Fix type derivation of kernel buffers (nest#936)

pynestml/codegeneration/nest_code_generator.py

@@ -846,10 +846,8 @@ def get_spike_update_expressions(self, neuron: ASTNeuron, kernel_buffers, solver
 
  for kernel_var in kernel.get_variables():
  for var_order in range(ASTUtils.get_kernel_var_order_from_ode_toolbox_result(kernel_var.get_name(), solver_dicts)):
- kernel_spike_buf_name = ASTUtils.construct_kernel_X_spike_buf_name(
- kernel_var.get_name(), spike_input_port, var_order)
- expr = ASTUtils.get_initial_value_from_ode_toolbox_result(
- kernel_spike_buf_name, solver_dicts)
+ kernel_spike_buf_name = ASTUtils.construct_kernel_X_spike_buf_name(kernel_var.get_name(), spike_input_port, var_order)
+ expr = ASTUtils.get_initial_value_from_ode_toolbox_result(kernel_spike_buf_name, solver_dicts)
  assert expr is not None, "Initial value not found for kernel " + kernel_var
  expr = str(expr)
  if expr in ["0", "0.", "0.0"]:

pynestml/utils/ast_utils.py

@@ -28,6 +28,7 @@
 
 from pynestml.codegeneration.printers.ast_printer import ASTPrinter
 from pynestml.codegeneration.printers.cpp_variable_printer import CppVariablePrinter
+from pynestml.codegeneration.printers.nestml_printer import NESTMLPrinter
 from pynestml.generated.PyNestMLLexer import PyNestMLLexer
 from pynestml.meta_model.ast_assignment import ASTAssignment
 from pynestml.meta_model.ast_block import ASTBlock
@@ -994,7 +995,7 @@ def add_declarations_to_state_block(cls, neuron: ASTNeuron, variables: List, ini
  return neuron
 
  @classmethod
- def add_declaration_to_state_block(cls, neuron: ASTNeuron, variable: str, initial_value: str) -> ASTNeuron:
+ def add_declaration_to_state_block(cls, neuron: ASTNeuron, variable: str, initial_value: str, type_str: str = "real") -> ASTNeuron:
  """
  Adds a single declaration to an arbitrary state block of the neuron. The declared variable is of type real.
  :param neuron: a neuron
@@ -1007,7 +1008,7 @@ def add_declaration_to_state_block(cls, neuron: ASTNeuron, variable: str, initia
 
  tmp = ModelParser.parse_expression(initial_value)
  vector_variable = ASTUtils.get_vectorized_variable(tmp, neuron.get_scope())
- declaration_string = variable + ' real' + (
+ declaration_string = variable + " " + type_str + (
  '[' + vector_variable.get_vector_parameter() + ']'
  if vector_variable is not None and vector_variable.has_vector_parameter() else '') + ' = ' + initial_value
  ast_declaration = ModelParser.parse_declaration(declaration_string)
@@ -1619,12 +1620,13 @@ def update_initial_values_for_odes(cls, neuron: ASTNeuron, solver_dicts: List[di
 
  @classmethod
  def create_initial_values_for_kernels(cls, neuron: ASTNeuron, solver_dicts: List[dict], kernels: List[ASTKernel]) -> None:
- """
+ r"""
  Add the variables used in kernels from the ode-toolbox result dictionary as ODEs in NESTML AST
  """
  for solver_dict in solver_dicts:
  if solver_dict is None:
  continue
+
  for var_name in solver_dict["initial_values"].keys():
  if cls.variable_in_kernels(var_name, kernels):
  # original initial value expressions should have been removed to make place for ode-toolbox results
@@ -1637,9 +1639,19 @@ def create_initial_values_for_kernels(cls, neuron: ASTNeuron, solver_dicts: List
  for var_name, expr in solver_dict["initial_values"].items():
  # overwrite is allowed because initial values might be repeated between numeric and analytic solver
  if cls.variable_in_kernels(var_name, kernels):
- expr = "0" # for kernels, "initial value" returned by ode-toolbox is actually the increment value; the actual initial value is assumed to be 0
+ spike_in_port_name = var_name.split("__X__")[1]
+ spike_in_port_name = spike_in_port_name.split("__d")[0]
+ spike_in_port = ASTUtils.get_input_port_by_name(neuron.get_input_blocks(), spike_in_port_name)
+ if spike_in_port:
+ type_str = NESTMLPrinter().print_data_type(spike_in_port.data_type)
+ differential_order: int = len(re.findall("__d", var_name))
+ if differential_order:
+ type_str += "*s**-" + str(differential_order)
+ else:
+ type_str = "real"
+ expr = "0 " + type_str # for kernels, "initial value" returned by ode-toolbox is actually the increment value; the actual initial value is assumed to be 0
  if not cls.declaration_in_state_block(neuron, var_name):
- cls.add_declaration_to_state_block(neuron, var_name, expr)
+ cls.add_declaration_to_state_block(neuron, var_name, expr, type_str)
 
  @classmethod
  def transform_ode_and_kernels_to_json(cls, neuron: ASTNeuron, parameters_blocks: Sequence[ASTBlockWithVariables],

pynestml/utils/messages.py

@@ -1050,7 +1050,7 @@ def templated_arg_types_inconsistent(cls, function_name, failing_arg_idx, other_
  """
  message = 'In function \'' + function_name + '\': actual derived type of templated parameter ' + \
  str(failing_arg_idx + 1) + ' is \'' + failing_arg_type_str + '\', which is inconsistent with that of parameter(s) ' + \
- ', '.join([str(_ + 1) for _ in other_args_idx]) + ', which have type \'' + other_type_str + '\''
+ ', '.join([str(_ + 1) for _ in other_args_idx]) + ', which has/have type \'' + other_type_str + '\''
  return MessageCode.TEMPLATED_ARG_TYPES_INCONSISTENT, message
 
  @classmethod