[data] update searx.data - update_engine_descriptions.py

Bumps [docker/scout-action](https://github.com/docker/scout-action) from 1.21.0 to 1.22.0.
- [Release notes](https://github.com/docker/scout-action/releases)
- [Commits](cd72f264be...7520205ff6)

---
updated-dependencies:
- dependency-name: docker/scout-action
  dependency-version: 1.22.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

.github/workflows/container.yml

@@ -73,7 +73,7 @@ jobs:
       # yamllint enable rule:line-length
 
       - name: Setup Python
-        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405  # v6.2.0
+        uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1  # v6.3.0
         with:
           python-version: "${{ env.PYTHON_VERSION }}"
 
@@ -84,7 +84,7 @@ jobs:
           fetch-depth: "0"
 
       - name: Setup cache Python
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
           restore-keys: |
@@ -96,7 +96,7 @@ jobs:
         run: echo "date=$(date +'%Y%m%d')" >>$GITHUB_OUTPUT
 
       - name: Setup cache container
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "container-${{ matrix.arch }}-${{ steps.date.outputs.date }}-${{ hashFiles('./requirements*.txt') }}"
           restore-keys: |

.github/workflows/data-update.yml

@@ -41,7 +41,7 @@ jobs:
 
     steps:
       - name: Setup Python
-        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405  # v6.2.0
+        uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1  # v6.3.0
         with:
           python-version: "${{ env.PYTHON_VERSION }}"
 
@@ -51,7 +51,7 @@ jobs:
           persist-credentials: "false"
 
       - name: Setup cache Python
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
           restore-keys: |

.github/workflows/documentation.yml

@@ -32,7 +32,7 @@ jobs:
 
     steps:
       - name: Setup Python
-        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405  # v6.2.0
+        uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1  # v6.3.0
         with:
           python-version: "${{ env.PYTHON_VERSION }}"
 
@@ -43,7 +43,7 @@ jobs:
           fetch-depth: "0"
 
       - name: Setup cache Python
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
           restore-keys: |

.github/workflows/integration.yml

@@ -34,7 +34,7 @@ jobs:
 
     steps:
       - name: Setup Python
-        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405  # v6.2.0
+        uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1  # v6.3.0
         with:
           python-version: "${{ matrix.python-version }}"
 
@@ -44,7 +44,7 @@ jobs:
           persist-credentials: "false"
 
       - name: Setup cache Python
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "python-${{ matrix.python-version }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
           restore-keys: |
@@ -62,7 +62,7 @@ jobs:
     runs-on: ubuntu-24.04-arm
     steps:
       - name: Setup Python
-        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405  # v6.2.0
+        uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1  # v6.3.0
         with:
           python-version: "${{ env.PYTHON_VERSION }}"
 
@@ -77,13 +77,13 @@ jobs:
           node-version-file: "./.nvmrc"
 
       - name: Setup cache Node.js
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "nodejs-${{ runner.arch }}-${{ hashFiles('./.nvmrc', './package.json') }}"
           path: "./client/simple/node_modules/"
 
       - name: Setup cache Python
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
           restore-keys: |

.github/workflows/l10n.yml

@@ -35,7 +35,7 @@ jobs:
 
     steps:
       - name: Setup Python
-        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405  # v6.2.0
+        uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1  # v6.3.0
         with:
           python-version: "${{ env.PYTHON_VERSION }}"
 
@@ -46,7 +46,7 @@ jobs:
           fetch-depth: "0"
 
       - name: Setup cache Python
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
           restore-keys: |
@@ -83,7 +83,7 @@ jobs:
 
     steps:
       - name: Setup Python
-        uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405  # v6.2.0
+        uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1  # v6.3.0
         with:
           python-version: "${{ env.PYTHON_VERSION }}"
 
@@ -94,7 +94,7 @@ jobs:
           fetch-depth: "0"
 
       - name: Setup cache Python
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae  # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8  # v6.0.0
         with:
           key: "python-${{ env.PYTHON_VERSION }}-${{ runner.arch }}-${{ hashFiles('./requirements*.txt') }}"
           restore-keys: |

