policyengine 3.1.12__tar.gz → 3.1.14__tar.gz

{policyengine-3.1.12 → policyengine-3.1.14}/CHANGELOG.md

@@ -5,6 +5,18 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), 
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [3.1.14] - 2025-12-10 21:59:24
+
+### Fixed
+
+- Improvements to loading taxbenefitsystems.
+
+## [3.1.13] - 2025-12-10 19:29:28
+
+### Fixed
+
+- Naming improvements.
+
 ## [3.1.12] - 2025-12-10 14:43:52
 
 ### Fixed
@@ -269,6 +281,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 
 
+[3.1.14]: https://github.com/PolicyEngine/policyengine.py/compare/3.1.13...3.1.14
+[3.1.13]: https://github.com/PolicyEngine/policyengine.py/compare/3.1.12...3.1.13
 [3.1.12]: https://github.com/PolicyEngine/policyengine.py/compare/3.1.11...3.1.12
 [3.1.11]: https://github.com/PolicyEngine/policyengine.py/compare/3.1.10...3.1.11
 [3.1.10]: https://github.com/PolicyEngine/policyengine.py/compare/3.1.9...3.1.10

{policyengine-3.1.12 → policyengine-3.1.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: policyengine
-Version: 3.1.12
+Version: 3.1.14
 Summary: A package to conduct policy analysis using PolicyEngine tax-benefit models.
 Author-email: PolicyEngine <hello@policyengine.org>
 License:                     GNU AFFERO GENERAL PUBLIC LICENSE

{policyengine-3.1.12 → policyengine-3.1.14}/changelog.yaml

@@ -220,3 +220,13 @@
     fixed:
     - Bug where all datasets would be the last year.
   date: 2025-12-10 14:43:52
+- bump: patch
+  changes:
+    fixed:
+    - Naming improvements.
+  date: 2025-12-10 19:29:28
+- bump: patch
+  changes:
+    fixed:
+    - Improvements to loading taxbenefitsystems.
+  date: 2025-12-10 21:59:24

{policyengine-3.1.12 → policyengine-3.1.14}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "policyengine"
-version = "3.1.12"
+version = "3.1.14"
 description = "A package to conduct policy analysis using PolicyEngine tax-benefit models."
 readme = "README.md"
 authors = [

policyengine-3.1.14/src/policyengine/core/parameter.py

@@ -0,0 +1,65 @@
+from typing import TYPE_CHECKING, Any
+from uuid import uuid4
+
+from pydantic import BaseModel, Field, PrivateAttr
+
+from .parameter_value import ParameterValue
+from .tax_benefit_model_version import TaxBenefitModelVersion
+
+if TYPE_CHECKING:
+    from .parameter_value import ParameterValue
+
+
+class Parameter(BaseModel):
+    model_config = {"arbitrary_types_allowed": True}
+
+    id: str = Field(default_factory=lambda: str(uuid4()))
+    name: str
+    label: str | None = None
+    description: str | None = None
+    data_type: type | None = None
+    tax_benefit_model_version: TaxBenefitModelVersion
+    unit: str | None = None
+
+    # Lazy loading: store core param ref, build values on demand
+    _core_param: Any = PrivateAttr(default=None)
+    _parameter_values: list["ParameterValue"] | None = PrivateAttr(
+        default=None
+    )
+
+    def __init__(self, _core_param: Any = None, **data):
+        super().__init__(**data)
+        self._core_param = _core_param
+        self._parameter_values = None
+
+    @property
+    def parameter_values(self) -> list["ParameterValue"]:
+        """Lazily build parameter values on first access."""
+        if self._parameter_values is None:
+            self._parameter_values = []
+            if self._core_param is not None:
+                from policyengine.utils import parse_safe_date
+
+                for i in range(len(self._core_param.values_list)):
+                    param_at_instant = self._core_param.values_list[i]
+                    if i + 1 < len(self._core_param.values_list):
+                        next_instant = self._core_param.values_list[i + 1]
+                    else:
+                        next_instant = None
+                    pv = ParameterValue(
+                        parameter=self,
+                        start_date=parse_safe_date(
+                            param_at_instant.instant_str
+                        ),
+                        end_date=parse_safe_date(next_instant.instant_str)
+                        if next_instant
+                        else None,
+                        value=param_at_instant.value,
+                    )
+                    self._parameter_values.append(pv)
+        return self._parameter_values
+
+    @parameter_values.setter
+    def parameter_values(self, value: list["ParameterValue"]) -> None:
+        """Allow direct setting of parameter values."""
+        self._parameter_values = value

