TestWallFriction fails with pedantic check

[dietmarw]

Whilst testing the major SI unit PR I checked the MSL (4.0.0-alpha.1 (almost)) in Dymola in pedantic mode and got the following error on Modelica.Fluid.Fittings.BaseClasses.QuadraticTurbulent.TestWallFriction:

Check of Modelica.Fluid.Fittings.BaseClasses.QuadraticTurbulent.TestWallFriction:

Component type specifier MassFlowRate is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 13
Component context: port_a.m_flow
Component declared as MassFlowRate m_flow in Modelica.Media.Interfaces.PartialMedium

Component type specifier MassFlowRate is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 13
Component context: port_b.m_flow
Component declared as MassFlowRate m_flow in Modelica.Media.Interfaces.PartialMedium

Component type specifier MassFlowRate is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 289
Component context: m_flow_start
Component declared as MassFlowRate m_flow_start in Modelica.Media.Interfaces.PartialMedium

Component type specifier MassFlowRate is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 293
Component context: m_flow_small
Component declared as MassFlowRate m_flow_small in Modelica.Media.Interfaces.PartialMedium

Component type specifier MassFlowRate is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 306
Component context: m_flow
Component declared as MassFlowRate m_flow in Modelica.Media.Interfaces.PartialMedium

Component type specifier ThermodynamicState is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 332
Component context: state_a
Component declared as ThermodynamicState state_a in Modelica.Media.Interfaces.PartialMedium

Component type specifier ThermodynamicState is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 333
Component context: state_b
Component declared as ThermodynamicState state_b in Modelica.Media.Interfaces.PartialMedium

Component type specifier MassFlowRate is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 714
Component context: m_flow
Component declared as MassFlowRate m_flow in Modelica.Media.Interfaces.PartialMedium

Component type specifier MassFlowRate is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Interfaces.mo, line 726
Component context: m_flow_start
Component declared as MassFlowRate m_flow_start in Modelica.Media.Interfaces.PartialMedium

Component type specifier ThermodynamicState is from a partial package. That is not correct.
File: /home/dietmarw/.workspace/gh-modelica/Modelica/Modelica/Fluid/Fittings.mo, line 1570
Component context: state_nominal
Component declared as ThermodynamicState state_nominal in Modelica.Media.Interfaces.PartialMedium

ERRORS have been issued.

[beutlich]

This issue is independent of #3379 and already can observed with MSL v3.2.3 in Dymola 2020.

[HansOlsson]

That error message when checking the specific model is a bit excessive (and not exclusive to this model as far as I can tell). It does not occur when checking the entire package.

It should say "It must be redeclared before being used." (as similar messages do); and the underlying problem is that the types are in the partial package Medium that must be redeclared.

(Note: Just so that no-one tries to "correct" this by making the model partial: you can currently drag in this model and redeclare the Medium - and then it just works; similarly as if a parameter lacked a value.)

[wischhusen]

I do not understand the purpose of the model. What kind of tests were intended? Even if a suitable fluid model is chosen failures will remain due to the missing inner System model. I would rather say let us remove the model ...

[beutlich]

It is not used anywhere inside MSL or ModelicaTest.

[HansOlsson]

It is not used anywhere inside MSL or ModelicaTest.

It is used in ModelicaTest.Fluid.TestComponents.Fittings.TestWallFriction - that compares different pipes and also this one. Thus one possibility would be to move it to ModelicaTest, as it seems it isn't intended for actual use (according to description).

[beutlich]

Ah, sorry. I mixed it up with Modelica.Fluid.Fittings.BaseClasses.QuadraticTurbulent.BaseModel which is nowhere used except TestWallFriction.