TensorIndexing.h

#pragma once

#include <c10/util/Optional.h>
#include <ATen/core/TensorBody.h>
#include <ATen/ExpandUtils.h>
#include <ATen/Functions.h>
#include <ATen/ScalarOps.h>

// TODO: try to remove this
// There is some back story, see https://github.com/pytorch/pytorch/issues/48684
#include <ATen/NativeFunctions.h>

#include <ATen/core/List.h>

namespace at {
namespace indexing {

const int64_t INDEX_MAX = std::numeric_limits<int64_t>::max();
const int64_t INDEX_MIN = std::numeric_limits<int64_t>::min();

enum class TensorIndexType { None, Ellipsis, Integer, Boolean, Slice, Tensor };

constexpr c10::nullopt_t None = c10::nullopt;

struct TORCH_API EllipsisIndexType final { EllipsisIndexType() {} };
TORCH_API extern const EllipsisIndexType Ellipsis;

struct TORCH_API Slice final {
 public:
  // This mirrors `__PySlice_Unpack` in torch/csrc/utils/python_compat.h
  Slice(
    c10::optional<int64_t> start_index = c10::nullopt,
    c10::optional<int64_t> stop_index = c10::nullopt,
    c10::optional<int64_t> step_index = c10::nullopt) {
    if (!step_index.has_value()) {
      step_ = 1;
    } else {
      step_ = step_index.value();
      TORCH_CHECK_VALUE(step_ != 0, "slice step cannot be zero");

      // Here step might be -INDEX_MAX-1; in this case we replace it
      // with -INDEX_MAX.  This doesn't affect the semantics, and it
      // guards against later undefined behaviour resulting from code that
      // does "step = -step" as part of a slice reversal.
      if (step_ < -INDEX_MAX)
        step_ = -INDEX_MAX;
    }
    if (!start_index.has_value()) {
      start_ = step_ < 0 ? INDEX_MAX : 0;
    } else {
      start_ = start_index.value();
    }
    if (!stop_index.has_value()) {
      stop_ = step_ < 0 ? INDEX_MIN : INDEX_MAX;
    } else {
      stop_ = stop_index.value();
    }
  }

  inline int64_t start() const {
    return start_;
  }

  inline int64_t stop() const {
    return stop_;
  }

  inline int64_t step() const {
    return step_;
  }

 private:
  int64_t start_;
  int64_t stop_;
  int64_t step_;
};

TORCH_API std::ostream& operator<<(std::ostream& stream, const Slice& slice);

// `at::indexing::TensorIndex` is used for converting C++ tensor indices such as
// `{None, "...", Ellipsis, 0, true, Slice(1, None, 2), torch::tensor({1, 2})}`
// into its equivalent `std::vector<TensorIndex>`, so that further tensor indexing
// operations can be performed using the supplied indices.
//
// There is one-to-one correspondence between Python and C++ tensor index types:
// Python                  | C++
// -----------------------------------------------------
// `None`                  | `at::indexing::None`
// `Ellipsis`              | `at::indexing::Ellipsis`
// `...`                   | `"..."`
// `123`                   | `123`
// `True` / `False`        | `true` / `false`
// `:`                     | `Slice()` / `Slice(None, None)`
// `::`                    | `Slice()` / `Slice(None, None, None)`
// `1:`                    | `Slice(1, None)`
// `1::`                   | `Slice(1, None, None)`
// `:3`                    | `Slice(None, 3)`
// `:3:`                   | `Slice(None, 3, None)`
// `::2`                   | `Slice(None, None, 2)`
// `1:3`                   | `Slice(1, 3)`
// `1::2`                  | `Slice(1, None, 2)`
// `:3:2`                  | `Slice(None, 3, 2)`
// `1:3:2`                 | `Slice(1, 3, 2)`
// `torch.tensor([1, 2])`) | `torch::tensor({1, 2})`
struct TORCH_API TensorIndex final {
  // Case 1: `at::indexing::None`
  // NOLINTNEXTLINE(clang-analyzer-optin.cplusplus.UninitializedObject)
  TensorIndex(c10::nullopt_t) : type_(TensorIndexType::None) {}

