This PR makes a number of changes to how meshes and vertex attributes are handled, which the goal of enabling easy and flexible custom vertex attributes:
* Reworks the `Mesh` type to use the newly added `VertexAttribute` internally
* `VertexAttribute` defines the name, a unique `VertexAttributeId`, and a `VertexFormat`
* `VertexAttributeId` is used to produce consistent sort orders for vertex buffer generation, replacing the more expensive and often surprising "name based sorting"
* Meshes can be used to generate a `MeshVertexBufferLayout`, which defines the layout of the gpu buffer produced by the mesh. `MeshVertexBufferLayouts` can then be used to generate actual `VertexBufferLayouts` according to the requirements of a specific pipeline. This decoupling of "mesh layout" vs "pipeline vertex buffer layout" is what enables custom attributes. We don't need to standardize _mesh layouts_ or contort meshes to meet the needs of a specific pipeline. As long as the mesh has what the pipeline needs, it will work transparently.
* Mesh-based pipelines now specialize on `&MeshVertexBufferLayout` via the new `SpecializedMeshPipeline` trait (which behaves like `SpecializedPipeline`, but adds `&MeshVertexBufferLayout`). The integrity of the pipeline cache is maintained because the `MeshVertexBufferLayout` is treated as part of the key (which is fully abstracted from implementers of the trait ... no need to add any additional info to the specialization key).
* Hashing `MeshVertexBufferLayout` is too expensive to do for every entity, every frame. To make this scalable, I added a generalized "pre-hashing" solution to `bevy_utils`: `Hashed<T>` keys and `PreHashMap<K, V>` (which uses `Hashed<T>` internally) . Why didn't I just do the quick and dirty in-place "pre-compute hash and use that u64 as a key in a hashmap" that we've done in the past? Because its wrong! Hashes by themselves aren't enough because two different values can produce the same hash. Re-hashing a hash is even worse! I decided to build a generalized solution because this pattern has come up in the past and we've chosen to do the wrong thing. Now we can do the right thing! This did unfortunately require pulling in `hashbrown` and using that in `bevy_utils`, because avoiding re-hashes requires the `raw_entry_mut` api, which isn't stabilized yet (and may never be ... `entry_ref` has favor now, but also isn't available yet). If std's HashMap ever provides the tools we need, we can move back to that. Note that adding `hashbrown` doesn't increase our dependency count because it was already in our tree. I will probably break these changes out into their own PR.
* Specializing on `MeshVertexBufferLayout` has one non-obvious behavior: it can produce identical pipelines for two different MeshVertexBufferLayouts. To optimize the number of active pipelines / reduce re-binds while drawing, I de-duplicate pipelines post-specialization using the final `VertexBufferLayout` as the key. For example, consider a pipeline that needs the layout `(position, normal)` and is specialized using two meshes: `(position, normal, uv)` and `(position, normal, other_vec2)`. If both of these meshes result in `(position, normal)` specializations, we can use the same pipeline! Now we do. Cool!
To briefly illustrate, this is what the relevant section of `MeshPipeline`'s specialization code looks like now:
```rust
impl SpecializedMeshPipeline for MeshPipeline {
type Key = MeshPipelineKey;
fn specialize(
&self,
key: Self::Key,
layout: &MeshVertexBufferLayout,
) -> RenderPipelineDescriptor {
let mut vertex_attributes = vec![
Mesh::ATTRIBUTE_POSITION.at_shader_location(0),
Mesh::ATTRIBUTE_NORMAL.at_shader_location(1),
Mesh::ATTRIBUTE_UV_0.at_shader_location(2),
];
let mut shader_defs = Vec::new();
if layout.contains(Mesh::ATTRIBUTE_TANGENT) {
shader_defs.push(String::from("VERTEX_TANGENTS"));
vertex_attributes.push(Mesh::ATTRIBUTE_TANGENT.at_shader_location(3));
}
let vertex_buffer_layout = layout
.get_layout(&vertex_attributes)
.expect("Mesh is missing a vertex attribute");
```
Notice that this is _much_ simpler than it was before. And now any mesh with any layout can be used with this pipeline, provided it has vertex postions, normals, and uvs. We even got to remove `HAS_TANGENTS` from MeshPipelineKey and `has_tangents` from `GpuMesh`, because that information is redundant with `MeshVertexBufferLayout`.
