#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Analyse de la photo de façade (1/annonce) via Claude Haiku 4.5.
Attributs visuels OBJECTIFS. Parallèle, résumable (facade_specs.jsonl), retry.
"""
import json, os, time, threading, urllib.request, urllib.error
from concurrent.futures import ThreadPoolExecutor

SRC = "data/processed/maisons_avec_prix.json"
OUT = "data/processed/facade_specs.jsonl"
WORKERS = int(os.environ.get("WORKERS", "16"))
NSHARDS = int(os.environ.get("NSHARDS", "1"))
SHARD = int(os.environ.get("SHARD", "0"))
MODEL = "claude-haiku-4-5"

for line in open(".env", encoding="utf-8"):
    if line.startswith("ANTHROPIC_API_KEY="):
        KEY = line.split("=", 1)[1].strip()

SCHEMA = {
    "type": "object", "additionalProperties": False,
    "properties": {
        "is_exterior": {"type": "boolean", "description": "la photo montre-t-elle l'extérieur/façade"},
        "building_type_visual": {"type": "string", "enum": ["maison_detachee","jumele","rangee","condo_immeuble","chalet","terrain","autre"]},
        "condition_score": {"type": "integer", "description": "état visuel 1=délabré à 10=neuf/luxueux"},
        "curb_appeal": {"type": "integer", "description": "attrait extérieur 1-10"},
        "renovation_level": {"type": "string", "enum": ["a_renover","date","standard","renove","haut_de_gamme","inconnu"]},
        "exterior_materials": {"type": "array", "items": {"type": "string", "enum": ["brique","pierre","vinyle","bois","aluminium","stucco","crepi","beton","autre"]}},
        "landscaping": {"type": "string", "enum": ["aucun","minimal","entretenu","amenage_soigne"]},
        "has_garage_visible": {"type": "boolean"},
        "storeys_visual": {"type": "string", "enum": ["1","1.5","2","3+","inconnu"]},
        "setting": {"type": "string", "enum": ["urbain_dense","banlieue","rural","bord_eau","boise","inconnu"]},
        "photo_quality": {"type": "integer", "description": "qualité de la photo 1-10"},
        "curb_luxury": {"type": "boolean", "description": "signes de luxe (grande taille, matériaux nobles, prestige)"}
    },
    "required": ["is_exterior","building_type_visual","condition_score","curb_appeal",
                 "renovation_level","exterior_materials","landscaping","has_garage_visible",
                 "storeys_visual","setting","photo_quality","curb_luxury"]
}
SYS = ("Tu es un évaluateur immobilier objectif. Analyse la photo (censée être la "
       "façade d'une propriété) et renvoie des attributs FACTUELS et reproductibles "
       "au format JSON. Fonde les scores sur des critères concrets (état des "
       "surfaces, matériaux, aménagement), pas sur un goût personnel. Si la photo "
       "n'est pas un extérieur, mets is_exterior=false et remplis au mieux.")

ROOM_KW = ("kitchen","cuisine","bathroom","salle de bain","bedroom","chambre",
           "living","salon","basement","sous-sol","dining","salle à manger",
           "garage","office","bureau","laundry","closet","stair")

def pick_facade(d):
    imgs = d.get("images") or []
    if not imgs:
        return None
    if d.get("source") == "realtor":
        return imgs[0].get("url")  # 1re photo = extérieur
    # duproprio : caption = adresse (sans mot-clé de pièce) = photo extérieure
    for im in imgs:
        cap = (im.get("caption") or "").lower()
        if cap and not any(k in cap for k in ROOM_KW):
            return im.get("url")
    return imgs[0].get("url")

def call(url, tries=5):
    body = {"model": MODEL, "max_tokens": 400, "system": SYS,
            "messages": [{"role": "user", "content": [
                {"type": "image", "source": {"type": "url", "url": url}},
                {"type": "text", "text": "Analyse cette photo de propriété."}]}],
            "output_config": {"format": {"type": "json_schema", "schema": SCHEMA}}}
    data = json.dumps(body).encode()
    for t in range(tries):
        try:
            req = urllib.request.Request(
                "https://api.anthropic.com/v1/messages", data=data,
                headers={"x-api-key": KEY, "anthropic-version": "2023-06-01",
                         "content-type": "application/json"})
            with urllib.request.urlopen(req, timeout=120) as r:
                out = json.load(r)
            txt = next(b["text"] for b in out["content"] if b["type"] == "text")
            return json.loads(txt)
        except urllib.error.HTTPError as e:
            # 400 inclus : fetch d'URL image transitoire côté Anthropic sous charge
            if e.code in (400, 429, 500, 502, 503, 504, 529) and t < tries - 1:
                time.sleep(2 ** t + 1); continue
            raise
        except (urllib.error.URLError, TimeoutError, ValueError):
            if t < tries - 1:
                time.sleep(2 ** t + 1); continue
            raise
    return None

data = json.load(open(SRC, encoding="utf-8"))
done = set()
if os.path.exists(OUT):
    for line in open(OUT, encoding="utf-8"):
        try: done.add(json.loads(line)["uid"])
        except Exception: pass

todo = []
for d in data:
    uid = f"{d.get('source')}:{d.get('listingId')}"
    if uid in done:
        continue
    url = pick_facade(d)
    if url:
        todo.append((uid, url))

if NSHARDS > 1:
    todo = [t for i, t in enumerate(todo) if i % NSHARDS == SHARD]

print(f"[shard {SHARD}/{NSHARDS}, workers={WORKERS}] Total={len(data)}  "
      f"déjà fait={len(done)}  à traiter (ce shard)={len(todo)}", flush=True)

lock = threading.Lock()
c = {"n": 0, "err": 0}
fh = open(OUT, "a", encoding="utf-8")

def work(item):
    uid, url = item
    try:
        rec = {"uid": uid, "facade": call(url)}
    except Exception as e:
        rec = {"uid": uid, "error": str(e)[:200]}
    with lock:
        fh.write(json.dumps(rec, ensure_ascii=False) + "\n"); fh.flush()
        c["n"] += 1
        if "error" in rec: c["err"] += 1
        if c["n"] % 250 == 0:
            print(f"  {c['n']}/{len(todo)} (err={c['err']})", flush=True)

t0 = time.time()
with ThreadPoolExecutor(max_workers=WORKERS) as ex:
    list(ex.map(work, todo))
fh.close()
print(f"Terminé : {c['n']} traités, {c['err']} erreurs, {time.time()-t0:.0f}s", flush=True)