feat(thumbnails): add cover image support with rero-invenio-thumbnails

* Add rero-invenio-thumbnails dependency for book cover retrieval
* Configure RERO_INVENIO_THUMBNAILS_FILES_DIR in config.py
* Fix Flask endpoint name: api_thumbnails.get_thumbnail
* Fix URL parameter bug: weight → width

API Endpoints:
* Add /cover/<isbn> API endpoint for cover image retrieval
  - Supports ISBN-10 and ISBN-13 formats
  - Accepts cached, width, and height query parameters
  - Returns JSON with thumbnail URL and metadata

Template Filters:
* Add get_cover_art template filter for Jinja2 templates
  - Supports electronicLocator coverImage URLs
  - Falls back to ISBN-based thumbnail lookup
  - Handles multiple ISBNs in sorted order

CLI Commands:
* Add add_cover_urls command to bulk process documents
  - Supports --commit flag for database persistence
  - Supports --cached/--no-cached for thumbnail caching
  - Supports --scroll for Elasticsearch timeout configuration
  - Supports --pids option for processing specific documents
  - Excludes documents already having cover images
  - Sorts PIDs numerically for consistent processing

Document Extensions:
* Add AddCoverUrlExtension for automatic cover URL injection
  - Automatically adds cover URLs on document create/commit
  - Checks for existing cover images to avoid duplicates
  - Configurable caching support

Tasks:
* Add add_cover_urls shared task for batch processing
  - Used by CLI command and can be called programmatically
  - Supports filtering by PIDs or processing all documents
  - Updates timestamp tracking for monitoring

Co-Authored-by: Peter Weber <peter.weber@rero.ch>

Author: rerowep

data/thumbnails/9782607000423.jpg

@@ -52,6 +52,8 @@ dependencies = [
     ## RERO specific python modules
     "rero-invenio-base (>=0.3.0)",
     "rero-invenio-files (>=1.0.0,<2.0.0)",
+    # TODO: remove the git dependency when the module will be released on PyPI
+    "rero-invenio-thumbnails @ git+https://github.com/rero/rero-invenio-thumbnails.git",
     "flask-wiki (>=1.0.0)",
     ## RERO ILS specific python modules
     "PyYAML (>=5.3.1)",

pyproject.toml

@@ -3499,6 +3499,9 @@ def _(x):
 #: page is displayed on RERO-ILS frontpage.
 RERO_ILS_UI_GIT_HASH = None
 
+#: Thumbnails directory configuration
+RERO_INVENIO_THUMBNAILS_FILES_DIR = "./data/thumbnails"
+
 #: RERO_ILS MEF base url could be changed.
 RERO_ILS_MEF_REF_BASE_URL = os.environ.get("RERO_ILS_MEF_REF_BASE_URL", "mef.rero.ch")
 #: RERO_ILS MEF specific configurations.
@@ -3536,8 +3539,6 @@ def _(x):
 
 RERO_ILS_HELP_PAGE = "https://github.com/rero/rero-ils/wiki/Public-demo-help"
 
-#: Cover service
-RERO_ILS_THUMBNAIL_SERVICE_URL = "https://services.test.rero.ch/cover"
 
 #: Entities
 RERO_ILS_AGENTS_SOURCES = ["idref", "gnd", "rero"]

rero_ils/config.py

@@ -38,7 +38,6 @@
 import yaml
 from celery import current_app as current_celery
 from dojson.contrib.marc21.utils import create_record
-from elasticsearch_dsl.query import Q
 from flask import current_app
 from flask.cli import with_appcontext
 from invenio_db import db
@@ -55,9 +54,9 @@
 from werkzeug.local import LocalProxy
 from werkzeug.security import gen_salt
 
-from rero_ils.modules.documents.api import Document, DocumentsSearch
+from rero_ils.modules.documents.api import Document
 from rero_ils.modules.documents.dojson.contrib.marc21tojson.rero import marc21
-from rero_ils.modules.documents.views import get_cover_art
+from rero_ils.modules.documents.tasks import add_cover_urls as task_add_cover_urls
 from rero_ils.modules.entities.remote_entities.api import RemoteEntity
 from rero_ils.modules.files.cli import load_files
 from rero_ils.modules.items.api import Item
@@ -1579,26 +1578,25 @@ def token_create(name, user, scopes, internal, access_token):
         click.secho("No user found", fg="red")
 