{policyengine-3.1.12 → policyengine-3.1.14}/src/policyengine/core/tax_benefit_model_version.py

@@ -14,6 +14,8 @@ if TYPE_CHECKING:
 
 
 class TaxBenefitModelVersion(BaseModel):
+    model_config = {"arbitrary_types_allowed": True}
+
     id: str = Field(default_factory=lambda: str(uuid4()))
     model: TaxBenefitModel
     version: str
@@ -24,6 +26,14 @@ class TaxBenefitModelVersion(BaseModel):
     parameters: list["Parameter"] = Field(default_factory=list)
     parameter_values: list["ParameterValue"] = Field(default_factory=list)
 
+    # Lookup dicts for O(1) access (excluded from serialization)
+    variables_by_name: dict[str, "Variable"] = Field(
+        default_factory=dict, exclude=True
+    )
+    parameters_by_name: dict[str, "Parameter"] = Field(
+        default_factory=dict, exclude=True
+    )
+
     def run(self, simulation: "Simulation") -> "Simulation":
         raise NotImplementedError(
             "The TaxBenefitModel class must define a method to execute simulations."
@@ -39,40 +49,28 @@ class TaxBenefitModelVersion(BaseModel):
             "The TaxBenefitModel class must define a method to load simulations."
         )
 
-    def get_parameter(self, name: str) -> "Parameter":
-        """Get a parameter by name.
-
-        Args:
-            name: The parameter name (e.g., "gov.hmrc.income_tax.allowances.personal_allowance.amount")
+    def add_parameter(self, param: "Parameter") -> None:
+        """Add a parameter and index it for fast lookup."""
+        self.parameters.append(param)
+        self.parameters_by_name[param.name] = param
 
-        Returns:
-            Parameter: The matching parameter
+    def add_variable(self, var: "Variable") -> None:
+        """Add a variable and index it for fast lookup."""
+        self.variables.append(var)
+        self.variables_by_name[var.name] = var
 
-        Raises:
-            ValueError: If parameter not found
-        """
-        for param in self.parameters:
-            if param.name == name:
-                return param
+    def get_parameter(self, name: str) -> "Parameter":
+        """Get a parameter by name (O(1) lookup)."""
+        if name in self.parameters_by_name:
+            return self.parameters_by_name[name]
         raise ValueError(
             f"Parameter '{name}' not found in {self.model.id} version {self.version}"
         )
 
     def get_variable(self, name: str) -> "Variable":
-        """Get a variable by name.
-
-        Args:
-            name: The variable name (e.g., "income_tax", "household_net_income")
-
-        Returns:
-            Variable: The matching variable
-
-        Raises:
-            ValueError: If variable not found
-        """
-        for var in self.variables:
-            if var.name == name:
-                return var
+        """Get a variable by name (O(1) lookup)."""
+        if name in self.variables_by_name:
+            return self.variables_by_name[name]
         raise ValueError(
             f"Variable '{name}' not found in {self.model.id} version {self.version}"
         )

{policyengine-3.1.12 → policyengine-3.1.14}/src/policyengine/tax_benefit_models/uk/datasets.py

