| Time |
Nick |
Message |
| 03:52 |
|
esperantoman joined #minetest-dev |
| 03:55 |
|
esperantoman joined #minetest-dev |
| 04:00 |
|
esperantoman joined #minetest-dev |
| 04:02 |
|
esperantoman joined #minetest-dev |
| 08:00 |
|
ShadowNinja joined #minetest-dev |
| 08:22 |
|
twoelk joined #minetest-dev |
| 08:26 |
|
calcul0n joined #minetest-dev |
| 09:34 |
|
Fixer joined #minetest-dev |
| 10:49 |
|
Taoki joined #minetest-dev |
| 10:51 |
|
Fixer_ joined #minetest-dev |
| 13:17 |
|
calcul0n_ joined #minetest-dev |
| 13:22 |
|
Fixer joined #minetest-dev |
| 13:46 |
|
proller joined #minetest-dev |
| 14:00 |
|
lisac joined #minetest-dev |
| 14:18 |
|
proller joined #minetest-dev |
| 15:24 |
MTDiscord |
<appguru> Picked up #9907 again, should be ready for a merge now |
| 15:24 |
ShadowBot |
https://github.com/minetest/minetest/issues/9907 -- Shaders as media files (server-sent shaders) by appgurueu |
| 15:25 |
MTDiscord |
<appguru> Additional features can be addressed in a follow-up PR. |
| 16:21 |
|
hhmhm joined #minetest-dev |
| 17:06 |
|
hhmhm joined #minetest-dev |
| 17:08 |
|
twoelk left #minetest-dev |
| 17:09 |
|
Seirdy joined #minetest-dev |
| 17:27 |
|
pyrollo joined #minetest-dev |
| 17:30 |
|
homthack joined #minetest-dev |
| 17:38 |
|
fluxflux joined #minetest-dev |
| 17:55 |
|
fluxflux joined #minetest-dev |
| 18:06 |
|
NetherEran joined #minetest-dev |
| 18:08 |
pyrollo |
Mmm... is there an undocumented minimap Lua object ? |
| 18:08 |
pyrollo |
Oh, ok it's for CSM I guess |
| 18:17 |
MTDiscord |
<GreenXenith> There will be a minimap HUD object once #9079 is merged |
| 18:17 |
ShadowBot |
https://github.com/minetest/minetest/issues/9079 -- Minimap in HUD, minimap showing texure, control minimap modes from Lua by EvidenceBKidscode |
| 18:19 |
pyrollo |
Yes:) I'm working on it |
| 18:29 |
|
fluxflux joined #minetest-dev |
| 19:14 |
|
v-rob joined #minetest-dev |
| 19:51 |
v-rob |
I'm not happy with the heterogeneous container I have in my formspec refactor so far (DynamicSpec) |
| 19:51 |
v-rob |
To add a new type it can hold, you have to add something to the copy constructor, destructor, and a getter/setter macro |
| 19:52 |
v-rob |
Since we don't have C++17 with std::any (the optimal solution), I've tried to make something similar. |
| 19:52 |
v-rob |
I have little experience with templates, so I'd like someone to look over the code, if possible |
| 20:00 |
sfan5 |
I wouldn't know about a good design either but you can link me the code and I'll look |
| 20:01 |
v-rob |
I mostly want to make sure there are no memory leaks or whatnot |
| 20:01 |
v-rob |
https://gist.github.com/v-rob/935510aa7c02d1a489258815f05dcb4a |
| 20:02 |
Krock |
question is how it should be organized at the end. perhaps you don't need to implement this |
| 20:02 |
Krock |
base class + inherited classes for each element might work just fine as well? |
| 20:05 |
v-rob |
How would that work with dynamic changing of one single property? |
| 20:06 |
Krock |
you'd change it to a getter/setter class that takes/gives a base class, or you'd cast the object to the class you'd need |
| 20:06 |
v-rob |
Either re-change every property of an element when it is modified (not good, breaks interactions) or have a flag for every member for whether it's been changed |
| 20:07 |
Krock |
bool m_is_dirty = true/false doesn't suffice? |
| 20:08 |
v-rob |
Any single property can be changed out of all the properties. |
| 20:08 |
Krock |
going off now. checking tomorrow |
| 20:09 |
v-rob |
I just don't see a way to flag every property cleanly. A heterogeneous container seems the cleanest to me (and avoids making tens of classes) |
| 20:11 |
v-rob |
If there is an easier way than with the template magic that my `Any` is, I'll take it. |
| 20:11 |
v-rob |
At least it's better than the current DynamicSpec in my refactor |
| 20:17 |
sfan5 |
hm that Any class looks pretty complicated |
| 20:20 |
rubenwardy |
I feel like you're overcomplicating this |
| 20:21 |
rubenwardy |
I feel like we should move all the formspec code to src/client/donotopen |
| 20:21 |
rubenwardy |
and start new GUI anew |
| 20:21 |
rubenwardy |
well, with some code taken |
| 20:23 |
v-rob |
I do feel like I'm overcomplicating it |
| 20:25 |
v-rob |
I don't know where I'd start with entirely new code. |
| 20:25 |
v-rob |
If I had a design in front of me, I could probably program it no problem; I'm just not as good at the designing part |
| 20:30 |
v-rob |
At the very least, programming Any was good experience. It didn't even use RTTI which Irrlicht might not have |
| 20:32 |
v-rob |
Well, I'll see about at least familiarizing myself with the code that's above GUIFormSpecMenu |
| 20:40 |
|
appguru joined #minetest-dev |
| 20:46 |
v-rob |
As far as I can tell, GUIEngine is just for the main menu, correct? |
| 20:47 |
rubenwardy |
yeah |
| 20:47 |
rubenwardy |
it's bad design, I believe pyrollo has a redesign for it |
| 20:48 |
v-rob |
I'll leave it alone for now then |
| 20:48 |
v-rob |
Seems all formspec stuff takes place in Game and GameUI |
| 20:55 |
|
mizux joined #minetest-dev |
| 21:03 |
|
pyrollo joined #minetest-dev |
| 21:05 |
v-rob |
I don't think opinions have changed on using JSON as the serialization format, right? |
| 21:07 |
sfan5 |
for the future rewritten formspecs? |
| 21:08 |
v-rob |
Yes, possibly rewritten from the ground up like Rubenwardy wants |
| 21:08 |
rubenwardy |
JSON in the network probably makes sense |
| 21:08 |
rubenwardy |
the API should be a Lua DSL |
| 21:08 |
v-rob |
That would make it easy to serialize as well |
| 21:38 |
|
T4im joined #minetest-dev |
| 21:39 |
MTDiscord |
<appguru> Does everything in Minetest use a left-handed coordinate system? |
| 21:40 |
|
T4im joined #minetest-dev |
| 21:41 |
|
numzero joined #minetest-dev |
| 21:44 |
MTDiscord |
<appguru> set_bone_position? Are position and rotation left-handed? |
| 22:16 |
|
proller joined #minetest-dev |
| 22:30 |
|
Taoki joined #minetest-dev |
| 22:50 |
|
fluxflux joined #minetest-dev |
| 22:54 |
|
homthack joined #minetest-dev |
| 23:35 |
|
homthack joined #minetest-dev |