.github/workflows/security.yml

@@ -29,7 +29,7 @@ jobs:
           persist-credentials: "false"
 
       - name: Sync GHCS from Docker Scout
-        uses: docker/scout-action@cd72f264beff1cd72735de31148b9d3244a0234a  # v1.21.0
+        uses: docker/scout-action@7520205ff60037fdc436b40b6a1d1e55a839ec2d  # v1.22.0
         with:
           organization: "searxng"
           dockerhub-user: "${{ secrets.DOCKER_USER }}"

searx/enginelib/__init__.py

@@ -207,6 +207,7 @@ class EngineAbout(msgspec.Struct, kw_only=True):
 
     use_official_api: bool = False
     """SearXNG engine makes use of the official API or not"""
+
     require_api_key: bool = False
     """API requires a key or not."""
 
@@ -217,8 +218,7 @@ class EngineAbout(msgspec.Struct, kw_only=True):
     """Brief description of the engine and where it gets its data from.
 
     This value should only be set as long as no description of the data source
-    is available via a :py:obj:`EngineAbout.wikidata_id`.
+    is available via a :py:obj:`EngineAbout.wikidata_id`."""
-    """
 
     language: str = ""
     """Deprecated! Migrate your setting from `engine.about.language` to

searx/engines/chinaso.py

@@ -19,11 +19,9 @@ Configuration
 
 The engine has the following additional settings:
 
-- :py:obj:`chinaso_category` (:py:obj:`ChinasoCategoryType`)
 - :py:obj:`chinaso_news_source` (:py:obj:`ChinasoNewsSourceType`)
 
-In the example below, all three ChinaSO engines are using the :ref:`network
+In the example below, ChinaSO is configured for news search.
-<engine network>` from the ``chinaso news`` engine.
 
 .. code:: yaml
 
@@ -31,23 +29,8 @@ In the example below, all three ChinaSO engines are using the :ref:`network
      engine: chinaso
      shortcut: chinaso
      categories: [news]
-     chinaso_category: news
      chinaso_news_source: all
 
-   - name: chinaso images
-     engine: chinaso
-     network: chinaso news
-     shortcut: chinasoi
-     categories: [images]
-     chinaso_category: images
-
-   - name: chinaso videos
-     engine: chinaso
-     network: chinaso news
-     shortcut: chinasov
-     categories: [videos]
-     chinaso_category: videos
-
 
 Implementations
 ===============
@@ -78,19 +61,6 @@ results_per_page = 10
 categories = []
 language = "zh"
 
-ChinasoCategoryType = t.Literal['news', 'videos', 'images']
-"""ChinaSo supports news, videos, images search.
-
-- ``news``: search for news
-- ``videos``: search for videos
-- ``images``: search for images
-
-In the category ``news`` you can additionally filter by option
-:py:obj:`chinaso_news_source`.
-"""
-chinaso_category = 'news'
-"""Configure ChinaSo category (:py:obj:`ChinasoCategoryType`)."""
-
 ChinasoNewsSourceType = t.Literal['CENTRAL', 'LOCAL', 'BUSINESS', 'EPAPER', 'all']
 """Filtering ChinaSo-News results by source:
 
@@ -109,39 +79,24 @@ base_url = "https://www.chinaso.com"
 
 
 def init(_):
-    if chinaso_category not in ('news', 'videos', 'images'):
+    if chinaso_news_source not in t.get_args(ChinasoNewsSourceType):
-        raise ValueError(f"Unsupported category: {chinaso_category}")
-    if chinaso_category == 'news' and chinaso_news_source not in t.get_args(ChinasoNewsSourceType):
         raise ValueError(f"Unsupported news source: {chinaso_news_source}")
 
 
 def request(query, params):
-    query_params = {"q": query}
+    query_params = {'q': query, 'pn': params["pageno"], 'ps': results_per_page}
 
     if time_range_dict.get(params['time_range']):
         query_params["stime"] = time_range_dict[params['time_range']]
         query_params["etime"] = 'now'
 
