update rbx_loader

make invalid states unrepresentable!!!

.cargo/config.toml

@@ -0,0 +1,2 @@
+[registries.strafesnet]
+index = "sparse+https://git.itzana.me/api/packages/strafesnet/cargo/"

.gitignore

@@ -1,2 +1 @@
 /target
-/textures

Cargo.toml

@@ -1,27 +1,36 @@
 [package]
 name = "strafe-client"
-version = "0.8.0"
+version = "0.10.5"
 edition = "2021"
+repository = "https://git.itzana.me/StrafesNET/strafe-client"
+license = "Custom"
+description = "StrafesNET game client for bhop and surf."
+authors = ["Rhys Lloyd <krakow20@gmail.com>"]
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+[features]
+default = ["snf"]
+snf = ["dep:strafesnet_snf"]
+source = ["dep:strafesnet_deferred_loader", "dep:strafesnet_bsp_loader"]
+roblox = ["dep:strafesnet_deferred_loader", "dep:strafesnet_rbx_loader"]
 
 [dependencies]
 bytemuck = { version = "1.13.1", features = ["derive"] }
 configparser = "3.0.2"
 ddsfile = "0.5.1"
-glam = "0.24.1"
-lazy-regex = "3.0.2"
-obj = "0.10.2"
+glam = "0.29.0"
+id = { version = "0.1.0", registry = "strafesnet" }
 parking_lot = "0.12.1"
 pollster = "0.3.0"
-rbx_binary = "0.7.1"
-rbx_dom_weak = "2.5.0"
-rbx_reflection_database = "0.2.7"
-rbx_xml = "0.13.1"
-wgpu = "0.17.0"
-winit = "0.28.6"
+strafesnet_bsp_loader = { version = "0.2.1", registry = "strafesnet", optional = true }
+strafesnet_common = { version = "0.5.2", registry = "strafesnet" }
+strafesnet_deferred_loader = { version = "0.4.0", features = ["legacy"], registry = "strafesnet", optional = true }
+strafesnet_rbx_loader = { version = "0.5.1", registry = "strafesnet", optional = true }
+strafesnet_snf = { version = "0.2.0", registry = "strafesnet", optional = true }
+wgpu = "22.1.0"
+winit = "0.30.5"
 
-#[profile.release]
+[profile.release]
 #lto = true
-#strip = true
-#codegen-units = 1
+strip = true
+codegen-units = 1

LICENSE

@@ -1,5 +1,5 @@
 /*******************************************************
-* Copyright (C) 2023 Rhys Lloyd <krakow20@gmail.com>
+* Copyright (C) 2023-2024 Rhys Lloyd <krakow20@gmail.com>
 *
 * This file is part of the StrafesNET bhop/surf client.
 *

src/aabb.rs

@@ -1,89 +0,0 @@
-use crate::integer::Planar64Vec3;
-
-#[derive(Debug,Clone,Copy,Hash,Eq,PartialEq)]
-pub enum AabbFace{
-	Right,//+X
-	Top,
-	Back,
-	Left,
-	Bottom,
-	Front,
-}
-#[derive(Clone)]
-pub struct Aabb{
-	pub min:Planar64Vec3,
-	pub max:Planar64Vec3,
-}
-
-impl Default for Aabb {
-	fn default()->Self {
-		Self{min:Planar64Vec3::MAX,max:Planar64Vec3::MIN}
-	}
-}
-
-impl Aabb{
-	const VERTEX_DATA:[Planar64Vec3;8]=[
-		Planar64Vec3::int( 1,-1,-1),
-		Planar64Vec3::int( 1, 1,-1),
-		Planar64Vec3::int( 1, 1, 1),
-		Planar64Vec3::int( 1,-1, 1),
-		Planar64Vec3::int(-1,-1, 1),
-		Planar64Vec3::int(-1, 1, 1),
-		Planar64Vec3::int(-1, 1,-1),
-		Planar64Vec3::int(-1,-1,-1),
-	];
-
-	pub fn grow(&mut self,point:Planar64Vec3){
-		self.min=self.min.min(point);
-		self.max=self.max.max(point);
-	}
-	pub fn join(&mut self,aabb:&Aabb){
-		self.min=self.min.min(aabb.min);
-		self.max=self.max.max(aabb.max);
-	}
-	pub fn inflate(&mut self,hs:Planar64Vec3){
-		self.min-=hs;
-		self.max+=hs;
-	}
-	pub fn intersects(&self,aabb:&Aabb)->bool{
-		(self.min.cmplt(aabb.max)&aabb.min.cmplt(self.max)).all()
-	}
-	pub fn normal(face:AabbFace)->Planar64Vec3{
-		match face {
-			AabbFace::Right=>Planar64Vec3::int(1,0,0),
-			AabbFace::Top=>Planar64Vec3::int(0,1,0),
-			AabbFace::Back=>Planar64Vec3::int(0,0,1),
-			AabbFace::Left=>Planar64Vec3::int(-1,0,0),
-			AabbFace::Bottom=>Planar64Vec3::int(0,-1,0),
-			AabbFace::Front=>Planar64Vec3::int(0,0,-1),
-		}
-	}
-	pub fn unit_vertices()->[Planar64Vec3;8] {
-		return Self::VERTEX_DATA;
-	}
-	// pub fn face(&self,face:AabbFace)->Aabb {
-	// 	let mut aabb=self.clone();
-	// 	//in this implementation face = worldspace aabb face
-	// 	match face {
-	// 		AabbFace::Right => aabb.min.x=aabb.max.x,
-	// 		AabbFace::Top => aabb.min.y=aabb.max.y,
-	// 		AabbFace::Back => aabb.min.z=aabb.max.z,
-	// 		AabbFace::Left => aabb.max.x=aabb.min.x,
-	// 		AabbFace::Bottom => aabb.max.y=aabb.min.y,
-	// 		AabbFace::Front => aabb.max.z=aabb.min.z,
-	// 	}
-	// 	return aabb;
-	// }
-	pub fn center(&self)->Planar64Vec3{
-		return self.min.midpoint(self.max)
-	}
-	//probably use floats for area & volume because we don't care about precision
-	// pub fn area_weight(&self)->f32{
-	// 	let d=self.max-self.min;
-	// 	d.x*d.y+d.y*d.z+d.z*d.x
-	// }
-	// pub fn volume(&self)->f32{
-	// 	let d=self.max-self.min;
-	// 	d.x*d.y*d.z
-	// }
-}

src/bvh.rs

@@ -1,107 +0,0 @@
-use crate::aabb::Aabb;
-
-//da algaritum
-//lista boxens
-//sort by {minx,maxx,miny,maxy,minz,maxz} (6 lists)
-//find the sets that minimizes the sum of surface areas
-//splitting is done when the minimum split sum of surface areas is larger than the node's own surface area
-
-//start with bisection into octrees because a bad bvh is still 1000x better than no bvh
-//sort the centerpoints on each axis (3 lists)
-//bv is put into octant based on whether it is upper or lower in each list
-#[derive(Default)]
-pub struct BvhNode{
-	children:Vec<Self>,
-	models:Vec<usize>,
-	aabb:Aabb,
-}
-
-impl BvhNode{
-	pub fn the_tester<F:FnMut(usize)>(&self,aabb:&Aabb,f:&mut F){
-		for &model in &self.models{
-			f(model);
-		}
-		for child in &self.children{
-			if aabb.intersects(&child.aabb){
-				child.the_tester(aabb,f);
-			}
-		}
-	}
-}
-
-pub fn generate_bvh(boxen:Vec<Aabb>)->BvhNode{
-	generate_bvh_node(boxen.into_iter().enumerate().collect())
-}
-
-fn generate_bvh_node(boxen:Vec<(usize,Aabb)>)->BvhNode{
-	let n=boxen.len();
-	if n<20{
-		let mut aabb=Aabb::default();
-		let models=boxen.into_iter().map(|b|{aabb.join(&b.1);b.0}).collect();
-		BvhNode{
-			children:Vec::new(),
-			models,
-			aabb,
-		}
-	}else{
-		let mut octant=std::collections::HashMap::with_capacity(n);//this ids which octant the boxen is put in
-		let mut sort_x=Vec::with_capacity(n);
-		let mut sort_y=Vec::with_capacity(n);
-		let mut sort_z=Vec::with_capacity(n);
-		for (i,aabb) in boxen.iter(){
-			let center=aabb.center();
-			octant.insert(*i,0);
-			sort_x.push((*i,center.x()));
-			sort_y.push((*i,center.y()));
-			sort_z.push((*i,center.z()));
-		}
-		sort_x.sort_by(|tup0,tup1|tup0.1.partial_cmp(&tup1.1).unwrap());
-		sort_y.sort_by(|tup0,tup1|tup0.1.partial_cmp(&tup1.1).unwrap());
-		sort_z.sort_by(|tup0,tup1|tup0.1.partial_cmp(&tup1.1).unwrap());
-		let h=n/2;
-		let median_x=sort_x[h].1;
-		let median_y=sort_y[h].1;
-		let median_z=sort_z[h].1;
-		for (i,c) in sort_x{
-			if median_x<c{
-				octant.insert(i,octant[&i]+1<<0);
-			}
-		}
-		for (i,c) in sort_y{
-			if median_y<c{
-				octant.insert(i,octant[&i]+1<<1);
-			}
-		}
-		for (i,c) in sort_z{
-			if median_z<c{
-				octant.insert(i,octant[&i]+1<<2);
-			}
-		}
-		//generate lists for unique octant values
-		let mut list_list=Vec::with_capacity(8);
-		let mut octant_list=Vec::with_capacity(8);
-		for (i,aabb) in boxen.into_iter(){
-			let octant_id=octant[&i];
-			let list_id=if let Some(list_id)=octant_list.iter().position(|&id|id==octant_id){
-				list_id
-			}else{
-				let list_id=list_list.len();
-				octant_list.push(octant_id);
-				list_list.push(Vec::new());
-				list_id
-			};
-			list_list[list_id].push((i,aabb));
-		}
-		let mut aabb=Aabb::default();
-		let children=list_list.into_iter().map(|b|{
-			let node=generate_bvh_node(b);
-			aabb.join(&node.aabb);
-			node
-		}).collect();
-		BvhNode{
-			children,
-			models:Vec::new(),
-			aabb,
-		}
-	}
-}

src/compat_worker.rs

@@ -0,0 +1,21 @@
+pub type QNWorker<'a,Task>=CompatNWorker<'a,Task>;
+pub type INWorker<'a,Task>=CompatNWorker<'a,Task>;
+
+pub struct CompatNWorker<'a,Task>{
+	data:std::marker::PhantomData<Task>,
+	f:Box<dyn FnMut(Task)+Send+'a>,
+}
+
+impl<'a,Task> CompatNWorker<'a,Task>{
+	pub fn new(f:impl FnMut(Task)+Send+'a)->CompatNWorker<'a,Task>{
+		Self{
+			data:std::marker::PhantomData,
+			f:Box::new(f),
+		}
+	}
+
+	pub fn send(&mut self,task:Task)->Result<(),()>{
+		(self.f)(task);
+		Ok(())
+	}
+}

src/face_crawler.rs

@@ -0,0 +1,127 @@
+use crate::physics::Body;
+use crate::model_physics::{GigaTime,FEV,MeshQuery,DirectedEdge,MinkowskiMesh,MinkowskiFace,MinkowskiDirectedEdge,MinkowskiVert};
+use strafesnet_common::integer::{Time,Fixed,Ratio};
+
+#[derive(Debug)]
+enum Transition<F,E:DirectedEdge,V>{
+	Miss,
+	Next(FEV<F,E,V>,GigaTime),
+	Hit(F,GigaTime),
+}
+
+type MinkowskiFEV=FEV<MinkowskiFace,MinkowskiDirectedEdge,MinkowskiVert>;
+type MinkowskiTransition=Transition<MinkowskiFace,MinkowskiDirectedEdge,MinkowskiVert>;
+
+	fn next_transition(fev:&MinkowskiFEV,body_time:GigaTime,mesh:&MinkowskiMesh,body:&Body,mut best_time:GigaTime)->MinkowskiTransition{
+		//conflicting derivative means it crosses in the wrong direction.
+		//if the transition time is equal to an already tested transition, do not replace the current best.
+		let mut best_transition=MinkowskiTransition::Miss;
+		match fev{
+			&MinkowskiFEV::Face(face_id)=>{
+				//test own face collision time, ignoring roots with zero or conflicting derivative
+				//n=face.normal d=face.dot
+				//n.a t^2+n.v t+n.p-d==0
+				let (n,d)=mesh.face_nd(face_id);
+				//TODO: use higher precision d value?
+				//use the mesh transform translation instead of baking it into the d value.
+				for dt in Fixed::<4,128>::zeroes2((n.dot(body.position)-d)*2,n.dot(body.velocity)*2,n.dot(body.acceleration)){
+					if body_time.le_ratio(dt)&&dt.lt_ratio(best_time)&&n.dot(body.extrapolated_velocity_ratio_dt(dt)).is_negative(){
+						best_time=dt;
+						best_transition=MinkowskiTransition::Hit(face_id,dt);
+						break;
+					}
+				}
+				//test each edge collision time, ignoring roots with zero or conflicting derivative
+				for &directed_edge_id in mesh.face_edges(face_id).iter(){
+					let edge_n=mesh.directed_edge_n(directed_edge_id);
+					let n=n.cross(edge_n);
+					let verts=mesh.edge_verts(directed_edge_id.as_undirected());
+					//WARNING: d is moved out of the *2 block because of adding two vertices!
+					//WARNING: precision is swept under the rug!
+					for dt in Fixed::<4,128>::zeroes2(n.dot(body.position*2-(mesh.vert(verts[0])+mesh.vert(verts[1]))).fix_4(),n.dot(body.velocity).fix_4()*2,n.dot(body.acceleration).fix_4()){
+						if body_time.le_ratio(dt)&&dt.lt_ratio(best_time)&&n.dot(body.extrapolated_velocity_ratio_dt(dt)).is_negative(){
+							best_time=dt;
+							best_transition=MinkowskiTransition::Next(MinkowskiFEV::Edge(directed_edge_id.as_undirected()),dt);
+							break;
+						}
+					}
+				}
+				//if none:
+			},
+			&MinkowskiFEV::Edge(edge_id)=>{
+				//test each face collision time, ignoring roots with zero or conflicting derivative
+				let edge_n=mesh.edge_n(edge_id);
+				let edge_verts=mesh.edge_verts(edge_id);
+				let delta_pos=body.position*2-(mesh.vert(edge_verts[0])+mesh.vert(edge_verts[1]));
+				for (i,&edge_face_id) in mesh.edge_faces(edge_id).iter().enumerate(){
+					let face_n=mesh.face_nd(edge_face_id).0;
+					//edge_n gets parity from the order of edge_faces
+					let n=face_n.cross(edge_n)*((i as i64)*2-1);
+					//WARNING yada yada d *2
+					for dt in Fixed::<4,128>::zeroes2(n.dot(delta_pos).fix_4(),n.dot(body.velocity).fix_4()*2,n.dot(body.acceleration).fix_4()){
+						if body_time.le_ratio(dt)&&dt.lt_ratio(best_time)&&n.dot(body.extrapolated_velocity_ratio_dt(dt)).is_negative(){
+							best_time=dt;
+							best_transition=MinkowskiTransition::Next(MinkowskiFEV::Face(edge_face_id),dt);
+							break;
+						}
+					}
+				}
+				//test each vertex collision time, ignoring roots with zero or conflicting derivative
+				for (i,&vert_id) in edge_verts.iter().enumerate(){
+					//vertex normal gets parity from vert index
+					let n=edge_n*(1-2*(i as i64));
+					for dt in Fixed::<2,64>::zeroes2((n.dot(body.position-mesh.vert(vert_id)))*2,n.dot(body.velocity)*2,n.dot(body.acceleration)){
+						if body_time.le_ratio(dt)&&dt.lt_ratio(best_time)&&n.dot(body.extrapolated_velocity_ratio_dt(dt)).is_negative(){
+							let dt=Ratio::new(dt.num.fix_4(),dt.den.fix_4());
+							best_time=dt;
+							best_transition=MinkowskiTransition::Next(MinkowskiFEV::Vert(vert_id),dt);
+							break;
+						}
+					}
+				}
+				//if none:
+			},
+			&MinkowskiFEV::Vert(vert_id)=>{
+				//test each edge collision time, ignoring roots with zero or conflicting derivative
+				for &directed_edge_id in mesh.vert_edges(vert_id).iter(){
+					//edge is directed away from vertex, but we want the dot product to turn out negative
+					let n=-mesh.directed_edge_n(directed_edge_id);
+					for dt in Fixed::<2,64>::zeroes2((n.dot(body.position-mesh.vert(vert_id)))*2,n.dot(body.velocity)*2,n.dot(body.acceleration)){
+						if body_time.le_ratio(dt)&&dt.lt_ratio(best_time)&&n.dot(body.extrapolated_velocity_ratio_dt(dt)).is_negative(){
+							let dt=Ratio::new(dt.num.fix_4(),dt.den.fix_4());
+							best_time=dt;
+							best_transition=MinkowskiTransition::Next(MinkowskiFEV::Edge(directed_edge_id.as_undirected()),dt);
+							break;
+						}
+					}
+				}
+				//if none:
+			},
+		}
+		best_transition
+	}
+pub enum CrawlResult<F,E:DirectedEdge,V>{
+	Miss(FEV<F,E,V>),
+	Hit(F,GigaTime),
+}
+type MinkowskiCrawlResult=CrawlResult<MinkowskiFace,MinkowskiDirectedEdge,MinkowskiVert>;
+pub fn crawl_fev(mut fev:MinkowskiFEV,mesh:&MinkowskiMesh,relative_body:&Body,start_time:Time,time_limit:Time)->MinkowskiCrawlResult{
+	let mut body_time={
+		let r=(start_time-relative_body.time).to_ratio();
+		Ratio::new(r.num.fix_4(),r.den.fix_4())
+	};
+	let time_limit={
+		let r=(time_limit-relative_body.time).to_ratio();
+		Ratio::new(r.num.fix_4(),r.den.fix_4())
+	};
+	for _ in 0..20{
+		match next_transition(&fev,body_time,mesh,relative_body,time_limit){
+			Transition::Miss=>return CrawlResult::Miss(fev),
+			Transition::Next(next_fev,next_time)=>(fev,body_time)=(next_fev,next_time),
+			Transition::Hit(face,time)=>return CrawlResult::Hit(face,time),
+		}
+	}
+	//TODO: fix all bugs
+	//println!("Too many iterations!  Using default behaviour instead of crashing...");
+	CrawlResult::Miss(fev)
+}

src/file.rs

@@ -0,0 +1,144 @@
+use std::io::Read;
+
+#[derive(Debug)]
+pub enum ReadError{
+	#[cfg(feature="roblox")]
+	Roblox(strafesnet_rbx_loader::ReadError),
+	#[cfg(feature="source")]
+	Source(strafesnet_bsp_loader::ReadError),
+	#[cfg(feature="snf")]
+	StrafesNET(strafesnet_snf::Error),
+	#[cfg(feature="snf")]
+	StrafesNETMap(strafesnet_snf::map::Error),
+	Io(std::io::Error),
+	UnknownFileFormat,
+}
+impl std::fmt::Display for ReadError{
+	fn fmt(&self,f:&mut std::fmt::Formatter<'_>)->std::fmt::Result{
+		write!(f,"{self:?}")
+	}
+}
+impl std::error::Error for ReadError{}
+
+pub enum DataStructure{
+	#[cfg(feature="roblox")]
+	Roblox(strafesnet_rbx_loader::Model),
+	#[cfg(feature="source")]
+	Source(strafesnet_bsp_loader::Bsp),
+	#[cfg(feature="snf")]
+	StrafesNET(strafesnet_common::map::CompleteMap),
+}
+
+pub fn read<R:Read+std::io::Seek>(input:R)->Result<DataStructure,ReadError>{
+	let mut buf=std::io::BufReader::new(input);
+	let peek=std::io::BufRead::fill_buf(&mut buf).map_err(ReadError::Io)?;
+	match &peek[0..4]{
+		#[cfg(feature="roblox")]
+		b"<rob"=>Ok(DataStructure::Roblox(strafesnet_rbx_loader::read(buf).map_err(ReadError::Roblox)?)),
+		#[cfg(feature="source")]
+		b"VBSP"=>Ok(DataStructure::Source(strafesnet_bsp_loader::read(buf).map_err(ReadError::Source)?)),
+		#[cfg(feature="snf")]
+		b"SNFM"=>Ok(DataStructure::StrafesNET(
+			strafesnet_snf::read_map(buf).map_err(ReadError::StrafesNET)?
+			.into_complete_map().map_err(ReadError::StrafesNETMap)?
+		)),
+		_=>Err(ReadError::UnknownFileFormat),
+	}
+}
+
+#[derive(Debug)]
+pub enum LoadError{
+	ReadError(ReadError),
+	File(std::io::Error),
+	Io(std::io::Error),
+}
+impl std::fmt::Display for LoadError{
+	fn fmt(&self,f:&mut std::fmt::Formatter<'_>)->std::fmt::Result{
+		write!(f,"{self:?}")
+	}
+}
+impl std::error::Error for LoadError{}
+
+pub fn load<P:AsRef<std::path::Path>>(path:P)->Result<strafesnet_common::map::CompleteMap,LoadError>{
+	//blocking because it's simpler...
+	let file=std::fs::File::open(path).map_err(LoadError::File)?;
+	match read(file).map_err(LoadError::ReadError)?{
+		#[cfg(feature="snf")]
+		DataStructure::StrafesNET(map)=>Ok(map),
+		#[cfg(feature="roblox")]
+		DataStructure::Roblox(model)=>{
+			let mut place=model.into_place();
+			place.run_scripts();
+
+			let mut loader=strafesnet_deferred_loader::roblox_legacy();
+
+			let (texture_loader,mesh_loader)=loader.get_inner_mut();
+
+			let map_step1=strafesnet_rbx_loader::convert(
+				&place,
+				|name|texture_loader.acquire_render_config_id(name),
+				|name|mesh_loader.acquire_mesh_id(name),
+			);
+
+			let meshpart_meshes=mesh_loader.load_meshes().map_err(LoadError::Io)?;
+
+			let map_step2=map_step1.add_meshpart_meshes_and_calculate_attributes(
+				meshpart_meshes.into_iter().map(|(mesh_id,loader_model)|
+					(mesh_id,strafesnet_rbx_loader::data::RobloxMeshBytes::new(loader_model.get()))
+				)
+			);
+
+			let (textures,render_configs)=loader.into_render_configs().map_err(LoadError::Io)?.consume();
+
+			let map=map_step2.add_render_configs_and_textures(
+				render_configs.into_iter(),
+				textures.into_iter().map(|(texture_id,texture)|
+					(texture_id,match texture{
+						strafesnet_deferred_loader::texture::Texture::ImageDDS(data)=>data,
+					})
+				)
+			);
+
+			Ok(map)
+		},
+		#[cfg(feature="source")]
+		DataStructure::Source(bsp)=>{
+			let mut loader=strafesnet_deferred_loader::source_legacy();
+
+			let (texture_loader,mesh_loader)=loader.get_inner_mut();
+
+			let map_step1=strafesnet_bsp_loader::convert(
+				&bsp,
+				|name|texture_loader.acquire_render_config_id(name),
+				|name|mesh_loader.acquire_mesh_id(name),
+			);
+
+			let prop_meshes=mesh_loader.load_meshes(bsp.as_ref());
+
+			let map_step2=map_step1.add_prop_meshes(
+				//the type conflagulator 9000
+				prop_meshes.into_iter().map(|(mesh_id,loader_model)|
+					(mesh_id,strafesnet_bsp_loader::data::ModelData{
+						mdl:strafesnet_bsp_loader::data::MdlData::new(loader_model.mdl.get()),
+						vtx:strafesnet_bsp_loader::data::VtxData::new(loader_model.vtx.get()),
+						vvd:strafesnet_bsp_loader::data::VvdData::new(loader_model.vvd.get()),
+					})
+				),
+				|name|texture_loader.acquire_render_config_id(name),
+			);
+
+			let (textures,render_configs)=loader.into_render_configs().map_err(LoadError::Io)?.consume();
+
+			let map=map_step2.add_render_configs_and_textures(
+				render_configs.into_iter(),
+				textures.into_iter().map(|(texture_id,texture)|
+					(texture_id,match texture{
+						strafesnet_deferred_loader::texture::Texture::ImageDDS(data)=>data,
+					})
+				),
+			);
+
+			Ok(map)
+		},
+	}
+}

