Merge pull request #1617 from Hestia-Homes/fix/pashub-semi-exposed-floor

PasHub SAP accuracy: semi-exposed-unheated floor → Table 20 (+ Walsden/Paxton fixtures)

backend/documents_parser/tests/fixtures/pashub_accuracy/manifest.json

@@ -1091,8 +1091,8 @@
     {
       "deal_id": "507639151845",
       "uprn": null,
-      "pre_sap_raw": "C80",
-      "pre_sap_score": 80,
+      "pre_sap_raw": "C76",
+      "pre_sap_score": 76,
       "pdf": "507639151845.pdf"
     },
     {

backend/documents_parser/tests/test_pashub_sap_accuracy.py

@@ -115,7 +115,19 @@ _MANIFEST_PATH = _FIXTURES_DIR / "manifest.json"
 # built form near zero. Observed 79.5% here; vs the DB SAP-10.2 oracle
 # (property_baseline_performance.effective_sap_score, portfolio 838) the cohort
 # moves 62.1% -> 83.3% within-0.5. gov-API RdSAP corpus unchanged (78.9%).
-_MIN_WITHIN_HALF: float = 0.78
+# Ratcheted 0.78 -> 0.82 by the 2026-07-15e SEMI-EXPOSED-FLOOR fix: from_site_notes
+# mis-routed the surveyed "Semi Exposed (unheated)" floor label to RdSAP §5.14
+# (partially-heated constant U=0.7) when a floor semi-exposed to an enclosed
+# UNHEATED space belongs in §5.13 Table 20 `u_exposed_floor` (band B as-built
+# U=1.20) — the gov-API code-2 "To unheated space" and Elmhurst
+# `_is_floor_exposed_to_unheated_space` siblings both route it to Table 20. The
+# too-low U under-counted floor loss on the 9 label-carrying cohort fixtures (the
+# adjacent Ronald St mid-terrace pair among them) → a shared over-credit. A pure
+# data flip in `_PASHUB_FLOOR_EXPOSURE`, from_site_notes-only (gov-API corpus
+# untouched, still 78.9%). Observed 82.4% here; vs the DB SAP-10.2 oracle the
+# cohort moves 83.3% -> 86.9% within-0.5, no built form regresses, all 7 verified
+# dwellings unchanged (none carry the label).
+_MIN_WITHIN_HALF: float = 0.82
 # Ratcheted 2.71 -> 2.64 by #1590 bug 2 (roof "Insulation At: None" -> explicit
 # zero thickness): observed MAE 2.632 — the two ~7-SAP roof over-raters
 # (507665533138 / 507670893756) correcting is exactly 14/205 ≈ 0.07 of MAE.
@@ -187,7 +199,9 @@ _MIN_WITHIN_HALF: float = 0.78
 # (percent_draughtproofed + upper-floor +0.25 m + built-form sheltered_sides;
 # #1601 upper-floor RE-ADJUDICATED, see the within-0.5 note above): observed
 # MAE 0.389.
-_MAX_SAP_MAE: float = 0.40
+# Ratcheted 0.40 -> 0.37 by the 2026-07-15e semi-exposed-floor fix (see the
+# within-0.5 note above): observed MAE 0.367.
+_MAX_SAP_MAE: float = 0.37
 
