mvps: improve object move #367
Conversation
mesecons_mvps/init.lua
Outdated
| obj:setpos(np) | ||
| if ok then | ||
| local ent = obj:get_luaentity() | ||
| if not ent or not mesecon.is_mvps_unmov(ent.name) then |
mesecons_mvps/init.lua
Outdated
| local nn = minetest.get_node(np) | ||
| local node_def = minetest.registered_nodes[nn.name] | ||
| if (node_def and not node_def.walkable) or | ||
| math.abs(obj_pos[dir_k]) <= #nodestack - 0.5 + math.abs(pos[dir_k]) then |
There was a problem hiding this comment.
What if signs of pos[dir_k] and obj_pos[dir_k] are different?
mesecons_mvps/init.lua
Outdated
| local edge1, edge2 | ||
| if k ~= dir_k then | ||
| edge1 = v - 0.5 | ||
| edge2 = v + 0.5 |
There was a problem hiding this comment.
Objects lying on top of the nodestack aren't moved anymore. Bad?
You mean, when pushing horizontally? That’s acceptable, but may be fixed by increasing (slightly) constant here, if I understand the code correctly.
mesecons_mvps/init.lua
Outdated
| if edge1 > edge2 then | ||
| edge1, edge2 = edge2, edge1 | ||
| end | ||
| end |
There was a problem hiding this comment.
edges may be pre-calculated
There was a problem hiding this comment.
? This works well enough imo, I can't find a simpler way.
There was a problem hiding this comment.
This exact way, but before for id, obj in pairs(minetest.object_refs) do and not in the inner loop.
| local objects = minetest.get_objects_inside_radius(p_above, 1) | ||
| for _, obj in ipairs(objects) do | ||
| table.insert(objects_to_move, obj) | ||
| for id, obj in pairs(minetest.object_refs) do |
There was a problem hiding this comment.
Seems horribly, but that’s the only way; MT itself does the same, all the time...
|
Comments addressed. |
mesecons_mvps/init.lua
Outdated
| obj:setpos(np) | ||
| if ok then | ||
| local ent = obj:get_luaentity() | ||
| if not (ent and mesecon.is_mvps_unmov(ent.name)) then |
There was a problem hiding this comment.
Does this work at all?
The correct way seems to be if ent and not mesecon.is_mvps_unmov(ent.name) then
There was a problem hiding this comment.
This is correct, don't forget players.
There was a problem hiding this comment.
Oh, sorry.
Then the previous variant was better. But such clever code needs a comment, at least.
And, well, there is is_player.
There was a problem hiding this comment.
So, if obj:is_player() or not mesecon.is_mvps_unmov(obj:get_luaentity().name) then?
mesecons_mvps/init.lua
Outdated
| if edge1 > edge2 then | ||
| edge1, edge2 = edge2, edge1 | ||
| end | ||
| end |
There was a problem hiding this comment.
This exact way, but before for id, obj in pairs(minetest.object_refs) do and not in the inner loop.
mesecons_mvps/init.lua
Outdated
| local np = vector.add(obj_pos, dir) | ||
| -- Move only if destination is not solid or object is inside stack: | ||
| local nn = minetest.get_node(np) | ||
| local node_def = minetest.registered_nodes[nn.name] | ||
| local nodestack_front = (#nodestack - 0.5 + math.abs(pos[dir_k])) | ||
| nodestack_front = nodestack_front * pos[dir_k]/math.abs(pos[dir_k]) |
There was a problem hiding this comment.
Weird math. Can you write something simpler? Remember that you have dir_l.
Or let me do that (a bit later).
Make the check logic obvious. Also, it won't try to move unknown or bogus things, if any.
numberZero
force-pushed
the
mvps_move_obj
branch
from
ef9aef6 to
83f503b
Compare
Objects with big collision boxes are moved.
Objects in stack are moved, not only at front. Ergo fixes #361.
Objects lying on top of the nodestack aren't moved anymore. Bad?Edit: fixed.