src/graphics_context.rs

@@ -1,315 +0,0 @@
-fn optional_features() -> wgpu::Features {
-	wgpu::Features::empty()
-}
-fn required_features() -> wgpu::Features {
-	wgpu::Features::empty()
-}
-fn required_downlevel_capabilities() -> wgpu::DownlevelCapabilities {
-	wgpu::DownlevelCapabilities {
-		flags: wgpu::DownlevelFlags::empty(),
-		shader_model: wgpu::ShaderModel::Sm5,
-		..wgpu::DownlevelCapabilities::default()
-	}
-}
-pub fn required_limits() -> wgpu::Limits {
-	wgpu::Limits::downlevel_webgl2_defaults() // These downlevel limits will allow the code to run on all possible hardware
-}
-
-struct GraphicsContextPartial1{
-	backends:wgpu::Backends,
-	instance:wgpu::Instance,
-}
-fn create_instance()->GraphicsContextPartial1{
-	let backends=wgpu::util::backend_bits_from_env().unwrap_or_else(wgpu::Backends::all);
-	let dx12_shader_compiler=wgpu::util::dx12_shader_compiler_from_env().unwrap_or_default();
-	GraphicsContextPartial1{
-		backends,
-		instance:wgpu::Instance::new(wgpu::InstanceDescriptor{
-			backends,
-			dx12_shader_compiler,
-		}),
-	}
-}
-impl GraphicsContextPartial1{
-	fn create_surface(self,window:&winit::window::Window)->Result<GraphicsContextPartial2,wgpu::CreateSurfaceError>{
-		Ok(GraphicsContextPartial2{
-			backends:self.backends,
-			instance:self.instance,
-			surface:unsafe{self.instance.create_surface(window)}?
-		})
-	}
-}
-struct GraphicsContextPartial2{
-	backends:wgpu::Backends,
-	instance:wgpu::Instance,
-	surface:wgpu::Surface,
-}
-impl GraphicsContextPartial2{
-	fn pick_adapter(self)->GraphicsContextPartial3{
-		let adapter;
-
-		let optional_features=optional_features();
-		let required_features=required_features();
-
-		//no helper function smh gotta write it myself
-		let adapters=self.instance.enumerate_adapters(self.backends);
-
-		let mut chosen_adapter=None;
-		let mut chosen_adapter_score=0;
-		for adapter in adapters {
-			if !adapter.is_surface_supported(&self.surface) {
-				continue;
-			}
-
-			let score=match adapter.get_info().device_type{
-				wgpu::DeviceType::IntegratedGpu=>3,
-				wgpu::DeviceType::DiscreteGpu=>4,
-				wgpu::DeviceType::VirtualGpu=>2,
-				wgpu::DeviceType::Other|wgpu::DeviceType::Cpu=>1,
-			};
-
-			let adapter_features=adapter.features();
-			if chosen_adapter_score<score&&adapter_features.contains(required_features) {
-				chosen_adapter_score=score;
-				chosen_adapter=Some(adapter);
-			}
-		}
-
-		if let Some(maybe_chosen_adapter)=chosen_adapter{
-			adapter=maybe_chosen_adapter;
-		}else{
-			panic!("No suitable GPU adapters found on the system!");
-		}
-
-
-		let adapter_info=adapter.get_info();
-		println!("Using {} ({:?})", adapter_info.name, adapter_info.backend);
-
-		let required_downlevel_capabilities=required_downlevel_capabilities();
-		let downlevel_capabilities=adapter.get_downlevel_capabilities();
-		assert!(
-			downlevel_capabilities.shader_model >= required_downlevel_capabilities.shader_model,
-			"Adapter does not support the minimum shader model required to run this example: {:?}",
-			required_downlevel_capabilities.shader_model
-		);
-		assert!(
-			downlevel_capabilities
-				.flags
-				.contains(required_downlevel_capabilities.flags),
-			"Adapter does not support the downlevel capabilities required to run this example: {:?}",
-			required_downlevel_capabilities.flags - downlevel_capabilities.flags
-		);
-		GraphicsContextPartial3{
-			instance:self.instance,
-			surface:self.surface,
-			adapter,
-		}
-	}
-}
-struct GraphicsContextPartial3{
-	instance:wgpu::Instance,
-	surface:wgpu::Surface,
-	adapter:wgpu::Adapter,
-}
-impl GraphicsContextPartial3{
-	fn request_device(self)->GraphicsContextPartial4{
-		let optional_features=optional_features();
-		let required_features=required_features();
-
-		// Make sure we use the texture resolution limits from the adapter, so we can support images the size of the surface.
-		let needed_limits=required_limits().using_resolution(self.adapter.limits());
-
-		let trace_dir=std::env::var("WGPU_TRACE");
-		let (device, queue)=pollster::block_on(self.adapter
-			.request_device(
-				&wgpu::DeviceDescriptor {
-					label: None,
-					features: (optional_features & self.adapter.features()) | required_features,
-					limits: needed_limits,
-				},
-				trace_dir.ok().as_ref().map(std::path::Path::new),
-			))
-			.expect("Unable to find a suitable GPU adapter!");
-
-		GraphicsContextPartial4{
-			instance:self.instance,
-			surface:self.surface,
-			adapter:self.adapter,
-			device,
-			queue,
-		}
-	}
-}
-struct GraphicsContextPartial4{
-	instance:wgpu::Instance,
-	surface:wgpu::Surface,
-	adapter:wgpu::Adapter,
-	device:wgpu::Device,
-	queue:wgpu::Queue,
-}
-impl GraphicsContextPartial4{
-	fn configure_surface(self,size:&winit::dpi::PhysicalSize<u32>)->GraphicsContext{
-		let mut config=self.surface
-			.get_default_config(&self.adapter, size.width, size.height)
-			.expect("Surface isn't supported by the adapter.");
-		let surface_view_format=config.format.add_srgb_suffix();
-		config.view_formats.push(surface_view_format);
-		self.surface.configure(&self.device, &config);
-
-		GraphicsContext{
-			instance:self.instance,
-			surface:self.surface,
-			adapter:self.adapter,
-			device:self.device,
-			queue:self.queue,
-			config,
-		}
-	}
-}
-pub struct GraphicsContext{
-	pub instance:wgpu::Instance,
-	pub surface:wgpu::Surface,
-	pub adapter:wgpu::Adapter,
-	pub device:wgpu::Device,
-	pub queue:wgpu::Queue,
-	pub config:wgpu::SurfaceConfiguration,
-}
-
-pub fn setup(title:&str)->GraphicsContextSetup{
-	let event_loop=winit::event_loop::EventLoop::new();
-
-	let window=crate::window::WindowState::create_window(title,&event_loop).unwrap();
-
-	println!("Initializing the surface...");
-
-	let partial_1=create_instance();
-
-	let partial_2=partial_1.create_surface(&window).unwrap();
-
-	let partial_3=partial_2.pick_adapter();
-
-	let partial_4=partial_3.request_device();
-
-	GraphicsContextSetup{
-		window,
-		event_loop,
-		partial_graphics_context:partial_4,
-	}
-}
-
-struct GraphicsContextSetup{
-	window:winit::window::Window,
-	event_loop:winit::event_loop::EventLoop<()>,
-	partial_graphics_context:GraphicsContextPartial4,
-}
-
-impl GraphicsContextSetup{
-	fn into_split(self)->(winit::window::Window,winit::event_loop::EventLoop<()>,GraphicsContext){
-		let size=self.window.inner_size();
-		//Steal values and drop self
-		(
-			self.window,
-			self.event_loop,
-			self.partial_graphics_context.configure_surface(&size),
-		)
-	}
-	pub fn start(self,mut global_state:crate::GlobalState){
-		let (window,event_loop,graphics_context)=self.into_split();
-
-		println!("Entering render loop...");
-		event_loop.run(move |event,_,control_flow|{
-			*control_flow=if cfg!(feature="metal-auto-capture"){
-				winit::event_loop::ControlFlow::Exit
-			}else{
-				winit::event_loop::ControlFlow::Poll
-			};
-			match event{
-				winit::event::Event::RedrawEventsCleared=>{
-					window.request_redraw();
-				}
-				winit::event::Event::WindowEvent {
-					event:
-						winit::event::WindowEvent::Resized(size)
-						| winit::event::WindowEvent::ScaleFactorChanged {
-							new_inner_size:&mut size,
-							..
-						},
-					..
-				} => {
-					// Once winit is fixed, the detection conditions here can be removed.
-					// https://github.com/rust-windowing/winit/issues/2876
-					// this has been fixed if I update winit (remove the if statement and only use the else case)
-					//drop adapter when you delete this
-					let max_dimension=graphics_context.adapter.limits().max_texture_dimension_2d;
-					if max_dimension<size.width||max_dimension<size.height{
-						println!(
-							"The resizing size {:?} exceeds the limit of {}.",
-							size,
-							max_dimension
-						);
-					}else{
-						println!("Resizing to {:?}",size);
-						graphics_context.config.width=size.width.max(1);
-						graphics_context.config.height=size.height.max(1);
-						window.resize(&graphics_context);
-						graphics_context.surface.configure(&graphics_context.device,&graphics_context.config);
-					}
-				}
-				winit::event::Event::WindowEvent{event,..}=>match event{
-					winit::event::WindowEvent::KeyboardInput{
-						input:
-							winit::event::KeyboardInput{
-								virtual_keycode:Some(winit::event::VirtualKeyCode::Escape),
-								state: winit::event::ElementState::Pressed,
-								..
-							},
-						..
-					}
-					|winit::event::WindowEvent::CloseRequested=>{
-						*control_flow=winit::event_loop::ControlFlow::Exit;
-					}
-					winit::event::WindowEvent::KeyboardInput{
-						input:
-							winit::event::KeyboardInput{
-								virtual_keycode:Some(winit::event::VirtualKeyCode::Scroll),
-								state: winit::event::ElementState::Pressed,
-								..
-							},
-						..
-					}=>{
-						println!("{:#?}",graphics_context.instance.generate_report());
-					}
-					_=>{
-						global_state.update(event);
-					}
-				},
-				winit::event::Event::DeviceEvent{
-					event,
-					..
-				} => {
-					global_state.device_event(event);
-				},
-				winit::event::Event::RedrawRequested(_)=>{
-					let frame=match graphics_context.surface.get_current_texture(){
-						Ok(frame)=>frame,
-						Err(_)=>{
-							graphics_context.surface.configure(&graphics_context.device,&graphics_context.config);
-							graphics_context.surface
-								.get_current_texture()
-								.expect("Failed to acquire next surface texture!")
-						}
-					};
-					let view=frame.texture.create_view(&wgpu::TextureViewDescriptor{
-						format:Some(graphics_context.config.view_formats[0]),
-						..wgpu::TextureViewDescriptor::default()
-					});
-
-					graphics.render(&view,&graphics_context.device,&graphics_context.queue);
-
-					frame.present();
-				}
-				_=>{}
-			}
-		});
-	}
-}

src/graphics_worker.rs

@@ -0,0 +1,65 @@
+pub enum Instruction{
+	Render(crate::graphics::FrameState),
+	//UpdateModel(crate::graphics::GraphicsModelUpdate),
+	Resize(winit::dpi::PhysicalSize<u32>,crate::settings::UserSettings),
+	ChangeMap(strafesnet_common::map::CompleteMap),
+}
+
+//Ideally the graphics thread worker description is:
+/*
+WorkerDescription{
+	input:Immediate,
+	output:Realtime(PoolOrdering::Ordered(3)),
+}
+*/
+//up to three frames in flight, dropping new frame requests when all three are busy, and dropping output frames when one renders out of order
+
+pub fn new<'a>(
+	mut graphics:crate::graphics::GraphicsState,
+	mut config:wgpu::SurfaceConfiguration,
+	surface:wgpu::Surface<'a>,
+	device:wgpu::Device,
+	queue:wgpu::Queue,
+)->crate::compat_worker::INWorker<'a,Instruction>{
+	let mut resize=None;
+	crate::compat_worker::INWorker::new(move |ins:Instruction|{
+		match ins{
+			Instruction::ChangeMap(map)=>{
+				graphics.clear();
+				graphics.generate_models(&device,&queue,&map);
+			},
+			Instruction::Resize(size,user_settings)=>{
+				resize=Some((size,user_settings));
+			}
+			Instruction::Render(frame_state)=>{
+				if let Some((size,user_settings))=resize.take(){
+					println!("Resizing to {:?}",size);
+					let t0=std::time::Instant::now();
+					config.width=size.width.max(1);
+					config.height=size.height.max(1);
+					surface.configure(&device,&config);
+					graphics.resize(&device,&config,&user_settings);
+					println!("Resize took {:?}",t0.elapsed());
+				}
+				//this has to go deeper somehow
+				let frame=match surface.get_current_texture(){
+					Ok(frame)=>frame,
+					Err(_)=>{
+						surface.configure(&device,&config);
+						surface
+							.get_current_texture()
+							.expect("Failed to acquire next surface texture!")
+					}
+				};
+				let view=frame.texture.create_view(&wgpu::TextureViewDescriptor{
+					format:Some(config.view_formats[0]),
+					..wgpu::TextureViewDescriptor::default()
+				});
+
+				graphics.render(&view,&device,&queue,frame_state);
+
+				frame.present();
+			}
+		}
+	})
+}

src/instruction.rs

@@ -1,50 +0,0 @@
-use crate::integer::Time;
-
-#[derive(Debug)]
-pub struct TimedInstruction<I>{
-	pub time:Time,
-	pub instruction:I,
-}
-
-pub trait InstructionEmitter<I>{
-	fn next_instruction(&self,time_limit:Time)->Option<TimedInstruction<I>>;
-}
-pub trait InstructionConsumer<I>{
-	fn process_instruction(&mut self, instruction:TimedInstruction<I>);
-}
-
-//PROPER PRIVATE FIELDS!!!
-pub struct InstructionCollector<I>{
-	time:Time,
-	instruction:Option<I>,
-}
-impl<I> InstructionCollector<I>{
-	pub fn new(time:Time)->Self{
-		Self{
-			time,
-			instruction:None
-		}
-	}
-
-	pub fn collect(&mut self,instruction:Option<TimedInstruction<I>>){
-		match instruction{
-			Some(unwrap_instruction)=>{
-				if unwrap_instruction.time<self.time {
-					self.time=unwrap_instruction.time;
-					self.instruction=Some(unwrap_instruction.instruction);
-				}
-			},
-			None=>(),
-		}
-	}
-	pub fn instruction(self)->Option<TimedInstruction<I>>{
-		//STEAL INSTRUCTION AND DESTROY INSTRUCTIONCOLLECTOR
-		match self.instruction{
-			Some(instruction)=>Some(TimedInstruction{
-				time:self.time,
-				instruction
-			}),
-			None=>None,
-		}
-	}
-}

