Struct VisionRaster 

pub struct VisionRaster {
    pub width: u32,
    pub height: u32,
    pub format: VisionPixelFormat,
    pub data: Vec<u8>,
}

A page (or crop) rasterized for vision input, row-major.

Fields

width: u32

Width in pixels.

height: u32

Height in pixels.

format: VisionPixelFormat

Pixel layout of Self::data.

data: Vec<u8>

Pixel bytes, row-major, top-left origin.

Implementations

impl VisionRaster

pub fn crop_pixels( &self, x: u32, y: u32, width: u32, height: u32, ) -> Option<Self>

Crop a pixel rectangle out of this raster (top-left origin).

Returns None when the rectangle is empty or lies outside the raster.

pub fn crop_points( &self, bbox: [f64; 4], page_size_points: (f64, f64), dpi: u32, ) -> VisionResult<Self>

Crop a bbox in page points out of this raster.

bbox is [x0, y0, x1, y1] in PDF points with the origin at the bottom-left (the crate::PageBox convention). page_size_points is the page’s (width, height) in points, normally crate::PageGeometry::effective_dimensions. The mapping is pixels = points * dpi / 72, with the Y axis flipped to the raster’s top-left origin. The bbox is clamped to the page; an empty intersection is an error.

Trait Implementations

impl Clone for VisionRaster

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for VisionRaster

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.