[mod] Open Library engine: revision of the engine (Paper result)

Revision of the engine / use of the result type Paper as well as other
typifications.

Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>

docs/dev/engines/online/openlibrary.rst

@@ -0,0 +1,8 @@
+.. _openlibrary engine:
+
+============
+Open Library
+============
+
+.. automodule:: searx.engines.openlibrary
+   :members:

searx/engines/openlibrary.py

@@ -1,71 +1,109 @@
 # SPDX-License-Identifier: AGPL-3.0-or-later
-"""Open library (books)
+"""`Open Library`_ is an open, editable library catalog, building towards a web
-"""
+page for every book ever published.
-from urllib.parse import urlencode
-import re
 
+.. _Open Library: https://openlibrary.org
+
+Configuration
+=============
+
+The service sometimes takes a very long time to respond, the ``timeout`` may
+need to be adjusted.
+
+.. code:: yaml
+
+  - name: openlibrary
+    engine: openlibrary
+    shortcut: ol
+    timeout: 10
+
+
+Implementations
+===============
+
+"""
+
+from datetime import datetime
+import typing as t
+
+from urllib.parse import urlencode
 from dateutil import parser
 
+from searx.result_types import EngineResults
+
+if t.TYPE_CHECKING:
+    from searx.extended_types import SXNG_Response
+    from searx.search.processors import OnlineParams
+
 about = {
-    'website': 'https://openlibrary.org',
+    "website": "https://openlibrary.org",
-    'wikidata_id': 'Q1201876',
+    "wikidata_id": "Q1201876",
-    'require_api_key': False,
+    "require_api_key": False,
-    'use_official_api': False,
+    "use_official_api": False,
-    'official_api_documentation': 'https://openlibrary.org/developers/api',
+    "official_api_documentation": "https://openlibrary.org/developers/api",
 }
 
 paging = True
-categories = []
+categories = ["general", "books"]
 
 base_url = "https://openlibrary.org"
+search_api = "https://openlibrary.org/search.json"
+"""The engine uses the API at the endpoint search.json_.
+
+.. _search.json: https://openlibrary.org/dev/docs/api/search
+"""
 results_per_page = 10
 
 
-def request(query, params):
+def request(query: str, params: "OnlineParams") -> None:
     args = {
-        'q': query,
+        "q": query,
-        'page': params['pageno'],
+        "page": params["pageno"],
-        'limit': results_per_page,
+        "limit": results_per_page,
+        "fields": "*",
     }
-    params['url'] = f"{base_url}/search.json?{urlencode(args)}"
+    params["url"] = f"{search_api}?{urlencode(args)}"
-    return params
+    logger.debug("REST API: %s", params["url"])
 
 
-def _parse_date(date):
+def response(resp: "SXNG_Response") -> EngineResults:
-    try:
+    res = EngineResults()
-        return parser.parse(date)
+    json_data = resp.json()
-    except parser.ParserError:
-        return None
 
-
+    for item in json_data.get("docs", []):
-def response(resp):
+        cover = ""
-    results = []
+        if "lending_identifier_s" in item:
-
-    for item in resp.json().get("docs", []):
-        cover = None
-        if 'lending_identifier_s' in item:
             cover = f"https://archive.org/services/img/{item['lending_identifier_s']}"
 
-        published = item.get('publish_date')
+        published = item.get("publish_date")
         if published:
             published_dates = [date for date in map(_parse_date, published) if date]
             if published_dates:
                 published = min(published_dates)
 
         if not published:
-            published = parser.parse(str(item.get('first_published_year')))
+            published = _parse_date(str(item.get("first_publish_year")))
 
-        result = {
+        content = " / ".join(item.get("first_sentence", []))
-            'template': 'paper.html',
+        res.add(
-            'url': f"{base_url}{item['key']}",
+            res.types.Paper(
-            'title': item['title'],
+                url=f"{base_url}/{item['key']}",
-            'content': re.sub(r"\{|\}", "", item['first_sentence'][0]) if item.get('first_sentence') else '',
+                title=item["title"],
-            'isbn': item.get('isbn', [])[:5],
+                content=content,
-            'authors': item.get('author_name', []),
+                isbn=item.get("isbn", [])[:5],
-            'thumbnail': cover,
+                authors=item.get("author_name", []),
-            'publishedDate': published,
+                thumbnail=cover,
-            'tags': item.get('subject', [])[:10] + item.get('place', [])[:10],
+                publishedDate=published,
-        }
+                tags=item.get("subject", [])[:10] + item.get("place", [])[:10],
-        results.append(result)
+            )
+        )
+    return res
 
-    return results
+
+def _parse_date(date: str) -> datetime | None:
+    if not date:
+        return None
+    try:
+        return parser.parse(date)
+    except parser.ParserError:
+        return None

searx/settings.yml

@@ -1517,7 +1517,7 @@ engines:
   - name: openlibrary
     engine: openlibrary
     shortcut: ol
-    timeout: 5
+    timeout: 10
     disabled: true
 
   - name: openmeteo