-    category_config = {
-        'news': {'endpoint': '/v5/general/v1/web/search', 'params': {'pn': params["pageno"], 'ps': results_per_page}},
-        'images': {
-            'endpoint': '/v5/general/v1/search/image',
-            'params': {'start_index': (params["pageno"] - 1) * results_per_page, 'rn': results_per_page},
-        },
-        'videos': {
-            'endpoint': '/v5/general/v1/search/video',
-            'params': {'start_index': (params["pageno"] - 1) * results_per_page, 'rn': results_per_page},
-        },
-    }
     if chinaso_news_source != 'all':
         if chinaso_news_source == 'EPAPER':
-            category_config['news']['params']["type"] = 'EPAPER'
+            query_params["type"] = 'EPAPER'
         else:
-            category_config['news']['params']["cate"] = chinaso_news_source
+            query_params["cate"] = chinaso_news_source
 
-    query_params.update(category_config[chinaso_category]['params'])
+    params["url"] = f"{base_url}/v5/general/v1/web/search?{urlencode(query_params)}"
-
-    params["url"] = f"{base_url}{category_config[chinaso_category]['endpoint']}?{urlencode(query_params)}"
     cookie = {
         "uid": base64.b64encode(secrets.token_bytes(16)).decode("utf-8"),
     }
@@ -163,12 +118,6 @@ def response(resp):
     if not data["data"]:
         return []
 
-    parsers = {'news': parse_news, 'images': parse_images, 'videos': parse_videos}
-
-    return parsers[chinaso_category](data)
-
-
-def parse_news(data):
     results = []
     if not data.get("data", {}).get("data"):
         raise SearxEngineAPIException("Invalid response")
@@ -190,47 +139,3 @@ def parse_news(data):
             }
         )
     return results
-
-
-def parse_images(data):
-    results = []
-    if not data.get("data", {}).get("arrRes"):
-        raise SearxEngineAPIException("Invalid response")
-
-    for entry in data["data"]["arrRes"]:
-        results.append(
-            {
-                'url': entry["web_url"],
-                'title': html_to_text(entry["title"]),
-                'content': html_to_text(entry.get("ImageInfo", "")),
-                'template': 'images.html',
-                'img_src': entry["url"].replace("http://", "https://"),
-                'thumbnail_src': entry["largeimage"].replace("http://", "https://"),
-            }
-        )
-    return results
-
-
-def parse_videos(data):
-    results = []
-    if not data.get("data", {}).get("arrRes"):
-        raise SearxEngineAPIException("Invalid response")
-
-    for entry in data["data"]["arrRes"]:
-        published_date = None
-        if entry.get("VideoPubDate"):
-            try:
-                published_date = datetime.fromtimestamp(int(entry["VideoPubDate"]))
-            except (ValueError, TypeError):
-                pass
-
-        results.append(
-            {
-                'url': entry["url"],
-                'title': html_to_text(entry["raw_title"]),
-                'template': 'videos.html',
-                'publishedDate': published_date,
-                'thumbnail': entry["image_src"].replace("http://", "https://"),
-            }
-        )
-    return results

searx/engines/resulthunter.py

@@ -101,7 +101,7 @@ def _image_results(doc: "ElementBase") -> EngineResults:
                 res.types.Image(
                     url=extract_text(eval_xpath(result, "./@href")) or "",
                     title=extract_text(eval_xpath(result, "./img/@alt")) or "",
-                    thumbnail_src=extract_text(eval_xpath(result, "./img/@src")) or "",
+                    img_src=extract_text(eval_xpath(result, "./img/@src")) or "",
                 ),
             )
         )

searx/engines/tiger.py

@@ -73,7 +73,7 @@ def _obtain_session_code() -> str:
     if cached_session:
         return cached_session
 