src/integer.rs

@@ -1,953 +0,0 @@
-//integer units
-#[derive(Clone,Copy,Hash,PartialEq,PartialOrd,Debug)]
-pub struct Time(i64);
-impl Time{
-	pub const ZERO:Self=Self(0);
-	pub const ONE_SECOND:Self=Self(1_000_000_000);
-	pub const ONE_MILLISECOND:Self=Self(1_000_000);
-	pub const ONE_MICROSECOND:Self=Self(1_000);
-	pub const ONE_NANOSECOND:Self=Self(1);
-	#[inline]
-	pub fn from_secs(num:i64)->Self{
-		Self(Self::ONE_SECOND.0*num)
-	}
-	#[inline]
-	pub fn from_millis(num:i64)->Self{
-		Self(Self::ONE_MILLISECOND.0*num)
-	}
-	#[inline]
-	pub fn from_micros(num:i64)->Self{
-		Self(Self::ONE_MICROSECOND.0*num)
-	}
-	#[inline]
-	pub fn from_nanos(num:i64)->Self{
-		Self(Self::ONE_NANOSECOND.0*num)
-	}
-	//should I have checked subtraction? force all time variables to be positive?
-	#[inline]
-	pub fn nanos(&self)->i64{
-		self.0
-	}
-}
-impl From<Planar64> for Time{
-	#[inline]
-	fn from(value:Planar64)->Self{
-		Time((((value.0 as i128)*1_000_000_000)>>32) as i64)
-	}
-}
-impl std::fmt::Display for Time{
-	#[inline]
-	fn fmt(&self,f:&mut std::fmt::Formatter<'_>)->std::fmt::Result{
-		write!(f,"{}s+{:09}ns",self.0/Self::ONE_SECOND.0,self.0%Self::ONE_SECOND.0)
-	}
-}
-impl std::ops::Neg for Time{
-	type Output=Time;
-	#[inline]
-	fn neg(self)->Self::Output {
-		Time(-self.0)
-	}
-}
-impl std::ops::Add<Time> for Time{
-	type Output=Time;
-	#[inline]
-	fn add(self,rhs:Self)->Self::Output {
-		Time(self.0+rhs.0)
-	}
-}
-impl std::ops::Sub<Time> for Time{
-	type Output=Time;
-	#[inline]
-	fn sub(self,rhs:Self)->Self::Output {
-		Time(self.0-rhs.0)
-	}
-}
-impl std::ops::Mul<Time> for Time{
-	type Output=Time;
-	#[inline]
-	fn mul(self,rhs:Time)->Self::Output{
-		Self((((self.0 as i128)*(rhs.0 as i128))/1_000_000_000) as i64)
-	}
-}
-impl std::ops::Div<i64> for Time{
-	type Output=Time;
-	#[inline]
-	fn div(self,rhs:i64)->Self::Output {
-		Time(self.0/rhs)
-	}
-}
-
-#[inline]
-const fn gcd(mut a:u64,mut b:u64)->u64{
-	while b!=0{
-		(a,b)=(b,a.rem_euclid(b));
-	};
-	a
-}
-#[derive(Clone,Hash)]
-pub struct Ratio64{
-	num:i64,
-	den:u64,
-}
-impl Ratio64{
-	pub const ZERO:Self=Ratio64{num:0,den:1};
-	pub const ONE:Self=Ratio64{num:1,den:1};
-	#[inline]
-	pub const fn new(num:i64,den:u64)->Option<Ratio64>{
-		if den==0{
-			None
-		}else{
-			let d=gcd(num.unsigned_abs(),den);
-			Some(Self{num:num/d as i64,den:den/d})
-		}
-	}
-	#[inline]
-	pub fn mul_int(&self,rhs:i64)->i64{
-		rhs*self.num/self.den as i64
-	}
-	#[inline]
-	pub fn rhs_div_int(&self,rhs:i64)->i64{
-		rhs*self.den as i64/self.num
-	}
-	#[inline]
-	pub fn mul_ref(&self,rhs:&Ratio64)->Ratio64{
-		let (num,den)=(self.num*rhs.num,self.den*rhs.den);
-		let d=gcd(num.unsigned_abs(),den);
-		Self{
-			num:num/d as i64,
-			den:den/d,
-		}
-	}
-}
-//from num_traits crate
-#[inline]
-fn integer_decode_f32(f: f32) -> (u64, i16, i8) {
-	let bits: u32 = f.to_bits();
-	let sign: i8 = if bits >> 31 == 0 { 1 } else { -1 };
-	let mut exponent: i16 = ((bits >> 23) & 0xff) as i16;
-	let mantissa = if exponent == 0 {
-		(bits & 0x7fffff) << 1
-	} else {
-		(bits & 0x7fffff) | 0x800000
-	};
-	// Exponent bias + mantissa shift
-	exponent -= 127 + 23;
-	(mantissa as u64, exponent, sign)
-}
-#[inline]
-fn integer_decode_f64(f: f64) -> (u64, i16, i8) {
-	let bits: u64 = f.to_bits();
-	let sign: i8 = if bits >> 63 == 0 { 1 } else { -1 };
-	let mut exponent: i16 = ((bits >> 52) & 0x7ff) as i16;
-	let mantissa = if exponent == 0 {
-		(bits & 0xfffffffffffff) << 1
-	} else {
-		(bits & 0xfffffffffffff) | 0x10000000000000
-	};
-	// Exponent bias + mantissa shift
-	exponent -= 1023 + 52;
-	(mantissa, exponent, sign)
-}
-#[derive(Debug)]
-pub enum Ratio64TryFromFloatError{
-	Nan,
-	Infinite,
-	Subnormal,
-	HighlyNegativeExponent(i16),
-	HighlyPositiveExponent(i16),
-}
-const MAX_DENOMINATOR:u128=u64::MAX as u128;
-#[inline]
-fn ratio64_from_mes((m,e,s):(u64,i16,i8))->Result<Ratio64,Ratio64TryFromFloatError>{
-	if e< -127{
-		//this can also just be zero
-		Err(Ratio64TryFromFloatError::HighlyNegativeExponent(e))
-	}else if e< -63{
-		//approximate input ratio within denominator limit
-		let mut target_num=m as u128;
-		let mut target_den=1u128<<-e;
-
-		let mut num=1;
-		let mut den=0;
-		let mut prev_num=0;
-		let mut prev_den=1;
-
-		while target_den!=0{
-			let whole=target_num/target_den;
-			(target_num,target_den)=(target_den,target_num-whole*target_den);
-			let new_num=whole*num+prev_num;
-			let new_den=whole*den+prev_den;
-			if MAX_DENOMINATOR<new_den{
-				break;
-			}else{
-				(prev_num,prev_den)=(num,den);
-				(num,den)=(new_num,new_den);
-			}
-		}
-
-		Ok(Ratio64::new(num as i64,den as u64).unwrap())
-	}else if e<0{
-		Ok(Ratio64::new((m as i64)*(s as i64),1<<-e).unwrap())
-	}else if (64-m.leading_zeros() as i16)+e<64{
-		Ok(Ratio64::new((m as i64)*(s as i64)*(1<<e),1).unwrap())
-	}else{
-		Err(Ratio64TryFromFloatError::HighlyPositiveExponent(e))
-	}
-}
-impl TryFrom<f32> for Ratio64{
-	type Error=Ratio64TryFromFloatError;
-	#[inline]
-	fn try_from(value:f32)->Result<Self,Self::Error>{
-		match value.classify(){
-			std::num::FpCategory::Nan=>Err(Self::Error::Nan),
-			std::num::FpCategory::Infinite=>Err(Self::Error::Infinite),
-			std::num::FpCategory::Zero=>Ok(Self::ZERO),
-			std::num::FpCategory::Subnormal=>Err(Self::Error::Subnormal),
-			std::num::FpCategory::Normal=>ratio64_from_mes(integer_decode_f32(value)),
-		}
-	}
-}
-impl TryFrom<f64> for Ratio64{
-	type Error=Ratio64TryFromFloatError;
-	#[inline]
-	fn try_from(value:f64)->Result<Self,Self::Error>{
-		match value.classify(){
-			std::num::FpCategory::Nan=>Err(Self::Error::Nan),
-			std::num::FpCategory::Infinite=>Err(Self::Error::Infinite),
-			std::num::FpCategory::Zero=>Ok(Self::ZERO),
-			std::num::FpCategory::Subnormal=>Err(Self::Error::Subnormal),
-			std::num::FpCategory::Normal=>ratio64_from_mes(integer_decode_f64(value)),
-		}
-	}
-}
-impl std::ops::Mul<Ratio64> for Ratio64{
-	type Output=Ratio64;
-	#[inline]
-	fn mul(self,rhs:Ratio64)->Self::Output{
-		let (num,den)=(self.num*rhs.num,self.den*rhs.den);
-		let d=gcd(num.unsigned_abs(),den);
-		Self{
-			num:num/d as i64,
-			den:den/d,
-		}
-	}
-}
-impl std::ops::Mul<i64> for Ratio64{
-	type Output=Ratio64;
-	#[inline]
-	fn mul(self,rhs:i64)->Self::Output {
-		Self{
-			num:self.num*rhs,
-			den:self.den,
-		}
-	}
-}
-impl std::ops::Div<u64> for Ratio64{
-	type Output=Ratio64;
-	#[inline]
-	fn div(self,rhs:u64)->Self::Output {
-		Self{
-			num:self.num,
-			den:self.den*rhs,
-		}
-	}
-}
-#[derive(Clone,Hash)]
-pub struct Ratio64Vec2{
-	pub x:Ratio64,
-	pub y:Ratio64,
-}
-impl Ratio64Vec2{
-	pub const ONE:Self=Self{x:Ratio64::ONE,y:Ratio64::ONE};
-	#[inline]
-	pub fn new(x:Ratio64,y:Ratio64)->Self{
-		Self{x,y}
-	}
-	#[inline]
-	pub fn mul_int(&self,rhs:glam::I64Vec2)->glam::I64Vec2{
-		glam::i64vec2(
-			self.x.mul_int(rhs.x),
-			self.y.mul_int(rhs.y),
-		)
-	}
-}
-impl std::ops::Mul<i64> for Ratio64Vec2{
-	type Output=Ratio64Vec2;
-	#[inline]
-	fn mul(self,rhs:i64)->Self::Output {
-		Self{
-			x:self.x*rhs,
-			y:self.y*rhs,
-		}
-	}
-}
-
-///[-pi,pi) = [-2^31,2^31-1]
-#[derive(Clone,Copy,Hash)]
-pub struct Angle32(i32);
-impl Angle32{
-	pub const FRAC_PI_2:Self=Self(1<<30);
-	pub const PI:Self=Self(-1<<31);
-	#[inline]
-	pub fn wrap_from_i64(theta:i64)->Self{
-		//take lower bits
-		//note: this was checked on compiler explorer and compiles to 1 instruction!
-		Self(i32::from_ne_bytes(((theta&((1<<32)-1)) as u32).to_ne_bytes()))
-	}
-	#[inline]
-	pub fn clamp_from_i64(theta:i64)->Self{
-		//the assembly is a bit confusing for this, I thought it was checking the same thing twice
-		//but it's just checking and then overwriting the value for both upper and lower bounds.
-		Self(theta.clamp(i32::MIN as i64,i32::MAX as i64) as i32)
-	}
-	#[inline]
-	pub fn get(&self)->i32{
-		self.0
-	}
-	/// Clamps the value towards the midpoint of the range.
-	/// Note that theta_min can be larger than theta_max and it will wrap clamp the other way around
-	#[inline]
-	pub fn clamp(&self,theta_min:Self,theta_max:Self)->Self{
-		//((max-min as u32)/2 as i32)+min
-		let midpoint=((
-			u32::from_ne_bytes(theta_max.0.to_ne_bytes())
-			.wrapping_sub(u32::from_ne_bytes(theta_min.0.to_ne_bytes()))
-			/2
-		) as i32)//(u32::MAX/2) as i32 ALWAYS works
-		.wrapping_add(theta_min.0);
-		//(theta-mid).clamp(max-mid,min-mid)+mid
-		Self(
-			self.0.wrapping_sub(midpoint)
-			.max(theta_min.0.wrapping_sub(midpoint))
-			.min(theta_max.0.wrapping_sub(midpoint))
-			.wrapping_add(midpoint)
-		)
-	}
-	/*
-	#[inline]
-	pub fn cos(&self)->Unit32{
-		//TODO: fix this rounding towards 0
-		Unit32(unsafe{((self.0 as f64*ANGLE32_TO_FLOAT64_RADIANS).cos()*UNIT32_ONE_FLOAT64).to_int_unchecked()})
-	}
-	#[inline]
-	pub fn sin(&self)->Unit32{
-		//TODO: fix this rounding towards 0
-		Unit32(unsafe{((self.0 as f64*ANGLE32_TO_FLOAT64_RADIANS).sin()*UNIT32_ONE_FLOAT64).to_int_unchecked()})
-	}
-	*/
-}
-const ANGLE32_TO_FLOAT64_RADIANS:f64=std::f64::consts::PI/((1i64<<31) as f64);
-impl Into<f32> for Angle32{
-	#[inline]
-	fn into(self)->f32{
-		(self.0 as f64*ANGLE32_TO_FLOAT64_RADIANS) as f32
-	}
-}
-impl std::ops::Neg for Angle32{
-	type Output=Angle32;
-	#[inline]
-	fn neg(self)->Self::Output{
-		Angle32(self.0.wrapping_neg())
-	}
-}
-impl std::ops::Add<Angle32> for Angle32{
-	type Output=Angle32;
-	#[inline]
-	fn add(self,rhs:Self)->Self::Output {
-		Angle32(self.0.wrapping_add(rhs.0))
-	}
-}
-impl std::ops::Sub<Angle32> for Angle32{
-	type Output=Angle32;
-	#[inline]
-	fn sub(self,rhs:Self)->Self::Output {
-		Angle32(self.0.wrapping_sub(rhs.0))
-	}
-}
-impl std::ops::Mul<i32> for Angle32{
-	type Output=Angle32;
-	#[inline]
-	fn mul(self,rhs:i32)->Self::Output {
-		Angle32(self.0.wrapping_mul(rhs))
-	}
-}
-impl std::ops::Mul<Angle32> for Angle32{
-	type Output=Angle32;
-	#[inline]
-	fn mul(self,rhs:Self)->Self::Output {
-		Angle32(self.0.wrapping_mul(rhs.0))
-	}
-}
-
-/* Unit type unused for now, may revive it for map files
-///[-1.0,1.0] = [-2^30,2^30]
-pub struct Unit32(i32);
-impl Unit32{
-	#[inline]
-	pub fn as_planar64(&self) -> Planar64{
-		Planar64(4*(self.0 as i64))
-	}
-}
-const UNIT32_ONE_FLOAT64=((1<<30) as f64);
-///[-1.0,1.0] = [-2^30,2^30]
-pub struct Unit32Vec3(glam::IVec3);
-impl TryFrom<[f32;3]> for Unit32Vec3{
-	type Error=Unit32TryFromFloatError;
-	fn try_from(value:[f32;3])->Result<Self,Self::Error>{
-		Ok(Self(glam::ivec3(
-			Unit32::try_from(Planar64::try_from(value[0])?)?.0,
-			Unit32::try_from(Planar64::try_from(value[1])?)?.0,
-			Unit32::try_from(Planar64::try_from(value[2])?)?.0,
-		)))
-	}
-}
-*/
-
-///[-1.0,1.0] = [-2^32,2^32]
-#[derive(Clone,Copy,Hash,Eq,Ord,PartialEq,PartialOrd)]
-pub struct Planar64(i64);
-impl Planar64{
-	pub const ZERO:Self=Self(0);
-	pub const ONE:Self=Self(1<<32);
-	#[inline]
-	pub const fn int(num:i32)->Self{
-		Self(Self::ONE.0*num as i64)
-	}
-	#[inline]
-	pub const fn raw(num:i64)->Self{
-		Self(num)
-	}
-	#[inline]
-	pub const fn get(&self)->i64{
-		self.0
-	}
-	pub fn sqrt(&self)->Self{
-		Planar64(unsafe{(((self.0 as i128)<<32) as f64).sqrt().to_int_unchecked()})
-	}
-}
-const PLANAR64_ONE_FLOAT32:f32=(1u64<<32) as f32;
-const PLANAR64_CONVERT_TO_FLOAT32:f32=1.0/PLANAR64_ONE_FLOAT32;
-const PLANAR64_ONE_FLOAT64:f64=(1u64<<32) as f64;
-impl Into<f32> for Planar64{
-	#[inline]
-	fn into(self)->f32{
-		self.0 as f32*PLANAR64_CONVERT_TO_FLOAT32
-	}
-}
-impl From<Ratio64> for Planar64{
-	#[inline]
-	fn from(ratio:Ratio64)->Self{
-		Self((((ratio.num as i128)<<32)/ratio.den as i128) as i64)
-	}
-}
-#[derive(Debug)]
-pub enum Planar64TryFromFloatError{
-	Nan,
-	Infinite,
-	Subnormal,
-	HighlyNegativeExponent(i16),
-	HighlyPositiveExponent(i16),
-}
-#[inline]
-fn planar64_from_mes((m,e,s):(u64,i16,i8))->Result<Planar64,Planar64TryFromFloatError>{
-	let e32=e+32;
-	if e32<0&&(m>>-e32)==0{//shifting m will underflow to 0
-		Ok(Planar64::ZERO)
-		// println!("m{} e{} s{}",m,e,s);
-		// println!("f={}",(m as f64)*(2.0f64.powf(e as f64))*(s as f64));
-		// Err(Planar64TryFromFloatError::HighlyNegativeExponent(e))
-	}else if (64-m.leading_zeros() as i16)+e32<64{//shifting m will not overflow
-		if e32<0{
-			Ok(Planar64((m as i64)*(s as i64)>>-e32))
-		}else{
-			Ok(Planar64((m as i64)*(s as i64)<<e32))
-		}
-	}else{//if shifting m will overflow (prev check failed)
-		Err(Planar64TryFromFloatError::HighlyPositiveExponent(e))
-	}
-}
-impl TryFrom<f32> for Planar64{
-	type Error=Planar64TryFromFloatError;
-	#[inline]
-	fn try_from(value:f32)->Result<Self,Self::Error>{
-		match value.classify(){
-			std::num::FpCategory::Nan=>Err(Self::Error::Nan),
-			std::num::FpCategory::Infinite=>Err(Self::Error::Infinite),
-			std::num::FpCategory::Zero=>Ok(Self::ZERO),
-			std::num::FpCategory::Subnormal=>Err(Self::Error::Subnormal),
-			std::num::FpCategory::Normal=>planar64_from_mes(integer_decode_f32(value)),
-		}
-	}
-}
-impl TryFrom<f64> for Planar64{
-	type Error=Planar64TryFromFloatError;
-	#[inline]
-	fn try_from(value:f64)->Result<Self,Self::Error>{
-		match value.classify(){
-			std::num::FpCategory::Nan=>Err(Self::Error::Nan),
-			std::num::FpCategory::Infinite=>Err(Self::Error::Infinite),
-			std::num::FpCategory::Zero=>Ok(Self::ZERO),
-			std::num::FpCategory::Subnormal=>Err(Self::Error::Subnormal),
-			std::num::FpCategory::Normal=>planar64_from_mes(integer_decode_f64(value)),
-		}
-	}
-}
-impl std::fmt::Display for Planar64{
-	fn fmt(&self,f:&mut std::fmt::Formatter<'_>)->std::fmt::Result{
-		write!(f,"{:.3}",
-			Into::<f32>::into(*self),
-		)
-	}
-}
-impl std::ops::Neg for Planar64{
-	type Output=Planar64;
-	#[inline]
-	fn neg(self)->Self::Output{
-		Planar64(-self.0)
-	}
-}
-impl std::ops::Add<Planar64> for Planar64{
-	type Output=Planar64;
-	#[inline]
-	fn add(self, rhs: Self) -> Self::Output {
-		Planar64(self.0+rhs.0)
-	}
-}
-impl std::ops::Sub<Planar64> for Planar64{
-	type Output=Planar64;
-	#[inline]
-	fn sub(self, rhs: Self) -> Self::Output {
-		Planar64(self.0-rhs.0)
-	}
-}
-impl std::ops::Mul<i64> for Planar64{
-	type Output=Planar64;
-	#[inline]
-	fn mul(self, rhs: i64) -> Self::Output {
-		Planar64(self.0*rhs)
-	}
-}
-impl std::ops::Mul<Planar64> for Planar64{
-	type Output=Planar64;
-	#[inline]
-	fn mul(self, rhs: Self) -> Self::Output {
-		Planar64(((self.0 as i128*rhs.0 as i128)>>32) as i64)
-	}
-}
-impl std::ops::Mul<Time> for Planar64{
-	type Output=Planar64;
-	#[inline]
-	fn mul(self,rhs:Time)->Self::Output{
-		Planar64(((self.0 as i128*rhs.0 as i128)/1_000_000_000) as i64)
-	}
-}
-impl std::ops::Div<i64> for Planar64{
-	type Output=Planar64;
-	#[inline]
-	fn div(self, rhs: i64) -> Self::Output {
-		Planar64(self.0/rhs)
-	}
-}
-impl std::ops::Div<Planar64> for Planar64{
-	type Output=Planar64;
-	#[inline]
-	fn div(self, rhs: Planar64) -> Self::Output {
-		Planar64((((self.0 as i128)<<32)/rhs.0 as i128) as i64)
-	}
-}
-// impl PartialOrd<i64> for Planar64{
-// 	fn partial_cmp(&self, other: &i64) -> Option<std::cmp::Ordering> {
-// 		self.0.partial_cmp(other)
-// 	}
-// }
-
-
-///[-1.0,1.0] = [-2^32,2^32]
-#[derive(Clone,Copy,Default,Hash,Eq,PartialEq)]
-pub struct Planar64Vec3(glam::I64Vec3);
-impl Planar64Vec3{
-	pub const ZERO:Self=Planar64Vec3(glam::I64Vec3::ZERO);
-	pub const ONE:Self=Self::int(1,1,1);
-	pub const X:Self=Self::int(1,0,0);
-	pub const Y:Self=Self::int(0,1,0);
-	pub const Z:Self=Self::int(0,0,1);
-	pub const NEG_X:Self=Self::int(-1,0,0);
-	pub const NEG_Y:Self=Self::int(0,-1,0);
-	pub const NEG_Z:Self=Self::int(0,0,-1);
-	pub const MIN:Self=Planar64Vec3(glam::I64Vec3::MIN);
-	pub const MAX:Self=Planar64Vec3(glam::I64Vec3::MAX);
-	#[inline]
-	pub const fn int(x:i32,y:i32,z:i32)->Self{
-		Self(glam::i64vec3((x as i64)<<32,(y as i64)<<32,(z as i64)<<32))
-	}
-	#[inline]
-	pub const fn raw(x:i64,y:i64,z:i64)->Self{
-		Self(glam::i64vec3(x,y,z))
-	}
-	#[inline]
-	pub fn x(&self)->Planar64{
-		Planar64(self.0.x)
-	}
-	#[inline]
-	pub fn y(&self)->Planar64{
-		Planar64(self.0.y)
-	}
-	#[inline]
-	pub fn z(&self)->Planar64{
-		Planar64(self.0.z)
-	}
-	#[inline]
-	pub fn min(&self,rhs:Self)->Self{
-		Self(glam::i64vec3(
-			self.0.x.min(rhs.0.x),
-			self.0.y.min(rhs.0.y),
-			self.0.z.min(rhs.0.z),
-		))
-	}
-	#[inline]
-	pub fn max(&self,rhs:Self)->Self{
-		Self(glam::i64vec3(
-			self.0.x.max(rhs.0.x),
-			self.0.y.max(rhs.0.y),
-			self.0.z.max(rhs.0.z),
-		))
-	}
-	#[inline]
-	pub fn midpoint(&self,rhs:Self)->Self{
-		Self((self.0+rhs.0)/2)
-	}
-	#[inline]
-	pub fn cmplt(&self,rhs:Self)->glam::BVec3{
-		self.0.cmplt(rhs.0)
-	}
-	#[inline]
-	pub fn dot(&self,rhs:Self)->Planar64{
-		Planar64(((
-			(self.0.x as i128)*(rhs.0.x as i128)+
-			(self.0.y as i128)*(rhs.0.y as i128)+
-			(self.0.z as i128)*(rhs.0.z as i128)
-		)>>32) as i64)
-	}
-	#[inline]
-	pub fn length(&self)->Planar64{
-		let radicand=(self.0.x as i128)*(self.0.x as i128)+(self.0.y as i128)*(self.0.y as i128)+(self.0.z as i128)*(self.0.z as i128);
-		Planar64(unsafe{(radicand as f64).sqrt().to_int_unchecked()})
-	}
-	#[inline]
-	pub fn with_length(&self,length:Planar64)->Self{
-		let radicand=(self.0.x as i128)*(self.0.x as i128)+(self.0.y as i128)*(self.0.y as i128)+(self.0.z as i128)*(self.0.z as i128);
-		let self_length:i128=unsafe{(radicand as f64).sqrt().to_int_unchecked()};
-		//self.0*length/self_length
-		Planar64Vec3(
-			glam::i64vec3(
-				((self.0.x as i128)*(length.0 as i128)/self_length) as i64,
-				((self.0.y as i128)*(length.0 as i128)/self_length) as i64,
-				((self.0.z as i128)*(length.0 as i128)/self_length) as i64,
-			)
-		)
-	}
-}
-impl Into<glam::Vec3> for Planar64Vec3{
-	#[inline]
-	fn into(self)->glam::Vec3{
-		glam::vec3(
-			self.0.x as f32,
-			self.0.y as f32,
-			self.0.z as f32,
-		)*PLANAR64_CONVERT_TO_FLOAT32
-	}
-}
-impl TryFrom<[f32;3]> for Planar64Vec3{
-	type Error=Planar64TryFromFloatError;
-	#[inline]
-	fn try_from(value:[f32;3])->Result<Self,Self::Error>{
-		Ok(Self(glam::i64vec3(
-			Planar64::try_from(value[0])?.0,
-			Planar64::try_from(value[1])?.0,
-			Planar64::try_from(value[2])?.0,
-		)))
-	}
-}
-impl TryFrom<glam::Vec3A> for Planar64Vec3{
-	type Error=Planar64TryFromFloatError;
-	#[inline]
-	fn try_from(value:glam::Vec3A)->Result<Self,Self::Error>{
-		Ok(Self(glam::i64vec3(
-			Planar64::try_from(value.x)?.0,
-			Planar64::try_from(value.y)?.0,
-			Planar64::try_from(value.z)?.0,
-		)))
-	}
-}
-impl std::fmt::Display for Planar64Vec3{
-	fn fmt(&self,f:&mut std::fmt::Formatter<'_>)->std::fmt::Result{
-		write!(f,"{:.3},{:.3},{:.3}",
-			Into::<f32>::into(self.x()),Into::<f32>::into(self.y()),Into::<f32>::into(self.z()),
-		)
-	}
-}
-impl std::ops::Neg for Planar64Vec3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn neg(self)->Self::Output{
-		Planar64Vec3(-self.0)
-	}
-}
-impl std::ops::Add<Planar64Vec3> for Planar64Vec3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn add(self,rhs:Planar64Vec3) -> Self::Output {
-		Planar64Vec3(self.0+rhs.0)
-	}
-}
-impl std::ops::AddAssign<Planar64Vec3> for Planar64Vec3{
-	#[inline]
-	fn add_assign(&mut self,rhs:Planar64Vec3){
-		*self=*self+rhs
-	}
-}
-impl std::ops::Sub<Planar64Vec3> for Planar64Vec3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn sub(self,rhs:Planar64Vec3) -> Self::Output {
-		Planar64Vec3(self.0-rhs.0)
-	}
-}
-impl std::ops::SubAssign<Planar64Vec3> for Planar64Vec3{
-	#[inline]
-	fn sub_assign(&mut self,rhs:Planar64Vec3){
-		*self=*self-rhs
-	}
-}
-impl std::ops::Mul<Planar64Vec3> for Planar64Vec3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn mul(self, rhs: Planar64Vec3) -> Self::Output {
-		Planar64Vec3(glam::i64vec3(
-			(((self.0.x as i128)*(rhs.0.x as i128))>>32) as i64,
-			(((self.0.y as i128)*(rhs.0.y as i128))>>32) as i64,
-			(((self.0.z as i128)*(rhs.0.z as i128))>>32) as i64
-		))
-	}
-}
-impl std::ops::Mul<Planar64> for Planar64Vec3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn mul(self, rhs: Planar64) -> Self::Output {
-		Planar64Vec3(glam::i64vec3(
-			(((self.0.x as i128)*(rhs.0 as i128))>>32) as i64,
-			(((self.0.y as i128)*(rhs.0 as i128))>>32) as i64,
-			(((self.0.z as i128)*(rhs.0 as i128))>>32) as i64
-		))
-	}
-}
-impl std::ops::Mul<i64> for Planar64Vec3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn mul(self,rhs:i64)->Self::Output {
-		Planar64Vec3(glam::i64vec3(
-			self.0.x*rhs,
-			self.0.y*rhs,
-			self.0.z*rhs
-		))
-	}
-}
-impl std::ops::Mul<Time> for Planar64Vec3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn mul(self,rhs:Time)->Self::Output{
-		Planar64Vec3(glam::i64vec3(
-			(((self.0.x as i128)*(rhs.0 as i128))/1_000_000_000) as i64,
-			(((self.0.y as i128)*(rhs.0 as i128))/1_000_000_000) as i64,
-			(((self.0.z as i128)*(rhs.0 as i128))/1_000_000_000) as i64
-		))
-	}
-}
-impl std::ops::Div<i64> for Planar64Vec3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn div(self,rhs:i64)->Self::Output{
-		Planar64Vec3(glam::i64vec3(
-			self.0.x/rhs,
-			self.0.y/rhs,
-			self.0.z/rhs,
-		))
-	}
-}
-
-///[-1.0,1.0] = [-2^32,2^32]
-#[derive(Clone,Copy)]
-pub struct Planar64Mat3{
-	x_axis:Planar64Vec3,
-	y_axis:Planar64Vec3,
-	z_axis:Planar64Vec3,
-}
-impl Default for Planar64Mat3{
-	#[inline]
-	fn default() -> Self {
-		Self{
-			x_axis:Planar64Vec3::X,
-			y_axis:Planar64Vec3::Y,
-			z_axis:Planar64Vec3::Z,
-		}
-	}
-}
-impl std::ops::Mul<Planar64Vec3> for Planar64Mat3{
-	type Output=Planar64Vec3;
-	#[inline]
-	fn mul(self,rhs:Planar64Vec3) -> Self::Output {
-		self.x_axis*rhs.x()
-		+self.y_axis*rhs.y()
-		+self.z_axis*rhs.z()
-	}
-}
-
-impl Planar64Mat3{
-	#[inline]
-	pub fn from_cols(x_axis:Planar64Vec3,y_axis:Planar64Vec3,z_axis:Planar64Vec3)->Self{
-		Self{
-			x_axis,
-			y_axis,
-			z_axis,
-		}
-	}
-	pub const fn int_from_cols_array(array:[i32;9])->Self{
-		Self{
-			x_axis:Planar64Vec3::int(array[0],array[1],array[2]),
-			y_axis:Planar64Vec3::int(array[3],array[4],array[5]),
-			z_axis:Planar64Vec3::int(array[6],array[7],array[8]),
-		}
-	}
-	#[inline]
-	pub fn from_rotation_yx(yaw:Angle32,pitch:Angle32)->Self{
-		let xtheta=yaw.0 as f64*ANGLE32_TO_FLOAT64_RADIANS;
-		let (xs,xc)=xtheta.sin_cos();
-		let (xc,xs)=(xc*PLANAR64_ONE_FLOAT64,xs*PLANAR64_ONE_FLOAT64);
-		let ytheta=pitch.0 as f64*ANGLE32_TO_FLOAT64_RADIANS;
-		let (ys,yc)=ytheta.sin_cos();
-		let (yc,ys)=(yc*PLANAR64_ONE_FLOAT64,ys*PLANAR64_ONE_FLOAT64);
-		//TODO: fix this rounding towards 0
-		let (xc,xs):(i64,i64)=(unsafe{xc.to_int_unchecked()},unsafe{xs.to_int_unchecked()});
-		let (yc,ys):(i64,i64)=(unsafe{yc.to_int_unchecked()},unsafe{ys.to_int_unchecked()});
-		Self::from_cols(
-			Planar64Vec3(glam::i64vec3(xc,0,-xs)),
-			Planar64Vec3(glam::i64vec3(((xs as i128*ys as i128)>>32) as i64,yc,((xc as i128*ys as i128)>>32) as i64)),
-			Planar64Vec3(glam::i64vec3(((xs as i128*yc as i128)>>32) as i64,-ys,((xc as i128*yc as i128)>>32) as i64)),
-		)
-	}
-	#[inline]
-	pub fn from_rotation_y(angle:Angle32)->Self{
-		let theta=angle.0 as f64*ANGLE32_TO_FLOAT64_RADIANS;
-		let (s,c)=theta.sin_cos();
-		let (c,s)=(c*PLANAR64_ONE_FLOAT64,s*PLANAR64_ONE_FLOAT64);
-		//TODO: fix this rounding towards 0
-		let (c,s):(i64,i64)=(unsafe{c.to_int_unchecked()},unsafe{s.to_int_unchecked()});
-		Self::from_cols(
-			Planar64Vec3(glam::i64vec3(c,0,-s)),
-			Planar64Vec3::Y,
-			Planar64Vec3(glam::i64vec3(s,0,c)),
-		)
-	}
-}
-impl Into<glam::Mat3> for Planar64Mat3{
-	#[inline]
-	fn into(self)->glam::Mat3{
-		glam::Mat3::from_cols(
-			self.x_axis.into(),
-			self.y_axis.into(),
-			self.z_axis.into(),
-		)
-	}
-}
-impl TryFrom<glam::Mat3A> for Planar64Mat3{
-	type Error=Planar64TryFromFloatError;
-	#[inline]
-	fn try_from(value:glam::Mat3A)->Result<Self,Self::Error>{
-		Ok(Self{
-			x_axis:Planar64Vec3::try_from(value.x_axis)?,
-			y_axis:Planar64Vec3::try_from(value.y_axis)?,
-			z_axis:Planar64Vec3::try_from(value.z_axis)?,
-		})
-	}
-}
-impl std::fmt::Display for Planar64Mat3{
-	fn fmt(&self,f:&mut std::fmt::Formatter<'_>)->std::fmt::Result{
-		write!(f,"\n{:.3},{:.3},{:.3}\n{:.3},{:.3},{:.3}\n{:.3},{:.3},{:.3}",
-			Into::<f32>::into(self.x_axis.x()),Into::<f32>::into(self.x_axis.y()),Into::<f32>::into(self.x_axis.z()),
-			Into::<f32>::into(self.y_axis.x()),Into::<f32>::into(self.y_axis.y()),Into::<f32>::into(self.y_axis.z()),
-			Into::<f32>::into(self.z_axis.x()),Into::<f32>::into(self.z_axis.y()),Into::<f32>::into(self.z_axis.z()),
-		)
-	}
-}
-impl std::ops::Div<i64> for Planar64Mat3{
-	type Output=Planar64Mat3;
-	#[inline]
-	fn div(self,rhs:i64)->Self::Output{
-		Planar64Mat3{
-			x_axis:self.x_axis/rhs,
-			y_axis:self.y_axis/rhs,
-			z_axis:self.z_axis/rhs,
-		}
-	}
-}
-
-///[-1.0,1.0] = [-2^32,2^32]
-#[derive(Clone,Copy,Default)]
-pub struct Planar64Affine3{
-	pub matrix3:Planar64Mat3,//includes scale above 1
-	pub translation:Planar64Vec3,
-}
-
-impl Planar64Affine3{
-	#[inline]
-	pub fn new(matrix3:Planar64Mat3,translation:Planar64Vec3)->Self{
-		Self{matrix3,translation}
-	}
-	#[inline]
-	pub fn transform_point3(&self,point:Planar64Vec3) -> Planar64Vec3{
-		Planar64Vec3(
-			self.translation.0
-			+(self.matrix3.x_axis*point.x()).0
-			+(self.matrix3.y_axis*point.y()).0
-			+(self.matrix3.z_axis*point.z()).0
-		)
-	}
-}
-impl Into<glam::Mat4> for Planar64Affine3{
-	#[inline]
-	fn into(self)->glam::Mat4{
-		glam::Mat4::from_cols_array(&[
-			self.matrix3.x_axis.0.x as f32,self.matrix3.x_axis.0.y as f32,self.matrix3.x_axis.0.z as f32,0.0,
-			self.matrix3.y_axis.0.x as f32,self.matrix3.y_axis.0.y as f32,self.matrix3.y_axis.0.z as f32,0.0,
-			self.matrix3.z_axis.0.x as f32,self.matrix3.z_axis.0.y as f32,self.matrix3.z_axis.0.z as f32,0.0,
-			self.translation.0.x as f32,self.translation.0.y as f32,self.translation.0.z as f32,PLANAR64_ONE_FLOAT32
-		])*PLANAR64_CONVERT_TO_FLOAT32
-	}
-}
-impl TryFrom<glam::Affine3A> for Planar64Affine3{
-	type Error=Planar64TryFromFloatError;
-	fn try_from(value: glam::Affine3A)->Result<Self, Self::Error> {
-		Ok(Self{
-			matrix3:Planar64Mat3::try_from(value.matrix3)?,
-			translation:Planar64Vec3::try_from(value.translation)?
-		})
-	}
-}
-impl std::fmt::Display for Planar64Affine3{
-	fn fmt(&self,f:&mut std::fmt::Formatter<'_>)->std::fmt::Result{
-		write!(f,"translation: {:.3},{:.3},{:.3}\nmatrix3:\n{:.3},{:.3},{:.3}\n{:.3},{:.3},{:.3}\n{:.3},{:.3},{:.3}",
-			Into::<f32>::into(self.translation.x()),Into::<f32>::into(self.translation.y()),Into::<f32>::into(self.translation.z()),
-			Into::<f32>::into(self.matrix3.x_axis.x()),Into::<f32>::into(self.matrix3.x_axis.y()),Into::<f32>::into(self.matrix3.x_axis.z()),
-			Into::<f32>::into(self.matrix3.y_axis.x()),Into::<f32>::into(self.matrix3.y_axis.y()),Into::<f32>::into(self.matrix3.y_axis.z()),
-			Into::<f32>::into(self.matrix3.z_axis.x()),Into::<f32>::into(self.matrix3.z_axis.y()),Into::<f32>::into(self.matrix3.z_axis.z()),
-		)
-	}
-}
-
-#[test]
-fn test_sqrt(){
-	let r=Planar64::int(400);
-	println!("r{}",r.get());
-	let s=r.sqrt();
-	println!("s{}",s.get());
-}

