Function initcap

datafusion/functions/src/unicode/initcap.rs:151–187

Converts the first letter of each word to uppercase and the rest to lowercase. Words are sequences of alphanumeric characters separated by non-alphanumeric characters. Example: ```sql initcap('hi THOMAS') = 'Hi Thomas' ```

(args: &[ArrayRef])

/// initcap('hi THOMAS') = 'Hi Thomas'
/// ```
fn initcap<T: OffsetSizeTrait>(args: &[ArrayRef]) -> Result<ArrayRef> {
    let string_array = as_generic_string_array::<T>(&args[0])?;

    if string_array.is_ascii() {
        return Ok(initcap_ascii_array(string_array));
    }

    let len = string_array.len();
    let mut builder = GenericStringArrayBuilder::<T>::with_capacity(
        len,
        string_array.value_data().len(),
    );

    let mut container = String::new();
    let nulls = string_array.nulls().cloned();
    if let Some(ref n) = nulls {
        for i in 0..len {
            if n.is_null(i) {
                builder.append_placeholder();
            } else {
                // SAFETY: not null per check above.
                let s = unsafe { string_array.value_unchecked(i) };
                initcap_string(s, &mut container);
                builder.append_value(&container);
            }
        }
    } else {
        for i in 0..len {
            // SAFETY: no null buffer means every index is valid.
            let s = unsafe { string_array.value_unchecked(i) };
            initcap_string(s, &mut container);
            builder.append_value(&container);
        }
    }

    Ok(Arc::new(builder.finish(nulls)?) as ArrayRef)
}

/// Fast path for `Utf8` or `LargeUtf8` arrays that are ASCII-only. We can use a
/// single pass over the buffer and operate directly on bytes.