This is still a draft because I still need to:
* Add more docs
* Experiment with adding error handling to mesh pipeline specialization (which would print errors at runtime when a mesh is missing a vertex attribute required by a pipeline). If it doesn't tank perf, we'll keep it.
* Consider breaking out the PreHash / hashbrown changes into a separate PR.
* Add an example illustrating this change
* Verify that the "mesh-specialized pipeline de-duplication code" works properly
Please dont yell at me for not doing these things yet :) Just trying to get this in peoples' hands asap.
Alternative to #3120Fixes#3030
Co-authored-by: Carter Anderson <mcanders1@gmail.com>
# Objective
`all_tuples` panics when the start count is set to anything other than 0 or 1. Fix this bug.
## Solution
Originally part of #2381, this PR fixes the slice indexing used by the proc macro.
# Objective
- Fixes#4005
## Solution
- Include the `near` and `far` clipping values from the perspective projection in the `Camera` struct; before that, they were both being defaulted to 0.
# Context
I wanted to add a `texture` to my `ColorMaterial` without explicitly adding a `color`. To do this I used `..Default::default()` which in turn gave me unexpected results. I was expecting that my texture would render without any color modifications, but to my surprise it got rendered in a purple tint (`Color::rgb(1.0, 0.0, 1.0)`). To fix this I had to explicitly define the `color` using `color: Color::WHITE`.
## What I wanted to use
```rust
commands
.spawn_bundle(MaterialMesh2dBundle {
mesh: mesh_handle.clone().into(),
transform: Transform::default().with_scale(Vec3::splat(8.)),
material: materials.add(ColorMaterial {
texture: Some(texture_handle.clone()),
..Default::default() // here
}),
..Default::default()
})
```
## What I had to use instead
```rust
commands
.spawn_bundle(MaterialMesh2dBundle {
mesh: mesh_handle.clone().into(),
transform: Transform::default().with_scale(Vec3::splat(8.)),
material: materials.add(ColorMaterial {
texture: Some(texture_handle.clone()),
color: Color::WHITE, // here
}),
..Default::default()
})
```
Adds "hot reloading" of internal assets, which is normally not possible because they are loaded using `include_str` / direct Asset collection access.
This is accomplished via the following:
* Add a new `debug_asset_server` feature flag
* When that feature flag is enabled, create a second App with a second AssetServer that points to a configured location (by default the `crates` folder). Plugins that want to add hot reloading support for their assets can call the new `app.add_debug_asset::<T>()` and `app.init_debug_asset_loader::<T>()` functions.
* Load "internal" assets using the new `load_internal_asset` macro. By default this is identical to the current "include_str + register in asset collection" approach. But if the `debug_asset_server` feature flag is enabled, it will also load the asset dynamically in the debug asset server using the file path. It will then set up a correlation between the "debug asset" and the "actual asset" by listening for asset change events.
This is an alternative to #3673. The goal was to keep the boilerplate and features flags to a minimum for bevy plugin authors, and allow them to home their shaders near relevant code.
This is a draft because I haven't done _any_ quality control on this yet. I'll probably rename things and remove a bunch of unwraps. I just got it working and wanted to use it to start a conversation.
Fixes#3660
This enables shaders to (optionally) define their import path inside their source. This has a number of benefits:
1. enables users to define their own custom paths directly in their assets
2. moves the import path "close" to the asset instead of centralized in the plugin definition, which seems "better" to me.
3. makes "internal hot shader reloading" way more reasonable (see #3966)
4. logically opens the door to importing "parts" of a shader by defining "import_path blocks".
