tonhowtf/omniget · error

escolha ao menos uma imagem

Error message

escolha ao menos uma imagem

What it means

img_bg's run_blocking collects the input image paths from opts.inputs and opts.input_dir before starting background removal. If the resulting list is empty it refuses to proceed, since removing a background from nothing is meaningless. The library throws this early validation error instead of creating an ONNX session or spawning work on empty input.

Solutions

1. Populate opts.inputs with at least one image path before calling run
2. Point opts.input_dir at a directory that exists and contains image files
3. Check collect_inputs results (file existence, extension filters) for why inputs were dropped

Example fix

// before
BgOptions { inputs: vec![], input_dir: None, .. }
// after
BgOptions { inputs: vec!["photo.png".into()], input_dir: None, .. }

let files: Vec<_> = collect_inputs(&opts.inputs, &opts.input_dir);
if files.is_empty() {
    return Err(anyhow!("escolha ao menos uma imagem"));
}

When it happens

Trigger: Calling run (via run_blocking) with an opts where both opts.inputs is empty/absent and opts.input_dir points to a nonexistent or empty directory, so collect_inputs returns an empty Vec.

Common situations: CLI/Tauri frontend launched without selecting files; input_dir typo or empty folder; user cleared the file list in the UI before submitting.

Understand the failure class

Background: "must not be empty", "cannot be empty" — required-field validation errors across open-source libraries — this error's family across 41 libraries.

Related errors

• informe um appid, um link da loja ou marque a biblioteca…
• nenhuma imagem
• escolha a pasta de destino
• nenhum link de vídeo do TikTok na entrada
• escolha a pasta de destino para organizar

AI-assisted analysis of tonhowtf/omniget@8600b91f42 (2026-09-12). Data as JSON: /api/errors/39bb430d7e10fde8. Report an issue: GitHub.

Appendix: source

Thrown at src-tauri/omniget-core/src/core/tools/img_bg.rs:413

    let (ow, oh) = (img.width(), img.height());
    Ok(image::imageops::resize(
        &small,
        ow,
        oh,
        FilterType::Lanczos3,
    ))
}

fn run_blocking(
    opts: &BgOptions,
    model_id: &str,
    progress: &ProgressFn,
) -> anyhow::Result<BgResult> {
    let params = *params_for(model_id)
        .ok_or_else(|| anyhow!("o modelo {model_id} não serve para remover fundo"))?;
    let files = collect_inputs(&opts.inputs, &opts.input_dir);
    if files.is_empty() {
        return Err(anyhow!("escolha ao menos uma imagem"));
    }
    let background = parse_hex_color(&opts.background)?;
    let ext = resolve_format(&opts.format, background.is_some(), opts.mask_only);

    let mut session = super::onnx::session_for(model_id)?;
    let total = files.len() as u64;
    let mut items: Vec<BgItem> = Vec::with_capacity(files.len());
    let mut failed = 0usize;

    for (i, path) in files.iter().enumerate() {
        super::report(
            progress,
            TOOL_ID,
            "progress",
            i as u64,
            Some(total),
            Some(path.to_string_lossy().to_string()),
        );

View on GitHub (pinned to 8600b91f42)