Merge pull request #1346 from Hestia-Homes/fix/elmhurst-summary-roof-window-classification

fix(elmhurst): party-ceiling Main BP window is vertical, not a rooflight

datatypes/epc/domain/mapper.py

@@ -6229,12 +6229,16 @@ def _elmhurst_orientation_int(orientation: str) -> int:
 # skylight per the worksheet's (27a) row.
 _ELMHURST_ROOF_WINDOW_U_THRESHOLD: Final[float] = 3.0
 
-# RdSAP 10 §8.2 (PDF p.50) — BPs whose roof type is "Another dwelling
-# above" (A) or "Non-residential space above" (NR) have their own
-# external roof structure with potential rooflights, distinct from a
-# pitched-roof dwelling. Windows lodged against these BPs are routed
-# to `sap_roof_windows` regardless of their U-value.
+# Elmhurst §8 roof types that route an EXTENSION's window to a rooflight:
+# "A Another dwelling above" / "NR Non-residential space above". On an
+# extension BP these score on the worksheet (27a) Roof Windows line (cert
+# 000565 Ext2 [NR] + Ext4 [A]). NOT applied to the Main BP — a Main roof
+# of "A"/"NR" means the whole dwelling sits under another dwelling (a party
+# ceiling with no external roof), so its windows are vertical (simulated
+# case 53 / cert 000565 re-keyed as a mid-floor flat: Main roof "A",
+# External-wall window U 2.0 = vertical, Elmhurst Type "Window").
 _ELMHURST_BP_ROOF_TYPES_WITH_ROOFLIGHTS: Final[tuple[str, ...]] = ("A ", "NR ")
+_ELMHURST_MAIN_BUILDING_PARTS: Final[tuple[str, ...]] = ("Main", "Main Property")
 
 
 def _elmhurst_bp_roof_type(
@@ -6245,7 +6249,7 @@ def _elmhurst_bp_roof_type(
     Returns None when the BP isn't found (single-bp cert with no
     matching extension)."""
     bp = w.building_part
-    if bp in ("Main", "Main Property"):
+    if bp in _ELMHURST_MAIN_BUILDING_PARTS:
         return survey.roof.roof_type
     for ext in survey.extensions:
         if ext.name == bp:
@@ -6279,12 +6283,16 @@ def _is_elmhurst_roof_window(
 
       1. Single-glazed windows are never rooflights — Part L 2006
          minimum glazing for any rooflight is double-glazed.
-      2. BP roof type ∈ {A, NR} → rooflight — the BP has its own
-         external roof structure with rooflights (worksheet (30)
-         External roof + (27a) Roof Windows treatment).
-      3. U > 3.0 W/m²K → rooflight — cohort backstop catching old
-         skylights on pitched roofs (cohort cert 000516 W6).
-      4. Otherwise vertical.
+      2. Explicit "Roof of Room" location → rooflight.
+      3. EXTENSION BP roof type ∈ {A, NR} → rooflight (worksheet (30)
+         External roof + (27a) Roof Windows treatment). EXCLUDES the Main
+         BP: a Main roof of "A"/"NR" is a party ceiling (whole dwelling
+         under another dwelling), so its windows are vertical — simulated
+         case 53 / cert 000565 mid-floor flat (Main roof "A", External-wall
+         window U 2.0, Elmhurst Type "Window").
+      4. U > 3.0 W/m²K AND the BP has a room-in-roof → rooflight — cohort
+         backstop catching old skylights on a sloping ceiling (cert 000516 W6).
+      5. Otherwise vertical.
     """
     if w.glazing_type.startswith("Single"):
         return False
@@ -6293,11 +6301,15 @@ def _is_elmhurst_roof_window(
     # regardless of BP roof type or U-value.
     if "roof of room" in (w.location or "").lower():
         return True
-    bp_roof_type = _elmhurst_bp_roof_type(w, survey)
-    if bp_roof_type is not None and bp_roof_type.startswith(
-        _ELMHURST_BP_ROOF_TYPES_WITH_ROOFLIGHTS
-    ):
-        return True
+    # EXTENSION BP whose roof type is a party ceiling (A / NR) scores its
+    # windows on (27a) — but never the Main BP (a Main "A"/"NR" roof is the
+    # dwelling's own party ceiling; case 53).
+    if w.building_part not in _ELMHURST_MAIN_BUILDING_PARTS:
+        bp_roof_type = _elmhurst_bp_roof_type(w, survey)
+        if bp_roof_type is not None and bp_roof_type.startswith(
+            _ELMHURST_BP_ROOF_TYPES_WITH_ROOFLIGHTS
+        ):
+            return True
     # U > 3.0 backstop, gated on the BP having a room-in-roof. Elmhurst routes a
     # high-U "Double pre 2002" unit through the worksheet's (27a) Roof Windows
     # line regardless of its lodged "External wall" location — but ONLY where the