```rust
#define_import_path bevy_pbr::mesh_struct
struct Mesh {
model: mat4x4<f32>;
inverse_transpose_model: mat4x4<f32>;
// 'flags' is a bit field indicating various options. u32 is 32 bits so we have up to 32 options.
flags: u32;
};
let MESH_FLAGS_SHADOW_RECEIVER_BIT: u32 = 1u;
```
For some keys, it is too expensive to hash them on every lookup. Historically in Bevy, we have regrettably done the "wrong" thing in these cases (pre-computing hashes, then re-hashing them) because Rust's built in hashed collections don't give us the tools we need to do otherwise. Doing this is "wrong" because two different values can result in the same hash. Hashed collections generally get around this by falling back to equality checks on hash collisions. You can't do that if the key _is_ the hash. Additionally, re-hashing a hash increase the odds of collision!
#3959 needs pre-hashing to be viable, so I decided to finally properly solve the problem. The solution involves two different changes:
1. A new generalized "pre-hashing" solution in bevy_utils: `Hashed<T>` types, which store a value alongside a pre-computed hash. And `PreHashMap<K, V>` (which uses `Hashed<T>` internally) . `PreHashMap` is just an alias for a normal HashMap that uses `Hashed<T>` as the key and a new `PassHash` implementation as the Hasher.
2. Replacing the `std::collections` re-exports in `bevy_utils` with equivalent `hashbrown` impls. Avoiding re-hashes requires the `raw_entry_mut` api, which isn't stabilized yet (and may never be ... `entry_ref` has favor now, but also isn't available yet). If std's HashMap ever provides the tools we need, we can move back to that. The latest version of `hashbrown` adds support for the `entity_ref` api, so we can move to that in preparation for an std migration, if thats the direction they seem to be going in. Note that adding hashbrown doesn't increase our dependency count because it was already in our tree.
In addition to providing these core tools, I also ported the "table identity hashing" in `bevy_ecs` to `raw_entry_mut`, which was a particularly egregious case.
The biggest outstanding case is `AssetPathId`, which stores a pre-hash. We need AssetPathId to be cheaply clone-able (and ideally Copy), but `Hashed<AssetPath>` requires ownership of the AssetPath, which makes cloning ids way more expensive. We could consider doing `Hashed<Arc<AssetPath>>`, but cloning an arc is still a non-trivial expensive that needs to be considered. I would like to handle this in a separate PR. And given that we will be re-evaluating the Bevy Assets implementation in the very near future, I'd prefer to hold off until after that conversation is concluded.
Features must be called with the crate, otherwise the following error is thrown:
> error: none of the selected packages contains these features: trace_chrome
# Objective
The `custom_material.vert` shader used by the `shader_material_glsl` example is missing a `mat4 View` in `CameraViewProj` (added in [#3885](https://github.com/bevyengine/bevy/pull/3885))
## Solution
Update the definition of `CameraViewProj`
# Objective
- `WgpuOptions` is mutated to be updated with the actual device limits and features, but this information is readily available to both the main and render worlds through the `RenderDevice` which has .limits() and .features() methods
- Information about the adapter in terms of its name, the backend in use, etc were not being exposed but have clear use cases for being used to take decisions about what rendering code to use. For example, if something works well on AMD GPUs but poorly on Intel GPUs. Or perhaps something works well in Vulkan but poorly in DX12.
## Solution
- Stop mutating `WgpuOptions `and don't insert the updated values into the main and render worlds
- Return `AdapterInfo` from `initialize_renderer` and insert it into the main and render worlds
- Use `RenderDevice` limits in the lighting code that was using `WgpuOptions.limits`.
- Renamed `WgpuOptions` to `WgpuSettings`
I wanted to try one of the new examples but it felt so clunky that I wanted to improve it.
It did make me feel like maybe some input axes abstraction like Unity has might be useful.
Also, eating cake should probably be a separate system from movement.
# Objective
- `SystemStates` rock for dealing with exclusive world access, but are hard to figure out how to use.
- Fixes#3341.
## Solution
- Clearly document how to use `SystemState`, and why they're useful as an end-user.
What is says on the tin.
This has got more to do with making `clippy` slightly more *quiet* than it does with changing anything that might greatly impact readability or performance.
that said, deriving `Default` for a couple of structs is a nice easy win
# Objective
- Support overriding wgpu features and limits that were calculated from default values or queried from the adapter/backend.
