Time Nick Message 02:44 MTDiscord https://cdn.discordapp.com/attachments/747163566800633906/1293403735069954100/image.png?ex=67073f98&is=6705ee18&hm=48925994c30ff5895fb83176ca642f61ab9b61e155efaabcb2493e74fd85eb4f& 02:45 MTDiscord do we really want to keep irrlicht? rofl 02:48 MTDiscord I was wondering how i would detect if the device was a laptop, or it had a precise scrolling trackpad, but i dont think i want to do it how defaultsettings does it, since i need to check this at runtime based on the device being used. Pretty gross 08:06 sfan5 psa: planning to merge #15257 later today 08:06 ShadowBot https://github.com/minetest/minetest/issues/15257 -- [no squash] Build fixes by sfan5 08:10 nekobit >after I add some overscroll clipping code, i'm curious if you want to take a look at my scrollbar PR 08:10 nekobit a lot of stuff is already done, besides the trackpad stuff, which i think i might move to another PR 08:10 nekobit it'll have to prod device info to see if the current "scroll"'d device is a trackpad, i'm sure i could intervene with some SDL (or is it still Irrlicht?) input code somehow but im not sure 08:11 nekobit atm im struggling to get an irrlicht elements box bounds for clipping 08:11 nekobit because it uses all this Implementation garbage 08:12 nekobit somietimes i forget my fancy IDE can get this stuff for me 08:24 MTDiscord Command sent from Discord by herowl: 08:24 MTDiscord !tell grorp Please take a look at #14343 08:24 ShadowBot MTDiscord: OK. 08:29 sfan5 nekobit: no time or interest, sorry 08:30 nekobit alright :( 08:30 nekobit who has a knack for gui stuff then? 08:33 nekobit i think im just gonna move the trackpad scrolling to another PR. I don't see anything readily accessible, not to mention how hacked up the touchscreen detection system is, being just a setting set at start. 08:33 nekobit would need to fix that too, setting it once as a default config i dont think is a smart idea (for touchscreen, not a big deal, but for trackpad, where user will switch a lot, then yes, terrible idea) 08:34 nekobit aka i dont want precision trackpad scrolling for 5.10 yet (if this even makes it past feature-freeze) 08:34 Desour trackpad scrolling would be nice, imo. but I doubt you have to explicitly check if there's a trackpad. SDL probably already exposes float scroll stuff 08:35 nekobit im not sure its quite float scroll, but i get your point 08:35 Desour I mean scroll events where the distance is float 08:35 Desour and not discrete 08:36 nekobit i dont think it'd be a float either, really. I think its already being reported as just an integer 08:36 nekobit but regardless, we'd need to turn off the scroll step in the scrollbar code if it happens to be a trackpad-like device, or anything with hi-res scrolling (as libinput calls it) 08:36 Desour you'll probably have to change some irr events, and convert the SDL event to irr event 08:36 nekobit it looks like the closest thing to checking this would be looking for the number of fingers 08:37 nekobit with MultiGesture 08:37 nekobit whole different strew of issues 08:37 nekobit just not worth it for this PR 08:37 nekobit my PR opens the door for it, certainly, at least to easily just snip out the trackpad accel, and get an inertia effect 08:38 nekobit s/trackpad accel/scroll step 08:40 Desour https://wiki.libsdl.org/SDL2/SDL_MouseWheelEvent try using preciseX 08:40 nekobit interesting 08:40 nekobit not sure if how SDL2 events are being mapped to the irrlicht events right now 08:41 nekobit probably just stubbed over, thus i'd have to expose preciseX manually 08:41 nekobit (and Y) 08:42 nekobit Alright, i'm calling quits 08:42 nekobit It is up for testing and I'll unmark as draft and edit the issue 08:42 Desour look into the sdl irr device 08:43 nekobit i will, but again, much nicer if i split