  // Case 2: "..." / `at::indexing::Ellipsis`
  // NOLINTNEXTLINE(clang-analyzer-optin.cplusplus.UninitializedObject)
  TensorIndex(at::indexing::EllipsisIndexType) : type_(TensorIndexType::Ellipsis) {}
  TensorIndex(const char *str) : TensorIndex(at::indexing::Ellipsis) {
    // NOLINTNEXTLINE(clang-analyzer-optin.cplusplus.UninitializedObject)
    TORCH_CHECK_VALUE(
      strcmp(str, "...") == 0,
      "Expected \"...\" to represent an ellipsis index, but got \"", str, "\"");
  }

  // Case 3: Integer value
  // NOLINTNEXTLINE(clang-analyzer-optin.cplusplus.UninitializedObject)
  TensorIndex(int64_t integer) : integer_(integer), type_(TensorIndexType::Integer) {}
  // NOLINTNEXTLINE(clang-analyzer-optin.cplusplus.UninitializedObject)
  TensorIndex(int integer) : TensorIndex((int64_t)integer) {}

  // Case 4: Boolean value
  template <class T,
            class = typename std::enable_if<std::is_same<bool, T>::value>::type >
  // NOLINTNEXTLINE(clang-analyzer-optin.cplusplus.UninitializedObject)
  TensorIndex(T boolean) : boolean_(boolean), type_(TensorIndexType::Boolean) {}

  // Case 5: Slice represented in `at::indexing::Slice` form
  // NOLINTNEXTLINE(clang-analyzer-optin.cplusplus.UninitializedObject)
  TensorIndex(Slice slice) : slice_(std::move(slice)), type_(TensorIndexType::Slice) {}

  // Case 6: Tensor value
  // NOLINTNEXTLINE(clang-analyzer-optin.cplusplus.UninitializedObject)
  TensorIndex(Tensor tensor) : tensor_(std::move(tensor)), type_(TensorIndexType::Tensor) {}

  inline bool is_none() const {
    return type_ == TensorIndexType::None;
  }

  inline bool is_ellipsis() const {
    return type_ == TensorIndexType::Ellipsis;
  }

  inline bool is_integer() const {
    return type_ == TensorIndexType::Integer;
  }

  inline int64_t integer() const {
    return integer_;
  }

  inline bool is_boolean() const {
    return type_ == TensorIndexType::Boolean;
  }

  inline bool boolean() const {
    return boolean_;
  }

  inline bool is_slice() const {
    return type_ == TensorIndexType::Slice;
  }

  inline const Slice& slice() const {
    return slice_;
  }

  inline bool is_tensor() const {
    return type_ == TensorIndexType::Tensor;
  }

  inline const Tensor& tensor() const {
    return tensor_;
  }

 private:
  int64_t integer_;
  bool boolean_;
  Slice slice_;
  Tensor tensor_;
  TensorIndexType type_;
};

TORCH_API std::ostream& operator<<(std::ostream& stream, const TensorIndex& tensor_index);
TORCH_API std::ostream& operator<<(std::ostream& stream, const std::vector<TensorIndex>& tensor_indices);

namespace impl {
static inline Tensor applySlice(
    const Tensor& self,
    int64_t dim,
    int64_t start,
    int64_t stop,
    int64_t step,
    bool disable_slice_optimization,
    const at::Device& self_device,
    const IntArrayRef& self_sizes) {
  // TODO: implement negative step
  TORCH_CHECK_VALUE(step > 0, "step must be greater than zero");