- Fixes#3686
## Solution
- Add `disabled_features: Option<wgpu::Features>` to `WgpuOptions`
- Add `constrained_limits: Option<wgpu::Limits>` to `WgpuOptions`
- After maybe obtaining updated features and limits from the adapter/backend in the case of `WgpuOptionsPriority::Functionality`, enable the `WgpuOptions` `features`, disable the `disabled_features`, and constrain the `limits` by `constrained_limits`.
- Note that constraining the limits means for `wgpu::Limits` members named `max_.*` we take the minimum of that which was configured/queried for the backend/adapter and the specified constrained limit value. This means the configured/queried value is used if the constrained limit is larger as that is as much as the device/API supports, or the constrained limit value is used if it is smaller as we are imposing an artificial constraint. For members named `min_.*` we take the maximum instead. For example, a minimum stride might be 256 but we set constrained limit value of 1024, then 1024 is the more conservative value. If the constrained limit value were 16, then 256 would be the more conservative.
On platforms like wasm (on mobile) the cursor can disappear suddenly (ex: the user releases their finger from the screen). This causes the undesirable behavior in #3752. These changes make the UI handler properly handle this case.
Fixes#3752
Alternative to #3599
# Objective
If a user attempts to `.add_render_command::<P, C>()` on a world that does not contain `DrawFunctions<P>`, the engine panics with a generic `Option::unwrap` message:
```
thread 'main' panicked at 'called `Option::unwrap()` on a `None` value', /[redacted]/bevy/crates/bevy_render/src/render_phase/draw.rs:318:76
```
## Solution
This PR adds a panic message describing the problem:
```
thread 'main' panicked at 'DrawFunctions<outline::MeshStencil> must be added to the world as a resource before adding render commands to it', /[redacted]/bevy/crates/bevy_render/src/render_phase/draw.rs:322:17
```
# Objective
The documentation was unclear but it seemed like it was intended to _only_ flip the texture coordinates of the quad. However, it was also swapping the vertex positions, which resulted in inverted winding order so the front became a back face, and the normal was pointing into the face instead of out of it.
## Solution
- This change makes the only difference the UVs being horizontally flipped.
# Objective
Fix `SetSpriteTextureBindGroup` to use index instead of hard coded 1.
Fixes#3895
## Solution
1 -> I
Co-authored-by: devjobe <git@devjobe.com>
# Objective
- Fixes#3078
- Fixes#1397
## Solution
- Implement Commands::init_resource.
- Also implement for World, for consistency and to simplify internal structure.
- While we're here, clean up some of the docs for Command and World resource modification.
# Objective
My attempt at fixing #2142. My very first attempt at contributing to Bevy so more than open to any feedback.
I borrowed heavily from the [Bevy Cheatbook page](https://bevy-cheatbook.github.io/patterns/generic-systems.html?highlight=generic#generic-systems).
## Solution
Fairly straightforward example using a clean up system to delete entities that are coupled with app state after exiting that state.
Co-authored-by: B-Janson <brandon@canva.com>
(cherry picked from commit de943381bd2a8b242c94db99e6c7bbd70006d7c3)
# Objective
The view uniform lacks view transform information. The inverse transform is currently provided but this is not sufficient if you do not have access to an `inverse` function (such as in WGSL).
## Solution
Grab the view transform, put it in the view uniform, use the same matrix to compute the inverse as well.
# Objective
When I use the latest winit with bevy main, I got this error.
```
error[E0308]: mismatched types
--> /Users/ryo/.cargo/git/checkouts/bevy-f7ffde730c324c74/b13f238/crates/bevy_winit/src/lib.rs:191:5
|
191 | event_loop.run_return(event_handler)
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ expected `()`, found `i32`
|
help: consider using a semicolon here
|
191 | event_loop.run_return(event_handler);
| +
help: try adding a return type
|
187 | -> i32 where
| ++++++
```
In [this commit](a52f755ce8), the signature of `run_return` was changed in winit.
## Solution
This tiny PR does not add support for exit code, but makes compilation successful.
# Objective
- `CoreStage::Startup` is unique in the `CoreStage` enum, in that it represents a `Schedule` and not a `SystemStage`.
