| Time |
Nick |
Message |
| 00:34 |
|
rubenwardy joined #luanti-dev |
| 01:48 |
|
wsor4035 joined #luanti-dev |
| 04:00 |
|
MTDiscord joined #luanti-dev |
| 06:35 |
|
Panquesito7_ joined #luanti-dev |
| 07:39 |
|
crazyR joined #luanti-dev |
| 07:41 |
|
crazyR_ joined #luanti-dev |
| 07:44 |
|
pgimeno joined #luanti-dev |
| 09:35 |
|
swags joined #luanti-dev |
| 10:03 |
|
Mantar joined #luanti-dev |
| 10:03 |
|
izzyb joined #luanti-dev |
| 12:05 |
|
rubenwardy joined #luanti-dev |
| 12:08 |
|
rubenwardy joined #luanti-dev |
| 12:22 |
|
rubenwardy joined #luanti-dev |
| 13:51 |
|
clavi joined #luanti-dev |
| 13:51 |
|
clavi joined #luanti-dev |
| 14:49 |
sfan5 |
people now knowing that they need to explain their use case for feature requests seems to be a recurring thing |
| 14:49 |
sfan5 |
not* |
| 14:52 |
Krock |
@.zenonseth Revising my proposal ... |
| 14:56 |
Krock |
https://pastebin.com/raw/jyZp9RAL |
| 14:57 |
Krock |
currently there's a recursive dependency on "enabled_packs". Makes more sense to fill out the fields after removing the invalid ones |
| 14:58 |
Krock |
s/invalid ones/invalid texture pack paths/ |
| 15:09 |
sfan5 |
merging #17676, |
| 15:09 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17676 -- Make the content tab's sub-tabheader be flush against the list below it by ZenonSeth |
| 15:10 |
sfan5 |
merging #17676, #17665, #17666, #17675, #17654, #17591 in 30m |
| 15:10 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17676 -- Make the content tab's sub-tabheader be flush against the list below it by ZenonSeth |
| 15:10 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17665 -- [no squash] Three bug fixes by sfan5 |
| 15:10 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17666 -- Fix possible invalid string_view access by paradust7 |
| 15:10 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17675 -- Misc renames from Minetest to Luanti by user333dev |
| 15:10 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17654 -- Bump Android build target SDK to 36 + opt out of predictive back by ZenonSeth |
| 15:10 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17591 -- Limit mod channel memberships per client by acts-1631 |
| 15:40 |
MTDiscord |
<.zenonseth> Krock: I'll have a look in a bit, |
| 16:07 |
MTDiscord |
<.zenonseth> hmm so Krock: tested the new patch. Fixes numbering but it has some strange side effects that I think are unintentional? When a game is selected, on startup only the game's music plays, but the theme goes back to the luanti default. Also after messing with texture packs (move/uninstall/install), the game's music will keep playing in other menus - guessing because init() drops the music handle. These didn't happen in my pr or in plain |
| 16:07 |
MTDiscord |
5.17. Also list used to show folder name, now its texturepack name, but i assume that was intentional. Let me see if i can rewrite the code I have to the approach of setting the correct list from only what's found, avoiding the recursive concern hopefully, but without messing too much with the theme handling |
| 16:21 |
|
crazyR_ joined #luanti-dev |
| 16:30 |
Krock |
mhm tricky |
| 16:33 |
Krock |
so perhaps mm_game_theme cannot be called in any case |
| 16:34 |
Krock |
changing 'mm_game_theme.texturepack' alone might be sufficient? |
| 16:36 |
MTDiscord |
<.zenonseth> i gotta say the music keeping playing is actually nice, but it's best if its a feature not a side-effect of this bug-fix PR :P |
| 16:36 |
Krock |
I also noticed that after changing the texture pack order, the loading screen fixes itself after a few rejoins |
| 16:37 |
Krock |
as in: the progress bar does not update immediately. |
| 16:37 |
MTDiscord |
<.zenonseth> for only setting mm_game_theme.texturepack - yea maybe, i'll do the rewrite after dinner, maybe it is, just gotta test |
| 16:38 |
MTDiscord |
<.zenonseth> > changing the texture pack order, the loading screen fixes itself after a few rejoins yea possibly, i dind't test that |
| 17:16 |
|
Panquesito7 joined #luanti-dev |
| 17:50 |
MTDiscord |
<.zenonseth> Krock: PR updated, building list fresh from what's on disk |
| 17:55 |
Sheriff_U3 |
Would you guys be interested in a AppImage for v5.18-rc1? I'm currently building some from rollerozxa's scripts. Was thinking that it might be nice as another option for people to test the rc with. |
| 18:12 |
Krock |
Does https://github.com/luanti-org/luanti/blob/5.17.0-rc1/src/script/lua_api/l_env.cpp#L509-L528 actually run the destructor on box_type? I found that on MinGW it probably won't. |
| 18:13 |
Krock |
(I mean: in the case where `luaL_error` is called) |
| 18:16 |
Krock |
hmmm #12778 |
| 18:16 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/12778 -- unit test "testLuaDestructors" failure on many architecures other than amd64/i386. |
| 18:24 |
sfan5 |
i think it was working on mingw but not msvc? |
| 18:32 |
Krock |
I found #14329 > can't be fixed. largely inconsequential, maybe a few memory leaks |
| 18:32 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/14329 -- Switch to LLVM-based MinGW toolchain by sfan5 |
| 18:34 |
Krock |
there was a topic on Thread Local Storage which did not work on some MinGW flavours, I think. |
| 18:34 |
sfan5 |
right |
| 18:36 |
Krock |
at least on mingw64 and ucrt64 (https://packages.msys2.org/repos), a simple luaL_error() or similar check functions use a longjmp, skipping C++ destructors. |
| 18:36 |
Krock |
the Debian/Ubuntu guys must have done something special to get that to work on GCC/Linux |
| 18:44 |
Krock |
correction: they didn't. LuaJIT from master simply happens to work correctly despite all build with GCC (not G++) from what I can tell. |
| 19:32 |
|
Warr1024 joined #luanti-dev |
| 20:16 |
MTDiscord |
<.zenonseth> left a changes-requested review on #17523 which had one approval. if anyone thinks the issues raised aren't blockers, lmk |
| 20:16 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17523 -- Correct positioning of the player camera base when attached by Andrey2470T |
| 20:26 |
sfan5 |
it won't be possible to merge this within the freeze period anyway |
| 20:29 |
MTDiscord |
<.zenonseth> ok - any other PRs worth looking at for the 5.18 release? i just see one in the milestone - "Fix for objects (that are marked for deactivation)" - which you alrady left a request for changes fo |
| 21:23 |
|
vampirefrog joined #luanti-dev |
| 21:28 |
sfan5 |
nothing else in the milestone |
| 21:31 |
MTDiscord |
<.zenonseth> if #17674 gets approved might be good as a bugfix to incl in 5.18's new feature, but it's also not vital, since nothing gamebreaking happens if user deletes an active texturepack, numbers just look iffy |
| 21:31 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17674 -- Ensure all txtpacks in enabled list are present on reload, remove those that aren't by ZenonSeth |
| 21:41 |
sfan5 |
bug fixes are okay of course |
| 22:28 |
|
panwolfram joined #luanti-dev |
