[Shibi]

Hello,

I would like to code a new fixedValue boundary condition where I assign the value of each face center to the its corresponding x-coordinate.

I would like to know what is the correct procedure for implementing the boundary condition within openFoam.

Currently, I have:

void Foam::test::updateCoeffs()
{
    if (updated())
    {
        return;
    }
  
    const vectorField& Cf = patch().Cf();
 
    scalarField result (Cf.size(), Zero);
 
    forAll(result, faceI)
    {
        const scalar x = Cf[faceI].x();        
        result[faceI] = x;
    }
 
    operator==(result);
 
    fixedValueFvPatchScalarField::updateCoeffs();
}

But I believe this can also be achieved with:

void Foam::test::updateCoeffs()
{
    if (updated())
    {
        return;
    }
  
    const vectorField& Cf = patch().Cf();
 
    scalarField& field = *this; 
 
    forAll(field, faceI)
    {
        const scalar x = Cf[faceI].x();        
        field[faceI] = x;
    }
 
    fixedValueFvPatchScalarField::updateCoeffs();
}

Which version is more correct in OpenFOAM? overloading the == operator, or change directly the values of the field?

[s1291]

Do you need to create a new BC just for that? You can use the codedFixedValue BC:

https://www.openfoam.com/documentati...xed-value.html

[Shibi]

Originally Posted by s1291

Do you need to create a new BC just for that? You can use the codedFixedValue BC:

https://www.openfoam.com/documentati...xed-value.html

Hi,
I am aware of those coded boundaries. But I would like to create a new one for coding practice.
If both approaches are valid what is the advantage of overloading the == operator? Isn't it cleaner to just change the value of the field?

[s1291]

In this case, you should use the version with operator==

I think you will find the detailed answer in the following thread post #2:

Difference between = and ==

[Shibi]

Originally Posted by s1291

In this case, you should use the version with operator==

I think you will find the detailed answer in the following thread post #2:

Difference between = and ==

Thank you for the reply.
Just one further comment. You can see in here: https://cfd-training.com/2018/05/07/...dedfixedvalue/ and http://www.wolfdynamics.com/wiki/pro...streamINIT.pdf (I know that they are not official OF sources)

That the authors use the approach with *this.

Is the == doing anything else apart form assigning the value inside the parenthesis to the field?

If not, is it worth having an extra vector on the boundary to just store calculation and afterwards assign it to the field? Isn't this more inefficient?

[olesen]

Originally Posted by s1291

In this case, you should use the version with operator==

I think you will find the detailed answer in the following thread post #2:

Difference between = and ==

Careful. That discussion about == vs = pertains to GeometricFields, not to plain Fields!


For patch fields. Compare the implementation of '==' vs '=':
https://develop.openfoam.com/Develop...chField.C#L378
https://develop.openfoam.com/Develop...chField.C#L554


I don't think that you will see any difference in performance or overhead, or its meaning.

[olesen]

Originally Posted by Shibi

void Foam::test::updateCoeffs()
{
    if (updated())
    {
        return;
    }
  
    const vectorField& Cf = patch().Cf();
 
    scalarField& field = *this; 
 
    forAll(field, faceI)
    {
        const scalar x = Cf[faceI].x();        
        field[faceI] = x;
    }
 
    fixedValueFvPatchScalarField::updateCoeffs();
}

Which version is more correct in OpenFOAM? overloading the == operator, or change directly the values of the field?

If you are setting up the fields, there is no reason not to set them directly, like in your second code snippet.

[s1291]

Originally Posted by olesen

Careful. That discussion about == vs = pertains to GeometricFields, not to plain Fields!


For patch fields. Compare the implementation of '==' vs '=':
https://develop.openfoam.com/Develop...chField.C#L378
https://develop.openfoam.com/Develop...chField.C#L554


I don't think that you will see any difference in performance or overhead, or its meaning.

Dear Mark,

Thank you very much for your clarification.