Function hash_string_view_array_inner

datafusion/common/src/hash_utils.rs:312–366

(
    array: &GenericByteViewArray<T>,
    random_state: &RandomState,
    hashes_buffer: &mut [u64],
)

#[cfg(not(feature = "force_hash_collisions"))]
#[inline(never)]
fn hash_string_view_array_inner<
    T: ByteViewType,
    const HAS_NULLS: bool,
    const HAS_BUFFERS: bool,
    const REHASH: bool,
>(
    array: &GenericByteViewArray<T>,
    random_state: &RandomState,
    hashes_buffer: &mut [u64],
) {
    assert_eq!(
        hashes_buffer.len(),
        array.len(),
        "hashes_buffer and array should be of equal length"
    );

    let buffers = array.data_buffers();
    let view_bytes = |view_len: u32, view: u128| {
        let view = ByteView::from(view);
        let offset = view.offset as usize;
        // SAFETY: view is a valid view as it came from the array
        unsafe {
            let data = buffers.get_unchecked(view.buffer_index as usize);
            data.get_unchecked(offset..offset + view_len as usize)
        }
    };

    let hashes_and_views = hashes_buffer.iter_mut().zip(array.views().iter());
    for (i, (hash, &v)) in hashes_and_views.enumerate() {
        if HAS_NULLS && array.is_null(i) {
            continue;
        }
        let view_len = v as u32;
        // all views are inlined, no need to access external buffers
        if !HAS_BUFFERS || view_len <= 12 {
            if REHASH {
                let mut hasher = seeded_state(*hash).build_hasher();
                v.hash_write(&mut hasher);
                *hash = hasher.finish();
            } else {
                *hash = v.hash_one(random_state);
            }
            continue;
        }
        // view is not inlined, so we need to hash the bytes as well
        let value = view_bytes(view_len, v);
        if REHASH {
            let mut hasher = seeded_state(*hash).build_hasher();
            value.hash_write(&mut hasher);
            *hash = hasher.finish();
        } else {
            *hash = value.hash_one(random_state);
        }
    }
}

/// Builds hash values for array views and writes them into `hashes_buffer`
/// If `rehash==true` this combines the previous hash value in the buffer