Enum ConversionPrecision 

#[non_exhaustive]
pub enum ConversionPrecision {
    Standard,
    High,
}

The swscale precision tier for the conversion stage.

The extractor runs any resize plus the pixel-format conversion as ONE swscale pass, and this tier sets that pass’s scaler flags — so it affects resized output too, not only the YUV → RGB step. Color interpretation — which matrix and range are applied (ColorPolicy) — is identical in both tiers.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

Standard

The FFmpeg CLI’s default swscale configuration (flags=bicubic, no extra precision flags). This keeps swscale eligible for its fastest converters — including the unscaled yuv420p → rgb24 special case — and matches the bytes an ffmpeg -vf "scale=..,format=.." run with default flags produces. This is the default.

High

Adds accurate_rnd+full_chroma_int: accurate rounding plus full horizontal chroma interpolation. Relative to Standard the mean difference is a couple of steps per 8-bit channel; individual pixels on sharp chroma edges can differ by more, because the two tiers reconstruct subsampled chroma differently — and accurate_rnd also changes rounding on resized output, including single-plane formats like gray. The flags disable swscale’s unscaled fast-path converters, which multiplies the conversion cost several-fold (at 1080p the conversion, not the decode, becomes the bottleneck). Opt in when the last bit of rounding and chroma reconstruction matters more than throughput — e.g. quality metrics or golden-reference pipelines. v0.14.0 applied exactly these flags unconditionally, so High also reproduces v0.14.0 output, byte-identical when linked against the same libswscale build.

Trait Implementations

impl Clone for ConversionPrecision

fn clone(&self) -> ConversionPrecision

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for ConversionPrecision

impl Debug for ConversionPrecision

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

Formats the value using the given formatter.

impl Default for ConversionPrecision

fn default() -> ConversionPrecision

Returns the “default value” for a type.

impl Eq for ConversionPrecision

impl PartialEq for ConversionPrecision

fn eq(&self, other: &ConversionPrecision) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for ConversionPrecision