import json
import os
from datetime import datetime
from typing import List, Dict, Any, Optional


def _project_root() -> str:
    # modules/crm/search.py -> remonter à la racine projet
    return os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))


def _prospects_dir() -> str:
    return os.path.join(_project_root(), "Data", "prospects")


def _read_json(path: str) -> Optional[Dict[str, Any]]:
    try:
        with open(path, "r", encoding="utf-8") as f:
            return json.load(f)
    except Exception:
        return None


def _norm_date(date_str: Optional[str]) -> Optional[str]:
    if not date_str:
        return None
    # Tente de normaliser vers YYYY-MM-DD
    s = str(date_str)
    # Si déjà au format ISO ou contient "T"
    if len(s) >= 10:
        return s[:10]
    return None


def _extract_fields(data: Dict[str, Any], fallback_id: str) -> Dict[str, Any]:
    # Essaye plusieurs variantes de clés
    name = data.get("name") or data.get("nom") or data.get("full_name") or data.get("contact_name") or ""
    email = data.get("email") or data.get("mail") or ""
    company = data.get("company") or data.get("societe") or data.get("entreprise") or data.get("organization") or ""
    status = data.get("status") or data.get("statut") or ""
    city = data.get("city") or data.get("ville") or ""
    tags = data.get("tags") or []
    if isinstance(tags, str):
        # transformer "a, b, c" -> ["a","b","c"]
        tags = [t.strip() for t in tags.split(",") if t.strip()]
    created_at = data.get("created_at") or data.get("created") or data.get("date_created") or ""
    created_at = _norm_date(created_at) or ""
    pid = data.get("id") or fallback_id

    # Liaisons potentielles
    client_id = data.get("client_id") or data.get("clientId") or data.get("client")
    project_ids = data.get("project_ids") or data.get("projects") or data.get("projectId") or []
    if isinstance(project_ids, (str, int)):
        project_ids = [str(project_ids)]
    elif isinstance(project_ids, list):
        project_ids = [str(x) for x in project_ids]

    return {
        "id": pid,
        "entity_type": "prospect",
        "name": name,
        "email": email,
        "company": company,
        "status": status,
        "city": city,
        "tags": tags,
        "created_at": created_at,
        "client_id": client_id,
        "project_ids": project_ids,
    }


def load_all_prospects() -> List[Dict[str, Any]]:
    base_dir = _prospects_dir()
    os.makedirs(base_dir, exist_ok=True)
    results: List[Dict[str, Any]] = []
    for fn in os.listdir(base_dir):
        if not fn.endswith(".json"):
            continue
        path = os.path.join(base_dir, fn)
        data = _read_json(path)
        if not data:
            continue
        pid = os.path.splitext(fn)[0]
        results.append(_extract_fields(data, pid))
    return results


def _contains(hay: str, needle: str) -> bool:
    return needle in hay


def _safe_lower(s: Optional[str]) -> str:
    return (s or "").lower()


def filter_prospects(
    *,
    q: Optional[str] = None,
    name: Optional[str] = None,
    email: Optional[str] = None,
    company: Optional[str] = None,
    status: Optional[str] = None,
    city: Optional[str] = None,
    tags: Optional[List[str]] = None,
    date_from: Optional[str] = None,  # YYYY-MM-DD
    date_to: Optional[str] = None,    # YYYY-MM-DD
) -> List[Dict[str, Any]]:
    items = load_all_prospects()

    q = _safe_lower(q)
    name = _safe_lower(name)
    email = _safe_lower(email)
    company = _safe_lower(company)
    status = _safe_lower(status)
    city = _safe_lower(city)
    tags = [t.strip().lower() for t in (tags or []) if t.strip()]

    df = None
    dt = None
    try:
        if date_from:
            df = datetime.strptime(date_from, "%Y-%m-%d").date()
    except Exception:
        df = None
    try:
        if date_to:
            dt = datetime.strptime(date_to, "%Y-%m-%d").date()
    except Exception:
        dt = None

    out: List[Dict[str, Any]] = []
    for it in items:
        iname = _safe_lower(it.get("name"))
        iemail = _safe_lower(it.get("email"))
        icompany = _safe_lower(it.get("company"))
        istatus = _safe_lower(it.get("status"))
        icity = _safe_lower(it.get("city"))
        itags = [str(t).lower() for t in (it.get("tags") or [])]
        icreated = it.get("created_at") or ""
        icreated_d = None
        try:
            if icreated:
                icreated_d = datetime.strptime(icreated[:10], "%Y-%m-%d").date()
        except Exception:
            icreated_d = None

        # Filtre global q
        if q and not (
            _contains(iname, q) or _contains(iemail, q) or _contains(icompany, q) or _contains(icity, q)
        ):
            continue

        # Filtres spécifiques
        if name and not _contains(iname, name):
            continue
        if email and not _contains(iemail, email):
            continue
        if company and not _contains(icompany, company):
            continue
        if status and istatus != status:
            continue
        if city and not _contains(icity, city):
            continue

        if tags:
            # exiger que tous les tags demandés soient présents
            itags_set = set(itags)
            if not all(tag in itags_set for tag in tags):
                continue

        if df and (not icreated_d or icreated_d < df):
            continue
        if dt and (not icreated_d or icreated_d > dt):
            continue

        out.append(it)

    # On peut trier par date de création desc, puis nom
    out.sort(key=lambda d: (d.get("created_at") or "", d.get("name") or ""), reverse=True)
    return out