"""Search engine spider / bot detection based on User-Agent headers.

Provides utilities to identify well-known search engine crawlers so the
application can make informed decisions (e.g., serving pre-rendered HTML
for SEO, rate-limiting bot traffic, or tracking crawl statistics).

Usage:
    from app.spider_detection import is_spider, get_spider_info

    if is_spider(user_agent):
        spider = get_spider_info(user_agent)
        print(f\"{spider.name} ({spider.engine}) detected\")
"""

from __future__ import annotations

import re
from dataclasses import dataclass
from enum import Enum
from typing import Optional


class SpiderEngine(str, Enum):
    """Well-known search engine crawlers."""

    GOOGLE = "Google"
    BING = "Bing"
    YAHOO = "Yahoo"
    DUCKDUCKGO = "DuckDuckGo"
    BINGBOT = "Bing"  # alias kept for naming clarity — maps to Bing engine
    YANDEX = "Yandex"
    BAIDU = "Baidu"
    SOCK = "Sogou"
    FACEBOT = "Facebook"
    TWITTERBOT = "Twitter"
    APPLEBOT = "Apple"
    PETLEO = "PetleO"
    IA_archiver = "IA Archiver"
    OTHER = "Other"


@dataclass(frozen=True)
class SpiderInfo:
    """Identified spider details.

    Attributes:
        engine:   Human-friendly name of the search engine / platform.
        name:     Exact bot name from the User-Agent (e.g. 'Googlebot').
        category: High-level category — 'search', 'social', or 'other'.
    """

    engine: str
    name: str
    category: str  # "search" | "social" | "other"

    @property
    def is_search_spider(self) -> bool:
        return self.category == "search"

    @property
    def is_social_bot(self) -> bool:
        return self.category == "social"


# ── Spider patterns ──────────────────────────────────────────
# Each tuple: (compiled-regex, engine-label, category)
# Order matters: more specific patterns first to avoid false positives.

_SPIDER_PATTERNS: list[tuple[re.Pattern[str], str, str]] = [
    # ── Google family ─────────────────────────────────────
    # Main crawler (desktop)
    (re.compile(r"Googlebot/[\d.]+", re.IGNORECASE), "Google", "search"),
    # Googlebot for mobile
    (re.compile(r"Googlebot-Mobile", re.IGNORECASE), "Google", "search"),
    # Google Images
    (re.compile(r"Googlebot-Image", re.IGNORECASE), "Google", "search"),
    # Google News
    (re.compile(r"Googlebot-News", re.IGNORECASE), "Google", "search"),
    # Google Ads/AdSense
    (re.compile(r"Mediapartners-Google", re.IGNORECASE), "Google", "search"),
    # Google Storebot
    (re.compile(r"Storebot-Google", re.IGNORECASE), "Google", "search"),
    # Google Other (GoogleOther)
    (re.compile(r"Google-AdSense", re.IGNORECASE), "Google", "search"),

    # ── Bing / Microsoft ──────────────────────────────────
    (re.compile(r"bingbot/[\d.]+", re.IGNORECASE), "Bing", "search"),
    # MSNBot (legacy Bing crawler)
    (re.compile(r"msnbot/[\d.]+", re.IGNORECASE), "Bing", "search"),
    # MSNBot Media
    (re.compile(r"msnbot-media/[\d.]+", re.IGNORECASE), "Bing", "search"),
    # bingadsbot
    (re.compile(r"bingadsbot/[\d.]+", re.IGNORECASE), "Bing", "search"),
    # WebPreview / Slurp variants from Microsoft
    (re.compile(r"webpreview", re.IGNORECASE), "Bing", "search"),

    # ── Yahoo / Slurp ─────────────────────────────────────
    (re.compile(r"Yahoo! Slurp", re.IGNORECASE), "Yahoo", "search"),
    (re.compile(r"Yahoo! Slurp China", re.IGNORECASE), "Yahoo", "search"),
    (re.compile(r"Yahoo Deer Spider", re.IGNORECASE), "Yahoo", "search"),
    (re.compile(r"YandexBot", re.IGNORECASE), "Yandex", "search"),

    # ── DuckDuckGo ────────────────────────────────────────
    (re.compile(r"DuckDuckBot/[\d.]+", re.IGNORECASE), "DuckDuckGo", "search"),

    # ── Yandex (RU) ───────────────────────────────────────
    (re.compile(r"YandexBot/[\d.]+", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexImages/[\d.]+", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexAccessibilityBot", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexBlogs", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexDirect", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexFavicons", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexMarket", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexMobileBot", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexMetrika", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexSpravochnik", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexSitelinks", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexTracker", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexVertis", re.IGNORECASE), "Yandex", "search"),
    (re.compile(r"YandexWebmaster", re.IGNORECASE), "Yandex", "search"),

