The `dynamic_offsets` and `is_dirty` flags only make sense when the
slot is occupied, so they should be inside the `Option`. This makes
`State::bind` into an `ArrayVec<Option<BindState>>`, and cleans up
various other bits.
In parking_lot 0.12 and parking_lot_core 0.9.0, those crates switched
from the winapi crate to the official Microsoft windows-sys crate.
This is fine, except that windows-sys and its dependencies are even
larger than winapi. Some users may wish to stick with winapi for the
time being; this change allows wgpu to accommodate them.
It's very odd to have almost all the render pass and compute pass ffi
functions in `wgpu` except for the `set_index_buffer` functions, which
live in Firefox. I'd like to remove these from Firefox and put them
back next to their companions.
These functions were originally removed from wgpu in #1077, because
wgpu-native has its own incompatible version of IndexFormat (see that
PR for details). However, with wgpu-native#85, that code was removed,
so having these functions in `wgpu` should be no longer be a problem
for wgpu-native.
* fix: don't panic on invalid id in Storage::get
* formatting
* removed double matches
* more match removal
* fix formatting
* add fix to Storage::label_for_invalid_id
Lifetime names like `<'a, 'b>` mean that the reader must look at how
they are used to understand them. By changing them to `'tex` (lifetime
of borrows of the texture views the documentation describes) and `'desc`
(everything else), and mentioning them in the documentation, it's more
obvious which role each one plays.
For consistency, I also changed `begin_render_pass()`,
`RenderPassColorAttachment`, and `RenderPassDepthStencilAttachment` to
use matching lifetime names.
Also, this made it clear that the `Label` had the wrong lifetime --
the docs say that the texture views have a different lifetime from
“everything else”, but the `Label` in fact had the same lifetime as
the texture views, so I changed it to the `'desc` (formerly `'b`)
lifetime. (On review of change history, this mismatch was previously
introduced in commit 632f828eda9af1f6f18c74398b5f40644cf02f03.)
This change is definitely safe because I followed the data flow down
to `BasePass::new` which promptly calls `to_string()` on the label,
thus converting it to owned data.
Added:
* WebGPU specification links.
* How to create an instance, if it is not all-public-fields.
* What function or enclosing type this type is used for.
(The latter two are not present for `wgpu-types` items due to the
dependency running the wrong direction for that — I saw existing
`../wgpu/` links, but those do not work on https://docs.rs/wgpu/.)
This commit specifically does not:
* substantially revise existing language, except where necessary to fit
the new information, or when there was an obvious trivial improvement
like adding a link to support "how to use this".
* add similar information to the documentation of types not directly
corresponding to the WebGPU specification
The versions of wasm-bindgen-cli run by `.github/workflows/publish.yml` must
exactly match the version of `wasm-bindgen` used by `wgpu`. At the moment,
`wgpu/Cargo.toml` specifies `0.2.76`, and our `Cargo.lock` says `0.2.78`, so
the latter is what `publish.yml` should mention.
