}];

  let arguments = (ins
    StrAttr:$node_def,
    I32Attr:$op_key
  );

  let assemblyFormat = "attr-dict";
}

def ConstOp: TensorflowMlrt_Op<"constop", []> {
  let summary = "The tf_mlrt ConstOp";

  let description = [{
    The ConstOp creates a constant tensorflow::Tensor from serialized proto.
  }];

  let arguments = (ins
    StrAttr:$tensor_proto
  );

  let results = (outs

// PadOp and CastOp.
std::optional<BlockArgumentInfo> GetInputBlockArgNum(Value input) {
  auto block_arg_num = GetBlockArgNum(input);
  if (block_arg_num.has_value()) return block_arg_num;

  Value next_input = input;
  auto pad_op = dyn_cast_or_null<TF::PadOp>(next_input.getDefiningOp());
  auto cast_op = dyn_cast_or_null<TF::CastOp>(next_input.getDefiningOp());

namespace tflite {
namespace {

bool IsConst(mlir::Operation* op) {
  return llvm::isa<mlir::arith::ConstantOp, mlir::TF::ConstOp,
                   mlir::TFL::ConstOp, mlir::TFL::QConstOp>(op);
}

bool IsOpSupported(mlir::Operation* op, const std::string& hardware) {
  auto* devce_hardware = mlir::TFL::tac::GetTargetHardware(hardware);
  if (devce_hardware == nullptr) return {};

#include "tensorflow/compiler/mlir/tensorflow/ir/tf_ops.h"

namespace mlir {
namespace quant {

std::optional<TF::PartitionedCallOp> ApplyUniformQuantization(
    PatternRewriter& rewriter, TF::ConstOp op,
    tensorflow::quantization::QuantizationComponentSpec& weight_spec);

}  // namespace quant
}  // namespace mlir

  tensorflow::Algorithm tensorflow_rng =
      tensorflow::ToTensorflowAlgorithm(xla_rng);

  OpBuilder opbuilder(get_alg_op);

  auto tf_const = opbuilder.create<TF::ConstOp>(
      get_alg_op->getLoc(), opbuilder.getI32IntegerAttr((int)tensorflow_rng));

  get_alg_op->replaceAllUsesWith(tf_const);
  get_alg_op->erase();
  return success();
}

namespace TFL {
namespace tac {

// Returns true if 'op' is non const op. Returns false otherwise or if
// 'op' is null.
inline bool IsNonConstOp(Operation* op) {
  if (!op) return false;
  if (llvm::isa<arith::ConstantOp, mlir::func::ConstantOp>(op)) return false;
  if (op->hasTrait<OpTrait::ConstantLike>()) return false;
  if (llvm::isa<TFL::ConstOp, TFL::QConstOp>(op)) return false;
  return true;
}

include "tensorflow/compiler/mlir/tensorflow/ir/tf_ops.td"
include "mlir/IR/OpBase.td"
include "mlir/Dialect/Arith/IR/ArithOps.td"

def ConstDenseElementsI32ZeroAttr
  : NativeCodeCall<"$_builder.create<TFL::ConstOp>($_loc, DenseElementsAttr::get(RankedTensorType::get({}, $_builder.getI32Type()), {0}))">;

def Size1InputRange : NativeCodeCall<
  "SmallVector<Value, 1>{$0}">;

def Size2InputRange : NativeCodeCall<

    return success();
  }
};

class ConvertTFConstOpToArithConstOp : public OpRewritePattern<TF::ConstOp> {
 public:
  using OpRewritePattern::OpRewritePattern;
  LogicalResult matchAndRewrite(TF::ConstOp op,
                                PatternRewriter& rewriter) const override {
    rewriter.replaceOpWithNewOp<arith::ConstantOp>(op, op.getValue());
    return success();
  }
};

      ShapedType shaped_type =
          RankedTensorType::get({1}, TF::StringType::get(builder.getContext()));
      auto const_op = builder.create<TF::ConstOp>(
          asset.getLoc(),
          DenseStringElementsAttr::get(shaped_type, {filename}));
      for (auto init_op : init_table_from_text_file_ops_to_erase) {
        // Replace the InitializeTableFromTextFileV2Op to use the saved model's
        // asset filepath.

  } else {
    // Convert I64 const array to I32.
    llvm::SmallVector<int32_t> const_i32_vec;
    for (auto element : const_value) {
      const_i32_vec.push_back(static_cast<int32_t>(element));
    }
    const_value_raw = rewriter.getI32TensorAttr(const_i32_vec);
  }
  Value result_const = builder.create<TF::ConstOp>(const_value_raw);
  return result_const;
}