[books] Fix metadata scrapping

It's not perfect, but at least we get covers now

PROJECT.org

@@ -88,7 +88,7 @@ fetching and simple saving.
 *** Metadata sources
 **** Scraper
 
-* Backlog [0/14] :vrobbler:project:personal:
+* Backlog [1/15] :vrobbler:project:personal:
 ** TODO [#C] Create small utility to clean up tracks scrobbled with wonky playback times :vrobbler:personal:bug:music:scrobbles:
 :PROPERTIES:
 :ID:       702462cf-d54b-48c6-8a7c-78b8de751deb
@@ -442,15 +442,6 @@ displayed in the template.
 
 ** TODO [#B] Add CSV endpoint for book scrobbles that LibraryThing can ingest :personal:project:books:feature:export:
 https://app.todoist.com/app/task/add-a-csv-endpoint-for-users-book-reads-that-library-thing-can-ingest-6X7QPMRp265xMXqg#comment-6X7QrXq6gJjMP4hg
-** TODO [#B] Scrape ComicBookRoundUp ratings for comic book metadata :vrobbler:books:feature:comicbook:personal:project:
-
-- Note taken on [2025-09-25 Thu 10:51]
-
-  As an example https://comicbookroundup.com/comic-books/reviews/humanoids-publishing/the-history-of-science-fiction
-** TODO [#B] Find page numbers for comic books from ComicVine :feature:books:
-:PROPERTIES:
-:ID:       79f867c3-1288-4143-b6bf-2a452983ee9f
-:END:
 ** TODO [#B] Make IMAP and WebDAV configurable :webdav:feature:imap:importers:
 :PROPERTIES:
 :ID:       b1426d92-2feb-4d15-9738-d5b7b0594f96
@@ -518,6 +509,20 @@ log a warning and move on.
 We should have a global view `/favorites/` that shows the logged in users's
 favorited media objects.
 
+** TODO [#B] Scrape ComicBookRoundUp ratings for comic book metadata :vrobbler:books:feature:comicbook:personal:project:
+
+- Note taken on [2025-09-25 Thu 10:51]
+
+  As an example https://comicbookroundup.com/comic-books/reviews/humanoids-publishing/the-history-of-science-fiction
+** TODO [#B] Find page numbers for comic books from ComicVine :feature:books:
+:PROPERTIES:
+:ID:       79f867c3-1288-4143-b6bf-2a452983ee9f
+:END:
+** DONE [#B] Fix metadata scraping for books :books:metadata:
+:PROPERTIES:
+:ID:       ea416a69-a8a8-4d05-b7d4-0a3470820e34
+:END:
+
 * Version 51.1 [1/1]
 ** DONE [#A] Fix scrobbling comic books :books:scrobbles:bug:
 :PROPERTIES:

vrobbler/apps/books/management/commands/cleanup_book_metadata.py

@@ -0,0 +1,282 @@
+import logging
+import time
+
+from django.core.management.base import BaseCommand
+from django.db import transaction
+
+from books.constants import READCOMICSONLINE_URL
+
+logger = logging.getLogger(__name__)
+
+MISSING_ALL = [
+    "cover",
+    "summary",
+    "isbn",
+    "pages",
+    "language",
+    "publisher",
+    "publish_year",
+]
+
+MISSING_GROUPS = {
+    "cover": lambda b: not bool(b.cover),
+    "summary": lambda b: not b.summary,
+    "isbn": lambda b: not b.isbn_13 and not b.isbn_10,
+    "pages": lambda b: b.pages is None,
+    "language": lambda b: not b.language,
+    "publisher": lambda b: not b.publisher,
+    "publish_year": lambda b: b.first_publish_year is None,
+}
+
+
+def _book_matches(book, flags):
+    if not flags:
+        return False
+    for flag in flags:
+        fn = MISSING_GROUPS.get(flag)
+        if fn and fn(book):
+            return True
+    return False
+
+
+class Command(BaseCommand):
+    help = "Backfill missing metadata on books from Google Books, OpenLibrary, and ComicVine"
+
+    def add_arguments(self, parser):
+        parser.add_argument(
+            "--commit",
+            action="store_true",
+            help="Commit changes to the database",
+        )
+        parser.add_argument(
+            "--batch-size",
+            type=int,
+            default=100,
+            help="Number of books to process per batch (default: 100)",
+        )
+        parser.add_argument(
+            "--sleep",
+            type=float,
+            default=0.5,
+            help="Seconds to sleep between API calls (default: 0.5)",
+        )
+        for flag in MISSING_ALL:
+            parser.add_argument(
+                f"--missing-{flag}",
+                dest="missing_flags",
+                action="append_const",
+                const=flag,
+                help=f"Process books missing {flag}",
+            )
+        parser.add_argument(
+            "--comics-only",
+            action="store_true",
+            help="Only process books with a readcomicsonline.ru URL",
+        )
+        parser.add_argument(
+            "--all",
+            action="store_true",
+            dest="all_missing",
+            help="Process books missing any metadata field",
+        )
+
+    def handle(self, *args, **options):
+        from books.models import Book
+
+        commit = options["commit"]
+        batch_size = options["batch_size"]
+        sleep_secs = options["sleep"]
+        flags = options.get("missing_flags") or []
+        comics_only = options["comics_only"]
+        all_missing = options["all_missing"]
+
+        if all_missing:
+            flags = MISSING_ALL
+
+        if not flags and not comics_only:
+            self.stdout.write(
+                "No filters specified. Use --all, --missing-*, or --comics-only."
+            )
+            return
+
+        qs = Book.objects.all()
+        if comics_only:
+            qs = qs.filter(readcomics_url__isnull=False)
+
+        if flags:
+            qs = [b for b in qs.iterator() if _book_matches(b, flags)]
+        else:
+            qs = list(qs)
+
+        total = len(qs)
+        self.stdout.write(f"Found {total} books to process")
+
+        if not commit:
+            self.stdout.write(
+                "Dry run — no API calls will be made. Use --commit to run lookups."
+            )
+            return
+
+        enriched = 0
+        skipped = 0
+        stats = {
+            "cover_fixed": 0,
+            "summary_fixed": 0,
+            "isbn_fixed": 0,
+            "pages_fixed": 0,
+            "language_fixed": 0,
+            "publisher_fixed": 0,
+            "publish_year_fixed": 0,
+        }
+
+        for batch_num, offset in enumerate(range(0, len(qs), batch_size)):
+            batch = qs[offset : offset + batch_size]
+            for book in batch:
+                result = self._enrich_book(book, sleep_secs)
+                if result:
+                    enriched += 1
+                    for key in stats:
+                        if result.get(key):
+                            stats[key] += 1
+                else:
+                    skipped += 1
+
+            self.stdout.write(
+                f"  Batch {batch_num + 1}: {offset + len(batch)}/{total} — "
+                f"enriched: {enriched}, skipped: {skipped}"
+            )
+
+        self.stdout.write(
+            f"\nResults (commit={commit}):\n"
+            f"  Books enriched: {enriched}\n"
+            f"  Books skipped:  {skipped}\n"
+            f"  Covers fixed:   {stats['cover_fixed']}\n"
+            f"  Summaries fixed:{stats['summary_fixed']}\n"
+            f"  ISBNs fixed:    {stats['isbn_fixed']}\n"
+            f"  Pages fixed:    {stats['pages_fixed']}\n"