    # ── Baidu (CN) ────────────────────────────────────────
    (re.compile(r"Baiduspider/[\d.]+", re.IGNORECASE), "Baidu", "search"),
    (re.compile(r"Baiduboxapp", re.IGNORECASE), "Baidu", "search"),

    # ── Sogou (CN) ────────────────────────────────────────
    (re.compile(r"Sogou [Ss]pider", re.IGNORECASE), "Sogou", "search"),
    (re.compile(r"Sogou web spider", re.IGNORECASE), "Sogou", "search"),

    # ── Apple ─────────────────────────────────────────────
    (re.compile(r"Applebot/[\d.]+", re.IGNORECASE), "Apple", "search"),
    (re.compile(r"Applebot-Extended", re.IGNORECASE), "Apple", "search"),

    # ── PetleO (Japan) ────────────────────────────────────
    (re.compile(r"petalbot", re.IGNORECASE), "PetleO", "search"),

    # ── Internet Archive ──────────────────────────────────
    (re.compile(r"ia_archiver", re.IGNORECASE), "IA Archiver", "other"),
    (re.compile(r"archive\.org_bot", re.IGNORECASE), "IA Archiver", "other"),

    # ── Social media bots ─────────────────────────────────
    (re.compile(r"facebookexternalhit/[\d.]+", re.IGNORECASE), "Facebook", "social"),
    (re.compile(r"facebookplatform", re.IGNORECASE), "Facebook", "social"),
    (re.compile(r"Twitterbot/[\d.]+", re.IGNORECASE), "Twitter", "social"),
    (re.compile(r"LinkedInBot/[\d.]+", re.IGNORECASE), "LinkedIn", "social"),
    (re.compile(r"W3C_Validator", re.IGNORECASE), "W3C", "other"),
    (re.compile(r"WhatsApp/[\d.]+", re.IGNORECASE), "WhatsApp", "social"),

    # ── Generic / catch-all bot signatures ────────────────────────
    # These are last to reduce false positives on real browsers that
    # happen to include the word "bot" in extension names.
    # Standard bot naming convention: SomeBot/X.Y
    (re.compile(r"Bot/[\d.]", re.IGNORECASE), "Other", "other"),
    # Semicolon-delimited bot markers: "; bot" or ";bot"
    (re.compile(r";\s*bot\s*;?", re.IGNORECASE), "Other", "other"),
    # Spider anywhere in the UA
    (re.compile(r"spider", re.IGNORECASE), "Other", "other"),
    # Crawler anywhere in the UA
    (re.compile(r"crawler", re.IGNORECASE), "Other", "other"),
]


# ── Public API ───────────────────────────────────────────────

def is_spider(user_agent: str) -> bool:
    """Return True if the User-Agent string belongs to a known spider/bot.

    This is a fast check — use `get_spider_info()` when you need details.

    Args:
        user_agent: The raw User-Agent header value.

    Returns:
        True if a spider pattern matches, False for normal browsers.
    """
    if not user_agent:
        return False
    for pattern, _, _ in _SPIDER_PATTERNS:
        if pattern.search(user_agent):
            return True
    return False


def get_spider_info(user_agent: str) -> Optional[SpiderInfo]:
    """Identify a spider from a User-Agent string.

    Returns the first matching SpiderInfo or None if the UA does not
    belong to a recognized bot.

    Args:
        user_agent: The raw User-Agent header value.

    Returns:
        SpiderInfo with engine/name/category, or None for regular browsers.
    """
    if not user_agent:
        return None
    for pattern, engine, category in _SPIDER_PATTERNS:
        match = pattern.search(user_agent)
        if match:
            return SpiderInfo(engine=engine, name=match.group(0), category=category)
    return None


def is_search_spider(user_agent: str) -> bool:
    """Return True only for search-engine crawlers (not social bots).

    Useful when you need SSR/pre-rendering only for SEO-relevant crawlers.

    Args:
        user_agent: The raw User-Agent header value.

    Returns:
        True only if a search-engine spider pattern matches.
    """
    info = get_spider_info(user_agent)
    return info.is_search_spider if info else False


# ── FastAPI middleware helper ────────────────────────────────

def get_user_agent(headers: dict) -> str:
    """Extract the User-Agent string from a headers dict (case-insensitive).

    Args:
        headers: A dict-like mapping of header names to values.

    Returns:
        The User-Agent value, empty string if not present.
    """
    for key, value in headers.items():
        if key.lower() == "user-agent":
            return value
    return ""