- This can lead to confusion about how `CoreStage::Startup` and the `StartupStage` enum are related.
- Beginners sometimes try `.add_system_to_stage(CoreStage::Startup, setup.system())` instead of `.add_startup_system(setup.system())`, which causes a Panic:
```
thread 'main' panicked at 'Stage 'Startup' does not exist or is not a SystemStage', crates\bevy_ecs\src\schedule\mod.rs:153:13
stack backtrace:
0: std::panicking::begin_panic_handler
at /rustc/53cb7b09b00cbea8754ffb78e7e3cb521cb8af4b\/library\std\src\panicking.rs:493
1: std::panicking::begin_panic_fmt
at /rustc/53cb7b09b00cbea8754ffb78e7e3cb521cb8af4b\/library\std\src\panicking.rs:435
2: bevy_ecs::schedule::{{impl}}::add_system_to_stage::stage_not_found
at .\crates\bevy_ecs\src\schedule\mod.rs:153
3: bevy_ecs::schedule::{{impl}}::add_system_to_stage::{{closure}}<tuple<bevy_ecs::system::function_system::IsFunctionSystem, tuple<bevy_ecs::system::commands::Commands, bevy_ecs::change_detection::ResMut<bevy_asset::assets::Assets<bevy_render::mesh::mesh::Me
at .\crates\bevy_ecs\src\schedule\mod.rs:161
4: core::option::Option<mut bevy_ecs::schedule::stage::SystemStage*>::unwrap_or_else<mut bevy_ecs::schedule::stage::SystemStage*,closure-0>
at C:\Users\scher\.rustup\toolchains\stable-x86_64-pc-windows-msvc\lib\rustlib\src\rust\library\core\src\option.rs:427
5: bevy_ecs::schedule::Schedule::add_system_to_stage<tuple<bevy_ecs::system::function_system::IsFunctionSystem, tuple<bevy_ecs::system::commands::Commands, bevy_ecs::change_detection::ResMut<bevy_asset::assets::Assets<bevy_render::mesh::mesh::Mesh>>, bevy_ec
at .\crates\bevy_ecs\src\schedule\mod.rs:159
6: bevy_app::app_builder::AppBuilder::add_system_to_stage<tuple<bevy_ecs::system::function_system::IsFunctionSystem, tuple<bevy_ecs::system::commands::Commands, bevy_ecs::change_detection::ResMut<bevy_asset::assets::Assets<bevy_render::mesh::mesh::Mesh>>, be
at .\crates\bevy_app\src\app_builder.rs:196
7: 3d_scene::main
at .\examples\3d\3d_scene.rs:4
8: core::ops::function::FnOnce::call_once<fn(),tuple<>>
at C:\Users\scher\.rustup\toolchains\stable-x86_64-pc-windows-msvc\lib\rustlib\src\rust\library\core\src\ops\function.rs:227
```
## Solution
- Replace the `CoreStage::Startup` Label with the new `StartupSchedule` unit type.
Resolves#2229
# Objective
The docs for `{VertexState, FragmentState}::entry_point` stipulate that the entry point function in the shader must return void. This seems to be specific to GLSL; WGSL has no `void` type and its entry point functions return values that describe their output.
## Solution
Remove the mention of the `void` return type.
# Objective
- Currently, when getting a diagnostic value, the oldest value is returned. This is not the best for a diagnostic with a large history, as you could get a value from several frames away
## Solution
- I changed the order in which the history is used to follow ["The “default” usage of this type as a queue is to use push_back to add to the queue, and pop_front to remove from the queue."](https://doc.rust-lang.org/std/collections/vec_deque/struct.VecDeque.html)
# Objective
- `serde_json` assumes that numbers being deserialized are either u64 or i64.
- `Entity` serializes and deserializes as a u32.
- Deserializing an `Entity` with `serde_json` fails with: `Error("invalid type: integer 10947, expected expected Entity"`
## Solution
- Implemented a visitor for u64 that allows an `Entity` to be deserialized in this case.