src/load_roblox.rs

@@ -1,504 +0,0 @@
-use crate::primitives;
-use crate::integer::{Planar64,Planar64Vec3,Planar64Mat3,Planar64Affine3};
-
-fn class_is_a(class: &str, superclass: &str) -> bool {
-	if class==superclass {
-		return true
-	}
-	let class_descriptor=rbx_reflection_database::get().classes.get(class);
-	if let Some(descriptor) = &class_descriptor {
-		if let Some(class_super) = &descriptor.superclass {
-			return class_is_a(&class_super, superclass)
-		}
-	}
-	return false
-}
-fn recursive_collect_superclass(objects: &mut std::vec::Vec<rbx_dom_weak::types::Ref>,dom: &rbx_dom_weak::WeakDom, instance: &rbx_dom_weak::Instance, superclass: &str){
-	for &referent in instance.children() {
-		if let Some(c) = dom.get_by_ref(referent) {
-			if class_is_a(c.class.as_str(), superclass) {
-				objects.push(c.referent());//copy ref
-			}
-			recursive_collect_superclass(objects,dom,c,superclass);
-		}
-	}
-}
-fn get_texture_refs(dom:&rbx_dom_weak::WeakDom) -> Vec<rbx_dom_weak::types::Ref>{
-	let mut objects = std::vec::Vec::new();
-	recursive_collect_superclass(&mut objects, dom, dom.root(),"Decal");
-	//get ids
-	//clear vec
-	//next class
-	objects
-}
-fn planar64_affine3_from_roblox(cf:&rbx_dom_weak::types::CFrame,size:&rbx_dom_weak::types::Vector3)->Planar64Affine3{
-	Planar64Affine3::new(
-		Planar64Mat3::from_cols(
-			Planar64Vec3::try_from([cf.orientation.x.x,cf.orientation.y.x,cf.orientation.z.x]).unwrap()
-			*Planar64::try_from(size.x/2.0).unwrap(),
-			Planar64Vec3::try_from([cf.orientation.x.y,cf.orientation.y.y,cf.orientation.z.y]).unwrap()
-			*Planar64::try_from(size.y/2.0).unwrap(),
-			Planar64Vec3::try_from([cf.orientation.x.z,cf.orientation.y.z,cf.orientation.z.z]).unwrap()
-			*Planar64::try_from(size.z/2.0).unwrap(),
-		),
-		Planar64Vec3::try_from([cf.position.x,cf.position.y,cf.position.z]).unwrap()
-	)
-}
-fn get_attributes(name:&str,can_collide:bool,velocity:Planar64Vec3,force_intersecting:bool)->crate::model::CollisionAttributes{
-	let mut general=crate::model::GameMechanicAttributes::default();
-	let mut intersecting=crate::model::IntersectingAttributes::default();
-	let mut contacting=crate::model::ContactingAttributes::default();
-	let mut force_can_collide=can_collide;
-	match name{
-		"Water"=>{
-			force_can_collide=false;
-			//TODO: read stupid CustomPhysicalProperties
-			intersecting.water=Some(crate::model::IntersectingWater{density:Planar64::ONE,viscosity:Planar64::ONE/10,current:velocity});
-		},
-		"Accelerator"=>{
-			//although the new game supports collidable accelerators, this is a roblox compatability map loader
-			force_can_collide=false;
-			general.accelerator=Some(crate::model::GameMechanicAccelerator{acceleration:velocity});
-		},
-		"SetVelocity"=>general.trajectory=Some(crate::model::GameMechanicSetTrajectory::Velocity(velocity)),
-		"MapFinish"=>{force_can_collide=false;general.zone=Some(crate::model::GameMechanicZone{mode_id:0,behaviour:crate::model::ZoneBehaviour::Finish})},
-		"MapAnticheat"=>{force_can_collide=false;general.zone=Some(crate::model::GameMechanicZone{mode_id:0,behaviour:crate::model::ZoneBehaviour::Anitcheat})},
-		"Platform"=>general.teleport_behaviour=Some(crate::model::TeleportBehaviour::StageElement(crate::model::GameMechanicStageElement{
-			mode_id:0,
-			stage_id:0,
-			force:false,
-			behaviour:crate::model::StageElementBehaviour::Platform,
-		})),
-		other=>{
-			if let Some(captures)=lazy_regex::regex!(r"^(Force)?(Spawn|SpawnAt|Trigger|Teleport|Platform)(\d+)$")
-			.captures(other){
-				general.teleport_behaviour=Some(crate::model::TeleportBehaviour::StageElement(crate::model::GameMechanicStageElement{
-					mode_id:0,
-					stage_id:captures[3].parse::<u32>().unwrap(),
-					force:match captures.get(1){
-						Some(m)=>m.as_str()=="Force",
-						None=>false,
-					},
-					behaviour:match &captures[2]{
-						"Spawn"|"SpawnAt"=>crate::model::StageElementBehaviour::SpawnAt,
-						//cancollide false so you don't hit the side
-						//NOT a decoration
-						"Trigger"=>{force_can_collide=false;crate::model::StageElementBehaviour::Trigger},
-						"Teleport"=>{force_can_collide=false;crate::model::StageElementBehaviour::Teleport},
-						"Platform"=>crate::model::StageElementBehaviour::Platform,
-						_=>panic!("regex1[2] messed up bad"),
-					}
-				}));
-			}else if let Some(captures)=lazy_regex::regex!(r"^(Force)?(Jump)(\d+)$")
-			.captures(other){
-				general.teleport_behaviour=Some(crate::model::TeleportBehaviour::StageElement(crate::model::GameMechanicStageElement{
-					mode_id:0,
-					stage_id:0,
-					force:match captures.get(1){
-						Some(m)=>m.as_str()=="Force",
-						None=>false,
-					},
-					behaviour:match &captures[2]{
-						"Jump"=>crate::model::StageElementBehaviour::JumpLimit(captures[3].parse::<u32>().unwrap()),
-						_=>panic!("regex4[1] messed up bad"),
-					}
-				}));
-			}else if let Some(captures)=lazy_regex::regex!(r"^Bonus(Finish|Anticheat)(\d+)$")
-			.captures(other){
-				force_can_collide=false;
-				match &captures[1]{
-					"Finish"=>general.zone=Some(crate::model::GameMechanicZone{mode_id:captures[2].parse::<u32>().unwrap(),behaviour:crate::model::ZoneBehaviour::Finish}),
-					"Anticheat"=>general.zone=Some(crate::model::GameMechanicZone{mode_id:captures[2].parse::<u32>().unwrap(),behaviour:crate::model::ZoneBehaviour::Anitcheat}),
-					_=>panic!("regex2[1] messed up bad"),
-				}
-			}else if let Some(captures)=lazy_regex::regex!(r"^(WormholeIn)(\d+)$")
-			.captures(other){
-				force_can_collide=false;
-				match &captures[1]{
-					"WormholeIn"=>general.teleport_behaviour=Some(crate::model::TeleportBehaviour::Wormhole(crate::model::GameMechanicWormhole{destination_model_id:captures[2].parse::<u32>().unwrap()})),
-					_=>panic!("regex3[1] messed up bad"),
-				}
-			}
-		}
-	}
-	//need some way to skip this
-	if velocity!=Planar64Vec3::ZERO{
-		general.booster=Some(crate::model::GameMechanicBooster::Velocity(velocity));
-	}
-	match force_can_collide{
-		true=>{
-			match name{
-				"Bounce"=>contacting.contact_behaviour=Some(crate::model::ContactingBehaviour::Elastic(u32::MAX)),
-				"Surf"=>contacting.contact_behaviour=Some(crate::model::ContactingBehaviour::Surf),
-				"Ladder"=>contacting.contact_behaviour=Some(crate::model::ContactingBehaviour::Ladder(crate::model::ContactingLadder{sticky:true})),
-				_=>(),
-			}
-			crate::model::CollisionAttributes::Contact{contacting,general}
-		},
-		false=>if force_intersecting
-		||general.any()
-		||intersecting.any()
-		{
-			crate::model::CollisionAttributes::Intersect{intersecting,general}
-		}else{
-			crate::model::CollisionAttributes::Decoration
-		},
-	}
-}
-
-struct RobloxAssetId(u64);
-struct RobloxAssetIdParseErr;
-impl std::str::FromStr for RobloxAssetId {
-	type Err=RobloxAssetIdParseErr;
-	fn from_str(s: &str) -> Result<Self, Self::Err>{
-		let regman=lazy_regex::regex!(r"(\d+)$");
-		if let Some(captures) = regman.captures(s) {
-			if captures.len()==2{//captures[0] is all captures concatenated, and then each individual capture
-				if let Ok(id) = captures[0].parse::<u64>() {
-					return Ok(Self(id));
-				}
-			}
-		}
-		Err(RobloxAssetIdParseErr)
-	}
-}
-#[derive(Clone,Copy,PartialEq)]
-struct RobloxTextureTransform{
-	offset_u:f32,
-	offset_v:f32,
-	scale_u:f32,
-	scale_v:f32,
-}
-impl std::cmp::Eq for RobloxTextureTransform{}//????
-impl std::default::Default for RobloxTextureTransform{
-    fn default() -> Self {
-        Self{offset_u:0.0,offset_v:0.0,scale_u:1.0,scale_v:1.0}
-    }
-}
-impl std::hash::Hash for RobloxTextureTransform {
-	fn hash<H: std::hash::Hasher>(&self, state: &mut H) {
-		self.offset_u.to_ne_bytes().hash(state);
-		self.offset_v.to_ne_bytes().hash(state);
-		self.scale_u.to_ne_bytes().hash(state);
-		self.scale_v.to_ne_bytes().hash(state);
-	}
-}
-#[derive(Clone,PartialEq)]
-struct RobloxFaceTextureDescription{
-	texture:u32,
-	color:glam::Vec4,
-	transform:RobloxTextureTransform,
-}
-impl std::cmp::Eq for RobloxFaceTextureDescription{}//????
-impl std::hash::Hash for RobloxFaceTextureDescription {
-	fn hash<H: std::hash::Hasher>(&self, state: &mut H) {
-		self.texture.hash(state);
-		self.transform.hash(state);
-        for &el in self.color.as_ref().iter() {
-            el.to_ne_bytes().hash(state);
-        }
-    }
-}
-impl RobloxFaceTextureDescription{
-	fn to_face_description(&self)->primitives::FaceDescription{
-		primitives::FaceDescription{
-			texture:Some(self.texture),
-			transform:glam::Affine2::from_translation(
-				glam::vec2(self.transform.offset_u,self.transform.offset_v)
-			)
-			*glam::Affine2::from_scale(
-				glam::vec2(self.transform.scale_u,self.transform.scale_v)
-			),
-			color:self.color,
-		}
-	}
-}
-type RobloxPartDescription=[Option<RobloxFaceTextureDescription>;6];
-type RobloxWedgeDescription=[Option<RobloxFaceTextureDescription>;5];
-type RobloxCornerWedgeDescription=[Option<RobloxFaceTextureDescription>;5];
-#[derive(Clone,Eq,Hash,PartialEq)]
-enum RobloxBasePartDescription{
-	Sphere,
-	Part(RobloxPartDescription),
-	Cylinder,
-	Wedge(RobloxWedgeDescription),
-	CornerWedge(RobloxCornerWedgeDescription),
-}
-pub fn generate_indexed_models(dom:rbx_dom_weak::WeakDom) -> crate::model::IndexedModelInstances{
-	//IndexedModelInstances includes textures
-	let mut spawn_point=Planar64Vec3::ZERO;
-
-	let mut indexed_models=Vec::new();
-	let mut model_id_from_description=std::collections::HashMap::<RobloxBasePartDescription,usize>::new();
-
-	let mut texture_id_from_asset_id=std::collections::HashMap::<u64,u32>::new();
-	let mut asset_id_from_texture_id=Vec::new();
-
-	let mut object_refs=Vec::new();
-	let mut temp_objects=Vec::new();
-	recursive_collect_superclass(&mut object_refs, &dom, dom.root(),"BasePart");
-	for object_ref in object_refs {
-		if let Some(object)=dom.get_by_ref(object_ref){
-			if let (
-					Some(rbx_dom_weak::types::Variant::CFrame(cf)),
-					Some(rbx_dom_weak::types::Variant::Vector3(size)),
-					Some(rbx_dom_weak::types::Variant::Vector3(velocity)),
-					Some(rbx_dom_weak::types::Variant::Float32(transparency)),
-					Some(rbx_dom_weak::types::Variant::Color3uint8(color3)),
-					Some(rbx_dom_weak::types::Variant::Bool(can_collide)),
-				) = (
-					object.properties.get("CFrame"),
-					object.properties.get("Size"),
-					object.properties.get("Velocity"),
-					object.properties.get("Transparency"),
-					object.properties.get("Color"),
-					object.properties.get("CanCollide"),
-				)
-			{
-				let model_transform=planar64_affine3_from_roblox(cf,size);
-
-				//push TempIndexedAttributes
-				let mut force_intersecting=false;
-				let mut temp_indexing_attributes=Vec::new();
-				if let Some(attr)=match &object.name[..]{
-					"MapStart"=>{
-						spawn_point=model_transform.transform_point3(Planar64Vec3::ZERO)+Planar64Vec3::Y*5/2;
-						Some(crate::model::TempIndexedAttributes::Start(crate::model::TempAttrStart{mode_id:0}))
-					},
-					"UnorderedCheckpoint"=>Some(crate::model::TempIndexedAttributes::UnorderedCheckpoint(crate::model::TempAttrUnorderedCheckpoint{mode_id:0})),
-					other=>{
-						let regman=lazy_regex::regex!(r"^(BonusStart|Spawn|ForceSpawn|OrderedCheckpoint|WormholeOut)(\d+)$");
-						if let Some(captures) = regman.captures(other) {
-							match &captures[1]{
-								"BonusStart"=>Some(crate::model::TempIndexedAttributes::Start(crate::model::TempAttrStart{mode_id:captures[2].parse::<u32>().unwrap()})),
-								"Spawn"|"ForceSpawn"=>Some(crate::model::TempIndexedAttributes::Spawn(crate::model::TempAttrSpawn{mode_id:0,stage_id:captures[2].parse::<u32>().unwrap()})),
-								"OrderedCheckpoint"=>Some(crate::model::TempIndexedAttributes::OrderedCheckpoint(crate::model::TempAttrOrderedCheckpoint{mode_id:0,checkpoint_id:captures[2].parse::<u32>().unwrap()})),
-								"WormholeOut"=>Some(crate::model::TempIndexedAttributes::Wormhole(crate::model::TempAttrWormhole{wormhole_id:captures[2].parse::<u32>().unwrap()})),
-								_=>None,
-							}
-						}else{
-							None
-						}
-					}
-				}{
-					force_intersecting=true;
-					temp_indexing_attributes.push(attr);
-				}
-
-				//TODO: also detect "CylinderMesh" etc here
-				let shape=match &object.class[..]{
-					"Part"=>{
-						if let Some(rbx_dom_weak::types::Variant::Enum(shape))=object.properties.get("Shape"){
-							match shape.to_u32(){
-								0=>primitives::Primitives::Sphere,
-								1=>primitives::Primitives::Cube,
-								2=>primitives::Primitives::Cylinder,
-								3=>primitives::Primitives::Wedge,
-								4=>primitives::Primitives::CornerWedge,
-								_=>panic!("Funky roblox PartType={};",shape.to_u32()),
-							}
-						}else{
-							panic!("Part has no Shape!");
-						}
-					},
-					"WedgePart"=>primitives::Primitives::Wedge,
-					"CornerWedgePart"=>primitives::Primitives::CornerWedge,
-					_=>{
-						println!("Unsupported BasePart ClassName={}; defaulting to cube",object.class);
-						primitives::Primitives::Cube
-					}
-				};
-
-				//use the biggest one and cut it down later...
-				let mut part_texture_description:RobloxPartDescription=[None,None,None,None,None,None];
-				temp_objects.clear();
-				recursive_collect_superclass(&mut temp_objects, &dom, object,"Decal");
-				for &decal_ref in &temp_objects{
-					if let Some(decal)=dom.get_by_ref(decal_ref){
-						if let (
-							Some(rbx_dom_weak::types::Variant::Content(content)),
-							Some(rbx_dom_weak::types::Variant::Enum(normalid)),
-							Some(rbx_dom_weak::types::Variant::Color3(decal_color3)),
-							Some(rbx_dom_weak::types::Variant::Float32(decal_transparency)),
-						) = (
-							decal.properties.get("Texture"),
-							decal.properties.get("Face"),
-							decal.properties.get("Color3"),
-							decal.properties.get("Transparency"),
-						) {
-							if let Ok(asset_id)=content.clone().into_string().parse::<RobloxAssetId>(){
-								let texture_id=if let Some(&texture_id)=texture_id_from_asset_id.get(&asset_id.0){
-									texture_id
-								}else{
-									let texture_id=asset_id_from_texture_id.len() as u32;
-									texture_id_from_asset_id.insert(asset_id.0,texture_id);
-									asset_id_from_texture_id.push(asset_id.0);
-									texture_id
-								};
-								let normal_id=normalid.to_u32();
-								if normal_id<6{
-									let (roblox_texture_color,roblox_texture_transform)=if decal.class=="Texture"{
-										//generate tranform
-										if let (
-												Some(rbx_dom_weak::types::Variant::Float32(ox)),
-												Some(rbx_dom_weak::types::Variant::Float32(oy)),
-												Some(rbx_dom_weak::types::Variant::Float32(sx)),
-												Some(rbx_dom_weak::types::Variant::Float32(sy)),
-											) = (
-												decal.properties.get("OffsetStudsU"),
-												decal.properties.get("OffsetStudsV"),
-												decal.properties.get("StudsPerTileU"),
-												decal.properties.get("StudsPerTileV"),
-											)
-										{
-											let (size_u,size_v)=match normal_id{
-												0=>(size.z,size.y),//right
-												1=>(size.x,size.z),//top
-												2=>(size.x,size.y),//back
-												3=>(size.z,size.y),//left
-												4=>(size.x,size.z),//bottom
-												5=>(size.x,size.y),//front
-												_=>panic!("unreachable"),
-											};
-											(
-												glam::vec4(decal_color3.r,decal_color3.g,decal_color3.b,1.0-*decal_transparency),
-												RobloxTextureTransform{
-													offset_u:*ox/(*sx),offset_v:*oy/(*sy),
-													scale_u:size_u/(*sx),scale_v:size_v/(*sy),
-												}
-											)
-										}else{
-											(glam::Vec4::ONE,RobloxTextureTransform::default())
-										}
-									}else{
-										(glam::Vec4::ONE,RobloxTextureTransform::default())
-									};
-									part_texture_description[normal_id as usize]=Some(RobloxFaceTextureDescription{
-										texture:texture_id,
-										color:roblox_texture_color,
-										transform:roblox_texture_transform,
-									});
-								}else{
-									println!("NormalId={} unsupported for shape={:?}",normal_id,shape);
-								}
-							}
-						}
-					}
-				}
-				//obscure rust syntax "slice pattern"
-				let [
-					f0,//Cube::Right
-					f1,//Cube::Top
-					f2,//Cube::Back
-					f3,//Cube::Left
-					f4,//Cube::Bottom
-					f5,//Cube::Front
-				]=part_texture_description;
-				let basepart_texture_description=match shape{
-					primitives::Primitives::Sphere=>RobloxBasePartDescription::Sphere,
-					primitives::Primitives::Cube=>RobloxBasePartDescription::Part([f0,f1,f2,f3,f4,f5]),
-					primitives::Primitives::Cylinder=>RobloxBasePartDescription::Cylinder,
-					//use front face texture first and use top face texture as a fallback
-					primitives::Primitives::Wedge=>RobloxBasePartDescription::Wedge([
-						f0,//Cube::Right->Wedge::Right
-						if f5.is_some(){f5}else{f1},//Cube::Front|Cube::Top->Wedge::TopFront
-						f2,//Cube::Back->Wedge::Back
-						f3,//Cube::Left->Wedge::Left
-						f4,//Cube::Bottom->Wedge::Bottom
-					]),
-					//TODO: fix Left+Back texture coordinates to match roblox when not overwridden by Top
-					primitives::Primitives::CornerWedge=>RobloxBasePartDescription::CornerWedge([
-						f0,//Cube::Right->CornerWedge::Right
-						if f2.is_some(){f2}else{f1.clone()},//Cube::Back|Cube::Top->CornerWedge::TopBack
-						if f3.is_some(){f3}else{f1},//Cube::Left|Cube::Top->CornerWedge::TopLeft
-						f4,//Cube::Bottom->CornerWedge::Bottom
-						f5,//Cube::Front->CornerWedge::Front
-					]),
-				};
-				//make new model if unit cube has not been created before
-				let model_id=if let Some(&model_id)=model_id_from_description.get(&basepart_texture_description){
-					//push to existing texture model
-					model_id
-				}else{
-					let model_id=indexed_models.len();
-					model_id_from_description.insert(basepart_texture_description.clone(),model_id);//borrow checker going crazy
-					indexed_models.push(match basepart_texture_description{
-						RobloxBasePartDescription::Sphere=>primitives::unit_sphere(),
-						RobloxBasePartDescription::Part(part_texture_description)=>{
-							let mut cube_face_description=primitives::CubeFaceDescription::new();
-							for (face_id,roblox_face_description) in part_texture_description.iter().enumerate(){
-								cube_face_description.insert(
-								match face_id{
-									0=>primitives::CubeFace::Right,
-									1=>primitives::CubeFace::Top,
-									2=>primitives::CubeFace::Back,
-									3=>primitives::CubeFace::Left,
-									4=>primitives::CubeFace::Bottom,
-									5=>primitives::CubeFace::Front,
-									_=>panic!("unreachable"),
-								},
-								match roblox_face_description{
-									Some(roblox_texture_transform)=>roblox_texture_transform.to_face_description(),
-									None=>primitives::FaceDescription::default(),
-								});
-							}
-							primitives::generate_partial_unit_cube(cube_face_description)
-						},
-						RobloxBasePartDescription::Cylinder=>primitives::unit_cylinder(),
-						RobloxBasePartDescription::Wedge(wedge_texture_description)=>{
-							let mut wedge_face_description=primitives::WedgeFaceDescription::new();
-							for (face_id,roblox_face_description) in wedge_texture_description.iter().enumerate(){
-								wedge_face_description.insert(
-								match face_id{
-									0=>primitives::WedgeFace::Right,
-									1=>primitives::WedgeFace::TopFront,
-									2=>primitives::WedgeFace::Back,
-									3=>primitives::WedgeFace::Left,
-									4=>primitives::WedgeFace::Bottom,
-									_=>panic!("unreachable"),
-								},
-								match roblox_face_description{
-									Some(roblox_texture_transform)=>roblox_texture_transform.to_face_description(),
-									None=>primitives::FaceDescription::default(),
-								});
-							}
-							primitives::generate_partial_unit_wedge(wedge_face_description)
-						},
-						RobloxBasePartDescription::CornerWedge(cornerwedge_texture_description)=>{
-							let mut cornerwedge_face_description=primitives::CornerWedgeFaceDescription::new();
-							for (face_id,roblox_face_description) in cornerwedge_texture_description.iter().enumerate(){
-								cornerwedge_face_description.insert(
-								match face_id{
-									0=>primitives::CornerWedgeFace::Right,
-									1=>primitives::CornerWedgeFace::TopBack,
-									2=>primitives::CornerWedgeFace::TopLeft,
-									3=>primitives::CornerWedgeFace::Bottom,
-									4=>primitives::CornerWedgeFace::Front,
-									_=>panic!("unreachable"),
-								},
-								match roblox_face_description{
-									Some(roblox_texture_transform)=>roblox_texture_transform.to_face_description(),
-									None=>primitives::FaceDescription::default(),
-								});
-							}
-							primitives::generate_partial_unit_cornerwedge(cornerwedge_face_description)
-						},
-					});
-					model_id
-				};
-				indexed_models[model_id].instances.push(crate::model::ModelInstance {
-					transform:model_transform,
-					color:glam::vec4(color3.r as f32/255f32, color3.g as f32/255f32, color3.b as f32/255f32, 1.0-*transparency),
-					attributes:get_attributes(&object.name,*can_collide,Planar64Vec3::try_from([velocity.x,velocity.y,velocity.z]).unwrap(),force_intersecting),
-					temp_indexing:temp_indexing_attributes,
-				});
-			}
-		}
-	}
-	crate::model::IndexedModelInstances{
-		textures:asset_id_from_texture_id.iter().map(|t|t.to_string()).collect(),
-		models:indexed_models,
-		spawn_point,
-		modes:Vec::new(),
-	}
-}

