bookinfo.py

"""
bookinfo.py — Fallback book metadata from Google Books and Open Library.

When Audible search fails entirely, these APIs can still provide:
  - Cover image
  - Description / synopsis
  - Publisher, publish date
  - ISBN, page count
  - Genres / categories

No API keys needed. Both are free and public.
"""

import logging
import re

import requests

log = logging.getLogger(__name__)

_s = requests.Session()
_s.headers["User-Agent"] = "AudiobookshelfEnricher/1.0"


# ============================================================
# Google Books
# ============================================================

def search_google_books(title: str, author: str = "", max_results: int = 5) -> list[dict]:
  """
  Search Google Books API. Returns list of normalized result dicts.
  Free, no API key needed (rate limited to ~1000/day).
  """
  query = title
  if author:
    query += f"+inauthor:{author}"

  try:
    r = _s.get(
        "https://www.googleapis.com/books/v1/volumes",
        params={"q": query, "maxResults": max_results, "printType": "books"},
        timeout=15,
    )
    r.raise_for_status()
    data = r.json()
  except Exception as e:
    log.warning(f"  Google Books error: {e}")
    return []

  results = []
  for item in data.get("items", []):
    info = item.get("volumeInfo", {})
    result = _parse_google_book(info)
    if result.get("title"):
      result["source"] = "google_books"
      result["google_id"] = item.get("id", "")
      results.append(result)

  return results


def _parse_google_book(info: dict) -> dict:
  """Parse a Google Books volumeInfo into our standard format."""
  r = {}

  if info.get("title"):
    r["title"] = info["title"]
    if info.get("subtitle"):
      r["title"] += f": {info['subtitle']}"

  if info.get("authors"):
    r["author"] = ", ".join(info["authors"])

  if info.get("description"):
    r["description"] = info["description"]

  if info.get("publisher"):
    r["publisher"] = info["publisher"]

  if info.get("publishedDate"):
    r["publishedDate"] = info["publishedDate"]

  if info.get("categories"):
    r["genres"] = info["categories"]

  if info.get("language"):
    r["language"] = info["language"]

  # Cover image — prefer large, fall back to smaller
  images = info.get("imageLinks", {})
  for size in ["extraLarge", "large", "medium", "small", "thumbnail"]:
    if images.get(size):
      # Google Books returns http URLs, upgrade to https
      # Also remove edge=curl parameter for cleaner image
      cover = images[size].replace("http://", "https://")
      cover = re.sub(r'&edge=curl', '', cover)
      # Request larger zoom
      cover = re.sub(r'zoom=\d', 'zoom=3', cover)
      r["cover_url"] = cover
      break

  # ISBNs
  for ident in info.get("industryIdentifiers", []):
    if ident.get("type") == "ISBN_13":
      r["isbn13"] = ident["identifier"]
    elif ident.get("type") == "ISBN_10":
      r["isbn10"] = ident["identifier"]

  return r


# ============================================================
# Open Library
# ============================================================

def search_open_library(title: str, author: str = "", max_results: int = 5) -> list[dict]:
  """
  Search Open Library API. Returns list of normalized result dicts.
  Completely free, no limits.
  """
  params = {"title": title, "limit": max_results}
  if author:
    params["author"] = author

  try:
    r = _s.get(
        "https://openlibrary.org/search.json",
        params=params,
        timeout=15,
    )
    r.raise_for_status()
    data = r.json()
  except Exception as e:
    log.warning(f"  Open Library error: {e}")
    return []

  results = []
  for doc in data.get("docs", [])[:max_results]:
    result = _parse_open_library(doc)
    if result.get("title"):
      result["source"] = "open_library"
      results.append(result)

  return results


def _parse_open_library(doc: dict) -> dict:
  """Parse an Open Library search doc into our standard format."""
  r = {}

  if doc.get("title"):
    r["title"] = doc["title"]

  if doc.get("author_name"):
    r["author"] = ", ".join(doc["author_name"])

  if doc.get("publisher"):
    r["publisher"] = doc["publisher"][0] if doc["publisher"] else ""

  if doc.get("first_publish_year"):
    r["publishedDate"] = str(doc["first_publish_year"])

  if doc.get("subject"):
    r["genres"] = doc["subject"][:5]  # Cap at 5

  if doc.get("language"):
    r["language"] = doc["language"][0] if doc["language"] else ""

  # Cover from Open Library covers API
  cover_id = doc.get("cover_i")
  if cover_id:
    r["cover_url"] = f"https://covers.openlibrary.org/b/id/{cover_id}-L.jpg"

  # ISBNs
  if doc.get("isbn"):
    for isbn in doc["isbn"]:
      if len(isbn) == 13:
        r.setdefault("isbn13", isbn)
      elif len(isbn) == 10:
        r.setdefault("isbn10", isbn)

  # Open Library key for detail fetch
  if doc.get("key"):
    r["ol_key"] = doc["key"]

  return r


def get_open_library_description(ol_key: str) -> str:
  """Fetch description from Open Library work/edition page."""
  try:
    r = _s.get(f"https://openlibrary.org{ol_key}.json", timeout=10)
    r.raise_for_status()
    data = r.json()
  except Exception:
    return ""

  desc = data.get("description", "")
  if isinstance(desc, dict):
    desc = desc.get("value", "")
  return desc


# ============================================================
# Combined Search
# ============================================================

def search_book_metadata(title: str, author: str = "") -> list[dict]:
  """
  Search both Google Books and Open Library, return combined results.
  Google Books results come first (usually better covers and descriptions).
  """
  results = []
  results.extend(search_google_books(title, author, max_results=3))
  results.extend(search_open_library(title, author, max_results=3))

  # Enrich Open Library results with descriptions if missing
  for r in results:
    if r.get("source") == "open_library" and not r.get("description") and r.get("ol_key"):
      desc = get_open_library_description(r["ol_key"])
      if desc:
        r["description"] = desc

  return results