+            f"  Languages fixed:{stats['language_fixed']}\n"
+            f"  Publishers fixed:{stats['publisher_fixed']}\n"
+            f"  Publish yrs fixed: {stats['publish_year_fixed']}"
+        )
+
+    def _enrich_book(self, book, sleep_secs):
+        from books.sources.comicvine import lookup_comic_from_comicvine
+        from books.sources.google import lookup_book_from_google
+        from books.sources.openlibrary import lookup_book_from_openlibrary as lookup_book_from_ol
+
+        title = book.original_title or book.title
+        author_name = book.author.name if book.author else None
+        book_dict = {}
+
+        is_comic = bool(book.readcomics_url) or (
+            book.issue_number is not None or book.volume_number is not None
+        )
+        if is_comic and READCOMICSONLINE_URL in (book.readcomics_url or ""):
+            cv_data = lookup_comic_from_comicvine(title)
+            if cv_data:
+                book_dict.update(cv_data)
+
+        ol_data = lookup_book_from_ol(title, author=author_name)
+        time.sleep(sleep_secs)
+        google_data = lookup_book_from_google(title)
+
+        if ol_data:
+            for k, v in ol_data.items():
+                book_dict.setdefault(k, v)
+
+        if google_data:
+            for k, v in google_data.items():
+                if v:
+                    book_dict.setdefault(k, v)
+
+        if not book_dict:
+            return None
+
+        changed = self._apply(book, book_dict, title)
+        return changed
+
+    def _apply(self, book, data, title):
+        changed = {
+            "cover_fixed": False,
+            "summary_fixed": False,
+            "isbn_fixed": False,
+            "pages_fixed": False,
+            "language_fixed": False,
+            "publisher_fixed": False,
+            "publish_year_fixed": False,
+        }
+
+        update_fields = []
+
+        cover_url = data.pop("cover_url", "")
+
+        if data.get("summary") and not book.summary:
+            book.summary = data["summary"]
+            update_fields.append("summary")
+            changed["summary_fixed"] = True
+
+        if data.get("isbn_13") and not book.isbn_13:
+            book.isbn_13 = data["isbn_13"]
+            update_fields.append("isbn_13")
+            changed["isbn_fixed"] = True
+
+        if data.get("isbn_10") and not book.isbn_10:
+            book.isbn_10 = data["isbn_10"]
+            update_fields.append("isbn_10")
+            changed["isbn_fixed"] = True
+
+        if data.get("pages") and book.pages is None:
+            book.pages = data["pages"]
+            update_fields.append("pages")
+            changed["pages_fixed"] = True
+
+        if data.get("language") and not book.language:
+            book.language = data["language"]
+            update_fields.append("language")
+            changed["language_fixed"] = True
+
+        if data.get("publisher") and not book.publisher:
+            book.publisher = data["publisher"]
+            update_fields.append("publisher")
+            changed["publisher_fixed"] = True
+
+        if data.get("first_publish_year") and book.first_publish_year is None:
+            book.first_publish_year = data["first_publish_year"]
+            update_fields.append("first_publish_year")
+            changed["publish_year_fixed"] = True
+
+        if data.get("openlibrary_id") and not book.openlibrary_id:
+            book.openlibrary_id = data["openlibrary_id"]
+            update_fields.append("openlibrary_id")
+
+        if data.get("comicvine_id") and not book.comicvine_id:
+            book.comicvine_id = data["comicvine_id"]
+            update_fields.append("comicvine_id")
+
+        if data.get("issue_number") and book.issue_number is None:
+            book.issue_number = data["issue_number"]
+            update_fields.append("issue_number")
+
+        if data.get("volume_number") and book.volume_number is None:
+            book.volume_number = data["volume_number"]
+            update_fields.append("volume_number")
+
+        if update_fields:
+            book.save(update_fields=update_fields)
+            self.stdout.write(f"  [ENRICHED] {book} — {', '.join(update_fields)}")
+
+        if cover_url and not book.cover:
+            book.save_image_from_url(cover_url)
+            if book.cover:
+                changed["cover_fixed"] = True
+                self.stdout.write(f"  [COVER] {book} — cover saved from source")
+
+        genres = data.pop("genres", data.pop("generes", []))
+        if genres:
+            existing = set(book.genre.names())
+            new_genres = [g for g in genres if g not in existing]
+            if new_genres:
+                book.genre.add(*new_genres)
+                self.stdout.write(f"  [GENRES] {book} — added {len(new_genres)} genres")
+
+        return changed if any(changed.values()) else None