  // Skip this optimization if we are tracing, as the trace may be polymorphic
  // over the shape of the `self` tensor, and we still want to record
  // the slice.
  int64_t length = (self_device == at::kCPU || self_device == at::kCUDA) ? self_sizes[dim] : self.size(dim);
  if (!disable_slice_optimization && start == 0 && stop == length && step == 1) {
    return self;
  }
  return self.slice(dim, start, stop, step);
}

static inline Tensor applySelect(
    const Tensor& self,
    int64_t dim,
    int64_t index,
    int64_t real_dim,
    const at::Device& self_device,
    const IntArrayRef& self_sizes) {
  TORCH_CHECK_INDEX(
    !(index == 0 && dim == 0 && self_sizes.size() == 0),
    "invalid index of a 0-dim tensor. ",
    "Use `tensor.item()` in Python or `tensor.item<T>()` in C++ to convert a 0-dim tensor to a number");

  int64_t size = self_sizes[dim];
  TORCH_CHECK_INDEX(
    index >= -size && index < size,
    "index ", index, " is out of bounds for dimension ", real_dim, " with size ", size);

  // if the index is negative, do not normalize it because that would fix the index
  // on the current tensor size in the tracer.
  // aten::select also works on negative indices
  return self.select(dim, index);
}

static inline Tensor boolToIndexingTensorCPUOrCUDA(const Tensor& self, bool value) {
  // booleans add a dimension of size 1. true indexes this dimension as if 0:, false as empty.
  if (value) {
    return at::empty({1}, {}, self.options().dtype(kLong)).fill_(0.);
  } else {
    return at::empty({0}, {}, self.options().dtype(kLong));
  }
}

static inline Tensor boolToIndexingTensorNonNativeDeviceType(const Tensor& self, bool value) {
  // booleans add a dimension of size 1. true indexes this dimension as if 0:, false as empty.
  if (value) {
    return at::zeros({1}, {}, self.options().dtype(kLong));
  } else {
    return at::empty({0}, {}, self.options().dtype(kLong));
  }
}

static inline Tensor boolToIndexingTensor(const Tensor& self, bool value, const at::Device& self_device) {
  if (self_device == at::kCPU || self_device == at::kCUDA) {
    return boolToIndexingTensorCPUOrCUDA(self, value);
  } else {
    return boolToIndexingTensorNonNativeDeviceType(self, value);
  }
}

static inline Tensor scalarToTensorNonNativeDeviceType(const Scalar& v, const TensorOptions& options) {
  return at::scalar_tensor(v, options);
}

static inline void recordTensorIndex(const Tensor& tensor, std::vector<Tensor>& outIndices, int64_t* dim_ptr) {
  // TODO: check scalarType
  outIndices.resize(*dim_ptr + 1);
  outIndices[*dim_ptr] = tensor;
  (*dim_ptr)++;
};

static inline c10::List<c10::optional<Tensor>> typeConvertIndices(const Tensor& self, std::vector<Tensor>&& indices) {
  c10::List<c10::optional<Tensor>> converted_inds;
  converted_inds.reserve(indices.size());
  for (size_t i = 0; i < indices.size(); ++i) {
    const auto &ind = indices[i];
    if (ind.defined()) {
      converted_inds.push_back(ind.to(ind.options().device(self.device())));
    } else {
      converted_inds.push_back(std::move(indices[i]));
    }
  }
  return converted_inds;
}

// NOTE: Why do we mirror instead of replace the `count_specified_dimensions` function
// in torch/csrc/autograd/python_variable_indexing.cpp? It's because
// `count_specified_dimensions` is on the hot path of Python tensor multi-dim indexing
// (i.e. it's called by `applySlicing` which is called by `THPVariable_getitem` /
// `THPVariable_setitem` when handling indexing of more than one dimension). If we were
// to merge the Python/C++ `count_specified_dimensions` function, on the Python side
// we would have to construct a `std::vector` container to be consumed by the C++
// `count_specified_dimensions` function, which adds 100s of nanoseconds overhead and
// is undesirable.
static inline int64_t count_specified_dimensions(const ArrayRef<TensorIndex>& indices) {
  // Count the number of indexed dimensions (everything but ellipsis and None)
  int64_t count = 0;
  for (auto& obj : indices) {
    if (obj.is_tensor()) {
      auto& tensor = obj.tensor();
      if (tensor.scalar_type() == kByte || tensor.scalar_type() == kBool) {
        count += tensor.dim();
      } else {
        count++;
      }
    } else if (!obj.is_none() && !obj.is_ellipsis() && !obj.is_boolean()) {
      count++;
    }
  }
  return count;
}
} // namespace impl

