Time Nick Message 09:26 shivajiva minetest/dev/src/script/common/c_types.h:23:10: fatal error: lua.h: No such file or directory 09:27 shivajiva where is it actually looking for lua.h ???????????????? 09:30 calcul0n in lua headers i guess, i have one in /usr/include/lua5.1/lua.h 09:30 shivajiva oddly enough so do I 09:32 shivajiva just trying to compile on ubutu 18.04 but it doesn't want to play 09:32 calcul0n hmm, are you sure it's 5.1? afaik minetest only works with this version 09:33 shivajiva I can see /usr/include/lua5.1/ with the header sat in there 09:34 calcul0n ho wait, it's also included in minetest sources, in src/minetest/lib/lua/src/lua.h 09:36 shivajiva yes I have that too 09:37 shivajiva it's not like I haven't done this many times, just never on 18.04 09:38 Krock VanessaE: https://krock-works.uk.to/u/patches/0001-Fix-wrong-inventory_image-and-wield_image-types.patch 09:38 Krock crash fix for 5.3.0-dev type checks 09:39 shivajiva any ideas anyone? 09:40 calcul0n why do you have a minetest/dev subdir? i just have minetest/src/... here 09:40 Krock shivajiva: checked the makefile paths? 09:40 Krock dev/ is ok, if dev is the git clone location 09:40 calcul0n ho, i see 09:41 Krock clear cmake cache & retry. maybe manual path adjustments for LUA_INCLUDE_DIR is needed 09:41 Krock cmake . -DLUA_INCLUDE_DIR=/usr/include/lua5.1 in case that's the right location 09:45 shivajiva dev is the cloned location within a minetest folder that holds all the versions I have compiled, I just use it for the head 09:50 shivajiva CMakeCache.txt removed and retry works, thanks Krock 09:50 Krock :) 09:50 Krock !next 09:50 MinetestBot Another satisfied customer. Next! 09:50 shivajiva lol 09:54 shivajiva I think I tried to build it before I installed 5.1-dev 09:55 sfan5 since the minetest source comes with lua 5.1, you should only need to install it if you want luajit 09:55 shivajiva yea that's what I thought 09:56 shivajiva it should have built though based on that 10:00 shivajiva I did see it was using system for luajit though whilst poking around 16:51 Krock Minetest be like: 16:52 Krock > Implement one feature, close the according issue 16:52 Krock > People open 2 new issues as feature requests 16:52 Krock > 1 or 2 new bug reports regarding the feature 16:53 Krock So in order to bring the issue count down, is to not add any features, and keep bugs. 17:01 rubenwardy my favourite is when feature requests are disguised as bug reports 17:44 Jordach Krock, lmao another bug discorvered 17:44 Jordach it's not quaternion or euler xyz angles 17:44 Jordach it behaves exactly like 5.2 but the anim mixing works 17:44 Krock staaph it. no more bugs 17:45 Jordach is there an bone limit in irrlicht 17:45 Krock probably a 2^n number 17:45 Krock question of the integer size 17:45 Jordach 49 bones in model so it;s under 8 bits 17:46 Jordach another mention that's related 17:46 Jordach attached entities with said model also completely lose their shit 17:46 Jordach like they flicker every frame to a new position 17:48 Krock so.. same as before but worse 17:48 Jordach basically it acts like #9794 but even worse 17:48 ShadowBot https://github.com/minetest/minetest/issues/9794 -- set_bone_control broken when using set_animation 17:48 Jordach it's like irrlicht isn't handling long armature chains properly 17:54 Jordach ah fuck 17:55 Jordach i rember a thread on the irrlicht forums mentioning that animated scenenodes aren't meant for entity like objects 17:56 Jordach instead iskinnedmesh 17:58 Jordach oh my god 17:58 Jordach JESUS WHY 17:58 Jordach https://github.com/minetest/minetest/blob/master/src/client/content_cao.cpp#L1383 17:58 Jordach any manually positioned bone breaks the bug fix 18:00 Jordach "the joint returned by this method may not exist" 18:20 Jordach time to have a whack with MSVC 18:29 Jordach right, the problem happens in EUJOR_CONTROL 18:31 Jordach HMM 18:57 Jordach hold the front door 18:57 Jordach wouldn't this patch only fix it for *local animations* ie in third person view 18:57 sfan5 why local? 18:58 sfan5 every object is a ContentCAO 18:58 sfan5 (on the client, that is) 18:58 sfan5 unless you're missing with set_local_animation which is indeed only local 18:58 sfan5 messing* 18:58 Jordach my theory relates to that, if it works with set_bone_rotation with set_local 18:59 Jordach because my model does not use set_local at all 18:59 Jordach the only thing that comes to mind is that the animation has to be *played* after the bone change 18:59 Jordach not before 19:06 Jordach uh...weird 19:10 Jordach ...what 19:10 Jordach sfan5, the model is flipped horizontally 19:12 Jordach holy shit 19:12 Jordach something is flipping the model in eujor_control 19:14 Jordach sfan5, https://i.imgur.com/luvSx3L.png 19:53 shivajiva 'Well it's high noon at the U.K. corral and it's high time I got myself back on the rails. I'm the lonesome cowboy, ridin across the range, with just a hand held radio--to keep me sane, ridin through the f.m. stations, the tumbleweed and the petrol stations. Will all aboard this Yankee station prepare themselves for battle stations' - from the album Infected :) 19:56 Krock OMG it's a Yankee with no brim 19:57 shivajiva 1986 lol 20:04 shivajiva https://youtu.be/Snup4XmKioU 20:23 Jordach it's not the model 20:23 Jordach well it is the model 20:24 Jordach pulled out the offsets and noticed a perculiar thing 20:24 Jordach https://gist.github.com/Jordach/e2a51d0444b40783562c5145ab2e5f71 20:28 Jordach the models bone scale is all positive, so it's not that 20:36 Jordach also, it's not irrlicht mistakenly rotating bones 20:37 Jordach it's that blender rotates the bones in edit mode via the roll param