Print properties contained in Seawater prop model #1686

adam-a-a · 2025-11-12T05:37:51Z

Fixes/Resolves:

NA

Summary/Motivation:

While preparing materials for Academy, realized it isn't very easy to list all properties on a property model.

One would need to dive into source code or technical documentation.

This PR attempts to add list_properties and list_properties_as_dataframe methods to the SeawaterParameterBlock, using a new utility helper function, print_property_metadata.

This also required defining a separate PropertySet for Seawater--otherwise, you'll get back a long list of all supported IDAES properties that aren't necessarily supported in the seawater prop model.

Can extend this to other property models once we cover this PR.

Example implementation:

m = ConcreteModel()
m.fs = FlowsheetBlock(dynamic=False)
m.fs.seawater_properties = SeawaterParameterBlock()
m.fs.seawater_properties.list_properties()

Output:

Property Description         Model Attribute          Units          
---------------------------------------------------------------------
Boiling point elevation      boiling_point_elevation  K              
Mass concentration           conc_mass                kg*m**(-3)     
Specific heat capacity       cp_mass                  J/(kg*K)       
Mass density of solution     dens_mass                kg*m**(-3)     
Mass density of pure water   dens_mass_solvent        kg*m**(-3)     
Latent heat of vaporization  dh_vap_mass              J*kg**(-1)     
Diffusivity                  diffus                   m**2/s         
Enthalpy flow                enth_flow                J/s            
Specific enthalpy            enth_mass                J*kg**(-1)     
Mass flow rate               flow_mass                kg/s           
Molar flowrate               flow_mol_phase_comp      mol/s          
Total volumetric flow rate   flow_vol                 m**3/s         
Mass fraction                mass_frac                dimensionless  
Molality                     molality                 mol/kg         
Mole fraction                mole_frac                dimensionless  
Osmotic coefficient          osm_coeff                dimensionless  
Pressure                     pressure                 Pa             
Osmotic pressure             pressure_osm             Pa             
Vapor pressure               pressure_sat             Pa             
Temperature                  temperature              K              
Thermal conductivity         therm_cond               W/(m*K)        
Dynamic viscosity            visc_d                   Pa*s

Changes proposed in this PR:

add helper function to print or return df of properties
add methods to seawater parameterblock that leverage the helper function
add tests for helper function
add new SeawaterPropertySet that inherits from PropertySetBase instead of StandardPropertySet, which will include properties that aren't in the seawater prop model when printing out

Legal Acknowledgement

By contributing to this software project, I agree to the following terms and conditions for my contribution:

I agree my contributions are submitted under the license terms described in the LICENSE.txt file at the top level of this directory.
I represent I am authorized to make the contributions and grant the license. If my employer has rights to intellectual property that includes these contributions, I represent that I have received permission to make contributions and grant the required license on behalf of that employer.

…names, descriptions, and units later

…ater prop model

dangunter · 2025-11-12T14:45:34Z

watertap/core/util/property_helpers.py

+import pandas as pd
+
+
+def print_property_metadata(prop_pkg, return_df=False):


I think a simpler interface would be get_property_metadata. Wrapping that in a print statement is very little burden on the caller and then you don't need this keyword.

do you mean make it a method on parameterblock so that I don't need prop_pkg keyword? Or did you mean remove the need for returning df in this function and focus on getting the metadata?

I think he means rename this get_property_metadata and return the df and then the user can do what they want?

I like the idea of this being a method on ParameterBlock. But could you not just use this function and then @classmethod to avoid adding this code to every property model?

dangunter · 2025-11-12T14:46:16Z

watertap/core/util/property_helpers.py

+        prop_pkg: The property model ParameterBlock (e.g., m.fs.properties)
+        return_df: If True, returns a Pandas DataFrame instead of printing.
+    """
+    metadata = prop_pkg.get_metadata()


can this fail?

Here's the method (below). I suppose we can add a check for the _metadata attribute and raise an exception if it doesn't exist. Do you know of a better check to ensure the property model is a property model? Another idea is to check whether the prop_pkg input is a ParameterBlock.

