Trait ImageLoad 

pub trait ImageLoad {
    // Required methods
    fn load_image(
        &mut self,
        decoder: &mut ImageDecoder,
        data: &[u8],
        opts: &DecodeOptions,
    ) -> Result<ImageInfo>;
    fn load_image_read<R: Read>(
        &mut self,
        decoder: &mut ImageDecoder,
        reader: R,
        opts: &DecodeOptions,
    ) -> Result<ImageInfo>;
    fn load_image_file(
        &mut self,
        decoder: &mut ImageDecoder,
        path: impl AsRef<Path>,
        opts: &DecodeOptions,
    ) -> Result<ImageInfo>;
}

Extension trait for loading images into pre-allocated tensors.

Import this trait to add load_image, load_image_read, and load_image_file methods to Tensor<T> and TensorDyn.

Performance

For best performance, allocate tensors via [ImageProcessor::create_image()] which selects the optimal memory backend (DMA → PBO → Mem) with GPU-aligned pitch. Free-standing tensors work but cannot use PBO and may not have GPU-aligned pitch.

Strided Buffers

The decoder writes directly into the tensor’s strided memory layout, respecting [effective_row_stride()]. Allocate the tensor at the maximum expected image size; smaller images decode into the top-left region.

Example

use edgefirst_codec::{ImageDecoder, ImageLoad, DecodeOptions};
use edgefirst_tensor::{Tensor, TensorTrait, TensorMemory, PixelFormat};

let mut tensor = Tensor::<u8>::image(1920, 1080, PixelFormat::Rgb, Some(TensorMemory::Mem))
    .expect("alloc");
let mut decoder = ImageDecoder::new();

// Decode from bytes
let jpeg = std::fs::read("image.jpg").unwrap();
let info = tensor.load_image(&mut decoder, &jpeg, &DecodeOptions::default()).unwrap();
println!("Decoded {}×{} {:?}", info.width, info.height, info.format);

// Decode from file
let info = tensor.load_image_file(&mut decoder, "image.jpg", &DecodeOptions::default()).unwrap();

Required Methods

fn load_image( &mut self, decoder: &mut ImageDecoder, data: &[u8], opts: &DecodeOptions, ) -> Result<ImageInfo>

Decode image bytes (&[u8]) into this tensor.

The image format (JPEG or PNG) is detected from magic bytes.

Errors

Returns CodecError::InsufficientCapacity if the decoded image dimensions exceed the tensor’s capacity.

fn load_image_read<R: Read>( &mut self, decoder: &mut ImageDecoder, reader: R, opts: &DecodeOptions, ) -> Result<ImageInfo>

Decode image data from a Read source into this tensor.

The input is buffered into the decoder’s internal scratch before decoding. For large inputs, prefer load_image with a pre-read &[u8] to avoid the copy.

fn load_image_file( &mut self, decoder: &mut ImageDecoder, path: impl AsRef<Path>, opts: &DecodeOptions, ) -> Result<ImageInfo>

Decode an image file into this tensor.

Convenience wrapper that opens the file, buffers it, and decodes.

Dyn Compatibility

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety", so this trait is not object safe.

Implementations on Foreign Types

impl ImageLoad for TensorDyn

fn load_image( &mut self, decoder: &mut ImageDecoder, data: &[u8], opts: &DecodeOptions, ) -> Result<ImageInfo>

fn load_image_read<R: Read>( &mut self, decoder: &mut ImageDecoder, reader: R, opts: &DecodeOptions, ) -> Result<ImageInfo>

fn load_image_file( &mut self, decoder: &mut ImageDecoder, path: impl AsRef<Path>, opts: &DecodeOptions, ) -> Result<ImageInfo>

impl<T: ImagePixel> ImageLoad for Tensor<T>

fn load_image( &mut self, decoder: &mut ImageDecoder, data: &[u8], opts: &DecodeOptions, ) -> Result<ImageInfo>

fn load_image_read<R: Read>( &mut self, decoder: &mut ImageDecoder, reader: R, opts: &DecodeOptions, ) -> Result<ImageInfo>

fn load_image_file( &mut self, decoder: &mut ImageDecoder, path: impl AsRef<Path>, opts: &DecodeOptions, ) -> Result<ImageInfo>

Implementors