// NOTE: Many functions below are only for consumption from Python indexing
// implementation, they include:
//
// - `Tensor scalarToTensor(...)`
// - `IntArrayRef slicePrefix1sSize(...)`
// - `void copy_to(...)`
// - `Tensor handleDimInMultiDimIndexing(...)`
// - `Tensor dispatch_index(...)`
// - `Tensor dispatch_index_put_(...)`
// - `Tensor get_item(...)`
// - `void set_item(...)`
//
// The rest of the functions are in `at::indexing::impl` namespace, signifying
// that they shouldn't be used from Python indexing implementation.
static inline Tensor scalarToTensor(const Scalar& v, const TensorOptions& options, const at::Device& self_device) {
  if (self_device == at::kCPU) {
    return at::detail::scalar_tensor_static(v, options.dtype_opt()->toScalarType(), self_device);
  } else {
    return impl::scalarToTensorNonNativeDeviceType(v, options);
  }
}

// To match numpy semantics:
// As a special case for backwards compatibility,
// strip away unit dimensions from the left of 'src'
static inline IntArrayRef slicePrefix1sSize(const IntArrayRef& sizes) {
  size_t first_non1_src = sizes.size();
  for (size_t i = 0; i < sizes.size(); ++i) {
    if (sizes[i] != 1) {
      first_non1_src = i;
      break;
    }
  }

  return sizes.slice(first_non1_src);
}

static inline void copy_to(const Tensor& dst, const Tensor& src) {
  if (dst.sizes().equals(src.sizes())) {
    // A shortcut to avoid generating hard-coded constant sizes during tracing.
    // This is not a perfect solution: when src & dst have different shapes, constants will still
    // appear. Users can workaround that case by dst[index..] = src.reshape(..)
    dst.copy_(src);
    return;
  }
  auto src_view = src.view(slicePrefix1sSize(src.sizes()));
  c10::MaybeOwned<Tensor> b_src = expand_inplace(dst, src_view, "setitem");
  dst.copy_(*b_src);
}

// See NOTE [ Setting `disable_slice_optimization` when calling C++ tensor indexing functions from Python ]
static inline Tensor handleDimInMultiDimIndexing(
    const Tensor& prev_dim_result,
    const Tensor& original_tensor,
    const TensorIndex& index,
    int64_t* dim_ptr,
    int64_t* specified_dims_ptr,
    int64_t real_dim,
    std::vector<Tensor>& outIndices,
    bool disable_slice_optimization,
    const at::Device& original_tensor_device,
    const IntArrayRef& prev_dim_result_sizes) {
  if (index.is_integer()) {
    return impl::applySelect(prev_dim_result, *dim_ptr, index.integer(), real_dim, original_tensor_device, prev_dim_result_sizes);
  } else if (index.is_slice()) {
    Tensor result = impl::applySlice(
      prev_dim_result,
      *dim_ptr,
      index.slice().start(),
      index.slice().stop(),
      index.slice().step(),
      /*disable_slice_optimization=*/disable_slice_optimization,
      original_tensor_device,
      prev_dim_result_sizes);
    (*dim_ptr)++;
    return result;
  } else if (index.is_ellipsis()) {
    (*dim_ptr) += original_tensor.dim() - (*specified_dims_ptr);
    return prev_dim_result;
  } else if (index.is_none()) {
    Tensor result = prev_dim_result.unsqueeze(*dim_ptr);
    (*dim_ptr)++;
    return result;
  } else if (index.is_boolean()) {
    Tensor result = prev_dim_result.unsqueeze(*dim_ptr);
    impl::recordTensorIndex(impl::boolToIndexingTensor(result, index.boolean(), original_tensor_device), outIndices, dim_ptr);
    return result;
  } else if (index.is_tensor()) {
    Tensor result = prev_dim_result;
    const Tensor& tensor = index.tensor();
    auto scalar_type = tensor.scalar_type();
    if (tensor.dim() == 0 && at::isIntegralType(scalar_type, /*includeBool=*/true)) {
      if (scalar_type != at::kByte && scalar_type != at::kBool) {
        result = impl::applySelect(result, *dim_ptr, tensor.item<int64_t>(), real_dim, original_tensor_device, prev_dim_result_sizes);
      } else {
        result = result.unsqueeze(*dim_ptr);
        if (scalar_type == at::kBool) {
          impl::recordTensorIndex(impl::boolToIndexingTensor(result, tensor.item<bool>() != 0, original_tensor_device), outIndices, dim_ptr);
        } else {
          impl::recordTensorIndex(impl::boolToIndexingTensor(result, tensor.item<uint8_t>() != 0, original_tensor_device), outIndices, dim_ptr);
        }
      }
    } else {
      impl::recordTensorIndex(tensor, outIndices, dim_ptr);
    }
    return result;
  } else {
    TORCH_INTERNAL_ASSERT(false, "Invalid TensorIndex type");
  }
}