@classmethod def get_metadata(cls): """Get property parameter metadata. If the metadata is not defined, this will instantiate a new metadata object and call `define_metadata()` to set it up. If the metadata is already defined, it will be simply returned. Returns: PropertyClassMetadata: The metadata """ if cls._metadata is None: pcm = PropertyClassMetadata() cls.define_metadata(pcm) cls._metadata = pcm # Check that the metadata was actually populated # Check requires looking at private attributes # pylint: disable-next=protected-access if pcm._properties is None or pcm._default_units is None: raise PropertyPackageError( "Property package did not populate all expected metadata." ) return cls._metadata

dangunter · 2025-11-12T14:48:30Z

watertap/core/util/property_helpers.py

+
+    if return_df:
+        return pd.DataFrame(
+            {"Property Description": docs, "Model Attribute": vars, "Units": units}


I would suggest column headers with no spaces. "Name", "Description", "Units". The prefix "Property" is implied in all cases.

ok will do!

Does it need to be a dataframe? Why not just a dict?

dangunter · 2025-11-12T14:55:44Z

watertap/core/util/property_helpers.py

+            {"Property Description": docs, "Model Attribute": vars, "Units": units}
+        )
+
+    # Pretty-print


so again, I would just return the data frame. People know how to print a dataframe in a basic way in a terminal (i.e. just pass obj to print()) and a notebook will do it for you. If you really must print for the user, I'd suggest a separate method print_property_metadata that consists of basically :

pprint.pprint(get_property_metadata(*args).to_string(index=False))

Is there an issue with the way it's printing now or any advantage to doing it differently?

I would argue we should not return a dataframe in the first place.

The reason is we don't want to maintain code specific to pandas for printing stuff.

I think we we do want to display/print information we should use native python display tools otherwise we might run into system quirks ... (I wounder if pyomo table printer would work for this - also can look in my PR for WaterTAP flowsheet block for displaying a table - it uses native pyomo printer... ?)

dangunter · 2025-11-12T14:57:36Z

watertap/core/util/tests/test_property_helpers.py

+    assert "kg/s" in df["Units"].values
+
+
+def test_print_property_metadata_pretty_print(capsys):


I would just test that there are, say, between 100 and 1000 characters in the output.

dangunter · 2025-11-12T14:58:46Z

watertap/property_models/seawater_prop_pack.py

        self.set_default_scaling("diffus_phase_comp", 1e9)
        self.set_default_scaling("boiling_point_elevation_phase", 1e0, index="Liq")

+    def list_properties(self, return_df=False):


these seem completely generic, so should be on a base class

yes, but I started with this first before attempting to add to a base class (which would probably be best on IDAES repo but would slow things down for us)

dangunter · 2025-11-12T15:00:03Z

watertap/property_models/seawater_prop_pack.py


        # transforming constraints
        transform_property_constraints(self)
+


I realize this is normal in IDAES but I find this way of creating data very verbose.

Me too, but given the current IDAES functionality and the way metadata is handled and reported back, this was the cleaner way of producing the list of property names actually supported on this property model, without the noise. Open to suggestions if there's a slicker approach, but this is meant a quick fix, with planned refinement in subsequent PRs (this quarter).

I feel like this is redundant, don't the pyomo vars in the property block contain all of the same information? Would it not be easier to make a list of the "vars" you want to display and simply grab it?

dangunter · 2025-11-12T15:00:47Z

watertap/property_models/tests/test_seawater_prop_pack.py

+    capsys.readouterr()
+    m.fs.props.list_properties()
+    captured = capsys.readouterr().out
+    expected_output = """


tets for string matches are not a good idea imho. at a minimum use a regex to look for some things least likely to change.