src/main.rs

@@ -1,187 +1,17 @@
-use std::time::Instant;
-use physics::PhysicsInstruction;
-use render_thread::InputInstruction;
-use instruction::{TimedInstruction, InstructionConsumer};
-
-mod bvh;
-mod aabb;
-mod model;
+mod file;
+mod setup;
 mod window;
 mod worker;
-mod zeroes;
-mod integer;
 mod physics;
 mod graphics;
 mod settings;
-mod primitives;
-mod instruction;
-mod load_roblox;
-mod render_thread;
+mod face_crawler;
+mod compat_worker;
+mod model_physics;
 mod model_graphics;
-mod graphics_context;
-
-
-pub struct GlobalState{
-	start_time: std::time::Instant,
-	manual_mouse_lock:bool,
-	mouse:std::sync::Arc<std::sync::Mutex<physics::MouseState>>,
-	user_settings:settings::UserSettings,
-	//Ideally the graphics thread worker description is:
-	/*
-	WorkerDescription{
-		input:Immediate,
-		output:Realtime(PoolOrdering::Ordered(3)),
-	}
-	*/
-	//up to three frames in flight, dropping new frame requests when all three are busy, and dropping output frames when one renders out of order
-	graphics_thread:worker::INWorker<graphics::GraphicsInstruction>,
-	physics_thread:worker::QNWorker<TimedInstruction<InputInstruction>>,
-}
-
-fn load_file(path: std::path::PathBuf)->Option<model::IndexedModelInstances>{
-	println!("Loading file: {:?}", &path);
-	//oh boy! let's load the map!
-	if let Ok(file)=std::fs::File::open(path){
-		let mut input = std::io::BufReader::new(file);
-		let mut first_8=[0u8;8];
-		//.rbxm roblox binary = "<roblox!"
-		//.rbxmx roblox xml = "<roblox "
-		//.bsp = "VBSP"
-		//.vmf = 
-		//.snf = "SNMF"
-		//.snf = "SNBF"
-		if let (Ok(()),Ok(()))=(std::io::Read::read_exact(&mut input, &mut first_8),std::io::Seek::rewind(&mut input)){
-			match &first_8[0..4]{
-				b"<rob"=>{
-					match match &first_8[4..8]{
-						b"lox!"=>rbx_binary::from_reader(input).map_err(|e|format!("{:?}",e)),
-						b"lox "=>rbx_xml::from_reader(input,rbx_xml::DecodeOptions::default()).map_err(|e|format!("{:?}",e)),
-						other=>Err(format!("Unknown Roblox file type {:?}",other)),
-					}{
-						Ok(dom)=>Some(load_roblox::generate_indexed_models(dom)),
-						Err(e)=>{
-							println!("Error loading roblox file:{:?}",e);
-							None
-						},
-					}
-				},
-				//b"VBSP"=>Some(load_bsp::generate_indexed_models(input)),
-				//b"SNFM"=>Some(sniffer::generate_indexed_models(input)),
-				//b"SNFB"=>Some(sniffer::load_bot(input)),
-				other=>{
-					println!("loser file {:?}",other);
-					None
-				},
-			}
-		}else{
-			println!("Failed to read first 8 bytes and seek back to beginning of file.");
-			None
-		}
-	}else{
-		println!("Could not open file");
-		None
-	}
-}
-
-fn default_models()->model::IndexedModelInstances{
-	let mut indexed_models = Vec::new();
-	indexed_models.append(&mut model::generate_indexed_model_list_from_obj(obj::ObjData::load_buf(&include_bytes!("../models/teslacyberv3.0.obj")[..]).unwrap(),glam::Vec4::ONE));
-	indexed_models.push(primitives::unit_sphere());
-	indexed_models.push(primitives::unit_cylinder());
-	indexed_models.push(primitives::unit_cube());
-	println!("models.len = {:?}", indexed_models.len());
-	indexed_models[0].instances.push(model::ModelInstance{
-		transform:integer::Planar64Affine3::try_from(glam::Affine3A::from_translation(glam::vec3(10.,0.,-10.))).unwrap(),
-		..Default::default()
-	});
-	//quad monkeys
-	indexed_models[1].instances.push(model::ModelInstance{
-		transform:integer::Planar64Affine3::try_from(glam::Affine3A::from_translation(glam::vec3(10.,5.,10.))).unwrap(),
-		..Default::default()
-	});
-	indexed_models[1].instances.push(model::ModelInstance{
-		transform:integer::Planar64Affine3::try_from(glam::Affine3A::from_translation(glam::vec3(20.,5.,10.))).unwrap(),
-		color:glam::vec4(1.0,0.0,0.0,1.0),
-		..Default::default()
-	});
-	indexed_models[1].instances.push(model::ModelInstance{
-		transform:integer::Planar64Affine3::try_from(glam::Affine3A::from_translation(glam::vec3(10.,5.,20.))).unwrap(),
-		color:glam::vec4(0.0,1.0,0.0,1.0),
-		..Default::default()
-	});
-	indexed_models[1].instances.push(model::ModelInstance{
-		transform:integer::Planar64Affine3::try_from(glam::Affine3A::from_translation(glam::vec3(20.,5.,20.))).unwrap(),
-		color:glam::vec4(0.0,0.0,1.0,1.0),
-		..Default::default()
-	});
-	//decorative monkey
-	indexed_models[1].instances.push(model::ModelInstance{
-		transform:integer::Planar64Affine3::try_from(glam::Affine3A::from_translation(glam::vec3(15.,10.,15.))).unwrap(),
-		color:glam::vec4(0.5,0.5,0.5,0.5),
-		attributes:model::CollisionAttributes::Decoration,
-		..Default::default()
-	});
-	//teapot
-	indexed_models[2].instances.push(model::ModelInstance{
-		transform:integer::Planar64Affine3::try_from(glam::Affine3A::from_scale_rotation_translation(glam::vec3(0.5, 1.0, 0.2),glam::quat(-0.22248298016985793,-0.839457167990537,-0.05603504040830783,-0.49261857546227916),glam::vec3(-10.,7.,10.))).unwrap(),
-		..Default::default()
-	});
-	//ground
-	indexed_models[3].instances.push(model::ModelInstance{
-		transform:integer::Planar64Affine3::try_from(glam::Affine3A::from_translation(glam::vec3(0.,0.,0.))*glam::Affine3A::from_scale(glam::vec3(160.0, 1.0, 160.0))).unwrap(),
-		..Default::default()
-	});
-	model::IndexedModelInstances{
-		textures:Vec::new(),
-		models:indexed_models,
-		spawn_point:integer::Planar64Vec3::Y*50,
-		modes:Vec::new(),
-	}
-}
-
-impl GlobalState {
-	fn init() -> Self {
-		//wee
-		let user_settings=settings::read_user_settings();
-
-		let mut graphics=GraphicsState::new();
-
-		graphics.load_user_settings(&user_settings);
-
-		//how to multithread
-
-		//1. build
-		physics.generate_models(&indexed_model_instances);
-
-		//2. move
-		let physics_thread=physics.into_worker();
-
-		//3. forget
-
-		let mut state=GlobalState{
-			start_time:Instant::now(),
-			manual_mouse_lock:false,
-			mouse:physics::MouseState::default(),
-			user_settings,
-			graphics,
-			physics_thread,
-		};
-		state.generate_model_graphics(&device,&queue,indexed_model_instances);
-
-		let args:Vec<String>=std::env::args().collect();
-		if args.len()==2{
-			let indexed_model_instances=load_file(std::path::PathBuf::from(&args[1]));
-			state.render_thread=RenderThread::new(user_settings,indexed_model_instances);
-		}
-
-		return state;
-	}
-}
+mod physics_worker;
+mod graphics_worker;
 
 fn main(){
-	let title=format!("Strafe Client v{}",env!("CARGO_PKG_VERSION")).as_str();
-	let context=graphics_context::setup(title);
-	let global_state=GlobalState::init();//new
-	global_state.replace_models(&context,default_models());
-	context.start(global_state);
+	setup::setup_and_start(format!("Strafe Client v{}",env!("CARGO_PKG_VERSION")));
 }

src/model.rs

@@ -1,294 +0,0 @@
-use crate::integer::{Time,Planar64,Planar64Vec3,Planar64Affine3};
-pub type TextureCoordinate=glam::Vec2;
-pub type Color4=glam::Vec4;
-#[derive(Clone,Hash,PartialEq,Eq)]
-pub struct IndexedVertex{
-	pub pos:u32,
-	pub tex:u32,
-	pub normal:u32,
-	pub color:u32,
-}
-pub struct IndexedPolygon{
-	pub vertices:Vec<u32>,
-}
-pub struct IndexedGroup{
-	pub texture:Option<u32>,//RenderPattern? material/texture/shader/flat color
-	pub polys:Vec<IndexedPolygon>,
-}
-pub struct IndexedModel{
-	pub unique_pos:Vec<Planar64Vec3>,
-	pub unique_normal:Vec<Planar64Vec3>,
-	pub unique_tex:Vec<TextureCoordinate>,
-	pub unique_color:Vec<Color4>,
-	pub unique_vertices:Vec<IndexedVertex>,
-	pub groups: Vec<IndexedGroup>,
-	pub instances:Vec<ModelInstance>,
-}
-pub struct ModelInstance{
-	//pub id:u64,//this does not actually help with map fixes resimulating bots, they must always be resimulated
-	pub transform:Planar64Affine3,
-	pub color:Color4,//transparency is in here
-	pub attributes:CollisionAttributes,
-	pub temp_indexing:Vec<TempIndexedAttributes>,
-}
-impl std::default::Default for ModelInstance{
-	fn default() -> Self {
-		Self{
-			color:Color4::ONE,
-			transform:Default::default(),
-			attributes:Default::default(),
-			temp_indexing:Default::default(),
-		}
-	}
-}
-pub struct IndexedModelInstances{
-	pub textures:Vec<String>,//RenderPattern
-	pub models:Vec<IndexedModel>,
-	//may make this into an object later.
-	pub modes:Vec<ModeDescription>,
-	pub spawn_point:Planar64Vec3,
-}
-//stage description referencing flattened ids is spooky, but the map loading is meant to be deterministic.
-pub struct ModeDescription{
-	pub start:usize,//start=model_id
-	pub spawns:Vec<usize>,//spawns[spawn_id]=model_id
-	pub ordered_checkpoints:Vec<usize>,//ordered_checkpoints[checkpoint_id]=model_id
-	pub unordered_checkpoints:Vec<usize>,//unordered_checkpoints[checkpoint_id]=model_id
-	pub spawn_from_stage_id:std::collections::HashMap::<u32,usize>,
-	pub ordered_checkpoint_from_checkpoint_id:std::collections::HashMap::<u32,usize>,
-}
-impl ModeDescription{
-	pub fn get_spawn_model_id(&self,stage_id:u32)->Option<&usize>{
-		self.spawns.get(*self.spawn_from_stage_id.get(&stage_id)?)
-	}
-	pub fn get_ordered_checkpoint_model_id(&self,checkpoint_id:u32)->Option<&usize>{
-		self.ordered_checkpoints.get(*self.ordered_checkpoint_from_checkpoint_id.get(&checkpoint_id)?)
-	}
-}
-//I don't want this code to exist!
-#[derive(Clone)]
-pub struct TempAttrStart{
-	pub mode_id:u32,
-}
-#[derive(Clone)]
-pub struct TempAttrSpawn{
-	pub mode_id:u32,
-	pub stage_id:u32,
-}
-#[derive(Clone)]
-pub struct TempAttrOrderedCheckpoint{
-	pub mode_id:u32,
-	pub checkpoint_id:u32,
-}
-#[derive(Clone)]
-pub struct TempAttrUnorderedCheckpoint{
-	pub mode_id:u32,
-}
-#[derive(Clone)]
-pub struct TempAttrWormhole{
-	pub wormhole_id:u32,
-}
-pub enum TempIndexedAttributes{
-	Start(TempAttrStart),
-	Spawn(TempAttrSpawn),
-	OrderedCheckpoint(TempAttrOrderedCheckpoint),
-	UnorderedCheckpoint(TempAttrUnorderedCheckpoint),
-	Wormhole(TempAttrWormhole),
-}
-
-//you have this effect while in contact
-#[derive(Clone)]
-pub struct ContactingLadder{
-	pub sticky:bool
-}
-#[derive(Clone)]
-pub enum ContactingBehaviour{
-    Surf,
-    Ladder(ContactingLadder),
-    Elastic(u32),//[1/2^32,1] 0=None (elasticity+1)/2^32
-}
-//you have this effect while intersecting
-#[derive(Clone)]
-pub struct IntersectingWater{
-	pub viscosity:Planar64,
-	pub density:Planar64,
-	pub current:Planar64Vec3,
-}
-//All models can be given these attributes
-#[derive(Clone)]
-pub struct GameMechanicAccelerator{
-	pub acceleration:Planar64Vec3
-}
-#[derive(Clone)]
-pub enum GameMechanicBooster{
-	Affine(Planar64Affine3),//capable of SetVelocity,DotVelocity,normal booster,bouncy part,redirect velocity, and much more
-	Velocity(Planar64Vec3),//straight up boost velocity adds to your current velocity
-	Energy{direction:Planar64Vec3,energy:Planar64},//increase energy in direction
-}
-#[derive(Clone)]
-pub enum TrajectoryChoice{
-	HighArcLongDuration,//underhand lob at target: less horizontal speed and more air time
-	LowArcShortDuration,//overhand throw at target: more horizontal speed and less air time
-}
-#[derive(Clone)]
-pub enum GameMechanicSetTrajectory{
-	AirTime(Time),//air time (relative to gravity direction) is invariant across mass and gravity changes
-	Height(Planar64),//boost height (relative to gravity direction) is invariant across mass and gravity changes
-	TargetPointTime{//launch on a trajectory that will land at a target point in a set amount of time
-		target_point:Planar64Vec3,
-		time:Time,//short time = fast and direct, long time = launch high in the air, negative time = wrong way
-	},
-	TrajectoryTargetPoint{//launch at a fixed speed and land at a target point
-		target_point:Planar64Vec3,
-		speed:Planar64,//if speed is too low this will fail to reach the target.  The closest-passing trajectory will be chosen instead
-		trajectory_choice:TrajectoryChoice,
-	},
-	Velocity(Planar64Vec3),//SetVelocity
-	DotVelocity{direction:Planar64Vec3,dot:Planar64},//set your velocity in a specific direction without touching other directions
-}
-#[derive(Clone)]
-pub enum ZoneBehaviour{
-	//Start is indexed
-	//Checkpoints are indexed
-	Finish,
-	Anitcheat,
-}
-#[derive(Clone)]
-pub struct GameMechanicZone{
-	pub mode_id:u32,
-	pub behaviour:ZoneBehaviour,
-}
-// enum TrapCondition{
-// 	FasterThan(Planar64),
-// 	SlowerThan(Planar64),
-// 	InRange(Planar64,Planar64),
-// 	OutsideRange(Planar64,Planar64),
-// }
-#[derive(Clone)]
-pub enum StageElementBehaviour{
- 	//Spawn,//The behaviour of stepping on a spawn setting the spawnid
- 	SpawnAt,
- 	Trigger,
- 	Teleport,
- 	Platform,
- 	JumpLimit(u32),
- 	//Speedtrap(TrapCondition),//Acts as a trigger with a speed condition
-}
-#[derive(Clone)]
-pub struct GameMechanicStageElement{
-	pub mode_id:u32,
-	pub stage_id:u32,//which spawn to send to
-	pub force:bool,//allow setting to lower spawn id i.e. 7->3
-	pub behaviour:StageElementBehaviour
-}
-#[derive(Clone)]
-pub struct GameMechanicWormhole{
-	//destination does not need to be another wormhole
-	//this defines a one way portal to a destination model transform
-	//two of these can create a two way wormhole
-	pub destination_model_id:u32,
-	//(position,angles)*=origin.transform.inverse()*destination.transform
-}
-#[derive(Clone)]
-pub enum TeleportBehaviour{
-	StageElement(GameMechanicStageElement),
-	Wormhole(GameMechanicWormhole),
-}
-//attributes listed in order of handling
-#[derive(Default,Clone)]
-pub struct GameMechanicAttributes{
-	pub zone:Option<GameMechanicZone>,
-	pub booster:Option<GameMechanicBooster>,
-	pub trajectory:Option<GameMechanicSetTrajectory>,
-	pub teleport_behaviour:Option<TeleportBehaviour>,
-	pub accelerator:Option<GameMechanicAccelerator>,
-}
-impl GameMechanicAttributes{
-	pub fn any(&self)->bool{
-		self.booster.is_some()
-		||self.trajectory.is_some()
-		||self.zone.is_some()
-		||self.teleport_behaviour.is_some()
-		||self.accelerator.is_some()
-	}
-}
-#[derive(Default,Clone)]
-pub struct ContactingAttributes{
-	//friction?
-	pub contact_behaviour:Option<ContactingBehaviour>,
-}
-impl ContactingAttributes{
-	pub fn any(&self)->bool{
-		self.contact_behaviour.is_some()
-	}
-}
-#[derive(Default,Clone)]
-pub struct IntersectingAttributes{
-	pub water:Option<IntersectingWater>,
-}
-impl IntersectingAttributes{
-	pub fn any(&self)->bool{
-		self.water.is_some()
-	}
-}
-//Spawn(u32) NO! spawns are indexed in the map header instead of marked with attibutes
-pub enum CollisionAttributes{
-	Decoration,//visual only
-	Contact{//track whether you are contacting the object
-		contacting:ContactingAttributes,
-		general:GameMechanicAttributes,
-	},
-	Intersect{//track whether you are intersecting the object
-		intersecting:IntersectingAttributes,
-		general:GameMechanicAttributes,
-	},
-}
-impl std::default::Default for CollisionAttributes{
-	fn default() -> Self {
-		Self::Contact{
-			contacting:ContactingAttributes::default(),
-			general:GameMechanicAttributes::default()
-		}
-	}
-}
-
-pub fn generate_indexed_model_list_from_obj(data:obj::ObjData,color:Color4)->Vec<IndexedModel>{
-	let mut unique_vertex_index = std::collections::HashMap::<obj::IndexTuple,u32>::new();
-	return data.objects.iter().map(|object|{
-		unique_vertex_index.clear();
-		let mut unique_vertices = Vec::new();
-		let groups = object.groups.iter().map(|group|{
-			IndexedGroup{
-				texture:None,
-				polys:group.polys.iter().map(|poly|{
-					IndexedPolygon{
-						vertices:poly.0.iter().map(|&tup|{
-							if let Some(&i)=unique_vertex_index.get(&tup){
-								i
-							}else{
-								let i=unique_vertices.len() as u32;
-								unique_vertices.push(IndexedVertex{
-									pos: tup.0 as u32,
-									tex: tup.1.unwrap() as u32,
-									normal: tup.2.unwrap() as u32,
-									color: 0,
-								});
-								unique_vertex_index.insert(tup,i);
-								i
-							}
-						}).collect()
-					}
-				}).collect()
-			}
-		}).collect();
-		IndexedModel{
-			unique_pos: data.position.iter().map(|&v|Planar64Vec3::try_from(v).unwrap()).collect(),
-			unique_tex: data.texture.iter().map(|&v|TextureCoordinate::from_array(v)).collect(),
-			unique_normal: data.normal.iter().map(|&v|Planar64Vec3::try_from(v).unwrap()).collect(),
-			unique_color: vec![color],
-			unique_vertices,
-			groups,
-			instances:Vec::new(),
-		}
-	}).collect()
-}

src/model_graphics.rs

@@ -1,55 +1,48 @@
-use bytemuck::{Pod, Zeroable};
-use crate::model::{IndexedVertex,IndexedPolygon};
-#[derive(Clone, Copy, Pod, Zeroable)]
+use bytemuck::{Pod,Zeroable};
+use strafesnet_common::model::{IndexedVertex,PolygonGroup,RenderConfigId};
+#[derive(Clone,Copy,Pod,Zeroable)]
 #[repr(C)]
-pub struct GraphicsVertex {
-	pub pos: [f32; 3],
-	pub tex: [f32; 2],
-	pub normal: [f32; 3],
-	pub color: [f32; 4],
+pub struct GraphicsVertex{
+	pub pos:[f32;3],
+	pub tex:[f32;2],
+	pub normal:[f32;3],
+	pub color:[f32;4],
 }
-pub struct IndexedGroupFixedTexture{
-	pub polys:Vec<IndexedPolygon>,
-}
-pub struct IndexedModelGraphicsSingleTexture{
-	pub unique_pos:Vec<[f32; 3]>,
-	pub unique_tex:Vec<[f32; 2]>,
-	pub unique_normal:Vec<[f32; 3]>,
-	pub unique_color:Vec<[f32; 4]>,
+#[derive(Clone,Copy,id::Id)]
+pub struct IndexedGraphicsMeshOwnedRenderConfigId(u32);
+pub struct IndexedGraphicsMeshOwnedRenderConfig{
+	pub unique_pos:Vec<[f32;3]>,
+	pub unique_tex:Vec<[f32;2]>,
+	pub unique_normal:Vec<[f32;3]>,
+	pub unique_color:Vec<[f32;4]>,
 	pub unique_vertices:Vec<IndexedVertex>,
-	pub texture:Option<u32>,//RenderPattern? material/texture/shader/flat color
-	pub groups: Vec<IndexedGroupFixedTexture>,
-	pub instances:Vec<ModelGraphicsInstance>,
+	pub render_config:RenderConfigId,
+	pub polys:PolygonGroup,
+	pub instances:Vec<GraphicsModelOwned>,
 }
-pub struct ModelGraphicsSingleTexture{
-	pub instances: Vec<ModelGraphicsInstance>,
-	pub vertices: Vec<GraphicsVertex>,
-	pub entities: Vec<Vec<u16>>,
-	pub texture: Option<u32>,
+pub enum Indices{
+	U32(Vec<u32>),
+	U16(Vec<u16>),
 }
-#[derive(Clone,PartialEq)]
-pub struct ModelGraphicsColor4(glam::Vec4);
-impl ModelGraphicsColor4{
-	pub const fn get(&self)->glam::Vec4{
-		self.0
-	}
+pub struct GraphicsMeshOwnedRenderConfig{
+	pub vertices:Vec<GraphicsVertex>,
+	pub indices:Indices,
+	pub render_config:RenderConfigId,
+	pub instances:Vec<GraphicsModelOwned>,
 }
-impl From<glam::Vec4> for ModelGraphicsColor4{
-	fn from(value:glam::Vec4)->Self{
-		Self(value)
-	}
-}
-impl std::hash::Hash for ModelGraphicsColor4{
-	fn hash<H: std::hash::Hasher>(&self,state:&mut H) {
+#[derive(Clone,Copy,PartialEq,id::Id)]
+pub struct GraphicsModelColor4(glam::Vec4);
+impl std::hash::Hash for GraphicsModelColor4{
+	fn hash<H:std::hash::Hasher>(&self,state:&mut H) {
 		for &f in self.0.as_ref(){
 			bytemuck::cast::<f32,u32>(f).hash(state);
 		}
 	}
 }
-impl Eq for ModelGraphicsColor4{}
+impl Eq for GraphicsModelColor4{}
 #[derive(Clone)]
-pub struct ModelGraphicsInstance{
+pub struct GraphicsModelOwned{
 	pub transform:glam::Mat4,
 	pub normal_transform:glam::Mat3,
-	pub color:ModelGraphicsColor4,
+	pub color:GraphicsModelColor4,
 }

