The ADR-0036 guard only inspected EpcPropertyData's top-level fields, so a
dropped field on a NESTED object (the PV-array list, the floor heat-loss
flags) slipped straight through. Generalise it to walk every domain
dataclass reachable from EpcPropertyData and check each field is
reconstructed by a _compose/_to_* mapper or allow-listed (per-field or
whole-class), keyed by Class.field.
Surfaced 14 pre-existing nested gaps the old guard was blind to: 7 are
calculator-read with no FE column (scoring-relevant silent-drop, same class
as the PV bug — tracked follow-up), the rest dormant or awaiting FE tables.
Each is now explicit and justified.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
sap_energy_source.photovoltaic_arrays has no table, so every array is
dropped on save — worth ~12 SAP points on an electrically-heated dwelling
(persist != score). Inject two ordered arrays onto a PV-free fixture.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
is_exposed_floor / is_above_partially_heated_space have no
epc_floor_dimension column, so a True flag round-trips back to the False
default and silently flips the floor's heat-loss path (persist != score).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Even after batching the data writes, the handler still wrote to the DB per
property through the orchestrator's SubTask bookkeeping: create + start +
complete each self-committed, and _cascade re-listed every sibling and re-saved
the parent on every transition — ~5 writes per property plus an O(N^2) cascade.
- TaskOrchestrator.run_subtasks: create all children in one INSERT, run each
(failures isolated per child), then persist all terminal states in one bulk
save and cascade the parent once. Children go WAITING -> terminal; the
transient IN_PROGRESS row is never written.
- SubTaskRepository.create_many / save_many (bulk INSERT / bulk fetch + update).
- _cascade short-circuits when the Task is already FAILED (terminal) — skips the
sibling roll-up entirely.
- modelling_e2e handler fans out via run_subtasks instead of per-property
create_child_subtask + run_subtask.
Per N-property batch the SubTask bookkeeping drops from ~5N writes + an O(N^2)
cascade to ~2 writes + 1 cascade.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The handler fired ~2+2N read round-trips and N+N write transactions per
SQS batch, pinning RDS CPU under ~32 concurrent containers on pool_size=1.
Reads: merge the duplicate property query and add overrides_for_many /
SolarRepository.get_many so overrides, solar, and property rows each load
in one query (2+2N -> 3).
Writes: buffer each modelled property's persistence intent in memory
(_PropertyWrite) during the loop, then flush the whole batch in one
PostgresUnitOfWork with a single commit, and run the baseline orchestrator
once for all written ids (N+N -> 2 transactions). Per-property modelling
failures stay isolated in the loop; the batch write is all-or-nothing and
retried via SQS (saves are idempotent upserts).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The modelling_e2e Lambda runs on a single-connection pool (pool_size=1,
max_overflow=0) so one invocation uses one Postgres connection. But re-hydrating
a Property through PostgresUnitOfWork resolved its Landlord Overrides through a
PropertyOverridesPostgresReader built from the unit's session *factory* — which
opens a brand-new Session per call. While the unit's own read transaction was
still open (PropertyPostgresRepository.get_many had checked out the connection),
that second Session asked the pool for a second connection, found none, and timed
out after 30s:
QueuePool limit of size 1 overflow 0 reached, connection timed out, timeout 30.00
The baseline stage (PropertyBaselineOrchestrator.run -> uow.property.get_many ->
landlord overrides) hit this on every invocation.
Read the overrides on the unit's OWN session instead. property_overrides is
committed reference data, so reading it inside the unit's transaction sees the
same rows and keeps the invocation on one connection. Extract the query/mapping
into a shared helper and add OpenSessionPropertyOverridesReader (reads on a
caller-owned, already-open session without closing it) for the unit; the
standalone PropertyOverridesPostgresReader still opens its own short session for
use outside a unit.
Regression test pins the invariant with a real pool_size=1/max_overflow=0 engine:
without the fix it reproduces the exact QueuePool timeout.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The modelling_e2e Lambda held up to ~4 concurrent Postgres connections per
invocation: the read Session stayed open across the write loop (the catalogue
was queried live and overrides were read per-Property), each per-Property Unit
of Work opened a second, and the TaskOrchestrator ran on its own NullPool
engine — so the pool needed pool_size=2 + max_overflow=1 just for the modelling
work. Under 32 concurrent containers that approached RDS max_connections.