vrobbler/apps/books/models.py

@@ -229,36 +229,20 @@ class Book(LongPlayScrobblableMixin):
     ) -> "Book":
         book, created = cls.objects.get_or_create(title=title)
 
-        if not created:
+        if not created and not overwrite:
             return book
 
         book_dict = lookup_comic_from_comicvine(title)
+        if not book_dict:
+            return book
 
-        if created or overwrite:
+        for k, v in book_dict.items():
-            author_list = []
+            setattr(book, k, v)
-            author_dicts = book_dict.pop("author_dicts")
+        book.save()
-            if author_dicts:
-                for author_dict in author_dicts:
-                    if author_dict.get("authorId"):
-                        author, a_created = Author.objects.get_or_create(
-                            semantic_id=author_dict.get("authorId")
-                        )
-                        author_list.append(author)
-                        if a_created:
-                            author.name = author_dict.get("name")
-                            author.save()
-                            # TODO enrich author?
-                            ...
 
-            for k, v in book_dict.items():
+        genres = book_dict.get("genres", [])
-                setattr(book, k, v)
+        if genres:
-            book.save()
+            book.genre.add(*genres)
-
-            if author_list:
-                book.authors.add(*author_list)
-            genres = book_dict.pop("genres", [])
-            if genres:
-                book.genre.add(*genres)
         return book
 
     @classmethod
@@ -296,20 +280,27 @@ class Book(LongPlayScrobblableMixin):
             book_dict = lookup_comic_from_comicvine(title)
             if book_dict:
                 source_tag = MediaSourceTag.COMICVINE
-            book_dict["readcomics_url"] = get_comic_issue_url(url)
+                book_dict["readcomics_url"] = get_comic_issue_url(url)
-            book_dict["next_readcomics_url"] = next_url_if_exists(
+                book_dict["next_readcomics_url"] = next_url_if_exists(
-                book_dict["readcomics_url"]
+                    book_dict["readcomics_url"]
-            )
+                )
 
         if not book_dict:
-            book_dict = lookup_book_from_ol(title, author=author)
+            book_dict = {}
-            if book_dict:
+            ol_data = lookup_book_from_ol(title, author=author)
+            google_data = lookup_book_from_google(title)
+
+            if ol_data:
+                book_dict.update(ol_data)
                 source_tag = MediaSourceTag.OPENLIBRARY
 
-        if not book_dict:
+            if google_data:
-            book_dict = lookup_book_from_google(title)
+                for k, v in google_data.items():
-            if book_dict:
+                    if v:
+                        book_dict.setdefault(k, v)
                 source_tag = MediaSourceTag.GOOGLE_BOOKS
+                if ol_data and ol_data.get("cover_url"):
+                    book_dict["cover_url"] = ol_data["cover_url"]
 
         if not book_dict:
             logger.warning(
@@ -378,7 +369,6 @@ class Book(LongPlayScrobblableMixin):
 
             if not data and COMICVINE_API_KEY:
                 logger.warn(f"Checking ComicVine for {self.title}")
-                cv_client = ComicVineClient(api_key=COMICVINE_API_KEY)
                 data = lookup_comic_from_comicvine(str(self.title))
 
             if not data:

vrobbler/apps/books/sources/comicvine.py

@@ -18,7 +18,7 @@ class ComicVineClient(object):
     """
 
     # All API requests made by this client will be made to this URL.
-    API_URL = "https://www.comicvine.com/api/search/"
+    API_URL = "https://comicvine.gamespot.com/api/search/"
 
     # A valid User-Agent header must be set in order for our API requests to
     # be accepted, otherwise our request will be rejected with a
@@ -41,15 +41,12 @@ class ComicVineClient(object):
         "volume",
     }
 
-    def __init__(self, api_key, expire_after=300):
+    def __init__(self, api_key):
         """
-        Store the API key in a class variable, and install the requests cache,
+        Store the API key in a class variable.
-        configuring it using the ``expire_after`` parameter.
 