 
-@utils.command("add_cover_urls")
-@click.option("-v", "--verbose", "verbose", is_flag=True, default=False)
+@utils.command()
+@click.option("-c", "--commit", "commit", is_flag=True, default=False, help="Commit changes to database.")
+@click.option("-v", "--verbose", "verbose", is_flag=True, default=False, help="Verbose print.")
+@click.option("--cached/--no-cached", "cached", default=True, help="Use cached thumbnails.")
+@click.option("-s", "--scroll", "scroll", default="60m", help="Elasticsearch scroll timeout.")
+@click.option(
+    "-p", "--pids", "pids_file", type=click.File("r"), default=None, help="File with document pids to process."
+)
 @with_appcontext
-def add_cover_urls(verbose):
+def add_cover_urls(commit, verbose, cached, scroll, pids_file):
     """Add cover urls to all documents with isbns."""
-    click.secho("Add cover urls.", fg="green")
-    search = (
-        DocumentsSearch()
-        .filter("term", identifiedBy__type="bf:Isbn")
-        .filter("bool", must_not=[Q("term", electronicLocator__content="coverImage")])
-        .params(preserve_order=True)
-        .sort({"pid": {"order": "asc"}})
-        .source("pid")
-    )
-    for idx, hit in enumerate(search.scan()):
-        pid = hit.pid
-        record = Document.get_record_by_pid(pid)
-        url = get_cover_art(record=record, save_cover_url=True)
+
+    pids = None
+    if pids_file:
+        pids = [line.strip() for line in pids_file if line.strip()]
         if verbose:
-            click.echo(f"{idx}:\tdocument: {pid}\t{url}")
+            click.secho(f"Processing {len(pids)} documents from file", fg="green")
+
+    return task_add_cover_urls(commit=commit, verbose=verbose, cached=cached, scroll=scroll, pids=pids)
 
 
 @utils.command()

rero_ils/modules/cli/utils.py

@@ -42,6 +42,7 @@
 
 from .dumpers import document_indexer_dumper, document_replace_refs_dumper
 from .extensions import (
+    AddCoverUrlExtension,
     AddMEFPidExtension,
     EditionStatementExtension,
     ProvisionActivitiesExtension,
@@ -115,13 +116,14 @@ class Document(IlsRecord):
     enable_jsonref = False
 
     _extensions = [
-        OperationLogObserverExtension(),
+        AddCoverUrlExtension(),
         AddMEFPidExtension("subjects", "contribution", "genreForm"),
+        DeleteRelatedLocalFieldExtension(),
+        EditionStatementExtension(),
+        OperationLogObserverExtension(),
         ProvisionActivitiesExtension(),
         SeriesStatementExtension(),
-        EditionStatementExtension(),
         TitleExtension(),
-        DeleteRelatedLocalFieldExtension(),
     ]
 
     def _validate(self, **kwargs):
@@ -490,17 +492,40 @@ def document_types(self):
             document_types.append(main_type)
         return document_types or ["docmaintype_other"]
 
-    def add_cover_url(self, url, dbcommit=False, reindex=False):
-        """Adds electronicLocator with coverImage to document."""
-        electronic_locators = self.get("electronicLocator", [])
-        for electronic_locator in electronic_locators:
-            e_content = electronic_locator.get("content")
-            e_type = electronic_locator.get("type")
-            if e_content == "coverImage" and e_type == "relatedResource" and electronic_locator.get("url") == url:
+    def add_cover_url(self, url, provider=None, dbcommit=False, reindex=False, force=False):
+        """Add or replace a coverImage electronicLocator on the document.
+
+        - If a ``coverImage``/``relatedResource`` locator already exists and
+          ``force=False``: no-op, returns ``(self, False)``.
+        - If a ``coverImage``/``relatedResource`` locator already exists and
+          ``force=True``: replaces its URL and provider note in place.
+        - If no ``coverImage``/``relatedResource`` locator exists: appends one.
+
+        :param url: str - the cover image URL to store.
+        :param provider: str - optional provider/attribution note.
+        :param dbcommit: bool - commit the change to the database.
+        :param reindex: bool - reindex the document after update.
+        :param force: bool - overwrite an existing coverImage locator.
+        :returns: tuple (updated_record, changed) where changed is True if the
+            document was modified.
+        """
+        locators = self.get("electronicLocator", [])
+        if existing := next(
+            (loc for loc in locators if loc.get("content") == "coverImage" and loc.get("type") == "relatedResource"),
+            None,
+        ):
+            if not force:
                 return self, False
-        electronic_locators.append({"content": "coverImage", "type": "relatedResource", "url": url})
-        self["electronicLocator"] = electronic_locators
-        self = self.update(data=self, commit=True, dbcommit=dbcommit, reindex=reindex)
+            existing["url"] = url
+            if provider:
+                existing["note"] = provider
+        else:
+            locator = {"content": "coverImage", "type": "relatedResource", "url": url}
+            if provider:
+                locator["note"] = provider
+            locators.append(locator)
+        self["electronicLocator"] = locators
+        self.update(data=self, commit=True, dbcommit=dbcommit, reindex=reindex)
         return self, True
 