src/physics_worker.rs

@@ -0,0 +1,242 @@
+use strafesnet_common::mouse::MouseState;
+use strafesnet_common::physics::Instruction as PhysicsInputInstruction;
+use strafesnet_common::integer::Time;
+use strafesnet_common::instruction::TimedInstruction;
+use strafesnet_common::timer::{Scaled,Timer,TimerState};
+use mouse_interpolator::MouseInterpolator;
+
+#[derive(Debug)]
+pub enum InputInstruction{
+	MoveMouse(glam::IVec2),
+	MoveRight(bool),
+	MoveUp(bool),
+	MoveBack(bool),
+	MoveLeft(bool),
+	MoveDown(bool),
+	MoveForward(bool),
+	Jump(bool),
+	Zoom(bool),
+	ResetAndRestart,
+	ResetAndSpawn(strafesnet_common::gameplay_modes::ModeId,strafesnet_common::gameplay_modes::StageId),
+	PracticeFly,
+}
+pub enum Instruction{
+	Input(InputInstruction),
+	Render,
+	Resize(winit::dpi::PhysicalSize<u32>),
+	ChangeMap(strafesnet_common::map::CompleteMap),
+	//SetPaused is not an InputInstruction: the physics doesn't know that it's paused.
+	SetPaused(bool),
+	//Graphics(crate::graphics_worker::Instruction),
+}
+mod mouse_interpolator{
+	use super::*;
+	//TODO: move this or tab
+pub struct MouseInterpolator{
+	//"PlayerController"
+	user_settings:crate::settings::UserSettings,
+	//"MouseInterpolator"
+	timeline:std::collections::VecDeque<TimedInstruction<PhysicsInputInstruction>>,
+	last_mouse_time:Time,//this value is pre-transformed to simulation time
+	mouse_blocking:bool,
+	//"Simulation"
+	timer:Timer<Scaled>,
+	physics:crate::physics::PhysicsContext,
+
+}
+impl MouseInterpolator{
+	pub fn new(
+		physics:crate::physics::PhysicsContext,
+		user_settings:crate::settings::UserSettings,
+	)->MouseInterpolator{
+		MouseInterpolator{
+			mouse_blocking:true,
+			last_mouse_time:physics.get_next_mouse().time,
+			timeline:std::collections::VecDeque::new(),
+			timer:Timer::from_state(Scaled::identity(),false),
+			physics,
+			user_settings,
+		}
+	}
+	fn push_mouse_instruction(&mut self,ins:&TimedInstruction<Instruction>,m:glam::IVec2){
+		if self.mouse_blocking{
+			//tell the game state which is living in the past about its future
+			self.timeline.push_front(TimedInstruction{
+				time:self.last_mouse_time,
+				instruction:PhysicsInputInstruction::SetNextMouse(MouseState{time:self.timer.time(ins.time),pos:m}),
+			});
+		}else{
+			//mouse has just started moving again after being still for longer than 10ms.
+			//replace the entire mouse interpolation state to avoid an intermediate state with identical m0.t m1.t timestamps which will divide by zero
+			self.timeline.push_front(TimedInstruction{
+				time:self.last_mouse_time,
+				instruction:PhysicsInputInstruction::ReplaceMouse(
+					MouseState{time:self.last_mouse_time,pos:self.physics.get_next_mouse().pos},
+					MouseState{time:self.timer.time(ins.time),pos:m}
+				),
+			});
+			//delay physics execution until we have an interpolation target
+			self.mouse_blocking=true;
+		}
+		self.last_mouse_time=self.timer.time(ins.time);
+	}
+	fn push(&mut self,time:Time,phys_input:PhysicsInputInstruction){
+		//This is always a non-mouse event
+		self.timeline.push_back(TimedInstruction{
+			time:self.timer.time(time),
+			instruction:phys_input,
+		});
+	}
+	/// returns should_empty_queue
+	/// may or may not mutate internal state XD!
+	fn map_instruction(&mut self,ins:&TimedInstruction<Instruction>)->bool{
+		let mut update_mouse_blocking=true;
+		match &ins.instruction{
+			Instruction::Input(input_instruction)=>match input_instruction{
+				&InputInstruction::MoveMouse(m)=>{
+					if !self.timer.is_paused(){
+						self.push_mouse_instruction(ins,m);
+					}
+					update_mouse_blocking=false;
+				},
+				&InputInstruction::MoveForward(s)=>self.push(ins.time,PhysicsInputInstruction::SetMoveForward(s)),
+				&InputInstruction::MoveLeft(s)=>self.push(ins.time,PhysicsInputInstruction::SetMoveLeft(s)),
+				&InputInstruction::MoveBack(s)=>self.push(ins.time,PhysicsInputInstruction::SetMoveBack(s)),
+				&InputInstruction::MoveRight(s)=>self.push(ins.time,PhysicsInputInstruction::SetMoveRight(s)),
+				&InputInstruction::MoveUp(s)=>self.push(ins.time,PhysicsInputInstruction::SetMoveUp(s)),
+				&InputInstruction::MoveDown(s)=>self.push(ins.time,PhysicsInputInstruction::SetMoveDown(s)),
+				&InputInstruction::Jump(s)=>self.push(ins.time,PhysicsInputInstruction::SetJump(s)),
+				&InputInstruction::Zoom(s)=>self.push(ins.time,PhysicsInputInstruction::SetZoom(s)),
+				&InputInstruction::ResetAndSpawn(mode_id,stage_id)=>{
+					self.push(ins.time,PhysicsInputInstruction::Reset);
+					self.push(ins.time,PhysicsInputInstruction::SetSensitivity(self.user_settings.calculate_sensitivity()));
+					self.push(ins.time,PhysicsInputInstruction::Spawn(mode_id,stage_id));
+				},
+				InputInstruction::ResetAndRestart=>{
+					self.push(ins.time,PhysicsInputInstruction::Reset);
+					self.push(ins.time,PhysicsInputInstruction::SetSensitivity(self.user_settings.calculate_sensitivity()));
+					self.push(ins.time,PhysicsInputInstruction::Restart);
+				},
+				InputInstruction::PracticeFly=>self.push(ins.time,PhysicsInputInstruction::PracticeFly),
+			},
+			//do these really need to idle the physics?
+			//sending None dumps the instruction queue
+			Instruction::ChangeMap(_)=>self.push(ins.time,PhysicsInputInstruction::Idle),
+			Instruction::Resize(_)=>self.push(ins.time,PhysicsInputInstruction::Idle),
+			Instruction::Render=>self.push(ins.time,PhysicsInputInstruction::Idle),
+			&Instruction::SetPaused(paused)=>{
+				if let Err(e)=self.timer.set_paused(ins.time,paused){
+					println!("Cannot pause: {e}");
+				}
+				self.push(ins.time,PhysicsInputInstruction::Idle);
+			},
+		}
+		if update_mouse_blocking{
+			//this returns the bool for us
+			self.update_mouse_blocking(ins.time)
+		}else{
+			//do flush that queue
+			true
+		}
+	}
+	/// must check if self.mouse_blocking==true before calling!
+	fn unblock_mouse(&mut self,time:Time){
+		//push an event to extrapolate no movement from
+		self.timeline.push_front(TimedInstruction{
+			time:self.last_mouse_time,
+			instruction:PhysicsInputInstruction::SetNextMouse(MouseState{time:self.timer.time(time),pos:self.physics.get_next_mouse().pos}),
+		});
+		self.last_mouse_time=self.timer.time(time);
+		//stop blocking. the mouse is not moving so the physics does not need to live in the past and wait for interpolation targets.
+		self.mouse_blocking=false;
+	}
+	fn update_mouse_blocking(&mut self,time:Time)->bool{
+		if self.mouse_blocking{
+			//assume the mouse has stopped moving after 10ms.
+			//shitty mice are 125Hz which is 8ms so this should cover that.
+			//setting this to 100us still doesn't print even though it's 10x lower than the polling rate,
+			//so mouse events are probably not handled separately from drawing and fire right before it :(
+			if Time::from_millis(10)<self.timer.time(time)-self.physics.get_next_mouse().time{
+				self.unblock_mouse(time);
+				true
+			}else{
+				false
+			}
+		}else{
+			//keep this up to date so that it can be used as a known-timestamp
+			//that the mouse was not moving when the mouse starts moving again
+			self.last_mouse_time=self.timer.time(time);
+			true
+		}
+	}
+	fn empty_queue(&mut self){
+		while let Some(instruction)=self.timeline.pop_front(){
+			self.physics.run_input_instruction(instruction);
+		}
+	}
+	pub fn handle_instruction(&mut self,ins:&TimedInstruction<Instruction>){
+		let should_empty_queue=self.map_instruction(ins);
+		if should_empty_queue{
+			self.empty_queue();
+		}
+	}
+	pub fn get_frame_state(&self,time:Time)->crate::graphics::FrameState{
+		crate::graphics::FrameState{
+			body:self.physics.camera_body(),
+			camera:self.physics.camera(),
+			time:self.timer.time(time),
+		}
+	}
+	pub fn change_map(&mut self,time:Time,map:&strafesnet_common::map::CompleteMap){
+		//dump any pending interpolation state
+		if self.mouse_blocking{
+			self.unblock_mouse(time);
+		}
+		self.empty_queue();
+
+		//doing it like this to avoid doing PhysicsInstruction::ChangeMap(Rc<CompleteMap>)
+		self.physics.generate_models(&map);
+
+		//use the standard input interface so the instructions are written out to bots
+		self.handle_instruction(&TimedInstruction{
+			time:self.timer.time(time),
+			instruction:Instruction::Input(InputInstruction::ResetAndSpawn(
+				strafesnet_common::gameplay_modes::ModeId::MAIN,
+				strafesnet_common::gameplay_modes::StageId::FIRST,
+			)),
+		});
+	}
+	pub const fn user_settings(&self)->&crate::settings::UserSettings{
+		&self.user_settings
+	}
+}
+}
+
+pub fn new<'a>(
+	mut graphics_worker:crate::compat_worker::INWorker<'a,crate::graphics_worker::Instruction>,
+	user_settings:crate::settings::UserSettings,
+)->crate::compat_worker::QNWorker<'a,TimedInstruction<Instruction>>{
+	let physics=crate::physics::PhysicsContext::default();
+	let mut interpolator=MouseInterpolator::new(
+		physics,
+		user_settings
+	);
+	crate::compat_worker::QNWorker::new(move |ins:TimedInstruction<Instruction>|{
+		interpolator.handle_instruction(&ins);
+		match ins.instruction{
+			Instruction::Render=>{
+				let frame_state=interpolator.get_frame_state(ins.time);
+				graphics_worker.send(crate::graphics_worker::Instruction::Render(frame_state)).unwrap();
+			},
+			Instruction::Resize(size)=>{
+				graphics_worker.send(crate::graphics_worker::Instruction::Resize(size,interpolator.user_settings().clone())).unwrap();
+			},
+			Instruction::ChangeMap(map)=>{
+				interpolator.change_map(ins.time,&map);
+				graphics_worker.send(crate::graphics_worker::Instruction::ChangeMap(map)).unwrap();
+			},
+			Instruction::Input(_)=>(),
+			Instruction::SetPaused(_)=>(),
+		}
+	})
+}

src/primitives.rs

@@ -1,510 +0,0 @@
-use crate::model::{Color4,TextureCoordinate,IndexedModel,IndexedPolygon,IndexedGroup,IndexedVertex};
-use crate::integer::Planar64Vec3;
-
-#[derive(Debug)]
-pub enum Primitives{
-	Sphere,
-	Cube,
-	Cylinder,
-	Wedge,
-	CornerWedge,
-}
-#[derive(Hash,PartialEq,Eq)]
-pub enum CubeFace{
-	Right,
-	Top,
-	Back,
-	Left,
-	Bottom,
-	Front,
-}
-const CUBE_DEFAULT_TEXTURE_COORDS:[TextureCoordinate;4]=[
-	TextureCoordinate::new(0.0,0.0),
-	TextureCoordinate::new(1.0,0.0),
-	TextureCoordinate::new(1.0,1.0),
-	TextureCoordinate::new(0.0,1.0),
-];
-const CUBE_DEFAULT_VERTICES:[Planar64Vec3;8]=[
-	Planar64Vec3::int(-1,-1, 1),//0 left bottom back
-	Planar64Vec3::int( 1,-1, 1),//1 right bottom back
-	Planar64Vec3::int( 1, 1, 1),//2 right top back
-	Planar64Vec3::int(-1, 1, 1),//3 left top back
-	Planar64Vec3::int(-1, 1,-1),//4 left top front
-	Planar64Vec3::int( 1, 1,-1),//5 right top front
-	Planar64Vec3::int( 1,-1,-1),//6 right bottom front
-	Planar64Vec3::int(-1,-1,-1),//7 left bottom front
-];
-const CUBE_DEFAULT_NORMALS:[Planar64Vec3;6]=[
-	Planar64Vec3::int( 1, 0, 0),//CubeFace::Right
-	Planar64Vec3::int( 0, 1, 0),//CubeFace::Top
-	Planar64Vec3::int( 0, 0, 1),//CubeFace::Back
-	Planar64Vec3::int(-1, 0, 0),//CubeFace::Left
-	Planar64Vec3::int( 0,-1, 0),//CubeFace::Bottom
-	Planar64Vec3::int( 0, 0,-1),//CubeFace::Front
-];
-const CUBE_DEFAULT_POLYS:[[[u32;3];4];6]=[
-	// right (1, 0, 0)
-	[
-		[6,2,0],//[vertex,tex,norm]
-		[5,1,0],
-		[2,0,0],
-		[1,3,0],
-	],
-	// top (0, 1, 0)
-	[
-		[5,3,1],
-		[4,2,1],
-		[3,1,1],
-		[2,0,1],
-	],
-	// back (0, 0, 1)
-	[
-		[0,3,2],
-		[1,2,2],
-		[2,1,2],
-		[3,0,2],
-	],
-	// left (-1, 0, 0)
-	[
-		[0,2,3],
-		[3,1,3],
-		[4,0,3],
-		[7,3,3],
-	],
-	// bottom (0,-1, 0)
-	[
-		[1,1,4],
-		[0,0,4],
-		[7,3,4],
-		[6,2,4],
-	],
-	// front (0, 0,-1)
-	[
-		[4,1,5],
-		[5,0,5],
-		[6,3,5],
-		[7,2,5],
-	],
-];
-
-#[derive(Hash,PartialEq,Eq)]
-pub enum WedgeFace{
-	Right,
-	TopFront,
-	Back,
-	Left,
-	Bottom,
-}
-const WEDGE_DEFAULT_NORMALS:[Planar64Vec3;5]=[
-	Planar64Vec3::int( 1, 0, 0),//Wedge::Right
-	Planar64Vec3::int( 0, 1,-1),//Wedge::TopFront
-	Planar64Vec3::int( 0, 0, 1),//Wedge::Back
-	Planar64Vec3::int(-1, 0, 0),//Wedge::Left
-	Planar64Vec3::int( 0,-1, 0),//Wedge::Bottom
-];
-/*
-local cornerWedgeVerticies = {
-	Vector3.new(-1/2,-1/2,-1/2),7
-	Vector3.new(-1/2,-1/2, 1/2),0
-	Vector3.new( 1/2,-1/2,-1/2),6
-	Vector3.new( 1/2,-1/2, 1/2),1
-	Vector3.new( 1/2, 1/2,-1/2),5
-}
-*/
-#[derive(Hash,PartialEq,Eq)]
-pub enum CornerWedgeFace{
-	Right,
-	TopBack,
-	TopLeft,
-	Bottom,
-	Front,
-}
-const CORNERWEDGE_DEFAULT_NORMALS:[Planar64Vec3;5]=[
-	Planar64Vec3::int( 1, 0, 0),//CornerWedge::Right
-	Planar64Vec3::int( 0, 1, 1),//CornerWedge::BackTop
-	Planar64Vec3::int(-1, 1, 0),//CornerWedge::LeftTop
-	Planar64Vec3::int( 0,-1, 0),//CornerWedge::Bottom
-	Planar64Vec3::int( 0, 0,-1),//CornerWedge::Front
-];
-//HashMap fits this use case perfectly but feels like using a sledgehammer to drive a nail
-pub fn unit_sphere()->crate::model::IndexedModel{
-	let mut indexed_model=crate::model::generate_indexed_model_list_from_obj(obj::ObjData::load_buf(&include_bytes!("../models/suzanne.obj")[..]).unwrap(),Color4::ONE).remove(0);
-	for pos in indexed_model.unique_pos.iter_mut(){
-		*pos=*pos/2;
-	}
-	indexed_model
-}
-pub type CubeFaceDescription=std::collections::HashMap::<CubeFace,FaceDescription>;
-pub fn unit_cube()->crate::model::IndexedModel{
-	let mut t=CubeFaceDescription::new();
-	t.insert(CubeFace::Right,FaceDescription::default());
-	t.insert(CubeFace::Top,FaceDescription::default());
-	t.insert(CubeFace::Back,FaceDescription::default());
-	t.insert(CubeFace::Left,FaceDescription::default());
-	t.insert(CubeFace::Bottom,FaceDescription::default());
-	t.insert(CubeFace::Front,FaceDescription::default());
-	generate_partial_unit_cube(t)
-}
-const TEAPOT_TRANSFORM:crate::integer::Planar64Mat3=crate::integer::Planar64Mat3::int_from_cols_array([0,1,0, -1,0,0, 0,0,1]);
-pub fn unit_cylinder()->crate::model::IndexedModel{
-	let mut indexed_model=crate::model::generate_indexed_model_list_from_obj(obj::ObjData::load_buf(&include_bytes!("../models/teapot.obj")[..]).unwrap(),Color4::ONE).remove(0);
-	for pos in indexed_model.unique_pos.iter_mut(){
-		*pos=TEAPOT_TRANSFORM*(*pos)/10;
-	}
-	indexed_model
-}
-pub type WedgeFaceDescription=std::collections::HashMap::<WedgeFace,FaceDescription>;
-pub fn unit_wedge()->crate::model::IndexedModel{
-	let mut t=WedgeFaceDescription::new();
-	t.insert(WedgeFace::Right,FaceDescription::default());
-	t.insert(WedgeFace::TopFront,FaceDescription::default());
-	t.insert(WedgeFace::Back,FaceDescription::default());
-	t.insert(WedgeFace::Left,FaceDescription::default());
-	t.insert(WedgeFace::Bottom,FaceDescription::default());
-	generate_partial_unit_wedge(t)
-}
-pub type CornerWedgeFaceDescription=std::collections::HashMap::<CornerWedgeFace,FaceDescription>;
-pub fn unit_cornerwedge()->crate::model::IndexedModel{
-	let mut t=CornerWedgeFaceDescription::new();
-	t.insert(CornerWedgeFace::Right,FaceDescription::default());
-	t.insert(CornerWedgeFace::TopBack,FaceDescription::default());
-	t.insert(CornerWedgeFace::TopLeft,FaceDescription::default());
-	t.insert(CornerWedgeFace::Bottom,FaceDescription::default());
-	t.insert(CornerWedgeFace::Front,FaceDescription::default());
-	generate_partial_unit_cornerwedge(t)
-}
-
-#[derive(Clone)]
-pub struct FaceDescription{
-	pub texture:Option<u32>,
-	pub transform:glam::Affine2,
-	pub color:Color4,
-}
-impl std::default::Default for FaceDescription{
-	fn default()->Self {
-		Self{
-			texture:None,
-			transform:glam::Affine2::IDENTITY,
-			color:Color4::new(1.0,1.0,1.0,0.0),//zero alpha to hide the default texture
-		}
-	}
-}
-impl FaceDescription{
-	pub fn new(texture:u32,transform:glam::Affine2,color:Color4)->Self{
-		Self{texture:Some(texture),transform,color}
-	}
-	pub fn from_texture(texture:u32)->Self{
-		Self{
-			texture:Some(texture),
-			transform:glam::Affine2::IDENTITY,
-			color:Color4::ONE,
-		}
-	}
-}
-//TODO: it's probably better to use a shared vertex buffer between all primitives and use indexed rendering instead of generating a unique vertex buffer for each primitive.
-//implementation: put all roblox primitives into one model.groups <- this won't work but I forget why
-pub fn generate_partial_unit_cube(face_descriptions:CubeFaceDescription)->crate::model::IndexedModel{
-	let mut generated_pos=Vec::new();
-	let mut generated_tex=Vec::new();
-	let mut generated_normal=Vec::new();
-	let mut generated_color=Vec::new();
-	let mut generated_vertices=Vec::new();
-	let mut groups=Vec::new();
-	let mut transforms=Vec::new();
-	//note that on a cube every vertex is guaranteed to be unique, so there's no need to hash them against existing vertices.
-	for (face,face_description) in face_descriptions.into_iter(){
-		//assume that scanning short lists is faster than hashing.
-		let transform_index=if let Some(transform_index)=transforms.iter().position(|&transform|transform==face_description.transform){
-			transform_index
-		}else{
-			//create new transform_index
-			let transform_index=transforms.len();
-			transforms.push(face_description.transform);
-			for tex in CUBE_DEFAULT_TEXTURE_COORDS{
-				generated_tex.push(face_description.transform.transform_point2(tex));
-			}
-			transform_index
-		} as u32;
-		let color_index=if let Some(color_index)=generated_color.iter().position(|&color|color==face_description.color){
-			color_index
-		}else{
-			//create new color_index
-			let color_index=generated_color.len();
-			generated_color.push(face_description.color);
-			color_index
-		} as u32;
-		let face_id=match face{
-			CubeFace::Right => 0,
-			CubeFace::Top => 1,
-			CubeFace::Back => 2,
-			CubeFace::Left => 3,
-			CubeFace::Bottom => 4,
-			CubeFace::Front => 5,
-		};
-		//always push normal
-		let normal_index=generated_normal.len() as u32;
-		generated_normal.push(CUBE_DEFAULT_NORMALS[face_id]);
-		//push vertices as they are needed
-		groups.push(IndexedGroup{
-			texture:face_description.texture,
-			polys:vec![IndexedPolygon{
-				vertices:CUBE_DEFAULT_POLYS[face_id].map(|tup|{
-					let pos=CUBE_DEFAULT_VERTICES[tup[0] as usize];
-					let pos_index=if let Some(pos_index)=generated_pos.iter().position(|&p|p==pos){
-						pos_index
-					}else{
-						//create new pos_index
-						let pos_index=generated_pos.len();
-						generated_pos.push(pos);
-						pos_index
-					} as u32;
-					//always push vertex
-					let vertex=IndexedVertex{
-						pos:pos_index,
-						tex:tup[1]+4*transform_index,
-						normal:normal_index,
-						color:color_index,
-					};
-					let vert_index=generated_vertices.len();
-					generated_vertices.push(vertex);
-					vert_index as u32
-				}).to_vec(),
-			}],
-		});
-	}
-	IndexedModel{
-		unique_pos:generated_pos,
-		unique_tex:generated_tex,
-		unique_normal:generated_normal,
-		unique_color:generated_color,
-		unique_vertices:generated_vertices,
-		groups,
-		instances:Vec::new(),
-	}
-}
-//don't think too hard about the copy paste because this is all going into the map tool eventually...
-pub fn generate_partial_unit_wedge(face_descriptions:WedgeFaceDescription)->crate::model::IndexedModel{
-	let wedge_default_polys=vec![
-		// right (1, 0, 0)
-		vec![
-			[6,2,0],//[vertex,tex,norm]
-			[2,0,0],
-			[1,3,0],
-		],
-		// FrontTop (0, 1, -1)
-		vec![
-			[3,1,1],
-			[2,0,1],
-			[6,3,1],
-			[7,2,1],
-		],
-		// back (0, 0, 1)
-		vec![
-			[0,3,2],
-			[1,2,2],
-			[2,1,2],
-			[3,0,2],
-		],
-		// left (-1, 0, 0)
-		vec![
-			[0,2,3],
-			[3,1,3],
-			[7,3,3],
-		],
-		// bottom (0,-1, 0)
-		vec![
-			[1,1,4],
-			[0,0,4],
-			[7,3,4],
-			[6,2,4],
-		],
-	];
-	let mut generated_pos=Vec::new();
-	let mut generated_tex=Vec::new();
-	let mut generated_normal=Vec::new();
-	let mut generated_color=Vec::new();
-	let mut generated_vertices=Vec::new();
-	let mut groups=Vec::new();
-	let mut transforms=Vec::new();
-	//note that on a cube every vertex is guaranteed to be unique, so there's no need to hash them against existing vertices.
-	for (face,face_description) in face_descriptions.into_iter(){
-		//assume that scanning short lists is faster than hashing.
-		let transform_index=if let Some(transform_index)=transforms.iter().position(|&transform|transform==face_description.transform){
-			transform_index
-		}else{
-			//create new transform_index
-			let transform_index=transforms.len();
-			transforms.push(face_description.transform);
-			for tex in CUBE_DEFAULT_TEXTURE_COORDS{
-				generated_tex.push(face_description.transform.transform_point2(tex));
-			}
-			transform_index
-		} as u32;
-		let color_index=if let Some(color_index)=generated_color.iter().position(|&color|color==face_description.color){
-			color_index
-		}else{
-			//create new color_index
-			let color_index=generated_color.len();
-			generated_color.push(face_description.color);
-			color_index
-		} as u32;
-		let face_id=match face{
-			WedgeFace::Right => 0,
-			WedgeFace::TopFront => 1,
-			WedgeFace::Back => 2,
-			WedgeFace::Left => 3,
-			WedgeFace::Bottom => 4,
-		};
-		//always push normal
-		let normal_index=generated_normal.len() as u32;
-		generated_normal.push(WEDGE_DEFAULT_NORMALS[face_id]);
-		//push vertices as they are needed
-		groups.push(IndexedGroup{
-			texture:face_description.texture,
-			polys:vec![IndexedPolygon{
-				vertices:wedge_default_polys[face_id].iter().map(|tup|{
-					let pos=CUBE_DEFAULT_VERTICES[tup[0] as usize];
-					let pos_index=if let Some(pos_index)=generated_pos.iter().position(|&p|p==pos){
-						pos_index
-					}else{
-						//create new pos_index
-						let pos_index=generated_pos.len();
-						generated_pos.push(pos);
-						pos_index
-					} as u32;
-					//always push vertex
-					let vertex=IndexedVertex{
-						pos:pos_index,
-						tex:tup[1]+4*transform_index,
-						normal:normal_index,
-						color:color_index,
-					};
-					let vert_index=generated_vertices.len();
-					generated_vertices.push(vertex);
-					vert_index as u32
-				}).collect(),
-			}],
-		});
-	}
-	IndexedModel{
-		unique_pos:generated_pos,
-		unique_tex:generated_tex,
-		unique_normal:generated_normal,
-		unique_color:generated_color,
-		unique_vertices:generated_vertices,
-		groups,
-		instances:Vec::new(),
-	}
-}
-
-pub fn generate_partial_unit_cornerwedge(face_descriptions:CornerWedgeFaceDescription)->crate::model::IndexedModel{
-	let cornerwedge_default_polys=vec![
-		// right (1, 0, 0)
-		vec![
-			[6,2,0],//[vertex,tex,norm]
-			[5,1,0],
-			[1,3,0],
-		],
-		// BackTop (0, 1, 1)
-		vec![
-			[5,3,1],
-			[0,1,1],
-			[1,0,1],
-		],
-		// LeftTop (-1, 1, 0)
-		vec![
-			[5,3,2],
-			[7,2,2],
-			[0,1,2],
-		],
-		// bottom (0,-1, 0)
-		vec![
-			[1,1,3],
-			[0,0,3],
-			[7,3,3],
-			[6,2,3],
-		],
-		// front (0, 0,-1)
-		vec![
-			[5,0,4],
-			[6,3,4],
-			[7,2,4],
-		],
-	];
-	let mut generated_pos=Vec::new();
-	let mut generated_tex=Vec::new();
-	let mut generated_normal=Vec::new();
-	let mut generated_color=Vec::new();
-	let mut generated_vertices=Vec::new();
-	let mut groups=Vec::new();
-	let mut transforms=Vec::new();
-	//note that on a cube every vertex is guaranteed to be unique, so there's no need to hash them against existing vertices.
-	for (face,face_description) in face_descriptions.into_iter(){
-		//assume that scanning short lists is faster than hashing.
-		let transform_index=if let Some(transform_index)=transforms.iter().position(|&transform|transform==face_description.transform){
-			transform_index
-		}else{
-			//create new transform_index
-			let transform_index=transforms.len();
-			transforms.push(face_description.transform);
-			for tex in CUBE_DEFAULT_TEXTURE_COORDS{
-				generated_tex.push(face_description.transform.transform_point2(tex));
-			}
-			transform_index
-		} as u32;
-		let color_index=if let Some(color_index)=generated_color.iter().position(|&color|color==face_description.color){
-			color_index
-		}else{
-			//create new color_index
-			let color_index=generated_color.len();
-			generated_color.push(face_description.color);
-			color_index
-		} as u32;
-		let face_id=match face{
-			CornerWedgeFace::Right => 0,
-			CornerWedgeFace::TopBack => 1,
-			CornerWedgeFace::TopLeft => 2,
-			CornerWedgeFace::Bottom => 3,
-			CornerWedgeFace::Front => 4,
-		};
-		//always push normal
-		let normal_index=generated_normal.len() as u32;
-		generated_normal.push(CORNERWEDGE_DEFAULT_NORMALS[face_id]);
-		//push vertices as they are needed
-		groups.push(IndexedGroup{
-			texture:face_description.texture,
-			polys:vec![IndexedPolygon{
-				vertices:cornerwedge_default_polys[face_id].iter().map(|tup|{
-					let pos=CUBE_DEFAULT_VERTICES[tup[0] as usize];
-					let pos_index=if let Some(pos_index)=generated_pos.iter().position(|&p|p==pos){
-						pos_index
-					}else{
-						//create new pos_index
-						let pos_index=generated_pos.len();
-						generated_pos.push(pos);
-						pos_index
-					} as u32;
-					//always push vertex
-					let vertex=IndexedVertex{
-						pos:pos_index,
-						tex:tup[1]+4*transform_index,
-						normal:normal_index,
-						color:color_index,
-					};
-					let vert_index=generated_vertices.len();
-					generated_vertices.push(vertex);
-					vert_index as u32
-				}).collect(),
-			}],
-		});
-	}
-	IndexedModel{
-		unique_pos:generated_pos,
-		unique_tex:generated_tex,
-		unique_normal:generated_normal,
-		unique_color:generated_color,
-		unique_vertices:generated_vertices,
-		groups,
-		instances:Vec::new(),
-	}
-}

