I am doing a python project that needs a bit of a speed boost, so I am making rust do a bit of the heavy lifting. What my project requires is that I can take in a numpy array, convert it to a rust array, apply some function to it, and return it as a numpy array. My only problem right now is specifying the lifetime of the return type (it is giving me the error “missing lifetime specifier
this function’s return type contains a borrowed value, but the signature does not say whether it is borrowed from py or array” under this bit of code:

-> PyResult<Vec<&str>> 

).
All I really need right now is for that error to be fixed. Here is my full rust code:

use pyo3::wrap_pyfunction;
use numpy::PyArrayDyn;

#[pymodule]
fn dnd_images(py: Python, m: &PyModule) -> PyResult<()> {
    m.add_function(wrap_pyfunction!(get_shape_instructions, m)?)?;
    Ok(())
}

#[pyfunction]
fn get_shape_instructions(py: Python, array: &PyArrayDyn<u16>) -> PyResult<Vec<&str>> {
    let array = unsafe { array.as_array() };
    let mut result: Vec<&str>;
    for element in array{
        let s: String = element.to_string();
        result.push(&s);
    }
    
    Ok(result)
}

All I have tried, as I am not very experienced with rust, is mashing things together in different orders and functions with different types according forums, articles, tutorials, and GPT-4.