namespace impl {
// This mirrors `applySlicing` in torch/csrc/autograd/python_variable_indexing.cpp
static inline Tensor applySlicing(
    const Tensor& self,
    const ArrayRef<TensorIndex>& indices,
    std::vector<Tensor>& outIndices,
    bool disable_slice_optimization,
    const at::Device& self_device,
    const IntArrayRef& self_sizes) {
  int64_t dim = 0;
  int64_t specified_dims = impl::count_specified_dimensions(indices);

  TORCH_CHECK_INDEX(
    specified_dims <= (int64_t)self_sizes.size(),
    "too many indices for tensor of dimension ", (int)self_sizes.size());

  Tensor result = self;
  for (size_t i = 0; i < indices.size(); i++) {
    auto& obj = indices[i];
    result = handleDimInMultiDimIndexing(
      /*prev_dim_result=*/result,
      /*original_tensor=*/self,
      /*index=*/obj,
      /*dim=*/&dim,
      /*specified_dims=*/&specified_dims,
      /*real_dim=*/i,
      /*outIndices=*/outIndices,
      /*disable_slice_optimization=*/disable_slice_optimization,
      /*original_tensor_device=*/self_device,
      /*prev_dim_result_sizes=*/result.sizes());
  }
  return result;
}
} // namespace impl

static inline Tensor dispatch_index(const Tensor& self, std::vector<Tensor>&& indices) {
  return self.index(impl::typeConvertIndices(self, std::move(indices)));
}

static inline Tensor dispatch_index_put_(Tensor& self, std::vector<Tensor>&& indices, const Tensor& value) {
  return self.index_put_(impl::typeConvertIndices(self, std::move(indices)), value);
}

// NOTE [ Setting `disable_slice_optimization` when calling C++ tensor indexing functions from Python ]
//
// Question: When should we set `disable_slice_optimization` to `true` when calling C++ tensor indexing
// functions from Python indexing code?
//
// Answer: What "slice optimization" means: when we have a slicing expression like `x[0:5, 0]`, where the sliced tensor
// was of size 5 in dimension 0, we would skip dispatching the actual slice call as an optimization. However, here are
// the cases where we DON'T want this optimization:
//
// 1. When we are doing 1-D slicing (e.g. `tensor[:]`).
//    Reason: we always return a shallow copy for expressions such as `tensor[:]` / `tensor[...]` / `tensor[:, :]`.
//    (Note that for `tensor[:, :]`, we return an alias of `tensor` by doing the following:
//    ```
//    Tensor sliced = impl::applySlicing(self, indices, tensorIndices, disable_slice_optimization, self_device, self_sizes);
//    if (tensorIndices.empty()) {
//      if (sliced.is_same(self)) {
//        // ensure we return a shallow copy for things like x[...]
//        sliced = at::alias(sliced);
//      }
//      return sliced;
//    }
//    ```)
// 2. When we are doing JIT tracing.
//    Reason: JIT tracing needs the `self.slice(...)` call to properly trace the slice operation.

