}
    ]


def getProduct(product_id):
    products = getProducts()
    if product_id + 1 > len(products):
        return None
    else:
        return products[product_id]


def getProductDetails(product_id, headers):
    try:

                                                /*transpose_a=*/op.getAdjX(),
                                                /*transpose_b=*/op.getAdjY());
    matmuls.emplace_back(matmul.getProduct());
  }

  // Combine the result of each individual MatMul into a rank-3 tensor.
  Type packed_type = RankedTensorType::get(
      {bcast.output_batch_size(), rows, cols}, element_type);

    auto op = cast<SparseMatMulOp>(src_op);

    // Result type must be f32 for applying the pattern (currently this is
    // required by the op anyway but this might change).
    if (!mlir::cast<TensorType>(op.getProduct().getType())
             .getElementType()
             .isF32()) {
      return failure();
    }
    MLIRContext *context = rewriter.getContext();