#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""Test : extraction de specs via GPT-4o-mini (API REST, sans dépendance)."""
import json, os, urllib.request, urllib.error

# charge la clé depuis .env
for line in open(".env", encoding="utf-8"):
    if line.startswith("OPENAI_API_KEY="):
        os.environ["OPENAI_API_KEY"] = line.split("=", 1)[1].strip()
KEY = os.environ["OPENAI_API_KEY"]

SCHEMA = {
    "type": "object",
    "additionalProperties": False,
    "properties": {
        "condition":       {"type": "string", "enum": ["a_renover","original","bon","renove","refait_neuf","neuf","inconnu"]},
        "turnkey":         {"type": "boolean"},
        "recently_renovated": {"type": "boolean"},
        "flooring":        {"type": "array", "items": {"type": "string", "enum": ["bois_franc","ceramique","plancher_flottant","tapis","vinyle","beton","autre"]}},
        "countertop":      {"type": "string", "enum": ["granit","quartz","marbre","stratifie","bois","autre","inconnu"]},
        "heating":         {"type": "array", "items": {"type": "string", "enum": ["electrique","plinthes","thermopompe","bois","gaz","mazout","radiant","geothermie","autre"]}},
        "garage":          {"type": "boolean"},
        "pool":            {"type": "boolean"},
        "waterfront":      {"type": "boolean"},
        "view":            {"type": "string", "enum": ["eau","montagne","ville","aucune","inconnu"]},
        "finished_basement":{"type": "boolean"},
        "fireplace":       {"type": "boolean"},
        "central_ac":      {"type": "boolean"},
        "new_construction":{"type": "boolean"},
        "income_property": {"type": "boolean"},
        "highlights":      {"type": "array", "items": {"type": "string"}, "description": "3-6 mots-clés de vente saillants"}
    },
    "required": ["condition","turnkey","recently_renovated","flooring","countertop",
                 "heating","garage","pool","waterfront","view","finished_basement",
                 "fireplace","central_ac","new_construction","income_property","highlights"]
}

SYS = ("Tu es un extracteur d'attributs immobiliers. À partir de la description "
       "d'une propriété (français ou anglais), renvoie les specs au format JSON "
       "strict. Base-toi UNIQUEMENT sur le texte ; si une info est absente, mets "
       "false / 'inconnu' / [] selon le champ. highlights = 3 à 6 mots-clés.")

def extract(desc):
    body = {
        "model": "gpt-4o-mini",
        "temperature": 0,
        "messages": [{"role": "system", "content": SYS},
                     {"role": "user", "content": desc[:4000]}],
        "response_format": {"type": "json_schema", "json_schema":
            {"name": "specs", "strict": True, "schema": SCHEMA}},
    }
    req = urllib.request.Request(
        "https://api.openai.com/v1/chat/completions",
        data=json.dumps(body).encode(),
        headers={"Authorization": f"Bearer {KEY}",
                 "Content-Type": "application/json"})
    with urllib.request.urlopen(req, timeout=60) as r:
        out = json.load(r)
    return out["choices"][0]["message"]["content"], out.get("usage")

# 3 descriptions du dataset
data = json.load(open("data/processed/maisons_avec_prix.json", encoding="utf-8"))
descs = [d["description"] for d in data if d.get("description")][:3]
for i, dsc in enumerate(descs):
    try:
        content, usage = extract(dsc)
        print(f"\n=== Exemple {i+1} (tokens: {usage}) ===")
        print("DESC:", dsc[:160].replace("\n", " "), "...")
        print("SPECS:", json.dumps(json.loads(content), ensure_ascii=False))
    except urllib.error.HTTPError as e:
        print("HTTP", e.code, e.read().decode()[:400]); break