasing the unittest coverage on the core part are welcome too 16:31 Unarelith actually I often look to these to see if I can do anything, but there's not much I can do :/ 16:31 nerzhul yeah i have the same problem 16:31 nerzhul another point i must work on it the postgresql support on the auth part 17:10 VanessaE so... 17:10 VanessaE I've been thinking.. since my hotbar change went in, would mtg benefit from inclusion of my hotbar size mod also, to take advantage of that change? 17:10 VanessaE (for the 5.0.0 release) 17:42 p_gimeno VanessaE: is that something that can't wait until 5.0.0 is release? 17:42 nerzhul i don't know, what did you changed ? what are the new mtg features 17:42 p_gimeno +d 17:42 VanessaE nerzhul: it adds a chat command to change the hotbar to 8, 16, 23, 24, or 32 slots, with accompanying images. 17:42 VanessaE (the 23 is for 0.4.x compat) 17:43 VanessaE p_gimeno: if anything, it would ONLY go in with 5.0.0 if wanted 17:43 VanessaE since 0.4.x is no longer taking new feature 17:43 VanessaE s 17:44 VanessaE nerzhul: and it supposedly saves the player's selection between logins, but maybe that only works on a server (I have to have cheapie re-check that part, it's her code) 17:44 VanessaE https://gitlab.com/VanessaE/hotbar 17:44 VanessaE I'll submit a PR with the 0.4.x compat stuff removed, if desired. 17:44 p_gimeno VanessaE: I mean, can't it wait for 5.0.0 to be released, to include it in the 5.x series? from your description, it appears it can, and new features is not the main focus of 5.0.0 17:45 nerzhul does it you player attribute storage backend ? 17:45 rubenwardy I don't think it's particularly needed 17:45 VanessaE p_gimeno: oh, yeah. that's the idea. 17:45 rubenwardy Also, that name is already used 17:45 nerzhul p_gimeno is right, we really need to stabilize, i know i say that since 8 months, but really... 17:45 rubenwardy !mod hotbar 17:45 ShadowBot rubenwardy: 17:45 VanessaE nerzhul: it uses its own file. 17:45 VanessaE rubenwardy: well nothing stops it being renamed 17:46 VanessaE anyway, there ^ is the code. I've been using it for quite some time on a few of my servers. 17:46 nerzhul VanessaE, it's not good, player attribute backend is the recommended way, mtg should not have mods which doesn't respect the recommendations 17:47 VanessaE *shrug* ok. 17:47 nerzhul it's the advantage of a core game engine, it should use the latest features as it's released with core, it's also the way to produce good code base for mod developers to look for 17:48 VanessaE well this mod was written for 0.4.x and adapted for 5-dev. 17:48 rubenwardy The user facing name is player meta, nerzhul 17:48 nerzhul rubenwardy ty i forgot it :p 17:49 nerzhul the man who wrote the original code forgot it... 17:49 nerzhul or it was renamed after development can't remember :p 17:49 nerzhul i would the the AO refactor in engine then i will try to focus to finish to look for important changes like the android PR 17:49 nerzhul for me android PR is too huge 17:50 p_gimeno VanessaE: I suggest you postpone discussion until after 5.0.0 is released. New features that don't break compatibility can be implemented later. 17:50 nerzhul i think i will split it in multiple commits to separate changes 17:50 VanessaE p_gimeno: screw it. not going to bother now. 17:52 nerzhul oh the android pr is merged 17:52 nerzhul i will do a debug build to see the changes 17:53 rubenwardy Inb4 nerzhul accidentally releases 5.0 to Android 17:53 nerzhul rubenwardy ?? 17:54 nerzhul what do you mean ? 17:54 rubenwardy It's a joke 17:54 rubenwardy *on Android 17:54 rubenwardy *to play store 17:54 nerzhul i can push the apk on the play store as a beta program :p 17:55 nerzhul but yeah i know the community wait for me about that 17:55 nerzhul but community wait for 5.0.0 too 17:55 nerzhul all optimisations are good 17:56 nerzhul gg ruben for updating dev wiki :p 18:15 Krock rubenwardy: I'd prefer if you could make it disappear when there's only one game installed: #7940 18:15 ShadowBot https://github.com/minetest/minetest/issues/7940 -- Disable game bar and header on Android by rubenwardy 18:25 p_gimeno #6898 updated and rebased (on GitLab) 18:25 ShadowBot https://github.com/minetest/minetest/issues/6898 -- Allow distinguishing mods by modpack by pgimeno 18:26 p_gimeno https://gitlab.com/pgimeno/minetest/merge_requests/1 19:07 nerzhul just tested the 5.0.0 master build it's quite good, joystick and icon are better 19:08 nerzhul i push a little gradle build change to make gradle own an asset copy part instead of the shit makefile 19:08 nerzhul it also ensure we always have fresh assets in the build :) 19:18 Krock yes, the android devs did a good job with the build 19:20 nerzhul yep the joystick is a nice feature, old controls are ugly 19:20 nerzhul i think users will like it 19:20 nerzhul ingame icons menus are not very good, i think we should change this at a point 19:20 Krock oh well, the joystick is in the game for a long time already 19:20 nerzhul but too late for 5.0.0 release, we must keep them, not a huge problem 19:20 nerzhul it's not in the 0.4.17 :) 19:20 nerzhul but yeah it was merged long time ago in master 19:23 nerzhul i know the lambda can make the AOMgr more complex but it's just AO callback functions 19:23 nerzhul lambda is the better to keep functional without being too intrusive 19:24 nerzhul i cannot do better on this PR due to the current MT design, environment owns too many things