Restructure the handler to read everything up front — overrides, Scenario, an
in-memory catalogue snapshot, and stored Solar — through one short-lived read
Session, close it, then write each Property in a sequential Unit of Work. The
read and write Sessions no longer overlap, so the engine drops to pool_size=1,
max_overflow=0. Fold the orchestrator onto the same pooled engine: its repos
commit on every save, releasing the connection between bookkeeping calls, so it
holds none during the work. One invocation now uses one connection at a time.
The catalogue becomes a per-invocation snapshot (MaterialSnapshotRepository),
mirroring ProductPostgresRepository.get exactly — same drift mapping, lowest-id
pick, and errors — but priced after the Session closes. Transaction isolation
is preserved: per-Property writes and orchestrator bookkeeping keep their own
independent transactions, just drawn sequentially from a single connection.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Fail if any EpcPropertyData field is neither reconstructed by _compose nor on a
documented allow-list, turning latent persistence gaps into explicit decisions
(would have caught the conservatory and roof-window drops). ADR-0036.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Persist SapConservatory as five nullable conservatory_* columns on epc_property
(1:1 with the dwelling) and rebuild it in _compose, so the §6.1 fold survives
save -> reload -> score. Without this the scored (re-hydrated) EPC silently
dropped the conservatory (persist != score) — a latent gap shared with the
21.0.1 path. Adds a deep-equality round-trip test. ADR-0036.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
10 modelling_e2e properties failed with "unmapped SAP code in fuel_code: 10":
the billing layer (`sap_code_to_fuel`) had no carrier for Table-32 code 10
(dual fuel, mineral + wood) and raised rather than guess one.
SAP 10.2 treats dual fuel as its OWN fuel (its own Table-12 factors), so model
it as its own billing carrier rather than collapsing onto wood or coal:
- New `Fuel.DUAL_FUEL_MINERAL_AND_WOOD`.
- `_CODE_TO_FUEL[10]` -> that carrier.
- Fuel Rates snapshot prices it at 7.69 p/kWh — the midpoint of the COAL proxy
(7.13) and WOOD_LOGS (8.25). This mirrors SAP's own construction: Table-32
dual fuel (3.99) ~= midpoint of house coal (3.67) and wood logs (4.23).
Marked `derived` with a documented _note/_gap/_assumption (like the COAL and
HEAT_NETWORK proxies), since there is no retail blend price.
A dedicated carrier + rate (vs a one-line map to an existing carrier) keeps the
fuel identity faithful to SAP and avoids mispricing dual fuel as pure wood/coal.
Tests: code 10 -> DUAL_FUEL carrier; snapshot prices it at 7.69; grid-export
codes (36/60) still raise (the genuine no-carrier case).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
PostgresUnitOfWork built its PropertyPostgresRepository without an overrides
reader, so a Property re-hydrated through the unit silently dropped its
Landlord Overrides (ADR-0032). The Baseline orchestrator runs through the UoW,
so it scored the bare lodged EPC while the Plan modelled the override-folded
Effective EPC — the two diverged (e.g. baseline effective 71/C vs plan
baseline 62/D), producing "already at band C yet recommends reaching C".
Wire PropertyOverridesPostgresReader into the unit's property repo (uow-
independent committed reference data, read via the same session factory) so
every re-hydration folds overrides, matching the live modelling path.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Two reconciliations to make the modelling_e2e Lambda handler production-ready.
1. Price through the off-catalogue overlay, drop the workarounds
The handler priced through a plain ProductPostgresRepository and excluded
secondary_heating_removal / system_tune_up / system_tune_up_zoned to dodge
ProductNotFound (and a poisoning pgEnum DataError). Those measures are now
priced by catalogue_with_off_catalogue_overrides (already used by the e2e
runner and PostgresUnitOfWork), so the exclusions are removed and ALL measure
types are considered. This also fixes gas-boiler / single-glazed properties,
which Dan's handler never excluded and so still crashed (the standard
system_tune_up option is built unconditionally — the considered-measures
exclusion never actually gated it).
2. Broaden the EPC-Prediction cohort to nearby real postcodes (ADR-0031)
A property with no lodged EPC and no same-type comparable in its own postcode
(e.g. the only flat among houses) used to gate out and fail the subtask. The
gov EPC API cannot search by radius/outcode, so we resolve the real unit
postcodes physically nearest the target via postcodes.io (keyless; already a
trusted in-repo dependency) and walk them nearest-first until enough same-type
comparables surface. New PostcodesIoClient (transient-failure retry with
exponential backoff, soft-failing to the seed so broadening never breaks
prediction) and EpcComparablePropertiesRepository.candidates_near. Wired into
the handler and e2e runner; broadening is lazy (only on gate-out) and memoised
per (postcode, property_type).
