Struct OutputSpec 

pub struct OutputSpec {

    pub path: PathBuf,
    pub format: Option<String>,
    pub video_codec: Option<StreamCodec>,
    pub audio_codec: Option<StreamCodec>,
    pub subtitle_codec: Option<CodecId>,
    pub video_filters: Option<String>,
    pub filter_complex: Option<String>,
    pub maps: Vec<MapSpec>,
    pub overwrite: bool,
    pub max_video_frames: Option<u64>,
    pub trim_start: Option<f64>,
    pub trim_end: Option<f64>,
    pub frame_rate: Option<(u32, u32)>,
    pub audio_rate: Option<u32>,
    pub audio_channels: Option<u16>,
    pub audio_filters: Option<String>,
    pub metadata: Dictionary,
    pub format_options: Dictionary,
}

The output file for a job.

Fields

path: PathBuf

format: Option<String>

Force a container format instead of guessing from the extension (ffmpeg -f <fmt> ...).

video_codec: Option<StreamCodec>

Video codec for the output, if a video stream is produced.

audio_codec: Option<StreamCodec>

Audio codec for the output, if an audio stream is produced.

subtitle_codec: Option<CodecId>

Subtitle codec (-c:s subrip|webvtt). Text-subtitle packets share one contract (plain text + ms timing), so this relabels the stream for the muxer rather than re-encoding anything.

video_filters: Option<String>

Video filter graph (-vf), e.g. scale=320:240,crop=.... Applied to decoded video frames before re-encoding (transcode streams only).

filter_complex: Option<String>

Multi-input filter graph (-filter_complex). Currently models overlay: the last input is composited over input #0’s video.

maps: Vec<MapSpec>

Explicit stream selection (-map). Empty = default (all video + audio, in input/stream order).

overwrite: bool

Overwrite the output if it exists (-y); otherwise fail (-n).

max_video_frames: Option<u64>

Stop after this many VIDEO frames (-frames:v N). None = whole input.

This is a measurement-critical option: without it a harness that thinks it is encoding a 50-frame prefix silently encodes the whole clip, and any rate-vs-quality pairing computed against a prefix is then wrong.

trim_start: Option<f64>

Output trim window start, in seconds (-ss). Decoded frames (or copied packets) before this point are dropped; timestamps are shifted so the output starts at zero.

trim_end: Option<f64>

Output trim window end, in seconds (-to, or -ss + -t). Exclusive.

frame_rate: Option<(u32, u32)>

Constant output frame rate (-r), as (num, den) — e.g. (30000, 1001). Frames are duplicated/dropped to hit it (FFmpeg’s fps filter).

audio_rate: Option<u32>

Target audio sample rate (-ar). The encoder’s accepted-rate list still wins if it cannot take this exact rate (nearest accepted is used).

audio_channels: Option<u16>

Target audio channel count (-ac): 1↔2 downmix/upmix.

audio_filters: Option<String>

Audio filter chain (-af / -filter:a), e.g. volume=-6dB,atrim=....

metadata: Dictionary

Output metadata (-metadata title=...), handed to muxers that carry it.

format_options: Dictionary

Muxer options (-hls_time 6 → {"hls_time": "6"}), for formats that take them (HLS/DASH segmenting).

Trait Implementations

impl Clone for OutputSpec

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 OutputSpec

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

Formats the value using the given formatter.

impl Default for OutputSpec

fn default() -> Self

Returns the “default value” for a type.