// This mirrors `THPVariable_getitem` in torch/csrc/autograd/python_variable_indexing.cpp
// See NOTE [ Setting `disable_slice_optimization` when calling C++ tensor indexing functions from Python ]
static inline Tensor get_item(const Tensor& self, const ArrayRef<TensorIndex>& indices, bool disable_slice_optimization = false) {
  at::Device self_device = self.device();
  IntArrayRef self_sizes = self.sizes();

  // handle simple types: integers, slices, none, ellipsis, bool
  if (indices.size() == 1) {
    const TensorIndex& index = indices[0];
    if (index.is_integer()) {
      return impl::applySelect(self, 0, index.integer(), 0, self_device, self_sizes);
    } else if (index.is_slice()) {
      return impl::applySlice(
        self,
        0,
        index.slice().start(),
        index.slice().stop(),
        index.slice().step(),
        /*disable_slice_optimization=*/true,
        self_device,
        self_sizes);
    } else if (index.is_none()) {
      return self.unsqueeze(0);
    } else if (index.is_ellipsis()) {
      return at::alias(self);
    } else if (index.is_boolean()) {
      Tensor result = self.unsqueeze(0);
      return dispatch_index(
        result,
        std::vector<Tensor>{impl::boolToIndexingTensor(result, index.boolean(), self_device)}
      );
    }
  }

  std::vector<Tensor> tensorIndices;
  Tensor sliced = impl::applySlicing(self, indices, tensorIndices, disable_slice_optimization, self_device, self_sizes);
  if (tensorIndices.empty()) {
    if (sliced.is_same(self)) {
      // ensure we return a shallow copy for things like x[...]
      sliced = at::alias(sliced);
    }
    return sliced;
  }

  // indexing by tensors ("advanced" indexing)
  return dispatch_index(sliced, std::move(tensorIndices));
}

// This mirrors `THPVariable_setitem` in torch/csrc/autograd/python_variable_indexing.cpp
// for "the assigned value is a Tensor" case
// See NOTE [ Setting `disable_slice_optimization` when calling C++ tensor indexing functions from Python ]
static inline void set_item(const Tensor& self, const ArrayRef<TensorIndex>& indices, const Tensor& value, bool disable_slice_optimization = false) {
  at::Device self_device = self.device();
  IntArrayRef self_sizes = self.sizes();

  // handle simple types: integers, slices, ellipsis, bool
  if (indices.size() == 1) {
    const TensorIndex& index = indices[0];
    if (index.is_boolean() && !index.boolean()) {
      // do nothing for false (technically we should check the size, but we don't have
      // real 0-sized shapes.
      return;
    } else if (index.is_ellipsis()) {
      copy_to(self, value);
      return;
    } else if (index.is_none() || (index.is_boolean() && index.boolean())) {
      copy_to(self.unsqueeze(0), value);
      return;
    } else if (index.is_integer()) {
      copy_to(impl::applySelect(self, 0, index.integer(), 0, self_device, self_sizes), value);
      return;
    } else if (index.is_slice()) {
      copy_to(impl::applySlice(
        self,
        0,
        index.slice().start(),
        index.slice().stop(),
        index.slice().step(),
        /*disable_slice_optimization=*/disable_slice_optimization,
        self_device,
        self_sizes), value);
      return;
    }
  }

  std::vector<Tensor> tensorIndices;
  Tensor sliced = impl::applySlicing(self, indices, tensorIndices, disable_slice_optimization, self_device, self_sizes);
  if (tensorIndices.empty()) {
    copy_to(sliced, value);
    return;
  }

  IntArrayRef valueSizes = value.sizes();
  IntArrayRef slicedValueSizes = slicePrefix1sSize(valueSizes);
  Tensor valuesSliced;
  if (!valueSizes.equals(slicedValueSizes)) {
    valuesSliced = value.view(slicedValueSizes);
  } else {
    valuesSliced = value;
  }
  dispatch_index_put_(sliced, std::move(tensorIndices), valuesSliced);
  return;
}

} // namespace indexing
} // namespace at