scripts/summary_to_sap.py

@@ -0,0 +1,94 @@
+"""Elmhurst Summary PDF -> EpcPropertyData -> Sap10Calculator, with a dump of
+our SAP score + per-end-use kWh + the `intermediate` worksheet trail, for
+diffing against the accompanying Elmhurst worksheet PDF.
+
+Usage:
+    python -m scripts.summary_to_sap "<path to Summary_*.pdf>"
+
+Reuses the exact preprocessing the Summary->EpcPropertyData chain test uses
+(`backend/documents_parser/tests/test_summary_pdf_mapper_chain.py`):
+`pdftotext -layout` -> Textract-style label/value stream -> extractor ->
+`from_elmhurst_site_notes` mapper.
+"""
+
+from __future__ import annotations
+
+import re
+import subprocess
+import sys
+from pathlib import Path
+
+from backend.documents_parser.elmhurst_extractor import ElmhurstSiteNotesExtractor
+from datatypes.epc.domain.mapper import EpcPropertyDataMapper
+from domain.sap10_calculator.calculator import Sap10Calculator
+from domain.sap10_calculator.rdsap.cert_to_inputs import cert_to_inputs
+
+
+def _summary_pdf_to_textract_style_pages(pdf_path: Path) -> list[str]:
+    info = subprocess.run(
+        ["pdfinfo", str(pdf_path)], capture_output=True, text=True, check=True
+    ).stdout
+    m = re.search(r"Pages:\s+(\d+)", info)
+    if m is None:
+        raise RuntimeError(f"Could not parse page count from {pdf_path}")
+    page_count = int(m.group(1))
+    pages: list[str] = []
+    for i in range(1, page_count + 1):
+        layout = subprocess.run(
+            ["pdftotext", "-layout", "-f", str(i), "-l", str(i), str(pdf_path), "-"],
+            capture_output=True, text=True, check=True,
+        ).stdout
+        tokens: list[str] = []
+        for line in layout.splitlines():
+            if not line.strip():
+                tokens.append("")
+                continue
+            tokens.extend(p for p in re.split(r"\s{2,}", line.strip()) if p)
+        pages.append("\n".join(tokens))
+    return pages
+
+
+def main(pdf: str) -> None:
+    pdf_path = Path(pdf)
+    pages = _summary_pdf_to_textract_style_pages(pdf_path)
+    survey = ElmhurstSiteNotesExtractor(pages).extract()
+    epc = EpcPropertyDataMapper.from_elmhurst_site_notes(survey)
+    inp = cert_to_inputs(epc)
+    r = Sap10Calculator().calculate(epc)
+    p = epc.sap_building_parts[0] if epc.sap_building_parts else None
+
+    print(f"=== {pdf_path.name} ===")
+    print(f"dwelling_type={epc.dwelling_type!r} property_type={epc.property_type!r} "
+          f"age_band={p.construction_age_band if p else None} TFA={epc.total_floor_area_m2}")
+    print(f"OUR SAP = {r.sap_score} ({r.sap_score_continuous:.4f})  "
+          f"CO2={r.co2_kg_per_yr/1000:.3f} t/yr  PEUI={r.primary_energy_kwh_per_m2:.1f}")
+    print("--- per end use (kWh/yr) ---")
+    print(f"  space_heating useful = {r.space_heating_kwh_per_yr:.1f}")
+    print(f"  main_heating fuel    = {r.main_heating_fuel_kwh_per_yr:.1f}")
+    print(f"  secondary fuel       = {r.secondary_heating_fuel_kwh_per_yr:.1f}")
+    print(f"  hot_water            = {r.hot_water_kwh_per_yr:.1f}")
+    print(f"  lighting             = {r.lighting_kwh_per_yr:.1f}")
+    print(f"  pumps_fans           = {r.pumps_fans_kwh_per_yr:.1f}")
+    print(f"  delivered fuel total = {r.intermediate.get('delivered_fuel_kwh_per_yr', float('nan')):.1f}")
+    print("--- costs / rating ---")
+    for k in ("main_heating_cost_gbp", "secondary_heating_cost_gbp", "hot_water_cost_gbp",
+              "pumps_fans_cost_gbp", "lighting_cost_gbp", "ecf"):
+        print(f"  {k:28s} {r.intermediate.get(k, float('nan')):.4f}")
+    print(f"  is_off_peak={r.is_off_peak_meter} main_hrf={r.main_heating_high_rate_fraction} "
+          f"hw_hrf={r.hot_water_high_rate_fraction:.4f} other_hrf={r.other_electricity_high_rate_fraction}")
+    print(f"  space £/kWh={inp.space_heating_fuel_cost_gbp_per_kwh} "
+          f"hw £/kWh={inp.hot_water_fuel_cost_gbp_per_kwh} other £/kWh={inp.other_fuel_cost_gbp_per_kwh}")
+    print("--- heat balance (intermediate) ---")
+    for k in ("heat_transfer_coefficient_w_per_k", "heat_loss_parameter_w_per_m2k",
+              "walls_w_per_k", "roof_w_per_k", "floor_w_per_k", "party_walls_w_per_k",
+              "windows_w_per_k", "doors_w_per_k", "thermal_bridging_w_per_k",
+              "infiltration_w_per_k", "infiltration_ach", "internal_gains_annual_avg_w",
+              "mean_internal_temp_annual_avg_c", "useful_space_heating_kwh_per_yr"):
+        print(f"  {k:38s} {r.intermediate.get(k, float('nan')):.4f}")
+
+
+if __name__ == "__main__":
+    if len(sys.argv) != 2:
+        print(__doc__)
+        sys.exit(2)
+    main(sys.argv[1])