 _KNOWN_GAP_REASON = (
     "pashub `from_site_notes` mapper does not int-code a main-heating "
@@ -307,3 +321,48 @@ def test_pashub_sap_accuracy_aggregate(capsys: pytest.CaptureFixture[str]) -> No
     assert sap_mae <= _MAX_SAP_MAE, (
         f"SAP MAE {sap_mae:.3f} exceeded ceiling {_MAX_SAP_MAE}"
     )
+
+
+@pytest.mark.skipif(not _FIXTURES, reason="no pashub_accuracy fixtures/manifest")
+def test_walsden_pins_to_pashub_resurvey_sap_64() -> None:
+    """13 Walsden Street (deal 499630239947) has TWO distinct gov-EPC assessments
+    for the same UPRN: the PAS Hub site-note re-survey we parse (inspected
+    26 May 2026, accredited SAP **64**, band D, backed by `rd_sap_site_note`
+    files) and a SEPARATE later lodged cert (24 Jun 2026, SAP 68, registered, no
+    site-note). The portfolio-838 baseline latched onto the 24-Jun SAP-68 cert,
+    but our engine models the survey it was GIVEN and reproduces that survey's own
+    accredited 64 (~64.19). Confirmed against pashub (Khalim) and corroborated by
+    the manifest `pre_sap` D64 + hubspot D64; see the 2026-07-15e provenance
+    investigation in the campaign memory. The manifest is correctly 64, but the
+    fixture-builder can revert manifest values to a colliding DB row — so this pins
+    64 explicitly. Do NOT "correct" this target to 68: that is a different
+    assessment with no site-note backing, and 13 Walsden is a known
+    re-survey-diverges outlier, not an engine under-rate."""
+    outcome = _evaluate("499630239947")
+    assert outcome.sap_continuous is not None
+    assert abs(outcome.sap_continuous - 64) < 0.5, (
+        f"13 Walsden SAP {outcome.sap_continuous:.2f} diverged from the PAS Hub "
+        "re-survey's own accredited 64 (NOT the colliding 24-Jun cert's 68)"
+    )
+
+
+@pytest.mark.skipif(not _FIXTURES, reason="no pashub_accuracy fixtures/manifest")
+def test_paxton_fixture_is_v5_resurvey_sap_76() -> None:
+    """5 Paxton Place (deal 507639151845, uprn 10070864419) was RE-ISSUED: the
+    same PAS Hub survey (inspected 29 May/Jun 2026, report B74AA5A4-…) has two
+    accredited revisions — V2 (ventilation "Natural", SAP 80) and V5 (ventilation
+    "Mechanical Extract - Centralised", SAP 76, the current version). The
+    portfolio-838 oracle tracks the latest revision (76), so the fixture is
+    refreshed to the V5 PDF (2026-07-15e); our engine reproduces its accredited 76
+    (~75.6). Do NOT revert this fixture/manifest to the V2 SAP-80 version.
+    CAVEAT: `hubspot_deal_data.pre_sap` for this deal is still stale at "C80", so a
+    `scripts/build_pashub_accuracy_fixtures.py` rerun would regenerate the manifest
+    `pre_sap` back to 80 — this pin (asserting a literal 76 against the V5 fixture)
+    is the canary. Confirmed against pashub: April survey → SAP 76 (Khalim)."""
+    outcome = _evaluate("507639151845")
+    assert outcome.sap_continuous is not None
+    assert abs(outcome.sap_continuous - 76) < 0.5, (
+        f"5 Paxton SAP {outcome.sap_continuous:.2f} diverged from the current V5 "
+        "(MEV) re-survey's accredited 76 — has the fixture reverted to the V2 "
+        "(Natural, SAP 80) PDF?"
+    )

datatypes/epc/domain/mapper.py

@@ -6191,9 +6191,16 @@ _PASHUB_FLOOR_EXPOSURE: Final[Dict[str, tuple[bool, bool]]] = {
     # "Exposed Floor" → open to external air / over enclosed unheated space →
     # RdSAP 10 §5.13 Table 20 `u_exposed_floor` cascade (is_exposed_floor).
     "Exposed Floor": (True, False),
-    # "Semi Exposed (unheated)" → above a space heated to a lesser extent →
-    # RdSAP 10 §5.14 constant U=0.7 (is_above_partially_heated_space).
-    "Semi Exposed (unheated)": (False, True),
+    # "Semi Exposed (unheated)" → floor semi-exposed to an enclosed UNHEATED
+    # space → RdSAP 10 §5.13 Table 20 `u_exposed_floor` (is_exposed_floor),
+    # NOT §5.14. Table 20 "collapses exposed (to outside air) and semi-exposed
+    # (to enclosed unheated space) into the same lookup" (rdsap_uvalues.py:1416;
+    # band B as-built U=1.20). Mirrors the gov-API code-2 "To unheated space"
+    # (semi-exposed → Table 20, mapper.py:4797) and the Elmhurst
+    # `_is_floor_exposed_to_unheated_space` sibling (mapper.py:5970). §5.14's
+    # constant U=0.7 is reserved for a floor above a *partially* heated space
+    # (`is_above_partially_heated_space`); the #1602 routing transposed the two.
+    "Semi Exposed (unheated)": (True, False),
 }
 