src/render_thread.rs

@@ -1,129 +0,0 @@
-use crate::integer::Time;
-use crate::physics::{MouseState,PhysicsInputInstruction};
-use crate::instruction::{TimedInstruction,InstructionConsumer};
-
-#[derive(Debug)]
-pub enum InputInstruction {
-	MoveMouse(glam::IVec2),
-	MoveRight(bool),
-	MoveUp(bool),
-	MoveBack(bool),
-	MoveLeft(bool),
-	MoveDown(bool),
-	MoveForward(bool),
-	Jump(bool),
-	Zoom(bool),
-	Reset,
-	Render,
-		//Idle: there were no input events, but the simulation is safe to advance to this timestep
-		//for interpolation / networking / playback reasons, most playback heads will always want
-		//to be 1 instruction ahead to generate the next state for interpolation.
-}
-
-pub struct RenderState{
-	physics:crate::physics::PhysicsState,
-	graphics:crate::graphics::GraphicsState,
-}
-impl RenderState{
-	pub fn new(user_settings:&crate::settings::UserSettings,indexed_model_instances:crate::model::IndexedModelInstances){
-
-		let mut physics=crate::physics::PhysicsState::default();
-		physics.spawn(indexed_model_instances.spawn_point);
-		physics.load_user_settings(user_settings);
-		physics.generate_models(&indexed_model_instances);
-
-		let mut graphics=Self::new_graphics_state();
-		graphics.load_user_settings(user_settings);
-		graphics.generate_models(indexed_model_instances);
-		//manual reset
-	}
-	pub fn into_worker(mut self)->crate::worker::CNWorker<TimedInstruction<InputInstruction>>{
-		let mut mouse_blocking=true;
-		let mut last_mouse_time=self.physics.next_mouse.time;
-		let mut timeline=std::collections::VecDeque::new();
-		crate::worker::CNWorker::new(move |ins:TimedInstruction<InputInstruction>|{
-			let mut render=false;
-			if if let Some(phys_input)=match ins.instruction{
-				InputInstruction::MoveMouse(m)=>{
-					if mouse_blocking{
-						//tell the game state which is living in the past about its future
-						timeline.push_front(TimedInstruction{
-							time:last_mouse_time,
-							instruction:PhysicsInputInstruction::SetNextMouse(MouseState{time:ins.time,pos:m}),
-						});
-					}else{
-						//mouse has just started moving again after being still for longer than 10ms.
-						//replace the entire mouse interpolation state to avoid an intermediate state with identical m0.t m1.t timestamps which will divide by zero
-						timeline.push_front(TimedInstruction{
-							time:last_mouse_time,
-							instruction:PhysicsInputInstruction::ReplaceMouse(
-								MouseState{time:last_mouse_time,pos:self.physics.next_mouse.pos},
-								MouseState{time:ins.time,pos:m}
-							),
-						});
-						//delay physics execution until we have an interpolation target
-						mouse_blocking=true;
-					}
-					last_mouse_time=ins.time;
-					None
-				},
-				InputInstruction::MoveForward(s)=>Some(PhysicsInputInstruction::SetMoveForward(s)),
-				InputInstruction::MoveLeft(s)=>Some(PhysicsInputInstruction::SetMoveLeft(s)),
-				InputInstruction::MoveBack(s)=>Some(PhysicsInputInstruction::SetMoveBack(s)),
-				InputInstruction::MoveRight(s)=>Some(PhysicsInputInstruction::SetMoveRight(s)),
-				InputInstruction::MoveUp(s)=>Some(PhysicsInputInstruction::SetMoveUp(s)),
-				InputInstruction::MoveDown(s)=>Some(PhysicsInputInstruction::SetMoveDown(s)),
-				InputInstruction::Jump(s)=>Some(PhysicsInputInstruction::SetJump(s)),
-				InputInstruction::Zoom(s)=>Some(PhysicsInputInstruction::SetZoom(s)),
-				InputInstruction::Reset=>Some(PhysicsInputInstruction::Reset),
-				InputInstruction::Render=>{render=true;Some(PhysicsInputInstruction::Idle)},
-			}{
-				//non-mouse event
-				timeline.push_back(TimedInstruction{
-					time:ins.time,
-					instruction:phys_input,
-				});
-				
-				if mouse_blocking{
-					//assume the mouse has stopped moving after 10ms.
-					//shitty mice are 125Hz which is 8ms so this should cover that.
-					//setting this to 100us still doesn't print even though it's 10x lower than the polling rate,
-					//so mouse events are probably not handled separately from drawing and fire right before it :(
-					if Time::from_millis(10)<ins.time-self.physics.next_mouse.time{
-						//push an event to extrapolate no movement from
-						timeline.push_front(TimedInstruction{
-							time:last_mouse_time,
-							instruction:PhysicsInputInstruction::SetNextMouse(MouseState{time:ins.time,pos:self.physics.next_mouse.pos}),
-						});
-						last_mouse_time=ins.time;
-						//stop blocking. the mouse is not moving so the physics does not need to live in the past and wait for interpolation targets.
-						mouse_blocking=false;
-						true
-					}else{
-						false
-					}
-				}else{
-					//keep this up to date so that it can be used as a known-timestamp
-					//that the mouse was not moving when the mouse starts moving again
-					last_mouse_time=ins.time;
-					true
-				}
-			}else{
-				//mouse event
-				true
-			}{
-				//empty queue
-				while let Some(instruction)=timeline.pop_front(){
-					self.physics.run(instruction.time);
-					self.physics.process_instruction(TimedInstruction{
-						time:instruction.time,
-						instruction:crate::physics::PhysicsInstruction::Input(instruction.instruction),
-					});
-				}
-			}
-			if render{
-				self.graphics.render();
-			}
-		})
-	}
-}

src/settings.rs

@@ -1,11 +1,14 @@
-use crate::integer::{Ratio64,Ratio64Vec2};
+use strafesnet_common::integer::{Ratio64,Ratio64Vec2};
+#[derive(Clone)]
 struct Ratio{
 	ratio:f64,
 }
+#[derive(Clone)]
 enum DerivedFov{
 	FromScreenAspect,
 	FromAspect(Ratio),
 }
+#[derive(Clone)]
 enum Fov{
 	Exactly{x:f64,y:f64},
 	SpecifyXDeriveY{x:f64,y:DerivedFov},
@@ -16,9 +19,11 @@ impl Default for Fov{
 		Fov::SpecifyYDeriveX{x:DerivedFov::FromScreenAspect,y:1.0}
 	}
 }
+#[derive(Clone)]
 enum DerivedSensitivity{
 	FromRatio(Ratio64),
 }
+#[derive(Clone)]
 enum Sensitivity{
 	Exactly{x:Ratio64,y:Ratio64},
 	SpecifyXDeriveY{x:Ratio64,y:DerivedSensitivity},
@@ -30,7 +35,7 @@ impl Default for Sensitivity{
 	}
 }
 
-#[derive(Default)]
+#[derive(Default,Clone)]
 pub struct UserSettings{
 	fov:Fov,
 	sensitivity:Sensitivity,

src/setup.rs

@@ -0,0 +1,292 @@
+use crate::window::WindowInstruction;
+use strafesnet_common::instruction::TimedInstruction;
+use strafesnet_common::integer;
+
+fn optional_features()->wgpu::Features{
+	wgpu::Features::TEXTURE_COMPRESSION_ASTC
+	|wgpu::Features::TEXTURE_COMPRESSION_ETC2
+}
+fn required_features()->wgpu::Features{
+	wgpu::Features::TEXTURE_COMPRESSION_BC
+}
+fn required_downlevel_capabilities()->wgpu::DownlevelCapabilities{
+	wgpu::DownlevelCapabilities{
+		flags:wgpu::DownlevelFlags::empty(),
+		shader_model:wgpu::ShaderModel::Sm5,
+		..wgpu::DownlevelCapabilities::default()
+	}
+}
+pub fn required_limits()->wgpu::Limits{
+	wgpu::Limits::default()
+}
+
+struct SetupContextPartial1{
+	backends:wgpu::Backends,
+	instance:wgpu::Instance,
+}
+fn create_window(title:&str,event_loop:&winit::event_loop::EventLoop<()>)->Result<winit::window::Window,winit::error::OsError>{
+	let mut attr=winit::window::WindowAttributes::default();
+	attr=attr.with_title(title);
+	#[cfg(windows_OFF)] // TODO
+	{
+		use winit::platform::windows::WindowBuilderExtWindows;
+		builder=builder.with_no_redirection_bitmap(true);
+	}
+	event_loop.create_window(attr)
+}
+fn create_instance()->SetupContextPartial1{
+	let backends=wgpu::util::backend_bits_from_env().unwrap_or_else(wgpu::Backends::all);
+	let dx12_shader_compiler=wgpu::util::dx12_shader_compiler_from_env().unwrap_or_default();
+	SetupContextPartial1{
+		backends,
+		instance:wgpu::Instance::new(wgpu::InstanceDescriptor{
+			backends,
+			dx12_shader_compiler,
+			..Default::default()
+		}),
+	}
+}
+impl SetupContextPartial1{
+	fn create_surface<'a>(self,window:&'a winit::window::Window)->Result<SetupContextPartial2<'a>,wgpu::CreateSurfaceError>{
+		Ok(SetupContextPartial2{
+			backends:self.backends,
+			surface:self.instance.create_surface(window)?,
+			instance:self.instance,
+		})
+	}
+}
+struct SetupContextPartial2<'a>{
+	backends:wgpu::Backends,
+	instance:wgpu::Instance,
+	surface:wgpu::Surface<'a>,
+}
+impl<'a> SetupContextPartial2<'a>{
+	fn pick_adapter(self)->SetupContextPartial3<'a>{
+		let adapter;
+
+		//TODO: prefer adapter that implements optional features
+		//let optional_features=optional_features();
+		let required_features=required_features();
+
+		//no helper function smh gotta write it myself
+		let adapters=self.instance.enumerate_adapters(self.backends);
+
+		let mut chosen_adapter=None;
+		let mut chosen_adapter_score=0;
+		for adapter in adapters {
+			if !adapter.is_surface_supported(&self.surface) {
+				continue;
+			}
+
+			let score=match adapter.get_info().device_type{
+				wgpu::DeviceType::IntegratedGpu=>3,
+				wgpu::DeviceType::DiscreteGpu=>4,
+				wgpu::DeviceType::VirtualGpu=>2,
+				wgpu::DeviceType::Other|wgpu::DeviceType::Cpu=>1,
+			};
+
+			let adapter_features=adapter.features();
+			if chosen_adapter_score<score&&adapter_features.contains(required_features) {
+				chosen_adapter_score=score;
+				chosen_adapter=Some(adapter);
+			}
+		}
+
+		if let Some(maybe_chosen_adapter)=chosen_adapter{
+			adapter=maybe_chosen_adapter;
+		}else{
+			panic!("No suitable GPU adapters found on the system!");
+		}
+
+
+		let adapter_info=adapter.get_info();
+		println!("Using {} ({:?})", adapter_info.name, adapter_info.backend);
+
+		let required_downlevel_capabilities=required_downlevel_capabilities();
+		let downlevel_capabilities=adapter.get_downlevel_capabilities();
+		assert!(
+			downlevel_capabilities.shader_model >= required_downlevel_capabilities.shader_model,
+			"Adapter does not support the minimum shader model required to run this example: {:?}",
+			required_downlevel_capabilities.shader_model
+		);
+		assert!(
+			downlevel_capabilities
+				.flags
+				.contains(required_downlevel_capabilities.flags),
+			"Adapter does not support the downlevel capabilities required to run this example: {:?}",
+			required_downlevel_capabilities.flags - downlevel_capabilities.flags
+		);
+		SetupContextPartial3{
+			instance:self.instance,
+			surface:self.surface,
+			adapter,
+		}
+	}
+}
+struct SetupContextPartial3<'a>{
+	instance:wgpu::Instance,
+	surface:wgpu::Surface<'a>,
+	adapter:wgpu::Adapter,
+}
+impl<'a> SetupContextPartial3<'a>{
+	fn request_device(self)->SetupContextPartial4<'a>{
+		let optional_features=optional_features();
+		let required_features=required_features();
+
+		// Make sure we use the texture resolution limits from the adapter, so we can support images the size of the surface.
+		let needed_limits=required_limits().using_resolution(self.adapter.limits());
+
+		let trace_dir=std::env::var("WGPU_TRACE");
+		let (device, queue)=pollster::block_on(self.adapter
+			.request_device(
+				&wgpu::DeviceDescriptor {
+					label: None,
+					required_features: (optional_features & self.adapter.features()) | required_features,
+					required_limits: needed_limits,
+					memory_hints:wgpu::MemoryHints::Performance,
+				},
+				trace_dir.ok().as_ref().map(std::path::Path::new),
+			))
+			.expect("Unable to find a suitable GPU adapter!");
+
+		SetupContextPartial4{
+			instance:self.instance,
+			surface:self.surface,
+			adapter:self.adapter,
+			device,
+			queue,
+		}
+	}
+}
+struct SetupContextPartial4<'a>{
+	instance:wgpu::Instance,
+	surface:wgpu::Surface<'a>,
+	adapter:wgpu::Adapter,
+	device:wgpu::Device,
+	queue:wgpu::Queue,
+}
+impl<'a> SetupContextPartial4<'a>{
+	fn configure_surface(self,size:&'a winit::dpi::PhysicalSize<u32>)->SetupContext<'a>{
+		let mut config=self.surface
+			.get_default_config(&self.adapter, size.width, size.height)
+			.expect("Surface isn't supported by the adapter.");
+		let surface_view_format=config.format.add_srgb_suffix();
+		config.view_formats.push(surface_view_format);
+		config.present_mode=wgpu::PresentMode::AutoNoVsync;
+		self.surface.configure(&self.device, &config);
+
+		SetupContext{
+			instance:self.instance,
+			surface:self.surface,
+			device:self.device,
+			queue:self.queue,
+			config,
+		}
+	}
+}
+pub struct SetupContext<'a>{
+	pub instance:wgpu::Instance,
+	pub surface:wgpu::Surface<'a>,
+	pub device:wgpu::Device,
+	pub queue:wgpu::Queue,
+	pub config:wgpu::SurfaceConfiguration,
+}
+
+pub fn setup_and_start(title:String){
+	let event_loop=winit::event_loop::EventLoop::new().unwrap();
+
+	println!("Initializing the surface...");
+
+	let partial_1=create_instance();
+
+	let window=create_window(title.as_str(),&event_loop).unwrap();
+
+	let partial_2=partial_1.create_surface(&window).unwrap();
+
+	let partial_3=partial_2.pick_adapter();
+
+	let partial_4=partial_3.request_device();
+
+	let size=window.inner_size();
+
+	let setup_context=partial_4.configure_surface(&size);
+
+	//dedicated thread to ping request redraw back and resize the window doesn't seem logical
+
+	//the thread that spawns the physics thread
+	let mut window_thread=crate::window::worker(
+		&window,
+		setup_context,
+	);
+
+	if let Some(arg)=std::env::args().nth(1){
+		let path=std::path::PathBuf::from(arg);
+		window_thread.send(TimedInstruction{
+			time:integer::Time::ZERO,
+			instruction:WindowInstruction::WindowEvent(winit::event::WindowEvent::DroppedFile(path)),
+		}).unwrap();
+	};
+
+	println!("Entering event loop...");
+	let root_time=std::time::Instant::now();
+	run_event_loop(event_loop,window_thread,root_time).unwrap();
+}
+
+fn run_event_loop(
+	event_loop:winit::event_loop::EventLoop<()>,
+	mut window_thread:crate::compat_worker::QNWorker<TimedInstruction<WindowInstruction>>,
+	root_time:std::time::Instant
+	)->Result<(),winit::error::EventLoopError>{
+		event_loop.run(move |event,elwt|{
+			let time=integer::Time::from_nanos(root_time.elapsed().as_nanos() as i64);
+			// *control_flow=if cfg!(feature="metal-auto-capture"){
+			// 	winit::event_loop::ControlFlow::Exit
+			// }else{
+			// 	winit::event_loop::ControlFlow::Poll
+			// };
+			match event{
+				winit::event::Event::AboutToWait=>{
+					window_thread.send(TimedInstruction{time,instruction:WindowInstruction::RequestRedraw}).unwrap();
+				}
+				winit::event::Event::WindowEvent {
+					event:
+						// WindowEvent::Resized(size)
+						// | WindowEvent::ScaleFactorChanged {
+						// 	new_inner_size: &mut size,
+						// 	..
+						// },
+						winit::event::WindowEvent::Resized(size),//ignoring scale factor changed for now because mutex bruh
+					window_id:_,
+				} => {
+					window_thread.send(TimedInstruction{time,instruction:WindowInstruction::Resize(size)}).unwrap();
+				}
+				winit::event::Event::WindowEvent{event,..}=>match event{
+					winit::event::WindowEvent::KeyboardInput{
+						event:
+							winit::event::KeyEvent {
+							logical_key: winit::keyboard::Key::Named(winit::keyboard::NamedKey::Escape),
+								state: winit::event::ElementState::Pressed,
+								..
+							},
+						..
+					}
+					|winit::event::WindowEvent::CloseRequested=>{
+						elwt.exit();
+					}
+					winit::event::WindowEvent::RedrawRequested=>{
+						window_thread.send(TimedInstruction{time,instruction:WindowInstruction::Render}).unwrap();
+					}
+					_=>{
+						window_thread.send(TimedInstruction{time,instruction:WindowInstruction::WindowEvent(event)}).unwrap();
+					}
+				},
+				winit::event::Event::DeviceEvent{
+					event,
+					..
+				} => {
+					window_thread.send(TimedInstruction{time,instruction:WindowInstruction::DeviceEvent(event)}).unwrap();
+				},
+				_=>{}
+			}
+		})
+}

src/shader.wgsl

@@ -48,7 +48,7 @@ struct ModelInstance{
 //my fancy idea is to create a megatexture for each model that includes all the textures each intance will need
 //the texture transform then maps the texture coordinates to the location of the specific texture
 //group 1 is the model
-const MAX_MODEL_INSTANCES=4096;
+const MAX_MODEL_INSTANCES=512;
 @group(2)
 @binding(0)
 var<uniform> model_instances: array<ModelInstance, MAX_MODEL_INSTANCES>;

src/sweep.rs

@@ -1,8 +0,0 @@
-
-//something that implements body + hitbox + transform can predict collision
-impl crate::sweep::PredictCollision for Model {
-	fn predict_collision(&self,other:&Model) -> Option<crate::event::EventStruct> {
-		//math!
-		None
-	}
-}

src/window.rs

@@ -1,50 +1,71 @@
-pub struct WindowState{
-	//ok
-}
-impl WindowState{
-	fn resize(&mut self);
-	fn render(&self);
+use crate::physics_worker::InputInstruction;
+use strafesnet_common::integer;
+use strafesnet_common::instruction::TimedInstruction;
 
