-
Notifications
You must be signed in to change notification settings - Fork 141
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Migrate the repository to a virtual workspace. #590
Conversation
vello/src/shaders.rs
Outdated
@@ -62,7 +62,7 @@ pub fn full_shaders( | |||
#[cfg(feature = "hot_reload")] | |||
let mut shaders = vello_shaders::compile::ShaderInfo::from_dir(concat!( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can vello_shaders
instead export the path to directory the shaders are located in?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You mean the shader
directory name, or some more absolute path?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Alright I added a new function that returns an absolute path to the shader directory. It depends on using Cargo and the path is computed at compile time.
If someone has a build cache and then moves the directories around, they will need to cargo clean
or trigger a rebuild in another way.
/// The path is determined at compile time. | ||
pub fn shader_dir() -> &'static PathBuf { | ||
static SHADER_DIR: OnceLock<PathBuf> = OnceLock::new(); | ||
SHADER_DIR.get_or_init(|| Path::new(env!("CARGO_MANIFEST_DIR")).join("shader")) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If we used concat!
, I don't think the runtime string manipulation would be needed.
I guess this is about using the proper separator?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just using concat!
is buggy because we don't know if CARGO_MANIFEST_DIR
ends with a separator or not. Instead of doing some sort of .ends_with
dance and potentially missing some other edge cases, better to use the path
module.
Also, this is a singleton. The performance overhead here is trivial.
vello_shaders/src/compile/mod.rs
Outdated
|
||
/// Returns the absolute path to the directory containing the WGSL shaders. | ||
/// | ||
/// The path is determined at compile time. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
And may be invalid at runtime, especially if the program is deployed on a different machine
As a different point:
This is also a reproducible builds challenge, which I guess we should document?
Idk, maybe even making it doc(hidden)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sure, I added a note about the impact on reproducible builds.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Will give a final check on Monday if this hasn't landed by then
|
||
It is used as the rendering backend for [Xilem], a Rust GUI toolkit. | ||
|
||
> [!WARNING] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should change the formatting of this now, to use something which renders well on crates.io
That might just be using the
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Updating readme files is intentionally out of scope and planned as follow-up work to reduce the review friction here.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for getting this done. A few small nits, I don't think any are necessarily blocking
README.md
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'd add a big comment at the top that this is "temporarily" duplicated in the vello
folder, and updates should be preferred there.
|
||
at your option. | ||
|
||
In addition, all files in the [`vello_shaders/shader`](https://github.com/linebender/vello/tree/main/vello_shaders/shader) and [`vello_shaders/src/cpu`](https://github.com/linebender/vello/tree/main/vello_shaders/src/cpu) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These additional caveats are not important true for vello
itself. I imagine this has the same reasoning as above though?
The CI, VSCode config,
ARCHITECTURE.md
and someREADME.md
links have been updated to reflect the new directory structure.The WGSL shaders now live in the
vello_shaders
subdirectory. Vello hot reloading depends on our repo structure, which is fine as decided at last office hours due to this being a development feature.The examples continue to live in a root
examples
directory, as that is how it remained for Parley as well. I did this for consistency and don't really feel strongly either way if people think we should move them undervello/examples
instead. If we had examples withoutCargo.toml
that would make more sense, but we don't, so ..The repo
README.md
and thevello
sub-directoryREADME.md
are exact clones and both need updating. That is intentionally planned as follow-up work to reduce the review friction here.