 

datatypes/epc/domain/tests/test_from_site_notes.py

@@ -508,12 +508,16 @@ class TestWallDryLinedMapping:
 class TestFloorExposureMapping:
     """A surveyed "Floor type" of "Semi Exposed (unheated)" or "Exposed Floor"
     must set the SapFloorDimension exposure flags on the part's LOWEST storey
-    (floor==0), so the floor-U cascade routes through RdSAP §5.14 (U=0.7,
-    above partially heated space) / §5.13 Table 20 (u_exposed_floor) instead of
-    the default BS EN ISO 13370 ground-floor formula. `_map_floor_dimensions`
-    previously never set them — the exposure was silently dropped and every
-    semi-exposed/exposed floor billed as an on-soil ground floor (#1602). The
-    gov-API/Elmhurst siblings already flag the ground SapFloorDimension."""
+    (floor==0), so the floor-U cascade routes through RdSAP §5.13 Table 20
+    (`u_exposed_floor`) — which covers BOTH a floor exposed to outside air AND
+    one semi-exposed to an enclosed unheated space — instead of the default
+    BS EN ISO 13370 ground-floor formula. `_map_floor_dimensions` previously
+    never set them — the exposure was silently dropped and every semi-exposed/
+    exposed floor billed as an on-soil ground floor (#1602). §5.14's constant
+    U=0.7 is reserved for a floor above a *partially* heated space, which the
+    PAS Hub survey does not lodge; #1602 mis-routed "Semi Exposed (unheated)"
+    there and this corrects it to Table 20. The gov-API/Elmhurst siblings
+    already flag the ground SapFloorDimension to Table 20."""
 
     @staticmethod
     def _ground(part: SapBuildingPart) -> SapFloorDimension:
@@ -523,7 +527,13 @@ class TestFloorExposureMapping:
     def _upper(part: SapBuildingPart) -> SapFloorDimension:
         return next(fd for fd in part.sap_floor_dimensions if fd.floor == 1)
 
-    def test_semi_exposed_sets_above_partially_heated_on_ground_floor(self) -> None:
+    def test_semi_exposed_unheated_sets_exposed_floor_on_ground_floor(self) -> None:
+        # "Semi Exposed (unheated)" sits above an enclosed UNHEATED space →
+        # RdSAP 10 §5.13 Table 20 `u_exposed_floor` (is_exposed_floor), NOT
+        # §5.14's constant U=0.7 (reserved for a *partially* heated space
+        # below). Mirrors the gov-API code-2 "To unheated space" and the
+        # Elmhurst `_is_floor_exposed_to_unheated_space` siblings; corrects
+        # the #1602 routing that transposed the two branches.
         # Arrange
         data = load("pashub_rdsap_site_notes_example1.json")
         data["building_construction"]["floor"]["floor_type"] = "Semi Exposed (unheated)"
@@ -532,10 +542,10 @@ class TestFloorExposureMapping:
         # Act
         result = EpcPropertyDataMapper.from_site_notes(survey)
 
-        # Assert — §5.14 flag on the ground floor only, exposed flag clear.
+        # Assert — Table 20 exposed flag on the ground floor only.
         ground = self._ground(result.sap_building_parts[0])
-        assert ground.is_above_partially_heated_space is True
-        assert ground.is_exposed_floor is False
+        assert ground.is_exposed_floor is True
+        assert ground.is_above_partially_heated_space is False
 
     def test_exposed_floor_sets_exposed_on_ground_floor(self) -> None:
         # Arrange