I was expecting this sort of response! I was aiming for a quick test to make sure my seawater prop model was listing the correct properties. Can reconsider this, but I was hoping to get something merged asap and revise in subsequent PR(s)

MarcusHolly

If the intent is just to get something merged in preparation for Friday, then I think this is good as is. None of Dan's comments strike me as urgent, so these (and other) refinements can be made in future PRs.

into add_property_sets

avdudchenko

Overall looks good, I am not 100% sure about relying on pandas data frames for data storage and printing.

My biggest question is can't we pull most of the information here from already build pyomo vars defining the supported properties?

I almost wounder if this just need a registration class that lets you track vars you want to report and provide ways to both interact with?

avdudchenko · 2025-11-12T22:11:52Z

watertap/core/util/property_helpers.py

+
+    if return_df:
+        return pd.DataFrame(
+            {"Property Description": docs, "Model Attribute": vars, "Units": units}


Does it need to be a dataframe? Why not just a dict?

avdudchenko · 2025-11-12T22:14:24Z

watertap/core/util/property_helpers.py

+            {"Property Description": docs, "Model Attribute": vars, "Units": units}
+        )
+
+    # Pretty-print


I would argue we should not return a dataframe in the first place.

The reason is we don't want to maintain code specific to pandas for printing stuff.

I think we we do want to display/print information we should use native python display tools otherwise we might run into system quirks ... (I wounder if pyomo table printer would work for this - also can look in my PR for WaterTAP flowsheet block for displaying a table - it uses native pyomo printer... ?)

avdudchenko · 2025-11-13T21:04:10Z

watertap/property_models/seawater_prop_pack.py


        # transforming constraints
        transform_property_constraints(self)
+


I feel like this is redundant, don't the pyomo vars in the property block contain all of the same information? Would it not be easier to make a list of the "vars" you want to display and simply grab it?

adam-a-a added 7 commits November 11, 2025 23:51

define SeawaterPropertySet to make it easier to extract all property …

4bdc330

…names, descriptions, and units later

correct erroneous implementation of fix_initialization_states on seaw…

6f66821

…ater prop model

cleanup

e737d35

add list_properties and list_properties_as_dataframe to seawater model

1f60e87

add property_helpers and tests

4b2b128

add copyright to property_helpers

00a7880

run black

966a7cb

adam-a-a requested review from MarcusHolly, dangunter, kurbansitterley and tristantc November 12, 2025 05:37

adam-a-a requested a review from bknueven as a code owner November 12, 2025 05:37

adam-a-a added 3 commits November 12, 2025 00:41

tweak doc string for seawaterpropertyset and BEP

27a4cd2

add test for printout on sw prop model; black

b60568e

Merge branch 'main' into add_property_sets

46d0c29

adam-a-a added the Priority:High High Priority Issue or PR label Nov 12, 2025

adam-a-a self-assigned this Nov 12, 2025

dangunter reviewed Nov 12, 2025

View reviewed changes

MarcusHolly approved these changes Nov 12, 2025

View reviewed changes

adam-a-a and others added 4 commits November 12, 2025 17:57

address Dan's suggestions

7c7581b

Merge branch 'add_property_sets' of https://github.com/adam-a-a/watertap

eeff5ed

into add_property_sets

ensure rows won't be truncated. display all rows by default

81b7357

Run black

f1ac879

avdudchenko approved these changes Nov 13, 2025

View reviewed changes

		import pandas as pd


		def print_property_metadata(prop_pkg, return_df=False):

		assert "kg/s" in df["Units"].values


		def test_print_property_metadata_pretty_print(capsys):


		# transforming constraints
		transform_property_constraints(self)

Print properties contained in Seawater prop model #1686

Are you sure you want to change the base?

Print properties contained in Seawater prop model #1686

Uh oh!

Conversation

adam-a-a commented Nov 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Fixes/Resolves:

Summary/Motivation:

Changes proposed in this PR:

Legal Acknowledgement

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

adam-a-a Nov 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

MarcusHolly left a comment

Choose a reason for hiding this comment

Uh oh!

avdudchenko left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

adam-a-a commented Nov 12, 2025 •

edited

Loading

adam-a-a Nov 12, 2025 •

edited

Loading