"""
===============================================================================
FILE: services/scud_etl/merger.py
PROJECT: SCUD Orion AI (Unified Architecture)
MODULE: services / scud_etl
ROLE: Агрегация проходов, сопоставление исключений и мердж таблиц 1С:ЗУП и СКУД.

AI-CONTEXT-ANCHORS:
  - ANCHOR[MERGER_EXCEPTIONS]: Наложение флага исключений из exceptions.json.
  - ANCHOR[MERGER_AGGREGATION]: Агрегация множественных проходов до уникального ФИО.
  - ANCHOR[MERGER_BUILD_DATASET]: Сборка итогового датасета для сводок и отчетов.
===============================================================================
"""

import os
import json
import pandas as pd
from config import normalize_fio, DATA_DIR


# ANCHOR[MERGER_EXCEPTIONS]
def load_exceptions_config() -> dict:
    """Загружает exceptions.json из корня проекта."""
    root_dir = os.path.abspath(os.path.join(os.path.dirname(__file__), "../../"))
    json_path = os.path.join(root_dir, "exceptions.json")
    if not os.path.exists(json_path):
        return {}
    try:
        with open(json_path, 'r', encoding='utf-8') as f:
            return json.load(f)
    except Exception:
        return {}


def apply_exceptions_from_json(df: pd.DataFrame, exceptions_cfg: dict) -> pd.DataFrame:
    """Быстрая разметка флага is_excluded на основе exceptions.json."""
    if df is None or df.empty or not exceptions_cfg:
        if df is not None:
            df['is_excluded'] = False
        return df

    deps = [d.strip().lower() for d in exceptions_cfg.get("departments", []) if d]
    exact_pos = [p.strip().lower() for p in exceptions_cfg.get("positions", []) if p]
    pos_kw = [k.strip().lower() for k in exceptions_cfg.get("position_keywords", []) if k]
    exc_fios = [normalize_fio(f) for f in exceptions_cfg.get("fio", []) if f]

    df['is_excluded'] = False

    for idx, row in df.iterrows():
        fio_clean = row.get('fio_clean', '')
        dep_1c = str(row.get('Подразделение', '')).lower()
        dep_scud = str(row.get('department_scud', row.get('department', ''))).lower()
        pos = str(row.get('Должность', '')).lower()

        is_fio_exc = fio_clean in exc_fios
        is_pos_exc = (pos in exact_pos) or any(k in pos for k in pos_kw if k) if pos else False
        is_dep_exc = any(d in dep_1c or d in dep_scud for d in deps) if deps else False

        if is_fio_exc or is_dep_exc or is_pos_exc:
            df.at[idx, 'is_excluded'] = True

    return df


# ANCHOR[MERGER_AGGREGATION]
def aggregate_scud_by_employee(df: pd.DataFrame) -> pd.DataFrame:
    """Агрегирует проходы СКУД по уникальным сотрудникам."""
    if df is None or df.empty or 'fio_clean' not in df.columns:
        return df

    aggregated = []
    for fio_clean, group in df.groupby('fio_clean', sort=False):
        is_present = group['Пришел'].any() if 'Пришел' in group.columns else False
        if is_present and 'Пришел' in group.columns:
            present_rows = group[group['Пришел'] == True]
            best_row = present_rows.iloc[0].to_dict() if not present_rows.empty else group.iloc[0].to_dict()
        else:
            best_row = group.iloc[0].to_dict()

        best_row['Пришел'] = is_present
        aggregated.append(best_row)

    return pd.DataFrame(aggregated)


def filter_report_dataframe(merged_df: pd.DataFrame) -> pd.DataFrame:
    """Исключает подрядчиков и сотрудников без пропуска из детального отчета."""
    if merged_df is None or merged_df.empty:
        return merged_df

    has_1c_reason = (
        merged_df['Вид_отсутствия'].notna() & 
        (merged_df['Вид_отсутствия'].astype(str).str.strip() != '') & 
        (~merged_df['Вид_отсутствия'].astype(str).str.startswith('Исключение'))
    )
    is_not_excluded = merged_df.get('is_excluded', False) == False
    is_not_no_pass = merged_df.get('no_scud_pass', False) == False

    return merged_df[(is_not_excluded & is_not_no_pass) | has_1c_reason].copy()


def load_static_reason_workers() -> dict:
    """Загружает реестр удаленщиков из CSV."""
    static_path = os.path.join(DATA_DIR, "static_reason_workers.csv")
    if not os.path.exists(static_path):
        return {}
    try:
        df_static = pd.read_csv(static_path, encoding='utf-8')
        if 'fio' in df_static.columns and 'reason' in df_static.columns:
            df_static['fio_clean'] = df_static['fio'].apply(normalize_fio)
            return dict(zip(df_static['fio_clean'], df_static['reason']))
    except Exception:
        pass
    return {}