UFL conditions cannot be evaluated as bool in a Python context

dolfinx
1

Hello everybody,
I have a problem with the imposition of conditions when I activate the dolfin module. Specifically, in my script I have such conditions:

if (t1<(Day_after_surgery*86400)): 
        k_c=0
        k_r=0
    elif (t1>(Day_after_surgery*86400) and t1<((Day_after_surgery+42)*86400)): 
        k_r=0.0648/86400
        k_c=0.00735/86400

from which I get this output:

UFL conditions cannot be evaluated as bool in a Python context.
Traceback (most recent call last):
  File "/Users/lorenzomarta/Desktop/Script simulazioni/Test10/GBM_Besta_10.py", line 341, in <module>
    if (t1<(Day_after_surgery*86400)): #pre-inizio terapia
  File "/Applications/anaconda3/envs/fenicsproject/lib/python3.10/site-packages/ufl/conditional.py", line 46, in __bool__
    error("UFL conditions cannot be evaluated as bool in a Python context.")
  File "/Applications/anaconda3/envs/fenicsproject/lib/python3.10/site-packages/ufl/log.py", line 172, in error
    raise self._exception_type(self._format_raw(*message))
ufl.log.UFLException: UFL conditions cannot be evaluated as bool in a Python context.

How could I solve?
Thanks in advance

2

Without a minimal example reproducing the error it is hard to give any guidance as to how to solve your problem.

3

In my script I have inserted the following libraries:

from dolfin import *
import numpy as np
import sys

I used the following compiler options:

parameters["form_compiler"]["cpp_optimize"] = True
parameters["form_compiler"]["representation"] = "uflacs"
parameters["form_compiler"]["quadrature_degree"] = 4

After that I inserted the mesh, defined the variables, the function space and the function class:

# Finite Element function spaces
V = FiniteElement("Lagrange", mesh.ufl_cell(), 1) 
ME = FunctionSpace(mesh, V)

# FE-space to determine the initial condition of phi
element1 = MixedElement([V,V,V])
ME1 = FunctionSpace(mesh, element1)

# Trial functions at time n+1
solution1 = Function(ME1)
phi, mu, n = split(solution1)

# Trial functions at time n
solution1_old = Function(ME1)
phi_old, mu_old, n_old = split(solution1_old) 

# Trial functions for derivatives
d_solution = TrialFunction(ME1)
d_phi, d_mu, d_n = split(d_solution)

# Test functions
varphi, vv, q = TestFunctions(ME1)

# ---------------------- Cahn-Hilliard Class ----------------------------------------------- #

# Interface-class for the Newton solver
class CahnHilliardEquation(NonlinearProblem):
    #def __init__(self, a, L, bc):
    def __init__(self, a, L):
        NonlinearProblem.__init__(self)
        self.L = L
        self.a = a
        #self.bc = bc
        self.reset_sparsity = True
    def F(self, b, x):
        assemble(self.L, tensor=b)
        #self.bc.apply(b, x)
    def J(self, A, x):
        assemble(self.a, tensor=A)
        #self.bc.apply(A)
        self.reset_sparsity = False

At this point the initial condition has been defined and a time cycle has been started to update the solution. the conditions on the parameters k_c, k_s indicated above are found inside the loop.

4

In the code you have supplied, you have not defined
the variables in

or

which is the cause of your error message, it is still impossible to give you guidance.

5

Day_after_surgery is defined as follows:

Day_after_surgery = int(input('How many days after the surgery did the treatment start? '))
T_tot = 131 + Day_after_surgery

while t1 is defined as t1 = 0.0 before the start of the cycle and updated at each step with the addition of dt1. Specifically, we have that:

deltaT = (T_tot/301)*86400
dt1 = Constant(deltaT)

I made a mistake, the constraint on t1 is as follows:

if (t1<(Day_after_surgery*86400)): 
        k_c=0
        k_r=0
    elif (t1>(Day_after_surgery*86400) and t1<((Day_after_surgery+42)*86400)): 
        k_r=0.0648/86400
        k_c=0.00735/86400
6

Are you assiging t1 as
t1 += dt1?
You should convert use deltaT to update t and not dt1.

7

It would seem to work, thanks !!!

However I have another problem now, modifying inside the “for loop” my FO function passing by:

   F0 = phi * varphi * dx - phi_old * varphi * dx \
         + deltaT * (1/M_0) *  inner(mat_T * grad(mu), grad(varphi)) * dx \
         - deltaT * nu * (n - delta) * (1/2) * (1 + phi_old) * varphi * dx \

to

   F0 = phi * varphi * dx - phi_old * varphi * dx \
         + deltaT * (1/M_0) *  inner(mat_T * grad(mu), grad(varphi)) * dx \
         - deltaT * nu * (n - delta) * (1/2) * (1 + phi_old) * varphi * dx \
         + deltaT * (k_r + k_c) * phi_old * varphi * dx \

where phi is the function with respect to which the weak formulation is being solved, varphi the respective test function, deltaT the time step, k_s, k_c, nu, n and M_0 are constants, while mat_T a matrix.
I get the following output:

 File "/Users/lorenzomarta/Desktop/Script simulazioni/Test9/GBM_Besta_9.py", line 369, in <module>
    + dt1 * (k_r + k_c) * phi_old * varphi * dx \
  File "/Applications/anaconda3/envs/fenicsproject/lib/python3.10/site-packages/ufl/measure.py", line 441, in __rmul__
    error("This integral is missing an integration domain.")
  File "/Applications/anaconda3/envs/fenicsproject/lib/python3.10/site-packages/ufl/log.py", line 172, in error
    raise self._exception_type(self._format_raw(*message))
ufl.log.UFLException: This integral is missing an integration domain.

Since the differential term for integration is dx, I don’t understand what’s wrong, as before the last addition everything works fine.

8

You keep on just adding snippets of code that cannot reproduce the error.
This makes it very hard for anyone to give any guidance.
What I would suggest is overload dx with dx = Measure("dx", domain=mesh).