[mod] CORE 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>

searx/engines/core.py

@@ -5,6 +5,10 @@ research from repositories and journals.
 
 .. _CORE: https://core.ac.uk/about
 
+.. note::
+
+   The CORE engine requires an :py:obj:`API key <api_key>`.
+
 .. _core engine config:
 
 Configuration
@@ -17,135 +21,141 @@ The engine has the following additional settings:
 .. code:: yaml
 
   - name: core.ac.uk
-    engine: core
-    categories: science
-    shortcut: cor
     api_key: "..."
-    timeout: 5
+    inactive: false
 
 Implementations
 ===============
 
 """
-# pylint: disable=too-many-branches
+
+import typing as t
 
 from datetime import datetime
 from urllib.parse import urlencode
 
-from searx.exceptions import SearxEngineAPIException
+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://core.ac.uk',
+    "website": "https://core.ac.uk",
-    "wikidata_id": 'Q22661180',
+    "wikidata_id": "Q22661180",
-    "official_api_documentation": 'https://api.core.ac.uk/docs/v3',
+    "official_api_documentation": "https://api.core.ac.uk/docs/v3",
     "use_official_api": True,
     "require_api_key": True,
-    "results": 'JSON',
+    "results": "JSON",
 }
 
-api_key = 'unset'
+api_key = ""
 """For an API key register at https://core.ac.uk/services/api and insert
 the API key in the engine :ref:`core engine config`."""
 
-categories = ['science', 'scientific publications']
+categories = ["science", "scientific publications"]
 paging = True
 nb_per_page = 10
-base_url = 'https://api.core.ac.uk/v3/search/works/'
+base_url = "https://api.core.ac.uk/v3/search/works/"
 
 
-def request(query, params):
+def setup(engine_settings: dict[str, t.Any]) -> bool:
-    if api_key == 'unset':
+    """Initialization of the CORE_ engine, checks whether the :py:obj:`api_key`
-        raise SearxEngineAPIException('missing CORE API key')
+    is set, otherwise the engine is inactive.
+    """
+
+    key: str = engine_settings.get("api_key", "")
+    if key and key not in ("unset", "unknown", "..."):
+        return True
+    logger.error("CORE's API key is not set or invalid.")
+    return False
+
+
+def request(query: str, params: "OnlineParams") -> None:
 
     # API v3 uses different parameters
     search_params = {
-        'q': query,
+        "q": query,
-        'offset': (params['pageno'] - 1) * nb_per_page,
+        "offset": (params["pageno"] - 1) * nb_per_page,
-        'limit': nb_per_page,
+        "limit": nb_per_page,
-        'sort': 'relevance',
+        "sort": "relevance",
     }
 
-    params['url'] = base_url + '?' + urlencode(search_params)
+    params["url"] = base_url + "?" + urlencode(search_params)
-    params['headers'] = {'Authorization': f'Bearer {api_key}'}
+    params["headers"] = {"Authorization": f"Bearer {api_key}"}
-
-    return params
 
 
-def response(resp):
+def response(resp: "SXNG_Response") -> EngineResults:
-    results = []
+    # pylint: disable=too-many-branches
+    res = EngineResults()
     json_data = resp.json()
 
-    for result in json_data.get('results', []):
+    for result in json_data.get("results", []):
         # Get title
-        if not result.get('title'):
+        if not result.get("title"):
             continue
 
         # Get URL - try different options
-        url = None
+        url: str | None = None
 
         # Try DOI first
-        doi = result.get('doi')
+        doi: str = result.get("doi")
         if doi:
-            url = f'https://doi.org/{doi}'
+            url = f"https://doi.org/{doi}"
 
-        if url is None and result.get('doi'):
+        if url is None and result.get("doi"):
             # use the DOI reference
-            url = 'https://doi.org/' + str(result['doi'])
+            url = "https://doi.org/" + str(result["doi"])
-        elif result.get('id'):
+        elif result.get("id"):
-            url = 'https://core.ac.uk/works/' + str(result['id'])
+            url = "https://core.ac.uk/works/" + str(result["id"])
-        elif result.get('downloadUrl'):
+        elif result.get("downloadUrl"):
-            url = result['downloadUrl']
+            url = result["downloadUrl"]
-        elif result.get('sourceFulltextUrls'):
+        elif result.get("sourceFulltextUrls"):
-            url = result['sourceFulltextUrls']
+            url = result["sourceFulltextUrls"]
         else:
             continue
 
         # Published date
         published_date = None
 
-        raw_date = result.get('publishedDate') or result.get('depositedDate')
+        raw_date = result.get("publishedDate") or result.get("depositedDate")
         if raw_date:
             try:
-                published_date = datetime.fromisoformat(result['publishedDate'].replace('Z', '+00:00'))
+                published_date = datetime.fromisoformat(result["publishedDate"].replace("Z", "+00:00"))
             except (ValueError, AttributeError):
                 pass
 
         # Handle journals
         journals = []
-        if result.get('journals'):
+        if result.get("journals"):
-            journals = [j.get('title') for j in result['journals'] if j.get('title')]
+            journals = [j.get("title") for j in result["journals"] if j.get("title")]
 
         # Handle publisher
-        publisher = result.get('publisher', '').strip("'")
+        publisher = result.get("publisher", "").strip("'")
-        if publisher:
-            publisher = publisher.strip("'")
 
         # Handle authors
-        authors = set()
+        authors: set[str] = set()
-        for i in result.get('authors', []):
+        for i in result.get("authors", []):
-            name = i.get("name")
+            name: str | None = i.get("name")
             if name:
                 authors.add(name)
 
-        results.append(
+        res.add(
-            {
+            res.types.Paper(
-                'template': 'paper.html',
+                title=result.get("title"),
-                'title': result.get('title'),
+                url=url,
-                'url': url,
+                content=result.get("fullText", "") or "",
-                'content': result.get('fullText', '') or '',
+                tags=result.get("fieldOfStudy", []),
-                # 'comments': '',
+                publishedDate=published_date,
-                'tags': result.get('fieldOfStudy', []),
+                type=result.get("documentType", "") or "",
-                'publishedDate': published_date,
+                authors=authors,
-                'type': result.get('documentType', '') or '',
+                editor=", ".join(result.get("contributors", [])),
-                'authors': authors,
+                publisher=publisher,
-                'editor': ', '.join(result.get('contributors', [])),
+                journal=", ".join(journals),
-                'publisher': publisher,
+                doi=result.get("doi"),
-                'journal': ', '.join(journals),
+                pdf_url=result.get("downloadUrl", {}) or result.get("sourceFulltextUrls", {}),
-                'doi': result.get('doi'),
+            )
-                # 'issn' : ''
-                # 'isbn' : ''
-                'pdf_url': result.get('downloadUrl', {}) or result.get("sourceFulltextUrls", {}),
-            }
         )
 
-    return results
+    return res

searx/settings.yml

@@ -659,12 +659,12 @@ engines:
     timeout: 30
     disabled: true
 
-  # - name: core.ac.uk
+  - name: core.ac.uk
-  #   engine: core
+    engine: core
-  #   categories: science
+    shortcut: cor
-  #   shortcut: cor
+    # read https://docs.searxng.org/dev/engines/online/core.html
-  #   # get your API key from: https://core.ac.uk/api-keys/register/
+    api_key: ""
-  #   api_key: 'unset'
+    inactive: true
 
   - name: cppreference
     engine: cppreference