     def resolve(self):

rero_ils/modules/documents/api.py

@@ -24,21 +24,29 @@
 from flask import request as flask_request
 from invenio_jsonschemas import current_jsonschemas
 from invenio_jsonschemas.errors import JSONSchemaNotFound
+from rero_invenio_thumbnails import get_thumbnail_url
 
 from rero_ils.modules.decorators import check_logged_as_librarian
 
 from ..utils import cached
 from .api import Document
-from .utils import get_remote_cover
 
 api_blueprint = Blueprint("api_documents", __name__, url_prefix="/document")
 
 
 @api_blueprint.route("/cover/<isbn>")
-@cached(timeout=5 * 60, query_string=True)  # 5 minutes timeout
 def cover(isbn):
     """Document cover service."""
-    return jsonify(get_remote_cover(isbn))
+
+    use_cache = flask_request.args.get("cached", default="true").lower() != "false"
+    try:
+        url, provider = get_thumbnail_url(isbn, cached=use_cache)
+    except Exception as e:
+        current_app.logger.warning(f"Thumbnail lookup failed for ISBN {isbn}: {e}")
+        return jsonify({"success": False, "isbn": isbn})
+    if url:
+        return jsonify({"success": True, "image": url, "isbn": isbn, "provider": provider})
+    return jsonify({"success": False, "isbn": isbn})
 
 
 @api_blueprint.route("/<pid>/availability", methods=["GET"])

rero_ils/modules/documents/api_views.py

@@ -128,7 +128,7 @@ def json_to_descriptions(self, key, value):
         if key == "supplementaryContent":
             descriptions.append(data)
         elif key == "summary":
-            descriptions += [label["value"] for label in data.get("label", [])]
+            descriptions += [v for v in (label.get("value") for label in data.get("label", [])) if v is not None]
         elif label := data.get("label"):
             descriptions.append(label)
     if descriptions:
@@ -178,6 +178,8 @@ def json_to_dates(self, key, value):
 def json_to_types(self, key, value):
     """Get types data."""
     main_type = value.get("main_type")
+    if not main_type:
+        return None
     if subtype_type := value.get("subtype"):
         return " / ".join([_(main_type), _(subtype_type)])
     return _(main_type)

rero_ils/modules/documents/dojson/contrib/jsontodc/model.py

@@ -131,6 +131,8 @@ def do_contribution(contribution, source_order):
             return None, None, False, False
     if not (preferred_name := entity.get("preferred_name")):
         preferred_name = entity.get(f"authorized_access_point_{to_marc21.language}")
+    if not preferred_name:
+        return None, None, False, False
     result = {}
     conference = False
     surname = False
@@ -270,10 +272,10 @@ def get_name(resource, pid):
                 item_hits = ItemsSearch().filter("terms", pid=list(item_pids)).scan()
                 for item_hit in item_hits:
                     item_data = item_hit.to_dict()
-                    item_result = result
+                    item_result = dict(result)
                     item_result["item"] = {
                         "barcode": item_data.get("barcode"),
-                        "all_number": item_data.get("all_number"),
+                        "call_number": item_data.get("call_number"),
                         "second_call_number": item_data.get("second_call_number"),
                         "enumerationAndChronology": item_data.get("enumerationAndChronology"),
                         "url": item_data.get("url"),

rero_ils/modules/documents/dojson/contrib/jsontomarc21/model.py

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 # RERO ILS
-# Copyright (C) 2019-2022 RERO
+# Copyright (C) 2019-2026 RERO
 # Copyright (C) 2019-2022 UCLOUVAIN
 #
 # This program is free software: you can redistribute it and/or modify

rero_ils/modules/documents/dojson/contrib/marc21tojson/loc/model.py

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 #
 # RERO ILS
-# Copyright (C) 2019-2022 RERO
+# Copyright (C) 2019-2026 RERO
 # Copyright (C) 2019-2022 UCLOUVAIN
 #
 # This program is free software: you can redistribute it and/or modify