vortex-array 0.68.0

// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright the Vortex contributors

use std::hash::Hasher;
use std::sync::Arc;

use prost::Message;
use vortex_error::VortexExpect;
use vortex_error::VortexResult;
use vortex_error::vortex_bail;
use vortex_error::vortex_ensure;
use vortex_error::vortex_panic;
use vortex_session::VortexSession;

use crate::ArrayEq;
use crate::ArrayHash;
use crate::ArrayRef;
use crate::ExecutionCtx;
use crate::ExecutionResult;
use crate::IntoArray;
use crate::Precision;
use crate::array::Array;
use crate::array::ArrayId;
use crate::array::ArrayView;
use crate::array::VTable;
use crate::arrays::list::ListArrayExt;
use crate::arrays::list::ListData;
use crate::arrays::list::array::NUM_SLOTS;
use crate::arrays::list::array::SLOT_NAMES;
use crate::arrays::list::compute::PARENT_KERNELS;
use crate::arrays::list::compute::rules::PARENT_RULES;
use crate::arrays::listview::list_view_from_list;
use crate::buffer::BufferHandle;
use crate::dtype::DType;
use crate::dtype::Nullability;
use crate::dtype::PType;
use crate::serde::ArrayChildren;
use crate::validity::Validity;
mod operations;
mod validity;
/// A [`List`]-encoded Vortex array.
pub type ListArray = Array<List>;

#[derive(Clone, prost::Message)]
pub struct ListMetadata {
    #[prost(uint64, tag = "1")]
    elements_len: u64,
    #[prost(enumeration = "PType", tag = "2")]
    offset_ptype: i32,
}

impl ArrayHash for ListData {
    fn array_hash<H: Hasher>(&self, _state: &mut H, _precision: Precision) {}
}

impl ArrayEq for ListData {
    fn array_eq(&self, _other: &Self, _precision: Precision) -> bool {
        true
    }
}

impl VTable for List {
    type ArrayData = ListData;

    type OperationsVTable = Self;
    type ValidityVTable = Self;

    fn id(&self) -> ArrayId {
        Self::ID
    }

    fn nbuffers(_array: ArrayView<'_, Self>) -> usize {
        0
    }

    fn buffer(_array: ArrayView<'_, Self>, idx: usize) -> BufferHandle {
        vortex_panic!("ListArray buffer index {idx} out of bounds")
    }

    fn buffer_name(_array: ArrayView<'_, Self>, idx: usize) -> Option<String> {
        vortex_panic!("ListArray buffer_name index {idx} out of bounds")
    }

    fn reduce_parent(
        array: ArrayView<'_, Self>,
        parent: &ArrayRef,
        child_idx: usize,
    ) -> VortexResult<Option<ArrayRef>> {
        PARENT_RULES.evaluate(array, parent, child_idx)
    }

    fn serialize(
        array: ArrayView<'_, Self>,
        _session: &VortexSession,
    ) -> VortexResult<Option<Vec<u8>>> {
        Ok(Some(
            ListMetadata {
                elements_len: array.elements().len() as u64,
                offset_ptype: PType::try_from(array.offsets().dtype())? as i32,
            }
            .encode_to_vec(),
        ))
    }

    fn validate(
        &self,
        _data: &ListData,
        dtype: &DType,
        len: usize,
        slots: &[Option<ArrayRef>],
    ) -> VortexResult<()> {
        vortex_ensure!(
            slots.len() == NUM_SLOTS,
            "ListArray expected {NUM_SLOTS} slots, found {}",
            slots.len()
        );
        let elements = slots[crate::arrays::list::array::ELEMENTS_SLOT]
            .as_ref()
            .vortex_expect("ListArray elements slot");
        let offsets = slots[crate::arrays::list::array::OFFSETS_SLOT]
            .as_ref()
            .vortex_expect("ListArray offsets slot");
        vortex_ensure!(
            offsets.len().saturating_sub(1) == len,
            "ListArray length {} does not match outer length {}",
            offsets.len().saturating_sub(1),
            len
        );

        let actual_dtype = DType::List(Arc::new(elements.dtype().clone()), dtype.nullability());
        vortex_ensure!(
            &actual_dtype == dtype,
            "ListArray dtype {} does not match outer dtype {}",
            actual_dtype,
            dtype
        );

        Ok(())
    }

    fn deserialize(
        &self,
        dtype: &DType,
        len: usize,
        metadata: &[u8],

        _buffers: &[BufferHandle],
        children: &dyn ArrayChildren,
        _session: &VortexSession,
    ) -> VortexResult<crate::array::ArrayParts<Self>> {
        let metadata = ListMetadata::decode(metadata)?;
        let validity = if children.len() == 2 {
            Validity::from(dtype.nullability())
        } else if children.len() == 3 {
            let validity = children.get(2, &Validity::DTYPE, len)?;
            Validity::Array(validity)
        } else {
            vortex_bail!("Expected 2 or 3 children, got {}", children.len());
        };

        let DType::List(element_dtype, _) = &dtype else {
            vortex_bail!("Expected List dtype, got {:?}", dtype);
        };
        let elements = children.get(
            0,
            element_dtype.as_ref(),
            usize::try_from(metadata.elements_len)?,
        )?;

        let offsets = children.get(
            1,
            &DType::Primitive(metadata.offset_ptype(), Nullability::NonNullable),
            len + 1,
        )?;

        let data = ListData::try_build(elements.clone(), offsets.clone(), validity.clone())?;
        let slots = ListData::make_slots(&elements, &offsets, &validity, len);
        Ok(crate::array::ArrayParts::new(self.clone(), dtype.clone(), len, data).with_slots(slots))
    }

    fn slot_name(_array: ArrayView<'_, Self>, idx: usize) -> String {
        SLOT_NAMES[idx].to_string()
    }

    fn execute(array: Array<Self>, ctx: &mut ExecutionCtx) -> VortexResult<ExecutionResult> {
        Ok(ExecutionResult::done(
            list_view_from_list(array, ctx)?.into_array(),
        ))
    }

    fn execute_parent(
        array: ArrayView<'_, Self>,
        parent: &ArrayRef,
        child_idx: usize,
        ctx: &mut ExecutionCtx,
    ) -> VortexResult<Option<ArrayRef>> {
        PARENT_KERNELS.execute(array, parent, child_idx, ctx)
    }
}

#[derive(Clone, Debug)]
pub struct List;

impl List {
    pub const ID: ArrayId = ArrayId::new_ref("vortex.list");
}