# scraper/book_scraper.py

import requests
from bs4 import BeautifulSoup
from urllib.parse import urljoin, urlparse

from scraper.logger import log_debug
from scraper.utils import clean_text, load_replacements
from scraper.models.book_state import Chapter


class BookScraper:
    """
    Lightweight scraper: only metadata + chapter list.
    All downloading/parsing/saving is handled by Celery tasks.
    """

    def __init__(self, site, url):
        self.site = site
        self.url = url

        self.book_title = ""
        self.book_author = ""
        self.book_description = ""
        self.cover_url = ""

        self.chapters = []
        self.chapter_base = None

        # Load custom replacements
        extra = load_replacements("replacements.txt")
        self.site.replacements.update(extra)

    # ------------------------------------------------------------
    def parse_book_info(self):
        """Parse title, author, description, cover from the main page."""
        soup = self._fetch(self.url)

        self._parse_title(soup)
        self._parse_author(soup)
        self._parse_description(soup)
        self._parse_cover(soup)

        # Parse chapter list page + chapter links
        chapter_page = self.get_chapter_page(soup)
        self.parse_chapter_links(chapter_page)

    # ------------------------------------------------------------
    def _fetch(self, url):
        """Simple fetch (no retry), DownloadController handles errors."""
        log_debug(f"[BookScraper] Fetch: {url}")
        resp = requests.get(url, headers={"User-Agent": "Mozilla/5.0"}, timeout=10)
        resp.encoding = self.site.encoding
        return BeautifulSoup(resp.text, "lxml")

    # ------------------------------------------------------------
    def _parse_title(self, soup):
        h1 = soup.find("h1")
        self.book_title = h1.get_text(strip=True) if h1 else "UnknownTitle"
        log_debug(f"[BookScraper] Title = {self.book_title}")

    def _parse_author(self, soup):
        td = soup.find("td", string=lambda t: t and "作" in t)
        raw = td.get_text(strip=True) if td else ""
        self.book_author = raw.split("：")[1] if "：" in raw else "UnknownAuthor"
        log_debug(f"[BookScraper] Author = {self.book_author}")

    def _parse_description(self, soup):
        span = soup.find("span", string=lambda t: t and "内容简介" in t)
        if not span:
            self.book_description = ""
            log_debug("[BookScraper] Description not found")
            return

        parts = []
        for sib in span.next_siblings:
            # Stop when next book section begins
            if getattr(sib, "name", None) == "span":
                break

            text = (
                sib.get_text(strip=True)
                if hasattr(sib, "get_text")
                else str(sib).strip()
            )
            if text:
                parts.append(text)

        self.book_description = "\n".join(parts)
        log_debug(
            f"[BookScraper] Description length = {len(self.book_description)} characters"
        )

    # ------------------------------------------------------------
    def _parse_cover(self, soup):
        cover = soup.find("img", src=lambda v: v and "files/article/image" in v)
        if not cover:
            log_debug("[BookScraper] No cover found")
            return

        self.cover_url = urljoin(self.site.root, cover.get("src"))
        log_debug(f"[BookScraper] Cover URL = {self.cover_url}")

    # ------------------------------------------------------------
    def get_chapter_page(self, soup):
        """Return BeautifulSoup of the main chapter list page."""
        node = soup.select_one(
            "html > body > div:nth-of-type(6) > div:nth-of-type(2) > div > table"
        )
        href = node.select_one("a").get("href")
        url = urljoin(self.site.root, href)

        parsed = urlparse(url)
        bp = parsed.path.rsplit("/", 1)[0] + "/"
        self.chapter_base = f"{parsed.scheme}://{parsed.netloc}{bp}"

        return self._fetch(url)

    # ------------------------------------------------------------
    def parse_chapter_links(self, soup):
        cont = soup.select_one(self.site.chapter_list_selector)
        items = cont.select("ul li a[href]")

        self.chapters = []
        idx = 1

        for a in items:
            href = a.get("href")
            if not href.endswith(".html"):
                continue

            title = a.get_text(strip=True)
            full = urljoin(self.chapter_base, href)

            self.chapters.append(Chapter(idx, title, full))
            idx += 1

        log_debug(f"[BookScraper] Found {len(self.chapters)} chapters")

    # ------------------------------------------------------------
    def get_chapter_list(self):
        """Return the chapter list (DownloadController reads this)."""
        return self.chapters