# environment_loader.py
'''Utility module to load .xlsx files from environment-specific folders and optional Huggingface repositories.'''

import os
from pathlib import Path
import pandas as pd
from datetime import datetime, timedelta

# Optional Huggingface support
try:
    from huggingface_hub import hf_hub_download
except ImportError:
    hf_hub_download = None


def get_latest_file_in_directory(directory_path: str, pattern: str = '*.xlsx') -> Path:
    '''Return the Path to the latest modified file matching pattern in directory_path.'''
    dir_path = Path(directory_path)
    files = list(dir_path.glob(pattern))
    if not files:
        return None
    return max(files, key=lambda f: f.stat().st_mtime)


def get_file_by_date(directory_path: str, target_date: datetime.date, pattern: str = '*.xlsx') -> Path:
    '''Return Path to the file whose name contains target_date or whose modification date matches target_date.'''
    date_str = target_date.strftime('%Y-%m-%d')
    dir_path = Path(directory_path)
    # First try matching date string in filename
    candidates = [f for f in dir_path.glob(pattern) if date_str in f.name]
    if candidates:
        return max(candidates, key=lambda f: f.stat().st_mtime)
    # Fallback to checking file modification date
    files = [f for f in dir_path.glob(pattern) if datetime.fromtimestamp(f.stat().st_mtime).date() == target_date]
    if files:
        return max(files, key=lambda f: f.stat().st_mtime)
    return None


def load_latest_xlsx_for_env(env_code: str,
                             base_path: str = 'Q:/Selenium_Reports',
                             use_date: bool = False,
                             target_date: datetime.date = None) -> pd.DataFrame:
    '''Load the latest or date-specific .xlsx file for the given environment code.'''
    folder = Path(base_path) / env_code / 'XLSX'
    if not folder.exists():
        raise FileNotFoundError(f"Environment folder not found: {folder}")
    if use_date:
        if target_date is None:
            raise ValueError('target_date must be provided when use_date is True')
        file_path = get_file_by_date(folder, target_date)
    else:
        file_path = get_latest_file_in_directory(folder)
    if file_path is None:
        raise FileNotFoundError(f'No .xlsx files found for environment {env_code} in {folder}')
    return pd.read_excel(file_path)


def load_environments(env_codes: list,
                      base_path: str = 'Q:/Selenium_Reports',
                      by_date: bool = False,
                      days_ago: int = 1) -> dict:
    '''Load DataFrame for each environment code; by_date chooses file from days_ago days before.'''
    data = {}
    target_date = datetime.now().date() - timedelta(days=days_ago)
    for env in env_codes:
        df = load_latest_xlsx_for_env(
            env_code=env,
            base_path=base_path,
            use_date=by_date,
            target_date=target_date
        )
        data[env] = df
    return data


def load_from_huggingface(repo_id: str,
                          filenames: list,
                          revision: str = 'main') -> dict:
    '''Download files from a Huggingface repo and load as DataFrames.'''
    if hf_hub_download is None:
        raise ImportError('huggingface_hub is not installed. Please pip install huggingface_hub')
    data = {}
    for fname in filenames:
        local_path = hf_hub_download(repo_id=repo_id, filename=fname, revision=revision)
        data[fname] = pd.read_excel(local_path)
    return data


def get_latest_xlsx_path_for_env(env_code: str,
                                 base_path: str = 'Q:/Selenium_Reports',
                                 use_date: bool = False,
                                 target_date: datetime.date = None) -> Path:
    '''Return the Path to the desired .xlsx file for the given environment code without loading.'''
    folder = Path(base_path) / env_code / 'XLSX'
    if not folder.exists():
        raise FileNotFoundError(f"Environment folder not found: {folder}")
    if use_date:
        if target_date is None:
            raise ValueError('target_date must be provided when use_date is True')
        file_path = get_file_by_date(folder, target_date)
    else:
        file_path = get_latest_file_in_directory(folder)
    if file_path is None:
        raise FileNotFoundError(f'No .xlsx files found for environment {env_code} in {folder}')
    return file_path


def get_environments_paths(env_codes: list,
                           base_path: str = 'Q:/Selenium_Reports',
                           by_date: bool = False,
                           days_ago: int = 1) -> dict:
    '''Return file Paths for each environment code; by_date chooses file from days_ago days before.'''
    data = {}
    target_date = datetime.now().date() - timedelta(days=days_ago)
    for env in env_codes:
        path = get_latest_xlsx_path_for_env(
            env_code=env,
            base_path=base_path,
            use_date=by_date,
            target_date=target_date
        )
        data[env] = path
    return data


def get_huggingface_paths(repo_id: str,
                          filenames: list,
                          revision: str = 'main') -> dict:
    '''Download files from a Huggingface repo and return local Paths without loading.'''
    if hf_hub_download is None:
        raise ImportError('huggingface_hub is not installed. Please pip install huggingface_hub')
    data = {}
    for fname in filenames:
        local_path = hf_hub_download(repo_id=repo_id, filename=fname, revision=revision)
        data[fname] = Path(local_path)
    return data