-	fn update(&mut self, window: &winit::window::Window, event: winit::event::WindowEvent) {
-		let time=integer::Time::from_nanos(self.start_time.elapsed().as_nanos() as i64);
+pub enum WindowInstruction{
+	Resize(winit::dpi::PhysicalSize<u32>),
+	WindowEvent(winit::event::WindowEvent),
+	DeviceEvent(winit::event::DeviceEvent),
+	RequestRedraw,
+	Render,
+}
+
+//holds thread handles to dispatch to
+struct WindowContext<'a>{
+	manual_mouse_lock:bool,
+	mouse:strafesnet_common::mouse::MouseState,//std::sync::Arc<std::sync::Mutex<>>
+	screen_size:glam::UVec2,
+	window:&'a winit::window::Window,
+	physics_thread:crate::compat_worker::QNWorker<'a, TimedInstruction<crate::physics_worker::Instruction>>,
+}
+
+impl WindowContext<'_>{
+	fn get_middle_of_screen(&self)->winit::dpi::PhysicalPosition<f32>{
+		winit::dpi::PhysicalPosition::new(self.screen_size.x as f32/2.0,self.screen_size.y as f32/2.0)
+	}
+	fn window_event(&mut self,time:integer::Time,event: winit::event::WindowEvent) {
 		match event {
 			winit::event::WindowEvent::DroppedFile(path)=>{
-				std::thread::spawn(move ||{
-					let indexed_model_instances=load_file(path);
-					self.render_thread.send(Instruction::Die(indexed_model_instances));
-				});
+				match crate::file::load(path.as_path()){
+					Ok(map)=>self.physics_thread.send(TimedInstruction{time,instruction:crate::physics_worker::Instruction::ChangeMap(map)}).unwrap(),
+					Err(e)=>println!("Failed to load map: {e}"),
+				}
 			},
 			winit::event::WindowEvent::Focused(state)=>{
 				//pause unpause
+				self.physics_thread.send(TimedInstruction{
+					time,
+					instruction:crate::physics_worker::Instruction::SetPaused(!state),
+				}).unwrap();
 				//recalculate pressed keys on focus
 			},
-			winit::event::WindowEvent::KeyboardInput {
-				input:winit::event::KeyboardInput{state, virtual_keycode,..},
+			winit::event::WindowEvent::KeyboardInput{
+				event:winit::event::KeyEvent{state,logical_key,repeat:false,..},
 				..
 			}=>{
-				let s=match state {
-					winit::event::ElementState::Pressed => true,
-					winit::event::ElementState::Released => false,
+				let s=match state{
+					winit::event::ElementState::Pressed=>true,
+					winit::event::ElementState::Released=>false,
 				};
-				match virtual_keycode{
-					Some(winit::event::VirtualKeyCode::Tab)=>{
+				match logical_key{
+					winit::keyboard::Key::Named(winit::keyboard::NamedKey::Tab)=>{
 						if s{
 							self.manual_mouse_lock=false;
-							match window.set_cursor_position(winit::dpi::PhysicalPosition::new(self.graphics.camera.screen_size.x as f32/2.0, self.graphics.camera.screen_size.y as f32/2.0)){
+							match self.window.set_cursor_position(self.get_middle_of_screen()){
 								Ok(())=>(),
 								Err(e)=>println!("Could not set cursor position: {:?}",e),
 							}
-							match window.set_cursor_grab(winit::window::CursorGrabMode::None){
+							match self.window.set_cursor_grab(winit::window::CursorGrabMode::None){
 								Ok(())=>(),
 								Err(e)=>println!("Could not release cursor: {:?}",e),
 							}
 						}else{
 							//if cursor is outside window don't lock but apparently there's no get pos function
 							//let pos=window.get_cursor_pos();
-							match window.set_cursor_grab(winit::window::CursorGrabMode::Locked){
+							match self.window.set_cursor_grab(winit::window::CursorGrabMode::Locked){
 								Ok(())=>(),
 								Err(_)=>{
-									match window.set_cursor_grab(winit::window::CursorGrabMode::Confined){
+									match self.window.set_cursor_grab(winit::window::CursorGrabMode::Confined){
 										Ok(())=>(),
 										Err(e)=>{
 											self.manual_mouse_lock=true;
@@ -54,62 +75,67 @@ impl WindowState{
 								}
 							}
 						}
-						window.set_cursor_visible(s);
+						self.window.set_cursor_visible(s);
 					},
-					Some(winit::event::VirtualKeyCode::F11)=>{
+					winit::keyboard::Key::Named(winit::keyboard::NamedKey::F11)=>{
 						if s{
-							if window.fullscreen().is_some(){
-								window.set_fullscreen(None);
+							if self.window.fullscreen().is_some(){
+								self.window.set_fullscreen(None);
 							}else{
-								window.set_fullscreen(Some(winit::window::Fullscreen::Borderless(None)));
+								self.window.set_fullscreen(Some(winit::window::Fullscreen::Borderless(None)));
 							}
 						}
 					},
-					Some(winit::event::VirtualKeyCode::Escape)=>{
+					winit::keyboard::Key::Named(winit::keyboard::NamedKey::Escape)=>{
 						if s{
 							self.manual_mouse_lock=false;
-							match window.set_cursor_grab(winit::window::CursorGrabMode::None){
+							match self.window.set_cursor_grab(winit::window::CursorGrabMode::None){
 								Ok(())=>(),
 								Err(e)=>println!("Could not release cursor: {:?}",e),
 							}
-							window.set_cursor_visible(true);
+							self.window.set_cursor_visible(true);
 						}
 					},
-					Some(keycode)=>{
-						if let Some(input_instruction)=match keycode {
-							winit::event::VirtualKeyCode::W => Some(InputInstruction::MoveForward(s)),
-							winit::event::VirtualKeyCode::A => Some(InputInstruction::MoveLeft(s)),
-							winit::event::VirtualKeyCode::S => Some(InputInstruction::MoveBack(s)),
-							winit::event::VirtualKeyCode::D => Some(InputInstruction::MoveRight(s)),
-							winit::event::VirtualKeyCode::E => Some(InputInstruction::MoveUp(s)),
-							winit::event::VirtualKeyCode::Q => Some(InputInstruction::MoveDown(s)),
-							winit::event::VirtualKeyCode::Space => Some(InputInstruction::Jump(s)),
-							winit::event::VirtualKeyCode::Z => Some(InputInstruction::Zoom(s)),
-							winit::event::VirtualKeyCode::R => if s{Some(InputInstruction::Reset)}else{None},
-							_ => None,
+					keycode=>{
+						if let Some(input_instruction)=match keycode{
+							winit::keyboard::Key::Named(winit::keyboard::NamedKey::Space)=>Some(InputInstruction::Jump(s)),
+							winit::keyboard::Key::Character(key)=>match key.as_str(){
+								"w"=>Some(InputInstruction::MoveForward(s)),
+								"a"=>Some(InputInstruction::MoveLeft(s)),
+								"s"=>Some(InputInstruction::MoveBack(s)),
+								"d"=>Some(InputInstruction::MoveRight(s)),
+								"e"=>Some(InputInstruction::MoveUp(s)),
+								"q"=>Some(InputInstruction::MoveDown(s)),
+								"z"=>Some(InputInstruction::Zoom(s)),
+								"r"=>if s{
+									//mouse needs to be reset since the position is absolute
+									self.mouse=strafesnet_common::mouse::MouseState::default();
+									Some(InputInstruction::ResetAndRestart)
+								}else{None},
+								"f"=>if s{Some(InputInstruction::PracticeFly)}else{None},
+								_=>None,
+							},
+							_=>None,
 						}{
 							self.physics_thread.send(TimedInstruction{
 								time,
-								instruction:input_instruction,
+								instruction:crate::physics_worker::Instruction::Input(input_instruction),
 							}).unwrap();
 						}
 					},
-					_=>(),
 				}
 			},
 			_=>(),
 		}
 	}
 
-	fn device_event(&mut self, window: &winit::window::Window, event: winit::event::DeviceEvent) {
-		//there's no way this is the best way get a timestamp.
-		let time=integer::Time::from_nanos(self.start_time.elapsed().as_nanos() as i64);
+	fn device_event(&mut self,time:integer::Time,event: winit::event::DeviceEvent) {
 		match event {
 			winit::event::DeviceEvent::MouseMotion {
 			    delta,//these (f64,f64) are integers on my machine
 			} => {
 				if self.manual_mouse_lock{
-					match window.set_cursor_position(winit::dpi::PhysicalPosition::new(self.graphics.camera.screen_size.x as f32/2.0, self.graphics.camera.screen_size.y as f32/2.0)){
+					match self.window.set_cursor_position(self.get_middle_of_screen()){
 						Ok(())=>(),
 						Err(e)=>println!("Could not set cursor position: {:?}",e),
 					}
@@ -121,7 +147,7 @@ impl WindowState{
 				self.mouse.pos+=delta;
 				self.physics_thread.send(TimedInstruction{
 					time,
-					instruction:InputInstruction::MoveMouse(self.mouse.pos),
+					instruction:crate::physics_worker::Instruction::Input(InputInstruction::MoveMouse(self.mouse.pos)),
 				}).unwrap();
 			},
 			winit::event::DeviceEvent::MouseWheel {
@@ -131,22 +157,67 @@ impl WindowState{
 				if false{//self.physics.style.use_scroll{
 					self.physics_thread.send(TimedInstruction{
 						time,
-						instruction:InputInstruction::Jump(true),//activates the immediate jump path, but the style modifier prevents controls&CONTROL_JUMP bit from being set to auto jump
+						instruction:crate::physics_worker::Instruction::Input(InputInstruction::Jump(true)),//activates the immediate jump path, but the style modifier prevents controls&CONTROL_JUMP bit from being set to auto jump
 					}).unwrap();
 				}
 			}
 			_=>(),
 		}
 	}
-
-	pub fn create_window(title:&str,event_loop:&winit::event_loop::EventLoop<()>)->Result<winit::window::Window,winit::error::OsError>{
-		let mut builder = winit::window::WindowBuilder::new();
-		builder = builder.with_title(title);
-		#[cfg(windows_OFF)] // TODO
-		{
-			use winit::platform::windows::WindowBuilderExtWindows;
-			builder = builder.with_no_redirection_bitmap(true);
-		}
-		builder.build(event_loop)
-	}
+}
+pub fn worker<'a>(
+	window:&'a winit::window::Window,
+	setup_context:crate::setup::SetupContext<'a>,
+)->crate::compat_worker::QNWorker<'a,TimedInstruction<WindowInstruction>>{
+	// WindowContextSetup::new
+		let user_settings=crate::settings::read_user_settings();
+
+		let mut graphics=crate::graphics::GraphicsState::new(&setup_context.device,&setup_context.queue,&setup_context.config);
+		graphics.load_user_settings(&user_settings);
+
+	//WindowContextSetup::into_context
+		let screen_size=glam::uvec2(setup_context.config.width,setup_context.config.height);
+		let graphics_thread=crate::graphics_worker::new(graphics,setup_context.config,setup_context.surface,setup_context.device,setup_context.queue);
+		let mut window_context=WindowContext{
+			manual_mouse_lock:false,
+			mouse:strafesnet_common::mouse::MouseState::default(),
+			//make sure to update this!!!!!
+			screen_size,
+			window,
+			physics_thread:crate::physics_worker::new(
+				graphics_thread,
+				user_settings,
+			),
+		};
+
+	//WindowContextSetup::into_worker
+		crate::compat_worker::QNWorker::new(move |ins:TimedInstruction<WindowInstruction>|{
+			match ins.instruction{
+				WindowInstruction::RequestRedraw=>{
+					window_context.window.request_redraw();
+				}
+				WindowInstruction::WindowEvent(window_event)=>{
+					window_context.window_event(ins.time,window_event);
+				},
+				WindowInstruction::DeviceEvent(device_event)=>{
+					window_context.device_event(ins.time,device_event);
+				},
+				WindowInstruction::Resize(size)=>{
+					window_context.physics_thread.send(
+						TimedInstruction{
+							time:ins.time,
+							instruction:crate::physics_worker::Instruction::Resize(size)
+						}
+					).unwrap();
+				}
+				WindowInstruction::Render=>{
+					window_context.physics_thread.send(
+						TimedInstruction{
+							time:ins.time,
+							instruction:crate::physics_worker::Instruction::Render
+						}
+					).unwrap();
+				}
+			}
+		})
 }

src/worker.rs

@@ -1,6 +1,6 @@
 use std::thread;
 use std::sync::{mpsc,Arc};
-use parking_lot::{Mutex,Condvar};
+use parking_lot::Mutex;
 
 //WorkerPool
 struct Pool(u32);
@@ -104,17 +104,15 @@ QN = WorkerDescription{
 }
 */
 //None Output Worker does all its work internally from the perspective of the work submitter
-pub struct QNWorker<Task:Send> {
+pub struct QNWorker<'a,Task:Send>{
 	sender: mpsc::Sender<Task>,
+	handle:thread::ScopedJoinHandle<'a,()>,
 }
 
-impl<Task:Send+'static> QNWorker<Task>{
-	pub fn new<F:FnMut(Task)+Send+'static>(mut f:F)->Self{
+impl<'a,Task:Send+'a> QNWorker<'a,Task>{
+	pub fn new<F:FnMut(Task)+Send+'a>(scope:&'a thread::Scope<'a,'_>,mut f:F)->QNWorker<'a,Task>{
 		let (sender,receiver)=mpsc::channel::<Task>();
-		let ret=Self {
-			sender,
-		};
-		thread::spawn(move ||{
+		let handle=scope.spawn(move ||{
 			loop {
 				match receiver.recv() {
 					Ok(task)=>f(task),
@@ -125,7 +123,10 @@ impl<Task:Send+'static> QNWorker<Task>{
 				}
 			}
 		});
-		ret
+		Self{
+			sender,
+			handle,
+		}
 	}
 	pub fn send(&self,task:Task)->Result<(),mpsc::SendError<Task>>{
 		self.sender.send(task)
@@ -139,120 +140,57 @@ IN = WorkerDescription{
 }
 */
 //Inputs are dropped if the worker is busy
-pub struct INWorker<Task:Clone>{
-	input:Arc<(Mutex<Task>,Condvar)>,
+pub struct INWorker<'a,Task:Send>{
+	sender: mpsc::SyncSender<Task>,
+	handle:thread::ScopedJoinHandle<'a,()>,
 }
 
-impl<Task:Clone+Send+'static> INWorker<Task>{
-	pub fn new<F:FnMut(Task)+Send+'static>(task:Task,mut f:F)->Self{
-		let ret=Self {
-			input:Arc::new((Mutex::new(task),Condvar::new())),
-		};
-		let input=ret.input.clone();
-		thread::spawn(move ||{
+impl<'a,Task:Send+'a> INWorker<'a,Task>{
+	pub fn new<F:FnMut(Task)+Send+'a>(scope:&'a thread::Scope<'a,'_>,mut f:F)->INWorker<'a,Task>{
+		let (sender,receiver)=mpsc::sync_channel::<Task>(1);
+		let handle=scope.spawn(move ||{
 			loop {
-				input.1.wait(&mut input.0.lock());
-				f(input.0.lock().clone());
-			}
-		});
-		ret
-	}
-	pub fn send(&self,task:Task){
-		*self.input.0.lock()=task;
-        self.input.1.notify_one();
-	}
-}
-
-//worker pools work by cloning a mpsc and passing it into the thread, the thread sends its results through that
-//worker pools have a master thread that manages the pool so that the work submission thread does not need to implement async
-
-pub struct QQWorkerPool<Task:Send,Value:Send>{
-	sender:mpsc::Sender<Task>,
-	queue:Arc<Mutex<std::collections::VecDeque<Value>>>,
-}
-
-impl<Task:Send+'static,Value:Send+'static> QQWorkerPool<Task,Value>{
-	pub fn new<F:Fn(Task)->Value+Send+'static>(pool_size:usize,f:F)->Self{
-		let (task_sender,task_receiver)=mpsc::channel::<Task>();
-		let (value_sender,value_receiver)=mpsc::channel::<Value>();
-		let ret=Self{
-			sender:task_sender,
-			queue:Arc::new(Mutex::new(std::collections::VecDeque::new())),
-		};
-		let mut queue_collect=ret.queue.clone();
-		let mut worker_senders=Vec::with_capacity(pool_size);
-		let mut active_workers_collect=Arc::new(Mutex::new(0usize));
-		let mut active_workers_dispatch=active_workers_collect.clone();
-		let mut condvar_collect=Arc::new(Condvar::new());
-		let mut condvar_dispatch=condvar_collect.clone();
-		//task dispatch thread
-		thread::spawn(move ||{
-			loop{
-				//block if no workers are available, value collection thread will notify
-				let n=*active_workers_dispatch.lock();
-				if n==pool_size{
-					//wait for notifiy
-					condvar_dispatch.wait(&mut active_workers_dispatch.lock());
-				}
-				match task_receiver.recv(){
-					Ok(task)=>{
-						*active_workers_dispatch.lock()+=1;
-						//workers are never full here
-						//else if workers busy: spawn a new worker
-						//else: send task to an available worker
-					}
+				match receiver.recv() {
+					Ok(task)=>f(task),
 					Err(_)=>{
-						println!("Dispatch stopping.",);
+						println!("Worker stopping.",);
 						break;
 					}
 				}
 			}
 		});
-		//value collection thread (bad idea, put this logic in each thread)
-		thread::spawn(move ||{
-			loop{
-				match value_receiver.recv(){
-					Ok(value)=>{
-						//maybe I can be smart with this as a signal that a worker finished a task
-						let n=*active_workers_collect.lock();
-						*active_workers_collect.lock()-=1;
-						if n==pool_size{
-							condvar_collect.notify_one();
-						}
-						(*queue_collect.lock()).push_front(value);
-					}
-					Err(_)=>{
-						println!("Collection stopping.",);
-						break;
+		Self{
+			sender,
+			handle,
 		}
 	}
-			}
-		});
-		ret
-	}
-
-	pub fn send(&self,task:Task)->Result<(),mpsc::SendError<Task>>{
+	//blocking!
+	pub fn blocking_send(&self,task:Task)->Result<(), mpsc::SendError<Task>>{
 		self.sender.send(task)
 	}
-
-	pub fn get(&self)->Option<Value>{
-		(*self.queue.lock()).pop_back()
+	pub fn send(&self,task:Task)->Result<(), mpsc::TrySendError<Task>>{
+		self.sender.try_send(task)
 	}
 }
 
-#[test]//How to run this test with printing: cargo test --release -- --nocapture
-fn test_worker() {
-	println!("hiiiii");
+#[cfg(test)]
+mod test{
+	use super::{thread,QRWorker};
+	use crate::physics;
+	use strafesnet_common::{integer,instruction};
+	#[test]//How to run this test with printing: cargo test --release -- --nocapture
+	fn test_worker() {
 		// Create the worker thread
-	let worker=QRWorker::new(crate::physics::Body::with_pva(crate::integer::Planar64Vec3::ZERO,crate::integer::Planar64Vec3::ZERO,crate::integer::Planar64Vec3::ZERO),
-		|_|crate::physics::Body::with_pva(crate::integer::Planar64Vec3::ONE,crate::integer::Planar64Vec3::ONE,crate::integer::Planar64Vec3::ONE)
+		let test_body=physics::Body::new(integer::vec3::ONE,integer::vec3::ONE,integer::vec3::ONE,integer::Time::ZERO);
+		let worker=QRWorker::new(physics::Body::ZERO,
+			|_|physics::Body::new(integer::vec3::ONE,integer::vec3::ONE,integer::vec3::ONE,integer::Time::ZERO)
 		);
 
 		// Send tasks to the worker
 		for _ in 0..5 {
-		let task = crate::instruction::TimedInstruction{
-			time:crate::integer::Time::ZERO,
-			instruction:crate::physics::PhysicsInstruction::StrafeTick,
+			let task = instruction::TimedInstruction{
+				time:integer::Time::ZERO,
+				instruction:strafesnet_common::physics::Instruction::Idle,
 			};
 			worker.send(task).unwrap();
 		}
@@ -261,17 +199,18 @@ fn test_worker() {
 		// sender.send("STOP".to_string()).unwrap();
 
 		// Sleep to allow the worker thread to finish processing
-	thread::sleep(std::time::Duration::from_secs(2));
+		thread::sleep(std::time::Duration::from_millis(10));
 
 		// Send a new task
-	let task = crate::instruction::TimedInstruction{
-		time:crate::integer::Time::ZERO,
-		instruction:crate::physics::PhysicsInstruction::StrafeTick,
+		let task = instruction::TimedInstruction{
+			time:integer::Time::ZERO,
+			instruction:strafesnet_common::physics::Instruction::Idle,
 		};
 		worker.send(task).unwrap();
 
-	println!("value={}",worker.grab_clone());
+		//assert_eq!(test_body,worker.grab_clone());
 
 		// wait long enough to see print from final task
-	thread::sleep(std::time::Duration::from_secs(1));
+		thread::sleep(std::time::Duration::from_millis(10));
+	}
 }

src/zeroes.rs

@@ -1,32 +0,0 @@
-//find roots of polynomials
-use crate::integer::Planar64;
-
-#[inline]
-pub fn zeroes2(a0:Planar64,a1:Planar64,a2:Planar64) -> Vec<Planar64>{
-	if a2==Planar64::ZERO{
-		return zeroes1(a0, a1);
-	}
-	let radicand=a1.get() as i128*a1.get() as i128-a2.get() as i128*a0.get() as i128*4;
-	if 0<radicand {
-		//start with f64 sqrt
-		let planar_radicand=Planar64::raw(unsafe{(radicand as f64).sqrt().to_int_unchecked()});
-		//TODO: one or two newtons
-		if Planar64::ZERO<a2 {
-			return vec![(-a1-planar_radicand)/(a2*2),(-a1+planar_radicand)/(a2*2)];
-		} else {
-			return vec![(-a1+planar_radicand)/(a2*2),(-a1-planar_radicand)/(a2*2)];
-		}
-	} else if radicand==0 {
-		return vec![a1/(a2*-2)];
-	} else {
-		return vec![];
-	}
-}
-#[inline]
-pub fn zeroes1(a0:Planar64,a1:Planar64) -> Vec<Planar64> {
-	if a1==Planar64::ZERO{
-		return vec![];
-	} else {
-		return vec![-a0/a1];
-	}
-}

tools/arcane

@@ -0,0 +1 @@
+mangohud ../target/release/strafe-client bhop_maps/5692113331.snfm

tools/bhop_maps

@@ -0,0 +1 @@
+/run/media/quat/Files/Documents/map-files/verify-scripts/maps/bhop_snfm

tools/cross-compile.sh

@@ -0,0 +1 @@
+cargo build --release --target x86_64-pc-windows-gnu --all-features

tools/iso

@@ -0,0 +1 @@
+mangohud ../target/release/strafe-client bhop_maps/5692124338.snfm

tools/make-demo.sh

@@ -0,0 +1,4 @@
+mkdir -p ../target/demo
+mv ../target/x86_64-pc-windows-gnu/release/strafe-client.exe ../target/demo/strafe-client.exe
+rm ../target/demo.7z
+7z a -t7z -mx=9 -mfb=273 -ms -md=31 -myx=9 -mtm=- -mmt -mmtf -md=1536m -mmf=bt3 -mmc=10000 -mpb=0 -mlc=0 ../target/demo.7z ../target/demo

tools/run

@@ -0,0 +1 @@
+mangohud ../target/release/strafe-client "$1"

tools/settings.conf

@@ -0,0 +1,4 @@
+[camera]
+sensitivity_x=98384
+fov_y=1.0
+#fov_x_from_y_ratio=1.33333333333333333333333333333333

tools/surf_maps

@@ -0,0 +1 @@
+/run/media/quat/Files/Documents/map-files/verify-scripts/maps/surf_snfm

tools/toc

@@ -0,0 +1 @@
+mangohud ../target/release/strafe-client bhop_maps/5692152916.snfm

tools/utopia

@@ -0,0 +1 @@
+mangohud ../target/release/strafe-client surf_maps/5692145408.snfm