tonhowtf/omniget · error

ffmpeg

Error message

ffmpeg: {}

What it means

extract_audio runs ffmpeg as a child process to pull the audio track out of a downloaded Instagram media file. When ffmpeg exits with a nonzero status the library wraps ffmpeg's trimmed stderr into this anyhow error, so the actual diagnostic text comes straight from ffmpeg.

Solutions

1. Read the ffmpeg message after 'ffmpeg: ' — it names the real problem
2. Ensure the source file downloaded completely and contains an audio stream (ffprobe it)
3. Match the output file extension/codec to a supported one
4. Check that the installed ffmpeg version supports the required demuxers/codecs

Example fix

// before
ffmpeg -i clip.mp4 -vn -acodec aac out.m4a
// after (input may have no audio; probe first)
ffprobe -v error -select_streams a -show_entries stream=codec_type clip.mp4 && ffmpeg -i clip.mp4 -vn -acodec aac out.m4a

let ok = Command::new("ffprobe")
    .args(["-v", "error", "-select_streams", "a:0", "-show_entries",
           "stream=codec_type", "-of", "csv=p=0", input])
    .output()
    .await?
    .stdout
    .starts_with(b"audio");
if !ok { bail!("arquivo não tem faixa de áudio"); }

match extract_audio(&file, &out).await {
    Ok(audio) => use_audio(audio),
    Err(e) if e.to_string().starts_with("ffmpeg:") => {
        let ffmpeg_msg = e.to_string().trim_start_matches("ffmpeg: ").to_string();
        log::warn!("ffmpeg falhou: {ffmpeg_msg}");
    }
    Err(e) => return Err(e),
}

When it happens

Trigger: Calling extract_audio (via download_items) when ffmpeg exits nonzero: the input file has no audio stream, the file is corrupt or incomplete, the output path/codec combination is invalid, or the ffmpeg binary build lacks needed codecs.

Common situations: Instagram item is a video with no audio track; download was truncated by a network failure; output extension disagrees with the chosen codec; ffmpeg missing optional encoder.

Related errors

• {}
• Failed to run ffmpeg
• ffmpeg returned code
• Failed to convert thumbnail to JPEG
• ffmpeg nao iniciou

AI-assisted analysis of tonhowtf/omniget@8600b91f42 (2026-09-12). Data as JSON: /api/errors/aa3e862279e86371. Report an issue: GitHub.

Appendix: source

Thrown at src-tauri/omniget-core/src/core/tools/instagram/media.rs:305

    let out = video.with_extension(format);
    let mut cmd = crate::core::process::command(&ffmpeg);
    cmd.args(["-y", "-hide_banner", "-loglevel", "error", "-i"])
        .arg(video)
        .arg("-vn");
    if format == "mp3" {
        cmd.args(["-c:a", "libmp3lame", "-q:a", "2"]);
    } else {
        cmd.args(["-c:a", "copy"]);
    }
    cmd.arg(&out);
    let status = cmd
        .stdin(std::process::Stdio::null())
        .stdout(std::process::Stdio::null())
        .stderr(std::process::Stdio::piped())
        .output()
        .await?;
    if !status.status.success() {
        return Err(anyhow!(
            "ffmpeg: {}",
            String::from_utf8_lossy(&status.stderr).trim()
        ));
    }
    Ok(out)
}

/// Baixa uma lista de itens para `dest`, reportando `ig:<job>` com
/// done/total e o nome do arquivo atual.
pub async fn download_items(
    client: &IgClient,
    items: &[MediaItem],
    dest: &str,
    opts: &DownloadOptions,
    progress: &super::super::ProgressFn,
    job: &str,
    flag: &AtomicBool,
) -> anyhow::Result<DownloadResult> {

View on GitHub (pinned to 8600b91f42)