-    results_page = get(f"{base_url}/_internCode.aspx")
+    results_page = get(f"{base_url}/checkCode.aspx")
     doc = html.fromstring(results_page.text)
 
     extra_data: dict[str, str] = {}
@@ -107,7 +107,7 @@ def _obtain_session_code() -> str:
     }
 
     challenge_response = post(
-        f"{base_url}/_internCode.aspx",
+        f"{base_url}/checkCode.aspx",
         cookies=results_page.cookies,
         data=data,
     )
@@ -125,7 +125,9 @@ def request(query: str, params: "OnlineParams"):
     code = _obtain_session_code()
     args = {"w": query, "page": params["pageno"]}
     params["url"] = f"{base_url}/{tiger_category}?{urlencode(args)}"
-    params["cookies"]["Tiger.ch"] = f"Code={code}"
+    # Setting Checked=1 shows related search terms / suggestions
+    # Language and country could be set with Lng= and Land= in the future
+    params["cookies"]["Tiger.ch"] = f"Tiger.ch=&Code={code}&Checked=1"
 
 
 def response(resp: "SXNG_Response") -> EngineResults:
@@ -144,6 +146,9 @@ def response(resp: "SXNG_Response") -> EngineResults:
                     content=extract_text(eval_xpath(result, ".//*[contains(@class, 'webbodynopic')]")) or "",
                 )
             )
+
+        for suggestion in eval_xpath_list(doc, "//a[contains(@class, 'linkAnders')]"):
+            res.add(res.types.LegacyResult(suggestion=extract_text(suggestion)))
     elif tiger_category == "News":
         for result in eval_xpath_list(doc, "//div[@id='panNews']/div"):
             publishedDate = None

searx/result_types/image.py

@@ -14,6 +14,7 @@ template.
 # pylint: disable=too-few-public-methods
 __all__ = ["Image", "ImageRef"]
 
+import mimetypes
 import types
 import typing as t
 from collections.abc import Callable
@@ -71,7 +72,7 @@ class Image(MainResult, kw_only=True):
     """The resolution of the image (e.g. ``1920 x 1080`` pixel)"""
 
     img_format: str = ""
-    """The format of the image (e.g. ``png``)."""
+    """The format of the image :py:obj:`.MainResult.img_src` (e.g. ``png``)."""
 
     source: str = ""
     """Source of the image."""
@@ -83,6 +84,19 @@ class Image(MainResult, kw_only=True):
     formats: list[ImageRef] = []
     """List of links to alternative image formats."""
 
+    def __post_init__(self):
+        super().__post_init__()
+
+        if not self.img_format:
+            # automatically guess the image format based on the path of the image
+            mimetype = mimetypes.guess_type(self.img_src)[0]
+            if mimetype:
+                subtype = mimetype.split("/")[-1]
+                if subtype in MIMESUB:
+                    self.img_format = MIMESUB[subtype]
+                else:
+                    self.img_format = subtype.upper()
+
     def filter_urls(self, filter_func: "Callable[[Result | LegacyResult, str, str], str | bool ]"):
 
         for _ref in self.formats[:]:

searx/settings.yml

@@ -443,6 +443,12 @@ engines:
     timeout: 6.0
     shortcut: conda
     disabled: true
+    about:
+      website: https://anaconda.org/search
+      wikidata_id: Q18209310
+      official_api_documentation: https://api.anaconda.org/search
+      use_official_api: false
+      results: HTML
 
   - name: arch linux wiki
     engine: archlinux
@@ -644,29 +650,10 @@ engines:
     engine: chinaso
     shortcut: chinaso
     categories: [news]
-    chinaso_category: news
     chinaso_news_source: all
     disabled: true
     inactive: true
 
-  - name: chinaso images
-    engine: chinaso
-    network: chinaso news
-    shortcut: chinasoi
-    categories: [images]
-    chinaso_category: images
-    disabled: true
-    inactive: true
-
-  - name: chinaso videos
-    engine: chinaso
-    network: chinaso news
-    shortcut: chinasov
-    categories: [videos]
-    chinaso_category: videos
-    disabled: true
-    inactive: true
-
   - name: cl0q
     engine: json_engine
     shortcut: cl
@@ -1424,12 +1411,10 @@ engines:
 
