Function hash_embed

atomic-repository/src/repository/vault_embeddings.rs:283–296

Simple hash-based "embedding" for testing purposes. Produces a deterministic vector from text content using character frequency analysis. NOT suitable for production — use a real embedding model (MiniLM, OpenAI, etc.) instead.

(text: &str, dims: usize)

/// frequency analysis. NOT suitable for production — use a real
/// embedding model (MiniLM, OpenAI, etc.) instead.
pub fn hash_embed(text: &str, dims: usize) -> Vec<f32> {
    let mut vec = vec![0.0f32; dims];
    for (i, byte) in text.bytes().enumerate() {
        vec[i % dims] += byte as f32 / 255.0;
    }
    // Normalize
    let norm: f32 = vec.iter().map(|x| x * x).sum::<f32>().sqrt();
    if norm > 0.0 {
        for v in &mut vec {
            *v /= norm;
        }
    }
    vec
}

#[cfg(test)]
mod tests {