MISF-Planner-Download/src/config.py

"""OneDrive / Planner download configuration from Input/input.xlsx."""
from __future__ import annotations

import logging
import os
import re
import sys
from dataclasses import dataclass
from datetime import datetime
from pathlib import Path

import pandas as pd

from totp_utils import validate_totp_secret

if getattr(sys, "frozen", False):
    PROJECT_ROOT = Path(sys.executable).parent
else:
    PROJECT_ROOT = Path(__file__).resolve().parent.parent

CONFIG_FILE = PROJECT_ROOT / "Input" / "input.xlsx"
TODAY_ISO_DATE = datetime.now().strftime("%Y-%m-%d")
EXPECTED_FACILITY_COUNT = 54

DEFAULT_ONEDRIVE_URL = (
    "https://netorg393890-my.sharepoint.com/personal/joy_accesspsychology_org"
    "/_layouts/15/onedrive.aspx"
    "?id=%2Fpersonal%2Fjoy%5Faccesspsychology%5Forg%2FDocuments"
    "%2FAccess%20Facility%20Spreadsheets3"
)

DEFAULT_PLANNER_URL = (
    "https://planner.cloud.microsoft/webui/mytasks/assignedtome/view/grid"
    "?tid=0c15ad2c-19dd-4ded-9cd1-20d344386414"
)

# Rows like REPLACE_WITH_54TH_FACILITY_NAME are skipped when loading facilities.
_FACILITY_PLACEHOLDER_RE = re.compile(
    r"^REPLACE_WITH_.*",
    re.IGNORECASE,
)

_log = logging.getLogger(__name__)


@dataclass(frozen=True)
class AppConfig:
    proxy_ip: str
    proxy_port: str
    proxy_username: str
    proxy_password: str
    onedrive_username: str
    onedrive_password: str
    microsoft_totp_secret: str
    onedrive_url: str
    planner_url: str
    pdf_folder_path: str


def _load_dotenv_file() -> None:
    """Load .env from project root into os.environ (no-override: existing vars take priority)."""
    env_path = PROJECT_ROOT / ".env"
    if not env_path.is_file():
        return
    try:
        from dotenv import load_dotenv
        load_dotenv(dotenv_path=env_path, override=False)
    except ImportError:
        # Fallback: basic KEY=VALUE parser (no multi-line or comment support)
        for raw_line in env_path.read_text(encoding="utf-8").splitlines():
            line = raw_line.strip()
            if not line or line.startswith("#") or "=" not in line:
                continue
            key, _, value = line.partition("=")
            key = key.strip()
            value = value.strip().strip('"').strip("'")
            if key and key not in os.environ:
                os.environ[key] = value


def _env(name: str, *aliases: str) -> str:
    for key in (name, *aliases):
        val = os.environ.get(key, "").strip()
        if val:
            return val
    return ""


def _read_config_key_value_sheet(config_path: Path | None = None) -> dict[str, str]:
    path = config_path or CONFIG_FILE
    if not path.exists():
        raise FileNotFoundError(f"input.xlsx not found at {path}")

    try:
        df = pd.read_excel(str(path), sheet_name="Sheet1", header=None, usecols=[0, 1])
    except Exception as exc:
        raise RuntimeError(
            f"Could not read config from {path} (Sheet1): {exc}"
        ) from exc

    def _clean(val) -> str:
        s = str(val).strip()
        return "" if s.lower() in ("nan", "none", "") else s

    if len(df.columns) < 2:
        raise RuntimeError(
            f"Config sheet Sheet1 in {path} must have at least 2 columns (Key | Value)."
        )

    config: dict[str, str] = {}
    for _, row in df.iterrows():
        key = _clean(row[0])
        value = _clean(row[1])
        if key:
            config[key] = value
    return config


def missing_onedrive_credentials(
    onedrive_username: str,
    onedrive_password: str,
    microsoft_totp_secret: str,
) -> list[str]:
    missing = []
    if not onedrive_username:
        missing.append("OneDrive Username")
    if not onedrive_password:
        missing.append("OneDrive Password")
    if not microsoft_totp_secret:
        missing.append("Microsoft TOTP Secret")
    return missing


def credentials_setup_message(missing: list[str]) -> str:
    keys = ", ".join(missing)
    return (
        f"Missing: {keys}\n\n"
        f"Option A — edit {CONFIG_FILE} Sheet1, column B next to each key.\n"
        f"Option B — copy .env.example to .env in the project root and set:\n"
        "  ONEDRIVE_USERNAME, ONEDRIVE_PASSWORD, MICROSOFT_TOTP_SECRET\n"
        "  PLANNER_URL (optional)\n"
    )


