Struct Decoder

Source

pub struct Decoder { /* private fields */ }

Expand description

The main VP8/VP9 video decoder.

§Example

use std::num::NonZero;
use vpx_rs::{Decoder, DecoderConfig, Error, DecodedImageData};

// Initialize a new decoder
fn new_vp8_decoder(width: u32, height: u32,) -> Result<Decoder, Error> {
    // Prepare the configuration
    let config = DecoderConfig::new(
        vpx_rs::dec::CodecId::VP8,
        width,
        height,
    );
    // Instantiate the decoder
    Decoder::new(config)
}

// Passes a buffer to the decoder and extracts any completed frames
fn decode_vp8_buf(
    decoder: &mut Decoder,
    stream_data: &[u8],
) -> Result<(), Error> {
    for decoded_image in decoder.decode(stream_data)?.into_iter() {
        let (format, bit_depth) = match decoded_image.data() {
            DecodedImageData::Data8b(data) =>
                (format!("{:?}", data.format()), 8),
            DecodedImageData::Data16b(data) =>
                (format!("{:?}", data.format()), 16),
        };
        println!(
            "Decoded a {}x{} pixel image in {} format with bit depth {}",
            decoded_image.width(),
            decoded_image.height(),
            format,
            bit_depth,
        );
    }
    Ok(())
}

Implementations§

Source §

impl Decoder

Source

pub fn new(config: DecoderConfig) -> Result<Self>

Constructs a new decoder for the given configuration. After successful construction, you can start sending bitstream buffers to the decoder by calling Decoder::decode().

Source

pub fn decode(&mut self, data: &[u8]) -> Result<DecodedImageIterator<'_>>

Submits data to be decoded by the decoder, yielding a number of decoded images as a result. Please note that even if you only send data for a single frame at a time, due to internal bitstream frame rendering, the decoder might produce any number of frames per decode call. You should retrieve all decoded images from the returned iterator before trying to push more data to the decoder. The images will always be returned to you in presentation order, so there’s no need to perform image reordering on your end.

Source