audible.py

"""
audible.py — Pure Audible scraper.

Stateless functions. Takes region as parameter. No decisions, no title cleaning.

Functions:
    search(query, region)       -> list[dict]
    get_by_asin(asin, region)   -> dict | None
    get_details(url)            -> dict
"""

import re
from typing import Optional
from urllib.parse import quote_plus

import requests
from bs4 import BeautifulSoup

DOMAINS = {
    "us": "www.audible.com",
    "uk": "www.audible.co.uk",
    "de": "www.audible.de",
    "fr": "www.audible.fr",
}

REGIONS = list(DOMAINS.keys())

_UA = (
    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
    "AppleWebKit/537.36 (KHTML, like Gecko) "
    "Chrome/120.0.0.0 Safari/537.36"
)

_s = requests.Session()
_s.headers["User-Agent"] = _UA
_s.headers["Accept-Language"] = "en-US,en;q=0.9"


# ============================================================
# Search
# ============================================================

def search(query: str, region: str = "us", max_results: int = 5) -> list:
  """Search Audible. Returns list of result dicts."""
  domain = DOMAINS.get(region, DOMAINS["us"])
  url = f"https://{domain}/search?keywords={quote_plus(query)}"
  try:
    r = _s.get(url, timeout=15)
    r.raise_for_status()
  except Exception:
    return []

  soup = BeautifulSoup(r.text, "html.parser")
  results = []
  items = soup.select("li.productListItem, div.adbl-impression-container")
  if not items:
    items = soup.select("[data-widget='productList'] li")

  for el in items[:max_results]:
    try:
      parsed = _parse_result(el, domain)
      if parsed:
        results.append(parsed)
    except Exception:
      continue

  # Deduplicate by ASIN
  seen, out = set(), []
  for r in results:
    asin = r.get("asin", "")
    if asin and asin in seen:
      continue
    if asin:
      seen.add(asin)
    out.append(r)
  return out


# ============================================================
# ASIN Lookup
# ============================================================

def get_by_asin(asin: str, region: str = "us") -> Optional[dict]:
  """Fetch a book directly by ASIN."""
  domain = DOMAINS.get(region, DOMAINS["us"])
  url = f"https://{domain}/pd/{asin}"
  try:
    r = _s.get(url, timeout=15, allow_redirects=True)
    if r.status_code != 200:
      return None
  except Exception:
    return None

  details = _parse_detail_page(r.text)
  if details:
    details["asin"] = asin
    details["url"] = url
  return details if details else None


# ============================================================
# Book Details
# ============================================================

def get_details(url: str) -> dict:
  """Fetch full metadata from a book's Audible page."""
  try:
    r = _s.get(url, timeout=15)
    r.raise_for_status()
  except Exception:
    return {}
  return _parse_detail_page(r.text)


def _parse_detail_page(html: str) -> dict:
  """Parse an Audible book detail page."""
  soup = BeautifulSoup(html, "html.parser")
  d = {}

  # Description
  desc = soup.select_one(
      "#center-1 .bc-expander-content, "
      "div.productPublisherSummary span.bc-text"
  )
  if desc:
    d["description"] = desc.get_text(strip=True)

  # Genres
  genres = [g.get_text(strip=True)
            for g in soup.select("a.bc-chip, .categoriesLabel a")]
  if genres:
    d["genres"] = genres

  # Publisher, date, language
  for sel, key in [
      (".publisherLabel a, a[href*='publisher']", "publisher"),
      (".releaseDateLabel span.bc-text", "publishedDate"),
      (".languageLabel span.bc-text", "language"),
  ]:
    found = soup.select_one(sel)
    if found:
      d[key] = found.get_text(strip=True)

  # Cover image
  hero = soup.select_one(
      "img.bc-pub-block[src*='images'], "
      "img[alt*='cover'][src*='images']"
  )
  if hero:
    d["cover_url"] = re.sub(r'_SL\d+_', '_SL500_', hero.get("src", ""))

  # ASIN from page
  asin_el = soup.select_one("input[name='asin'], [data-asin]")
  if asin_el:
    d["asin"] = asin_el.get("value") or asin_el.get("data-asin")

  # Title from page
  title_el = soup.select_one("h1.bc-heading, h1")
  if title_el:
    t = title_el.get_text(strip=True)
    if t:
      d["title"] = t

  # Author from page
  author_el = soup.select_one(".authorLabel a, a[href*='author']")
  if author_el:
    d["author"] = author_el.get_text(strip=True)

  # Narrator from page
  narrator_el = soup.select_one(".narratorLabel a, a[href*='narrator']")
  if narrator_el:
    d["narrator"] = narrator_el.get_text(strip=True)

  return d


# ============================================================
# Internal Parsing
# ============================================================

def _parse_result(el, domain: str) -> Optional[dict]:
  """Parse a single search result element."""
  r = {}
  title_el = el.select_one("h3 a, .bc-heading a, a.bc-link[href*='/pd/']")
  if not title_el:
    return None

  r["title"] = title_el.get_text(strip=True)
  r["url"] = title_el.get("href", "")
  if r["url"] and not r["url"].startswith("http"):
    r["url"] = f"https://{domain}{r['url']}"

  # Extract ASIN from URL
  asin_match = re.search(r'/([A-Z0-9]{10})(?:\?|$)', r["url"])
  if asin_match:
    r["asin"] = asin_match.group(1)

  # Extract title from URL slug if element text was empty
  if not r["title"] and r["url"]:
    slug = re.search(r'/pd/([^/]+?)(?:-Audiobook)?/', r["url"])
    if slug:
      r["title"] = slug.group(1).replace('-', ' ')

  if not r["title"]:
    return None

  # Author, narrator
  for sel, key in [
      (".authorLabel a, .bc-color-secondary a[href*='author']", "author"),
      (".narratorLabel a, a[href*='narrator']", "narrator"),
  ]:
    found = el.select_one(sel)
    if found:
      r[key] = found.get_text(strip=True)

  # Series
  series_el = el.select_one(".seriesLabel a, a[href*='series']")
  if series_el:
    r["series"] = series_el.get_text(strip=True)
    parent = series_el.parent.get_text(
        strip=True) if series_el.parent else ""
    seq = re.search(r'Book\s+(\d+(?:\.\d+)?)', parent)
    if seq:
      r["series_sequence"] = seq.group(1)

  # Cover
  img = el.select_one("img[src*='images']")
  if img:
    r["cover_url"] = re.sub(r'_SL\d+_', '_SL500_', img.get("src", ""))

  # Runtime
  rt = el.select_one(".runtimeLabel, .bc-text:-soup-contains('hrs')")
  if rt:
    r["runtime"] = rt.get_text(strip=True)

  return r