#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""Fusionne les features GPT (specs.jsonl) aux données maison (maisons_avec_prix.json)."""
import json, csv

# 1) charger specs GPT
specs = {}
for line in open("data/processed/specs.jsonl", encoding="utf-8"):
    r = json.loads(line)
    if r.get("specs"):
        specs[r["uid"]] = r["specs"]
print(f"Specs GPT chargées : {len(specs)}")

# 2) charger maisons
data = json.load(open("data/processed/maisons_avec_prix.json", encoding="utf-8"))

# 3) JSON fusionné : record complet + gptSpecs imbriqué
merged = []
n_match = 0
for d in data:
    uid = f"{d.get('source')}:{d.get('listingId')}"
    d2 = dict(d)
    d2["uid"] = uid
    s = specs.get(uid)
    d2["gptSpecs"] = s          # None si absent
    if s:
        n_match += 1
    merged.append(d2)
json.dump(merged, open("data/processed/maisons_specs.json", "w", encoding="utf-8"),
          ensure_ascii=False)
print(f"JSON fusionné -> maisons_specs.json ({len(merged)} records, "
      f"{n_match} avec specs)")

# 4) CSV aplati pour analyse
BOOL = ["turnkey","recently_renovated","garage","pool","waterfront",
        "finished_basement","fireplace","central_ac","new_construction",
        "income_property","renovated_kitchen","renovated_bathroom",
        "large_lot","quiet_area"]
FLOOR = ["bois_franc","ceramique","plancher_flottant","tapis","vinyle","beton","autre"]
HEAT = ["electrique","plinthes","thermopompe","bois","gaz","mazout","radiant","geothermie","autre"]

cols = (["uid","source","listingId","price","currency","city","province",
         "ptype","sqft_raw","lot_raw","beds","baths","yearBuilt","parking",
         "has_gpt","g_condition","g_countertop","g_view"]
        + ["g_"+b for b in BOOL]
        + ["fl_"+f for f in FLOOR]
        + ["heat_"+h for h in HEAT]
        + ["g_highlights"])

with open("data/processed/maisons_specs.csv", "w", newline="", encoding="utf-8") as fh:
    w = csv.writer(fh)
    w.writerow(cols)
    for d in data:
        uid = f"{d.get('source')}:{d.get('listingId')}"
        p = d.get("property") or {}
        ad = d.get("address") or {}
        s = specs.get(uid)
        row = [uid, d.get("source"), d.get("listingId"), d.get("priceNumber"),
               d.get("currency"), ad.get("city"), ad.get("province"),
               (p.get("buildingType") or p.get("type")),
               p.get("squareFeet"), p.get("lotSize"),
               p.get("bedrooms"), p.get("bathrooms"), p.get("yearBuilt"),
               p.get("parkingSpaces"),
               1 if s else 0,
               (s or {}).get("condition",""),
               (s or {}).get("countertop",""),
               (s or {}).get("view","")]
        row += [int(bool((s or {}).get(b))) if s else "" for b in BOOL]
        fl = set((s or {}).get("flooring") or [])
        row += [int(f in fl) if s else "" for f in FLOOR]
        ht = set((s or {}).get("heating") or [])
        row += [int(h in ht) if s else "" for h in HEAT]
        row += ["; ".join((s or {}).get("highlights") or [])]
        w.writerow(row)
print(f"CSV aplati -> maisons_specs.csv ({len(cols)} colonnes)")

# 5) petit récap de distribution
import collections
cond = collections.Counter(s.get("condition") for s in specs.values())
print("\nRépartition 'condition' :", dict(cond))
for b in ["turnkey","recently_renovated","central_ac","pool","waterfront",
          "renovated_kitchen","income_property"]:
    n = sum(1 for s in specs.values() if s.get(b))
    print(f"  {b:20s}: {n:5d} ({n/len(specs)*100:.1f}%)")