@@ -161,8 +161,8 @@ def create_datasets(
 
             uk_dataset = PolicyEngineUKDataset(
                 id=f"{Path(dataset).stem}_year_{year}",
-                name=f"{dataset}-year-{year}",
-                description=f"UK Dataset for year {year} based on {dataset}",
+                name=f"{Path(dataset).stem}-year-{year}",
+                description=f"UK Dataset for year {year} based on {Path(dataset).stem}",
                 filepath=f"{data_folder}/{Path(dataset).stem}_year_{year}.h5",
                 year=int(year),
                 data=UKYearData(
@@ -196,8 +196,8 @@ def load_datasets(
         for year in years:
             filepath = f"{data_folder}/{Path(dataset).stem}_year_{year}.h5"
             uk_dataset = PolicyEngineUKDataset(
-                name=f"{dataset}-year-{year}",
-                description=f"UK Dataset for year {year} based on {dataset}",
+                name=f"{Path(dataset).stem}-year-{year}",
+                description=f"UK Dataset for year {year} based on {Path(dataset).stem}",
                 filepath=filepath,
                 year=int(year),
             )

{policyengine-3.1.12 → policyengine-3.1.14}/src/policyengine/tax_benefit_models/uk/model.py

@@ -118,7 +118,6 @@ class PolicyEngineUKLatest(TaxBenefitModelVersion):
 
         self.id = f"{self.model.id}@{self.version}"
 
-        self.variables = []
         for var_obj in system.variables.values():
             variable = Variable(
                 id=self.id + "-" + var_obj.name,
@@ -140,9 +139,8 @@ class PolicyEngineUKLatest(TaxBenefitModelVersion):
                         var_obj.possible_values._value2member_map_.values(),
                     )
                 )
-            self.variables.append(variable)
+            self.add_variable(variable)
 
-        self.parameters = []
         from policyengine_core.parameters import Parameter as CoreParameter
 
         for param_node in system.parameters.get_descendants():
@@ -153,33 +151,11 @@ class PolicyEngineUKLatest(TaxBenefitModelVersion):
                     label=param_node.metadata.get("label", param_node.name),
                     tax_benefit_model_version=self,
                     description=param_node.description,
-                    data_type=type(
-                        param_node(2025)
-                    ),  # Example year to infer type
+                    data_type=type(param_node(2025)),
                     unit=param_node.metadata.get("unit"),
+                    _core_param=param_node,  # Store for lazy value loading
                 )
-                self.parameters.append(parameter)
-
-                for i in range(len(param_node.values_list)):
-                    param_at_instant = param_node.values_list[i]
-                    if i + 1 < len(param_node.values_list):
-                        next_instant = param_node.values_list[i + 1]
-                    else:
-                        next_instant = None
-                    parameter_value = ParameterValue(
-                        parameter=parameter,
-                        start_date=parse_safe_date(
-                            param_at_instant.instant_str
-                        ),
-                        end_date=parse_safe_date(next_instant.instant_str)
-                        if next_instant
-                        else None,
-                        value=param_at_instant.value,
-                    )
-                    self.parameter_values.append(parameter_value)
-                    self.get_parameter(parameter.name).parameter_values.append(
-                        parameter_value
-                    )
+                self.add_parameter(parameter)
 
     def run(self, simulation: "Simulation") -> "Simulation":
         from policyengine_uk import Microsimulation

{policyengine-3.1.12 → policyengine-3.1.14}/src/policyengine/tax_benefit_models/us/datasets.py

@@ -266,8 +266,8 @@ def create_datasets(
 
             us_dataset = PolicyEngineUSDataset(
                 id=f"{Path(dataset).stem}_year_{year}",
-                name=f"{dataset}-year-{year}",
-                description=f"US Dataset for year {year} based on {dataset}",
+                name=f"{Path(dataset).stem}-year-{year}",
+                description=f"US Dataset for year {year} based on {Path(dataset).stem}",
                 filepath=f"{data_folder}/{Path(dataset).stem}_year_{year}.h5",
                 year=int(year),
                 data=USYearData(
@@ -317,8 +317,8 @@ def load_datasets(
         for year in years:
             filepath = f"{data_folder}/{Path(dataset).stem}_year_{year}.h5"
             us_dataset = PolicyEngineUSDataset(
-                name=f"{dataset}-year-{year}",
-                description=f"US Dataset for year {year} based on {dataset}",
+                name=f"{Path(dataset).stem}-year-{year}",
+                description=f"US Dataset for year {year} based on {Path(dataset).stem}",
                 filepath=filepath,
                 year=year,
             )

