Source code for energy_data_retrieval.wind_catalogue.repd_review
"""Bounded, offline review of saved REPD application identities, never farm import."""
from __future__ import annotations
import csv
import hashlib
import io
from typing import Any
REASON = "No unique explicit WRI crosswalk; review as a separate application/phase candidate"
FIELDS = (
"Site Name",
"Technology Type",
"Development Status",
"Planning Authority",
"Planning Application Reference",
"Appeal Reference",
"Secretary of State Reference",
"Are they re-applying (New REPD Ref)",
"Are they re-applying (Old REPD Ref) ",
)
[docs]
def review_unlinked_applications(
original: bytes,
decisions: list[dict[str, Any]],
*,
sha256: str,
offset: int,
limit: int,
reasons: tuple[str, ...] = (REASON,),
) -> list[dict[str, Any]]:
"""Return review facts in saved ledger order, retaining all identity holds.
The checksum pins the Windows-1252 release. Offsets apply only to the exact
selected reasons (the unresolved-crosswalk reason by default), not all
decisions. Explicit exception partitions must list their saved reasons.
Empty cells are unknown;
application IDs and replacement references do not establish physical farms.
No lifecycle, capacity, coordinate or equipment scalar is assigned.
"""
if hashlib.sha256(original).hexdigest() != sha256:
raise ValueError("REPD checksum mismatch")
if type(offset) is not int or type(limit) is not int or offset < 0 or limit <= 0:
raise ValueError("Invalid partition bounds")
if not reasons or any(not isinstance(reason, str) or not reason.strip() for reason in reasons):
raise ValueError("Explicit nonempty saved reasons required")
if len(reasons) != len(set(reasons)):
raise ValueError("Duplicate selected reason")
known_reasons = {d.get("reason") for d in decisions}
if reasons != (REASON,) and not set(reasons) <= known_reasons:
raise ValueError("Selected reason absent from saved decisions")
candidates = [d for d in decisions if d.get("reason") in reasons]
if offset > len(candidates):
raise ValueError("Partition offset exceeds saved candidates")
ids = [d["repd_id"] for d in candidates]
if len(ids) != len(set(ids)):
raise ValueError("Duplicate saved candidate ID")
reader = csv.DictReader(io.StringIO(original.decode("cp1252")))
if not {"Ref ID", *FIELDS} <= set(reader.fieldnames or []):
raise ValueError("Missing REPD review headers")
rows: dict[str, tuple[int, dict[str, str]]] = {}
for line, raw in enumerate(reader, 2):
if None in raw or any(value is None for value in raw.values()):
raise ValueError(f"Malformed REPD CSV record {line}")
ref = raw["Ref ID"]
if ref in rows:
raise ValueError("Duplicate release ID")
rows[ref] = (line, raw)
result = []
for decision in candidates[offset : offset + limit]:
ref = decision["repd_id"]
if ref not in rows:
raise ValueError("Saved candidate missing from pinned release")
line, raw = rows[ref]
if raw["Site Name"] != decision["name"] or raw["Technology Type"] not in {
"Wind Onshore",
"Wind Offshore",
}:
raise ValueError("Saved candidate differs from release identity")
result.append(
dict(
repd_id=ref,
locator=f"CSV record {line}; Ref ID={ref}",
source_id=decision["source_id"],
decision="deferred_identity",
reason=decision["reason"],
facts={field: raw[field].strip() or None for field in FIELDS},
scope="Planning application only; no farm identity or current-status inference",
)
)
return result