         :param api_key: Your personal ComicVine API key.
         :type  api_key: str
-        :param expire_after: The number of seconds to retain an entry in cache.
-        :type  expire_after: int or None
         """
 
         self.api_key = api_key
@@ -109,14 +106,17 @@ class ComicVineClient(object):
         :rtype:  dict
         """
 
-        return {
+        params = {
             "api_key": self.api_key,
             "format": "json",
             "limit": min(10, limit),  # hard limit of 10
             "offset": max(0, offset),  # cannot provide negative offset
             "query": query,
-            "resources": self._validate_resources(resources),
         }
+        validated = self._validate_resources(resources)
+        if validated:
+            params["resources"] = validated
+        return params
 
     def _validate_resources(self, resources):
         """
@@ -141,33 +141,35 @@ class ComicVineClient(object):
     def _query_api(self, params):
         """
         Query the ComicVine API's ``search`` resource, providing the required
-        headers and parameters with the request. Optionally allow the caller
+        headers and parameters with the request.
-        of the function to disable the request cache.
 
         If an error occurs during the request, handle it accordingly. Upon
         success, return the JSON from the response.
 
         :param params: Parameters to include with the request.
         :type  params: dict
-        :param use_cache: Toggle the use of requests_cache.
-        :type  use_cache: bool
 
         :return: The JSON contained in the response.
         :rtype:  dict
         """
 
-        # Since we're performing the identical action regardless of whether
+        response = requests.get(self.API_URL, headers=self.HEADERS, params=params)
-        # or not the request cache is to be used, store the procedure in a
-        # local function to avoid repetition.
-        def __httpget():
-            response = requests.get(self.API_URL, headers=self.HEADERS, params=params)
 
-            if not response.ok:
+        if not response.ok:
-                self._handle_http_error(response)
+            self._handle_http_error(response)
 
-            return response.json()
+        json_data = response.json()
 
-        return __httpget()
+        if json_data.get("status_code") != 1:
+            error_msg = json_data.get("error", "Unknown ComicVine API error")
+            logger.error(
+                "ComicVine API returned status_code %s: %s",
+                json_data.get("status_code"),
+                error_msg,
+            )
+            return {}
+
+        return json_data
 
     def _handle_http_error(self, response):
         """
@@ -200,10 +202,8 @@ def lookup_comic_from_comicvine(title: str) -> dict:
     original_title = title
 
     issue_number = None
-    volume_nubmer = None
     resource_type = "issue"
     if "Issue " in title:
-        resource_type = "issue"
         issue_number = title.split("Issue ")[1]
     volume_number = None
     if "Volume " in title:
@@ -215,48 +215,49 @@ def lookup_comic_from_comicvine(title: str) -> dict:
         logger.warning("No ComicVine API key configured, not looking anything up")
         return {}
 
-    client = ComicVineClient(api_key=getattr(settings, "COMICVINE_API_KEY", None))
+    client = ComicVineClient(api_key=api_key)
 
-    raw_results = client.search(title).get("results")
+    raw_results = client.search(title)
-    results = [r for r in raw_results if r.get("resource_type") == resource_type]
+    if not raw_results:
+        return {}
+    results = raw_results.get("results", [])
+    results = [r for r in results if r.get("resource_type") == resource_type]
     if not results:
         logger.warning("No comic found on ComicVine")
         return {}
 
     found_result = None
     for result in results:
-        if result.get("issue_number") == str(issue_number):
+        if issue_number is not None and result.get("issue_number") == str(issue_number):
             found_result = result
             break
-        if result.get("volume_number") == str(volume_number):
+        if volume_number is not None and result.get("volume_number") == str(volume_number):
             found_result = result
             break
 
     if not found_result:
         found_result = results[0]
 