Validated live: property 728476 (gas boiler) prices system_tune_up at GBP295;
property 718580 (lone flat in BR6 6BS) now predicts via nearby BR6 postcodes.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Make run_modelling_e2e the single script that does everything for a portfolio,
so the 291-property run needs one invocation with per-property recovery (no
all-or-nothing chunking):
- On --persist, a lodged-EPC Property now also gets its Baseline Performance
row written via PropertyBaselineOrchestrator (per Property, so one bad cert
does not abort the batch). Predicted (EPC-less) Properties have no lodged
figures, so they get a Plan but no baseline row.
- The run CSV gains api_sap (register) vs baseline_sap (calculator) + sap_delta,
so calculator-vs-API divergence is reviewable per property.
Fill the off-catalogue overlay for the measures the live material catalogue
cannot price, so they stop crashing the run:
- double_glazing (£550/window) and secondary_glazing (£400/window): priced
per window (the generator multiplies by single-glazed window count, matching
the legacy window_glazing). Grounded in 2025/26 UK installed costs; per-window
is the right unit for windows (fixed per-unit install dominates) — per-m2 fits
walls/floors, not glazing.
- gas_boiler_upgrade / system_tune_up / system_tune_up_zoned: these are priced
off the heating rate sheet (Products()), with get() reading the catalogue only
for an id — so the overlay entry exists to satisfy that lookup (material_id
stays None, as with ASHP); the rate sheet remains authoritative.
Validated on a 12-property sample (incl. a secondary-glazing case and a
SAP-Schema-16.2 cert): 12/12 baseline rows + plans, 0 errors.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Caught live writing property_overrides on portfolio 796: the Python
override_component SAEnum lagged the DB enum, so reading a new-component row
back threw LookupError. Guard it with a consistency test.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The FE-owned `material.type` pgEnum cannot carry `secondary_heating_removal`,
so pricing it through the DB catalogue raises a DataError that poisons the
session — the modelling pipeline crashed on any property with a lodged
secondary heater unless the measure was excluded on the Scenario.
Realise the `ProductRepository` docstring's intent (DB catalogue today, a JSON
file for costs the ETL does not yet supply, behind the same port): add a
`CompositeProductRepository` that resolves an override source first, then the
catalogue. Checking the override first keeps that Measure Type away from the DB
entirely; every other type misses the override and falls through unchanged.
- off_catalogue_costs.json prices it at £270 flat per-dwelling — the legacy
`Costs.heater_removal` ported to the new flat model (ADR-0028):
(£25 + £200 baseline) x 1.2 VAT, for the single fixed secondary a cert lodges.
Contingency (0.25) is joined from config, not the file.
- Wire the composite into PostgresUnitOfWork.product and run_modelling_e2e, so
the first-run pipeline and the local runner both honour the overlay.
- Integration test: drop the unrealistic seeded secondary_heating_removal DB
rows (the pgEnum can't hold the type) and assert it is JSON-sourced
(material_id is None, cost £270) end-to-end through a real Unit of Work.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds whole-dwelling property_type/built_form to EpcSimulation (folded by
apply_simulations) and maps those override components. property_type drives
party-wall heat loss + ASHP/solar/wall eligibility, so a landlord correction now
moves both the SAP calc and the measure menu; built_form has no calculator
consumer today (feeds the ML transform). Written as the landlord text value
(park-home check is text-only). Refines ADR-0032 dec-4.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Extends WallType coverage to timber/stone/system-built/cob/park-home/curtain and
adds RoofType "Pitched, N mm loft insulation" -> roof_insulation_thickness. The
"(assumed) insulated"/"partial" wall states stay deferred (ambiguous code, needs
Elmhurst validation per ADR-0032); property_type/built_form carry no SAP weight.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The new pipeline left no per-Property record of a run (the old engine set
property.has_recommendations and populated property_details_epc). Restore the
marker: PropertyRepository.mark_modelled sets has_recommendations (true when the
Plan carries measures, mirroring the old engine) and bumps updated_at, so a
first-run under the new process is identifiable as updated_at >= 2026-06-01.
ModellingOrchestrator marks each Property after its Scenarios (true if any
Scenario yielded a measure); run_modelling_e2e's --persist path marks it too
(its compute runs on in-memory fakes, so the DB UoW sets it directly). Adds the
has_recommendations/updated_at columns to the PropertyRow mirror.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The measures a run considers should come from the Scenario, not a CLI flag.