{policyengine-3.1.12 → policyengine-3.1.14}/src/policyengine/tax_benefit_models/us/model.py

@@ -111,7 +111,6 @@ class PolicyEngineUSLatest(TaxBenefitModelVersion):
 
         self.id = f"{self.model.id}@{self.version}"
 
-        self.variables = []
         for var_obj in system.variables.values():
             variable = Variable(
                 id=self.id + "-" + var_obj.name,
@@ -133,9 +132,8 @@ class PolicyEngineUSLatest(TaxBenefitModelVersion):
                         var_obj.possible_values._value2member_map_.values(),
                     )
                 )
-            self.variables.append(variable)
+            self.add_variable(variable)
 
-        self.parameters = []
         from policyengine_core.parameters import Parameter as CoreParameter
 
         for param_node in system.parameters.get_descendants():
@@ -143,33 +141,14 @@ class PolicyEngineUSLatest(TaxBenefitModelVersion):
                 parameter = Parameter(
                     id=self.id + "-" + param_node.name,
                     name=param_node.name,
+                    label=param_node.metadata.get("label"),
                     tax_benefit_model_version=self,
                     description=param_node.description,
                     data_type=type(param_node(2025)),
                     unit=param_node.metadata.get("unit"),
+                    _core_param=param_node,  # Store for lazy value loading
                 )
-                self.parameters.append(parameter)
-
-                for i in range(len(param_node.values_list)):
-                    param_at_instant = param_node.values_list[i]
-                    if i + 1 < len(param_node.values_list):
-                        next_instant = param_node.values_list[i + 1]
-                    else:
-                        next_instant = None
-                    parameter_value = ParameterValue(
-                        parameter=parameter,
-                        start_date=parse_safe_date(
-                            param_at_instant.instant_str
-                        ),
-                        end_date=parse_safe_date(next_instant.instant_str)
-                        if next_instant
-                        else None,
-                        value=param_at_instant.value,
-                    )
-                    self.parameter_values.append(parameter_value)
-                    self.get_parameter(parameter.name).parameter_values.append(
-                        parameter_value
-                    )
+                self.add_parameter(parameter)
 
     def run(self, simulation: "Simulation") -> "Simulation":
         from policyengine_us import Microsimulation

{policyengine-3.1.12 → policyengine-3.1.14}/src/policyengine.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: policyengine
-Version: 3.1.12
+Version: 3.1.14
 Summary: A package to conduct policy analysis using PolicyEngine tax-benefit models.
 Author-email: PolicyEngine <hello@policyengine.org>
 License:                     GNU AFFERO GENERAL PUBLIC LICENSE

{policyengine-3.1.12 → policyengine-3.1.14}/src/policyengine.egg-info/SOURCES.txt

@@ -79,4 +79,5 @@ src/policyengine/utils/plotting.py
 tests/test_aggregate.py
 tests/test_cache.py
 tests/test_change_aggregate.py
-tests/test_entity_mapping.py
+tests/test_entity_mapping.py
+tests/test_models.py