-    logger.info("ComicVine results", extra={"results": results})
-
-    if not found_result:
-        logger.warning("No matches found on ComicVine")
-        return {}
-
     title = found_result.get("name")
 
     if found_result.get("volume"):
         title = found_result.get("volume").get("name")
 
+    cover_url = None
+    if found_result.get("image"):
+        cover_url = found_result["image"].get("original_url")
+
     data_dict = {
         "title": title,
         "original_title": original_title,
         "issue_number": found_result.get("issue_number"),
         "volume_number": found_result.get("volume_number"),
-        "cover_url": found_result.get("image").get("original_url"),
+        "cover_url": cover_url,
         "comicvine_id": found_result.get("id"),
         "comicvine_data": found_result,
         "summary": found_result.get("description"),
         "publish_date": found_result.get("cover_date"),
-        "first_publish_year": found_result.get("cover_date", "")[:4],
+        "first_publish_year": (found_result.get("cover_date") or "")[:4],
     }
 
     return data_dict

vrobbler/apps/books/sources/google.py

@@ -26,8 +26,6 @@ def lookup_book_from_google(title: str) -> dict:
     if not google_result:
         return {}
 
-    publish_date = pendulum.parse(google_result.get("publishedDate"))
-
     isbn_13 = ""
     isbn_10 = ""
     for ident in google_result.get("industryIdentifiers", []):
@@ -35,25 +33,25 @@ def lookup_book_from_google(title: str) -> dict:
             isbn_13 = ident.get("identifier")
         if ident.get("type") == "ISBN_10":
             isbn_10 = ident.get("identifier")
-    # TODO this may lead to issues with the first get if Google changes our title
-    # book_metadata.title = google_result.get("title")
-    # if google_result.get("subtitle"):
-    #    book_metadata["title"] = ": ".join(
-    #        [google_result.get("title"), google_result.get("subtitle")]
-    #    )
-    # book_dict["subtitle"] = google_result.get("subtitle")
     book_dict["authors"] = google_result.get("authors")
     book_dict["publisher"] = google_result.get("publisher")
-    book_dict["first_publish_year"] = publish_date.year
     book_dict["pages"] = google_result.get("pageCount")
     book_dict["isbn_13"] = isbn_13
     book_dict["isbn_10"] = isbn_10
-    book_dict["publish_date"] = google_result.get("publishedDate")
-    if len(book_dict["publish_date"]) == 4:
-        book_dict["publish_date"] = f"{book_dict['publish_date']}-1-1"
     book_dict["language"] = google_result.get("language")
     book_dict["summary"] = google_result.get("description")
     book_dict["genres"] = google_result.get("categories")
+
+    raw_date = google_result.get("publishedDate")
+    if raw_date:
+        try:
+            publish_date = pendulum.parse(raw_date)
+            book_dict["first_publish_year"] = publish_date.year
+        except Exception:
+            pass
+        book_dict["publish_date"] = raw_date
+        if len(raw_date) == 4:
+            book_dict["publish_date"] = f"{raw_date}-1-1"
     book_dict["cover_url"] = (
         google_result.get("imageLinks", {})
         .get("thumbnail", "")

vrobbler/apps/scrobbles/scrobblers.py

@@ -357,10 +357,7 @@ def manual_scrobble_book(
 
     if action == "stop":
         if url:
-            if isinstance(scrobble.log, "BookLogData"):
+            scrobble.log["resume_url"] = next_url_if_exists(url)
-                scrobble.log.resume_url = next_url_if_exists(url)
-            else:
-                scrobble.log["resume_url"] = next_url_if_exists(url)
             scrobble.save(update_fields=["log"])
         scrobble.stop(force_finish=True)