Time Nick Message 12:37 Krock testing #8166 12:37 ShadowBot https://github.com/minetest/minetest/issues/8166 -- Fix (or workaround?) for 'Falling nodes with node inventory cause crash when unloaded' by Warr1024 13:38 sfan5 merging #8999 and #8166 in 5m 13:38 ShadowBot https://github.com/minetest/minetest/issues/8999 -- Fix unwanted detaching when damage = 0 by programmerjake 13:38 ShadowBot https://github.com/minetest/minetest/issues/8166 -- Fix (or workaround?) for 'Falling nodes with node inventory cause crash when unloaded' by Warr1024 13:47 sfan5 oh great github created a garbage shit merge commit again 13:48 sfan5 I forgot about github being a broken piece of shit 13:48 sfan5 even if you wait several minutes between merge 14:14 Krock well so.. no matter how hard I try, I cannot get the formspec to reach the screen edges with real coordinates 14:21 sfan5 is that needed for some builtin formspec or? 14:24 Krock regular modding 14:40 sfan5 #8240 is not essential for 5.1 is it? 14:40 ShadowBot https://github.com/minetest/minetest/issues/8240 -- Fix profiler crash since eea1fda by SmallJoker 14:43 Krock no, #8239 already fixed it, but in a very simple way 14:43 ShadowBot https://github.com/minetest/minetest/issues/8239 -- Fix profiler crash by pauloue 14:45 Krock > Operating system BeOS, Haiku. 14:45 Krock wrong channel 14:45 tomraceror LOL 14:46 Krock I probably don't understand what bgcolor[] is supposed to do at all. Why does fixing the true/false param not work? 14:46 sfan5 it would work 14:46 sfan5 but then you couldn't set both colors at the same time 14:47 Krock well, is that needed? 14:47 sfan5 oh and you couldn't disable both backgrounds (depending on how it's implemented) 14:47 sfan5 not really, but it's cleaner IMO 14:49 rubenwardy The true and false is whether to hide or show the full screen background 14:49 Krock true/false would be clearer params than a custom enum, but well then.. reviewing 14:49 rubenwardy The first parameter is always the formspec background color in 501 14:49 Krock bgcolor, the only element that you can remove again after creating it.. 14:51 Krock the additional param also requires noting that it's for formspec_version 2 or later 14:52 Krock either that or 5.0.1 clients will error 15:47 paramat apart from translations, which i do not understand and am unsure of, MTG is ready for release 15:48 paramat oh ... broken 9 slice 15:48 paramat that's the only milestone 15:49 paramat a job for the ruby warden 15:52 Krock what about optipng/pngcrush? 15:54 Krock will push MTG bg9 fix in 15 minutes: https://krock-works.uk.to/u/patches/0001-Use-background9-for-backwards-compatibility.patch 15:55 Krock err 15:56 Krock the patch included a testing texture. Ctrl+F5 for clean version 16:02 paramat work on texture optimisations was started too late, it's a big task, moved to after release 16:03 paramat thanks 16:08 sfan5 I wonder if it's fine to have the same background twice 16:08 Krock #9018 16:08 ShadowBot https://github.com/minetest/minetest/issues/9018 -- Revert "Fix the bgcolor formspec element (#8716)" by SmallJoker 16:08 sfan5 since newer clients will understand both background[] and background9[] 16:09 sfan5 merging #9017 in a few minutes 16:09 ShadowBot https://github.com/minetest/minetest/issues/9017 -- Fix warnings in guiButton.h by sfan5 16:10 sfan5 Krock: do you think #8996 isn't good for 5.1 or why the new PR? 16:10 ShadowBot https://github.com/minetest/minetest/issues/8996 -- Formspec: make bgcolor element less confusing by using 3 parameters by DS-Minetest 16:10 Krock please see new comment below 16:10 Krock adding new params is not backwards compatible 16:11 sfan5 can't we bump the formspec version? 16:11 sfan5 this is such a mess 16:11 Krock indeed it is 16:11 Krock after 5.0.1 is no longer to care about, this whole thing won't be an issue any more 16:11 Krock since formspec_version[x] can then used by modders to specify whatever they want to use 16:12 sfan5 wait does formspec_version not exist in 5.0? 16:12 sfan5 indeed 16:12 Krock well yes, but increasing it would probably not always result in the wanted formspec display in case something changes 16:13 Krock it's best to leave it to modders so they can decide what's best for their formspecs 16:14 Krock it's indeed an entire mess, but simply increasing it isn't necessary the best option since running such mods on a 5.0.1 client will cause warnings for connecting clients again 16:14 Krock unless the mods have a minimal Minetest version requirement 16:18 Krock relevant PR #8923 16:18 ShadowBot https://github.com/minetest/minetest/issues/8923 -- Formspec version by SmallJoker 16:22 Krock I wonder if it's fine to have the same background twice 16:23 Krock yes, the second will override the first in case it's supported 16:23 Krock texture packs can provide better inventory backgrounds this way 16:23 sfan5 https://github.com/minetest/minetest/blob/master/src/gui/guiFormSpecMenu.cpp#L815 this code looks more like it would just draw it twice 16:24 Krock one overriding the other 16:24 Krock not replace, but be drawn on top 16:24 sfan5 well, true 16:25 Krock different story if it had opacity 16:57 Krock well so, is it okay if I push that background9 patch? 16:57 Krock asking due to concerns 17:20 rubenwardy Lgtm 17:20 rubenwardy I've been quite busy 17:24 Krock ok. pushing 17:25 Krock 1c413ed 18:50 paramat don't forget ... we may need to regenerate conf.example before release 18:51 Krock PR is on hold. weblate needs to be fixed 18:52 paramat ok 18:52 paramat #9018 is approved, are we going to merge it? 18:52 ShadowBot https://github.com/minetest/minetest/issues/9018 -- Revert "Fix the bgcolor formspec element (#8716)" by SmallJoker 18:59 Krock so is the alternative PR 19:00 Krock dunno which path to chose 19:09 paramat ok, i wasn't sure if a decision had been made, maybe decide tomorrow 22:20 Wuzzy also don't forget to merge MTG translations ?