The live scenario table persists exclusions only (no inclusions column), as a
Postgres text-array of exact MeasureType values.
- Scenario gains `exclusions: frozenset[MeasureType]` + `considered_measures()`
(all measures minus the excluded ones, or None when none are excluded).
- ScenarioModel.to_domain parses the `{a,b,c}` exclusions array into
MeasureTypes, raising on a token that is not an exact MeasureType value
(no high-level category expansion), per the strict-enum convention.
- ModellingOrchestrator._plan_for derives the allowlist from the Scenario's
exclusions, combined (intersection) with any explicit considered_measures
override via the new `combine_considered_measures`.
- run_modelling_e2e sources the allowlist from the Scenario; --measures /
--exclude-measures become optional overlays (e.g. the technical
secondary_heating_removal exclusion the catalogue cannot yet stock).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Two review points from @dancafc:
1) Rename the `Comparable` dataclass → `ComparableProperty` (it models one
comparable *property*; the collection stays `ComparableProperties`). Applied
across domain, repositories, orchestration, harness, scripts, and tests with a
word-boundary rename so `ComparableProperties` is untouched.
2) Move `PredictionTarget` out of comparable_properties.py into prediction_target.py
(where `PredictionTargetAttributes` + `build_prediction_target` already live).
comparable_properties.py now imports it; no import cycle (prediction_target no
longer depends on comparable_properties). Importers updated.
92 tests pass across the touched suites; pyright strict clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add a `source` discriminator (lodged | predicted) to the EPC store so a Property
holds a lodged EPC and a predicted one (EPC Prediction gap-fill) at once
(ADR-0031). EpcRepository.save gains source="lodged"; idempotent delete is now
per-source (a predicted save no longer wipes lodged, and vice versa);
get_for_property/get_for_properties filter lodged; new get_predicted_for_property
/ get_predicted_for_properties read predicted. PropertyPostgresRepository.get +
get_many hydrate Property.predicted_epc, so the predicted picture reaches the
modelling read (both load via get_many). FakeEpcRepo mirrors the dual slot.
EpcPropertyModel gains `source` (default "lodged"); the test DB builds from the
SQLModel mirror so this is exercised without the prod migration. The matching
Drizzle change (column + per-(property_id,source) uniqueness) is the team's to
action before merge — docs/MIGRATION_NOTE_predicted_epc_source.md.
3 store tests (coexist, idempotent predicted re-save leaves lodged, lodged-only
has no predicted) + property-repo wiring; 85 pass across affected suites; new
code pyright-clean (2 pre-existing wwhrs errors in epc_property_table untouched).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Build the cohort IO port ADR-0029 deferred (ADR-0031 slice-5b):
`ComparablePropertiesRepository.candidates_for(postcode) -> list[Comparable]`,
with an EPC-API + geospatial adapter that lists the postcode's lodged certs
(search_by_postcode), fetches + maps each (get_by_certificate_number), and
resolves their UPRNs to coordinates in ONE batched read. Register metadata the
cert doesn't carry (address, registration date) is threaded off the search row;
a UPRN-less or unparseable-date cert is kept, just uncoordinated / unweighted.
The domain select_comparables then filters these candidates into the cohort.
Thin CohortEpcClient / CohortGeospatial Protocols keep the adapter testable
against fakes; EpcClientService + GeospatialS3Repository satisfy them
structurally (no changes). 3 tests; pyright strict clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adds GeospatialRepository.coordinates_for_uprns(uprns) -> dict — a batch
coordinate lookup returning only covered UPRNs. The S3 adapter overrides it
to read the meta once, group UPRNs by their covering partition, and read each
partition once for all the UPRNs it covers; co-located (closely-numbered)
UPRNs share a partition, so an EPC Prediction cohort is typically one or two
reads instead of one per neighbour. Default port impl is a per-UPRN loop.
Feeds the EPC Prediction geo-proximity work: a cohort's UPRNs resolve to
coordinates in a couple of reads (validated at corpus scale: 170 partition
reads for 2683 UPRNs).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Flat per-dwelling decommission price (sample_catalogue \£250) + 0.25 contingency
(covers unknown heater count / hard-wired-vs-plugged / repaint extent). The JSON
repo joins the contingency from config, proven by the new repo test. No composite
Products machinery — a lodged secondary is one roughly-fixed job, not room-scaled.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
