71 mooseError(
"I'm an AD object, so computeQpJacobian should never be called");
Eigen::Matrix< ADReal, Eigen::Dynamic, 1 > ADRealEigenVector
virtual void computeOffDiagJacobian(unsigned int jvar) override
Compute the off-diagonal Jacobian at one node.
ADRealEigenVector _work_vector
Work vector for residual.
const InputParameters & parameters() const
Get the parameters of the object.
virtual void computeJacobian() override
Compute the Jacobian at one node.
static InputParameters validParams()
Class constructor.
const MooseVariableFE< RealEigenVector > & variable() const override
Gets the variable this is active on.
virtual void computeResidual() override
Compute the residual at the current node.
virtual RealEigenVector computeQpJacobian()
Dummy method so we can make derived generic classes that template on <bool is_ad>=""> ...
const unsigned int _count
Number of components of the array variable.
Base class for creating nodal kernels with hand-coded Jacobians.
const ADArrayVariableValue & _u
Value of the unknown variable this is acting on.
MooseVariableFE< RealEigenVector > & _var
variable this works on
const Node * _my_node
Cache variable to make sure we don't do duplicate AD computations.
virtual void jacobianSetup() override
Gets called just before the Jacobian is computed and before this object is asked to do its job...
Base class for creating new types of nodal kernels.
Interface for objects that need to get values of MooseVariables.
void mooseError(Args &&... args) const
Emits an error prefixed with object name and type and optionally a file path to the top-level block p...
ADArrayNodalKernel(const InputParameters ¶meters)
Eigen::Matrix< Real, Eigen::Dynamic, 1 > RealEigenVector
virtual void computeQpResidual(ADRealEigenVector &residual)=0
The user can override this function to compute the residual at a node.