tests/datatypes/epc/domain/test_mapper_roof_window_classification.py

@@ -0,0 +1,89 @@
+"""Mapper boundary: Elmhurst §11 window vs roof window classification.
+
+`_is_elmhurst_roof_window` must NOT route a vertical window (lodged Location
+"External wall") to `sap_roof_windows` when the building part's roof is a
+PARTY ceiling — "A Another dwelling above" / "NR Non-residential space above"
+/ "S Same dwelling above". A party ceiling has no external roof structure, so
+it cannot host a rooflight (RdSAP 10 §3.7.1: a rooflight is glazing in a roof;
+party codes carry no heat-loss roof per `_ELMHURST_PARTY_ROOF_CODES`).
+
+Surfaced by "simulated case 53" (cert 000565 re-keyed as a mid-floor electric-
+storage flat, roof "A Another dwelling above"): its one External-wall window
+(U 2.00) was mis-routed to `sap_roof_windows`, so the window area was never
+deducted from the wall (wall over-counted ~7 W/K) and it was priced as a roof
+window — our SAP 74.0 vs Elmhurst worksheet 75. Elmhurst's own entry lodges it
+Type "Window", Location "External wall".
+"""
+
+from datatypes.epc.surveys.elmhurst_site_notes import (
+    RoofDetails,
+    Window as ElmhurstWindow,
+)
+from datatypes.epc.surveys.elmhurst_site_notes import ElmhurstSiteNotes
+from datatypes.epc.domain.mapper import (
+    _is_elmhurst_roof_window,  # pyright: ignore[reportPrivateUsage]
+)
+
+
+def _external_wall_window(*, glazing: str, u_value: float) -> ElmhurstWindow:
+    return ElmhurstWindow(
+        width_m=15.14,
+        height_m=1.0,
+        area_m2=15.14,
+        glazing_type=glazing,
+        frame_factor=0.7,
+        building_part="Main",
+        location="External wall",
+        orientation="North",
+        data_source="Manufacturer",
+        u_value=u_value,
+        g_value=0.72,
+        draught_proofed=True,
+        permanent_shutters="None",
+        frame_type="PVC",
+    )
+
+
+def _survey_with_main_roof(roof_type: str) -> ElmhurstSiteNotes:
+    """Partial `ElmhurstSiteNotes` carrying only what the roof-window
+    classifier reads: `roof`, `room_in_roof`, `extensions`."""
+    survey = object.__new__(ElmhurstSiteNotes)
+    survey.roof = RoofDetails(
+        roof_type=roof_type, insulation="", u_value_known=False
+    )
+    survey.room_in_roof = None
+    survey.extensions = []
+    return survey
+
+
+def test_external_wall_window_on_party_ceiling_is_vertical() -> None:
+    # Arrange — mid-floor flat: Main BP roof is "Another dwelling above"
+    # (party ceiling, no external roof). Window lodged on External wall.
+    window = _external_wall_window(
+        glazing="Double between 2002 and 2021", u_value=2.0
+    )
+    survey = _survey_with_main_roof("A Another dwelling above")
+
+    # Act / Assert — vertical, not a rooflight.
+    assert _is_elmhurst_roof_window(window, survey) is False
+
+
+def test_external_wall_window_under_non_residential_space_is_vertical() -> None:
+    window = _external_wall_window(
+        glazing="Double between 2002 and 2021", u_value=2.0
+    )
+    survey = _survey_with_main_roof("NR Non-residential space above")
+
+    assert _is_elmhurst_roof_window(window, survey) is False
+
+
+def test_roof_of_room_location_still_classified_as_rooflight() -> None:
+    # Guard: an explicit "Roof of Room" location stays a rooflight
+    # (unaffected by the party-ceiling fix).
+    window = _external_wall_window(
+        glazing="Double between 2002 and 2021", u_value=2.0
+    )
+    window.location = "Roof of Room"
+    survey = _survey_with_main_roof("A Another dwelling above")
+
+    assert _is_elmhurst_roof_window(window, survey) is True