it into another PR, since i need to ensure it even works well and doesnt act screwy when multiple devices are used 08:43 Desour CIrrDeviceSDL.h 08:43 nekobit this PR focuses on a smoother scrollbar and elastic scrolling 08:43 nekobit Thanks though 08:43 Desour sure, small PRs are good 08:44 nekobit alright good news 08:44 nekobit its exposed 08:44 nekobit hold on 08:44 nekobit hold all the thoughts 08:45 nekobit well, might have an issue, not sure how to see which one is used, since both are propagated, maybe i can check for both being set? 08:45 nekobit CIrrDeviceSDL.cpp:742 08:46 nekobit Ok, nevermind, this won't do. Gross 08:46 nekobit its only using the precise scroll one 08:46 nekobit which probably means SDL2 doesn't work nicely with it either 08:47 nekobit Sorry for the spam.. jotting throughts as I see this in live time. Im gonna sleep 10:21 [ nekobit: supporting multiple fingers isn't the same as supporting precision scrolling. there are trackpads that only support a single finger but still support scrolling, and non-trackpad devices that support precision scrolling (trackpoint) 10:27 [ Since this PR was closed: https://github.com/minetest/minetest/pull/15189, should I provide a patch that makes it use execvp? 10:34 Desour there's #15238 10:34 ShadowBot https://github.com/minetest/minetest/issues/15238 -- Use execvp in fs::RecursiveDelete() by sfan5 10:35 Desour (idk why sfan didn't mention the other PR, it makes navigating so much easier imo) 12:32 grorp merging #14542, #15231, #15248 in 15 min 12:32 ShadowBot https://github.com/minetest/minetest/issues/14542 -- [no squash] Auto-toggle TouchScreenGUI in-game when receiving touch/mouse input by grorp 12:32 ShadowBot https://github.com/minetest/minetest/issues/15231 -- Make bloom parameters server-controlled by grorp 12:32 ShadowBot https://github.com/minetest/minetest/issues/15248 -- ContentDB dialog: React to window resizing immediately by grorp 12:54 [MatrxMT] ping grorp 12:56 sfan5 he left 9 minutes ago 13:06 [MatrxMT] gnome-shell crashed with SIGABRT while testing the auto switching PR 🥲 13:06 [MatrxMT] *retesting 13:07 [MatrxMT] it still doesn't work on Linux for me, but it still works on Andeoid 13:16 [MatrxMT] still works on Windows too and it's disabled by default on Linux, so I guess it's fine? 13:18 [MatrxMT] definitely not going to investigate a gnome-shell crash now 14:11 sfan5 should be fine 14:12 sfan5 merging #15258, #15257 in 10m 14:12 ShadowBot https://github.com/minetest/minetest/issues/15258 -- Split log.h to speed up compilation by paradust7 14:12 ShadowBot https://github.com/minetest/minetest/issues/15257 -- [no squash] Build fixes by sfan5 16:08 [MatrxMT] alright, merging 16:31 jonadab There are still people who use Gnome, other than the Gnome devs? TIL. 16:35 grorp pushing https://github.com/grorp/minetest/commit/32409ed8e7d367a2c555ef7d441fae23847c4deb in 10 min (not a regression from the PR I just merged, it's an earlier one) 16:37 grorp jonadab: I love gnome 16:37 grorp I love *vanilla* gnome 20:54 MTDiscord Which status does #14343 have now? nauta-turbidus did a testing yesterday and confirmed it works. I did some testing also a few times in these days checking for how the ambient light is applied to nodes, objects, particles, wieldmesh, with/without enabled shaders, with/without dynamic shadows/bloom and with hardware-colored nodes and that entirely works correctly. What's necessary to do for it still to allow it being merged? 20:54 ShadowBot https://github.com/minetest/minetest/issues/14343 -- Ambient light and server control for it by Andrey2470T 21:07 MTDiscord @grorp needs to review it again, and maybe @Lars would have to confirm his approval too 21:07 MTDiscord Yeah I plan to do that very soon 21:43 [MatrxMT] I may not have time again before feature freeze