policyengine-3.1.14/tests/test_models.py

@@ -0,0 +1,68 @@
+"""Tests for UK and US tax-benefit model versions."""
+
+import pytest
+
+from policyengine.tax_benefit_models.uk import uk_latest
+from policyengine.tax_benefit_models.us import us_latest
+
+
+class TestUKModel:
+    """Tests for PolicyEngine UK model."""
+
+    def test_has_hundreds_of_parameters(self):
+        """UK model should have hundreds of parameters."""
+        assert len(uk_latest.parameters) >= 100
+
+    def test_has_hundreds_of_variables(self):
+        """UK model should have hundreds of variables."""
+        assert len(uk_latest.variables) >= 100
+
+    def test_parameters_have_values(self):
+        """Each parameter should have at least one parameter value."""
+        total_values = 0
+        for param in uk_latest.parameters[:100]:  # Check first 100 for speed
+            values = param.parameter_values
+            assert len(values) >= 1, f"Parameter {param.name} has no values"
+            total_values += len(values)
+
+        # Should have many parameter values in total
+        assert total_values >= 100
+
+    def test_parameter_values_have_required_fields(self):
+        """Parameter values should have start_date and value."""
+        for param in uk_latest.parameters[:50]:
+            for pv in param.parameter_values:
+                assert pv.start_date is not None
+                assert pv.value is not None
+                assert pv.parameter is param
+
+
+class TestUSModel:
+    """Tests for PolicyEngine US model."""
+
+    def test_has_hundreds_of_parameters(self):
+        """US model should have hundreds of parameters."""
+        assert len(us_latest.parameters) >= 100
+
+    def test_has_hundreds_of_variables(self):
+        """US model should have hundreds of variables."""
+        assert len(us_latest.variables) >= 100
+
+    def test_parameters_have_values(self):
+        """Each parameter should have at least one parameter value."""
+        total_values = 0
+        for param in us_latest.parameters[:100]:  # Check first 100 for speed
+            values = param.parameter_values
+            assert len(values) >= 1, f"Parameter {param.name} has no values"
+            total_values += len(values)
+
+        # Should have many parameter values in total
+        assert total_values >= 100
+
+    def test_parameter_values_have_required_fields(self):
+        """Parameter values should have start_date and value."""
+        for param in us_latest.parameters[:50]:
+            for pv in param.parameter_values:
+                assert pv.start_date is not None
+                assert pv.value is not None
+                assert pv.parameter is param

{policyengine-3.1.12 → policyengine-3.1.14}/uv.lock

@@ -1080,12 +1080,13 @@ wheels = [
 
 [[package]]
 name = "policyengine"
-version = "3.1.7"
+version = "3.1.12"
 source = { editable = "." }
 dependencies = [
     { name = "microdf-python" },
     { name = "pandas" },
     { name = "plotly" },
+    { name = "psutil" },
     { name = "pydantic" },
     { name = "requests" },
 ]
@@ -1133,6 +1134,7 @@ requires-dist = [
     { name = "policyengine-uk", marker = "extra == 'uk'", specifier = ">=2.51.0" },
     { name = "policyengine-us", marker = "extra == 'dev'", specifier = ">=1.213.1" },
     { name = "policyengine-us", marker = "extra == 'us'", specifier = ">=1.213.1" },
+    { name = "psutil", specifier = ">=5.9.0" },
     { name = "pydantic", specifier = ">=2.0.0" },
     { name = "pytest", marker = "extra == 'dev'" },
     { name = "pytest-asyncio", marker = "extra == 'dev'", specifier = ">=0.26.0" },

policyengine-3.1.12/src/policyengine/core/parameter.py

@@ -1,21 +0,0 @@
-from typing import TYPE_CHECKING
-from uuid import uuid4
-
-from pydantic import BaseModel, Field
-
-from .parameter_value import ParameterValue
-from .tax_benefit_model_version import TaxBenefitModelVersion
-
-if TYPE_CHECKING:
-    from .parameter_value import ParameterValue
-
-
-class Parameter(BaseModel):
-    id: str = Field(default_factory=lambda: str(uuid4()))
-    name: str
-    label: str | None = None
-    description: str | None = None
-    data_type: type | None = None
-    tax_benefit_model_version: TaxBenefitModelVersion
-    unit: str | None = None
-    parameter_values: list["ParameterValue"] = Field(default_factory=list)