- While I was here, also fixed the redundant "expected expected Entity" in the error message
- Tested the change in a local project which now correctly deserializes `Entity` structs with `serde_json` when it couldn't before
# Objective
- Fix#3559
- Avoid erasing existing resource `Assets<T>` when adding it twice
## Solution
- Before creating a new `Assets<T>`, check if it has already been added to the world
Co-authored-by: François <8672791+mockersf@users.noreply.github.com>
Co-authored-by: Aevyrie Roessler <aevyrie@gmail.com>
# Objective
- Bevy currently has no simple way to make an "empty" Entity work correctly in a Hierachy.
- The current Solution is to insert a Tuple instead:
```rs
.insert_bundle((Transform::default(), GlobalTransform::default()))
```
## Solution
* Add a `TransformBundle` that combines the Components:
```rs
.insert_bundle(TransformBundle::default())
```
* The code is based on #2331, except for missing the more controversial usage of `TransformBundle` as a Sub-bundle in preexisting Bundles.
Co-authored-by: MinerSebas <66798382+MinerSebas@users.noreply.github.com>
Co-authored-by: Carter Anderson <mcanders1@gmail.com>
Implements a new Queryable called AnyOf, which will return an item as long as at least one of it's requested Queryables returns something. For example, a `Query<AnyOf<(&A, &B, &C)>>` will return items with type `(Option<&A>, Option<&B>, Option<&C>)`, and will guarantee that for every element at least one of the option s is Some. This is a shorthand for queries like `Query<(Option<&A>, Option<&B>, Option<&C>), Or<(With<A>, With<B>, With&C>)>>`.
Co-authored-by: Carter Anderson <mcanders1@gmail.com>
Add two examples on how to communicate with a task that is running either in another thread or in a thread from `AsyncComputeTaskPool`.
Loosely based on https://github.com/bevyengine/bevy/discussions/1150
# Objective
Enable the user to specify any presentation modes (including `Mailbox`).
Fixes#3807
## Solution
I've added a new `PresentMode` enum in `bevy_window` that mirrors the `wgpu` enum 1:1. Alternatively, I could add a new dependency on `wgpu-types` if that would be preferred.
## Objective
The [`DrawMeshInstanced`] command in the example sets vertex buffer 0 twice, with two identical calls to:
```rs
pass.set_vertex_buffer(0, gpu_mesh.vertex_buffer.slice(..));
```
## Solution
Remove the second call as it is unecessary.
[`DrawMeshInstanced`]: f3de12bc5e/examples/shader/shader_instancing.rs (L217-L258)
## Objective
When print shader validation error messages, we didn't print the sources and error message text, which led to some confusing error messages.
```cs
error:
┌─ wgsl:15:11
│
15 │ return material.color + 1u;
│ ^^^^^^^^^^^^^^^^^^^^ naga::Expression [11]
```
## Solution
New error message:
```cs
error: Entry point fragment at Vertex is invalid
┌─ wgsl:15:11
│
15 │ return material.color + 1u;
│ ^^^^^^^^^^^^^^^^^^^^ naga::Expression [11]
│
= Expression [11] is invalid
= Operation Add can't work with [8] and [10]
```
# Objective
Add a simple way for user to get the size of a loaded texture in an Image object.
Aims to solve #3689
## Solution
Add a `size() -> Vec2` method
Add two simple tests for this method.
Updates:
. method named changed from `size_2d` to `size`
# Objective
- `asset_server.watch_for_changes().unwrap()` only watches changes for assets loaded **_after_** that call.
- Technically, the `hot_asset_reloading` example is racey as the watch on the asset path is set up in an async task scheduled from the asset `load()`, but the filesystem watcher is only constructed in a call that comes **_after_** the call to `load()`.
## Solution
- It feels safest to allow enabling watching the filesystem for changes on the asset server from the point of its construction. Therefore, adding such an option to `AssetServerSettings` seemed to be the correct solution.
- Fix `hot_asset_reloading` by inserting the `AssetServerSettings` resource with `watch_for_changes: true` instead of calling `asset_server.watch_for_changes().unwrap()`.
- Document the shortcomings of `.watch_for_changes()`
