Require policyengine-core 3.34.9

.github/test_outcome_allowlist.toml

@@ -19,18 +19,6 @@
 # Keep entries narrow. An entry that matches nothing in a run is reported as a
 # warning; remove it once its bug is fixed. Failures can never be accepted.
 
-[[allow]]
-nodeid = "policyengine_uk/tests/test_subsample.py::test_neutralized_inputs_and_model_values_survive_subsampling[[]*]"
-outcome = "xfailed"
-reason = "^needs policyengine-core holder resync$"
-why = "Issue #2254 / PR #2265 waits for core to resync restored holders after subsampling. Both ownership variants retain every assertion under strict xfail."
-
-[[allow]]
-nodeid = "policyengine_uk/tests/test_input_records.py::test_restoring_reform_can_replace_a_neutralized_input[[]subsampled-*]"
-outcome = "xfailed"
-reason = "^needs policyengine-core holder resync$"
-why = "Issue #2254 / PR #2265 waits for core to accept writes immediately after restoring a subsampled holder. Direct supplied/model controls remain unmarked."
-
 [[allow]]
 nodeid = "policyengine_uk/tests/test_uc_state_pension_properties.py::test_state_pension_does_not_change_partners_marriage_allowance"
 outcome = "xfailed"

changelog.d/core-3349.changed.md

@@ -0,0 +1 @@
+Require policyengine-core 3.34.9 or later. From that release, an input set on a variable that a reform has neutralised has no effect, as core documents, so a reform that restores the variable reads the input supplied before neutralisation.

policyengine_uk/tests/test_capital_gains_projection.py

@@ -1068,8 +1068,16 @@ def apply(self):
         simulation.apply_reform(
             {"gov.abolitions.capital_gains_before_response": {"2026": False}}
         )
+    # Core ignores inputs to a neutralized variable
+    # (TaxBenefitSystem.neutralize_variable), so it drops the 555 set under
+    # NeutralizeGains and the restored variable reads the situation's 120,000.
+    # The abolition parameter leaves the variable's definition in place, so
+    # core stores that 555.
+    expected = 120_000 if route == "neutralize_variable" else 555
     with pytest.warns(UserWarning, match="supplied"):
-        assert simulation.calculate("capital_gains_before_response", 2026)[0] == 555
+        assert (
+            simulation.calculate("capital_gains_before_response", 2026)[0] == expected
+        )
     restored_record = simulation.capital_gains_projection_metadata["application"][
         "by_variable"
     ]["capital_gains_before_response"]["2026"]

policyengine_uk/tests/test_input_records.py

@@ -189,18 +189,7 @@ def test_rebuilt_populations_forget_the_inputs_they_replace():
 @pytest.mark.parametrize("model", [False, True], ids=["supplied", "model"])
 @pytest.mark.parametrize(
     "subsampled",
-    [
-        pytest.param(False, id="direct"),
-        pytest.param(
-            True,
-            id="subsampled",
-            marks=pytest.mark.xfail(
-                strict=True,
-                raises=AssertionError,
-                reason="needs policyengine-core holder resync",
-            ),
-        ),
-    ],
+    [pytest.param(False, id="direct"), pytest.param(True, id="subsampled")],
 )
 def test_restoring_reform_can_replace_a_neutralized_input(model, subsampled):
     """A write after restoration replaces the hidden value and its ownership."""

policyengine_uk/tests/test_subsample.py

@@ -526,11 +526,6 @@ def test_a_model_value_stays_the_models():
 
 
 @pytest.mark.parametrize("model", [False, True], ids=["supplied", "model"])
-@pytest.mark.xfail(
-    strict=True,
-    raises=AssertionError,
-    reason="needs policyengine-core holder resync",
-)
 def test_neutralized_inputs_and_model_values_survive_subsampling(model):
     """Neutralisation hides stored inputs without discarding them or their owner."""
     data = dataset([(100.0, [1]), (100.0, [1])])

pyproject.toml

@@ -24,7 +24,7 @@ classifiers = [
 ]
 requires-python = ">=3.11"
 dependencies = [
-    "policyengine-core>=3.33.0",
+    "policyengine-core>=3.34.9",
     "microdf-python>=1.2.1",
     "pydantic>=2.11.7",
     "tables>=3.10.2",