use crate::ArrayData;
use crate::transform::_MutableArrayData;
use arrow_buffer::ArrowNativeType;
use arrow_schema::ArrowError;
use num_integer::Integer;
use num_traits::CheckedAdd;
pub(super) fn build_extend<T: ArrowNativeType + Integer + CheckedAdd>(
array: &ArrayData,
) -> crate::transform::Extend<'_> {
let offsets = array.buffer::<T>(0);
let sizes = array.buffer::<T>(1);
Box::new(
move |mutable: &mut _MutableArrayData, index: usize, start: usize, len: usize| {
let mut new_offset = T::usize_as(mutable.child_data[0].len());
for i in start..start + len {
mutable.buffer1.push(new_offset);
mutable.buffer2.push(sizes[i]);
new_offset = new_offset.checked_add(&sizes[i]).ok_or_else(|| {
ArrowError::InvalidArgumentError(
"offset overflow: data exceeds the capacity of the offset type. \
Try splitting into smaller batches or using a larger type \
(e.g. LargeListView instead of ListView)"
.to_string(),
)
})?;
let size = sizes[i].as_usize();
if size > 0 {
let child_start = offsets[i].as_usize();
mutable.child_data[0].try_extend(index, child_start, child_start + size)?;
}
}
Ok(())
},
)
}
pub(super) fn extend_nulls<T: ArrowNativeType>(
mutable: &mut _MutableArrayData,
len: usize,
) -> Result<(), ArrowError> {
let offset_buffer = &mut mutable.buffer1;
let sizes_buffer = &mut mutable.buffer2;
(0..len).for_each(|_| offset_buffer.push(T::default()));
(0..len).for_each(|_| sizes_buffer.push(T::default()));
Ok(())
}