def read_app_config(config_path: Path | None = None) -> AppConfig:
    """Load full application config from input.xlsx with .env fallbacks."""
    _load_dotenv_file()
    config = _read_config_key_value_sheet(config_path)

    proxy_ip = config.get("Proxy IP", "") or _env("PROXY_IP")
    proxy_port = config.get("Proxy Port", "") or _env("PROXY_PORT")
    proxy_username = config.get("Proxy Username", "") or _env("PROXY_USERNAME")
    proxy_password = config.get("Proxy Password", "") or _env("PROXY_PASSWORD")

    onedrive_username = config.get("OneDrive Username", "") or _env(
        "ONEDRIVE_USERNAME", "OneDrive Username"
    )
    onedrive_password = config.get("OneDrive Password", "") or _env(
        "ONEDRIVE_PASSWORD", "OneDrive Password"
    )
    microsoft_totp_secret = (
        config.get("Microsoft TOTP Secret", "")
        or config.get("OTP Secret", "")
        or _env("MICROSOFT_TOTP_SECRET", "OTP_SECRET", "TOTP_SECRET")
    )
    onedrive_url = config.get("OneDrive URL", "") or _env("ONEDRIVE_URL") or DEFAULT_ONEDRIVE_URL
    planner_url = (
        config.get("Planner URL", "")
        or config.get("Planner URL ", "")
        or _env("PLANNER_URL")
        or DEFAULT_PLANNER_URL
    )
    pdf_folder_path = config.get("pdf_folder_path", "") or _env("PDF_FOLDER_PATH")

    _log.info("Proxy              -> %s:%s", proxy_ip, proxy_port)
    _log.info("OneDrive User      -> %s", onedrive_username)
    _log.info("Planner URL        -> %s...", planner_url[:80])
    _log.info("PDF folder path    -> %s", pdf_folder_path or "(default Input/PDFs)")

    return AppConfig(
        proxy_ip=proxy_ip,
        proxy_port=proxy_port,
        proxy_username=proxy_username,
        proxy_password=proxy_password,
        onedrive_username=onedrive_username,
        onedrive_password=onedrive_password,
        microsoft_totp_secret=microsoft_totp_secret,
        onedrive_url=onedrive_url,
        planner_url=planner_url,
        pdf_folder_path=pdf_folder_path,
    )


def read_onedrive_config(config_path: Path | None = None) -> tuple:
    """
    Legacy tuple API. Prefer :func:`read_app_config`.

    Returns
    -------
    (proxy_ip, proxy_port, proxy_username, proxy_password,
     onedrive_username, onedrive_password, microsoft_totp_secret,
     onedrive_url, pdf_folder_path, planner_url)
    """
    cfg = read_app_config(config_path)
    return (
        cfg.proxy_ip,
        cfg.proxy_port,
        cfg.proxy_username,
        cfg.proxy_password,
        cfg.onedrive_username,
        cfg.onedrive_password,
        cfg.microsoft_totp_secret,
        cfg.onedrive_url,
        cfg.pdf_folder_path,
        cfg.planner_url,
    )


def _is_placeholder_facility(name: str) -> bool:
    return bool(_FACILITY_PLACEHOLDER_RE.match(name.strip()))


def read_facility_list(
    config_path: Path | None = None,
    *,
    skip_placeholders: bool = True,
    warn_if_not_expected_count: bool = True,
) -> list[str]:
    path = config_path or CONFIG_FILE
    if not path.exists():
        raise FileNotFoundError(f"input.xlsx not found at {path}")

    try:
        df = pd.read_excel(str(path), sheet_name="Facilities", usecols=[0])
    except ValueError as exc:
        raise RuntimeError(
            f"Sheet 'Facilities' not found in {path}. "
            "Add a sheet with column 'Facility Name' (one facility per row)."
        ) from exc

    # Always use the first column regardless of its header name.
    col = df.columns[0]

    names: list[str] = []
    skipped: list[str] = []
    for val in df[col]:
        s = str(val).strip()
        if not s or s.lower() in ("nan", "none", "facility name"):
            continue
        if skip_placeholders and _is_placeholder_facility(s):
            skipped.append(s)
            continue
        names.append(s)

    if skipped:
        _log.warning(
            "Skipped %d placeholder facility row(s): %s",
            len(skipped),
            ", ".join(skipped[:3]),
        )

    if not names:
        raise RuntimeError(f"No facility names found in 'Facilities' sheet of {path}.")

    if warn_if_not_expected_count and len(names) != EXPECTED_FACILITY_COUNT:
        _log.warning(
            "Expected %d facilities, found %d (check Facilities sheet names).",
            EXPECTED_FACILITY_COUNT,
            len(names),
        )

    _log.info("Loaded %d facility name(s) from Facilities sheet", len(names))
    return names