   - name: kozmonavt
     engine: xpath
-    paging: true
+    search_url: https://kozmonavt.su/s?q={query}
-    # pageno query param key is U+1D5D1 (looks like an "x")
-    search_url: https://kozmonavt.su/s?q={query}&%F0%9D%97%91={pageno}
     shortcut: koz
     disabled: true
-    # inactive: true
+    inactive: true
     results_xpath: //div[contains(@class, 'list')]/section
     url_xpath: concat('https://', substring-after(.//a/@href, '?q='))
     title_xpath: .//a

searx/translations/ru/LC_MESSAGES/messages.po

@@ -33,22 +33,24 @@
 # IcewindX <icewindx@noreply.codeberg.org>, 2025.
 # 0ko <0ko@noreply.codeberg.org>, 2025.
 # greatdng <greatdng@noreply.codeberg.org>, 2026.
+# code_gremlin <code_gremlin@noreply.codeberg.org>, 2026.
 msgid ""
 msgstr ""
 "Project-Id-Version: searx\n"
 "Report-Msgid-Bugs-To: EMAIL@ADDRESS\n"
 "POT-Creation-Date: 2026-06-13 11:31+0000\n"
-"PO-Revision-Date: 2026-05-19 12:08+0000\n"
+"PO-Revision-Date: 2026-06-24 10:07+0000\n"
-"Last-Translator: return42 <return42@noreply.codeberg.org>\n"
+"Last-Translator: code_gremlin <code_gremlin@noreply.codeberg.org>\n"
+"Language-Team: Russian <https://translate.codeberg.org/projects/searxng/"
+"searxng/ru/>\n"
 "Language: ru\n"
-"Language-Team: Russian "
-"<https://translate.codeberg.org/projects/searxng/searxng/ru/>\n"
-"Plural-Forms: nplurals=4; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && "
-"n%10<=4 && (n%100<12 || n%100>14) ? 1 : n%10==0 || (n%10>=5 && n%10<=9) "
-"|| (n%100>=11 && n%100<=14)? 2 : 3);\n"
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=utf-8\n"
 "Content-Transfer-Encoding: 8bit\n"
+"Plural-Forms: nplurals=4; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && "
+"n%10<=4 && (n%100<12 || n%100>14) ? 1 : n%10==0 || (n%10>=5 && n%10<=9) || "
+"(n%100>=11 && n%100<=14)? 2 : 3);\n"
+"X-Generator: Weblate 2026.6.1\n"
 "Generated-By: Babel 2.18.0\n"
 
 #. CONSTANT_NAMES['NO_SUBGROUPING']
@@ -1667,11 +1669,11 @@ msgstr "Разрешение"
 
 #: searx/templates/simple/result_templates/images.html:55
 msgid "Image formats"
-msgstr ""
+msgstr "Форматы изображений"
 
 #: searx/templates/simple/result_templates/images.html:56
 msgid "original format"
-msgstr ""
+msgstr "формат оригинала"
 
 #: searx/templates/simple/result_templates/images.html:63
 msgid "Engine"
@@ -2441,4 +2443,3 @@ msgstr "скрыть видео"
 
 #~ msgid "Format"
 #~ msgstr "Формат"
-

searxng_extra/update/update_engine_descriptions.py

@@ -15,25 +15,26 @@ from os.path import join
 
 from lxml.html import fromstring
 
+import searx.engines
 from searx.engines import wikidata, set_loggers
-from searx.utils import extract_text, searxng_useragent
+from searx.utils import extract_text
 from searx.locales import LOCALE_NAMES, locales_initialize, match_locale
 from searx import searx_dir
 from searx.utils import gen_useragent
 import searx.search
 import searx.network
-from searx.data import data_dir
+from searx.data import data_dir, ENGINE_DESCRIPTIONS
 
-DATA_FILE = data_dir / 'engine_descriptions.json'
+DATA_FILE = data_dir / "engine_descriptions.json"
 
-set_loggers(wikidata, 'wikidata')
+set_loggers(wikidata, "wikidata")
 locales_initialize()
 
 # you can run the query in https://query.wikidata.org
 # replace %IDS% by Wikidata entities separated by spaces with the prefix wd:
 # for example wd:Q182496 wd:Q1540899
 # replace %LANGUAGES_SPARQL% by languages
-SPARQL_WIKIPEDIA_ARTICLE = """
+SPARQL_WIKIPEDIA_ARTICLE: str = """
 SELECT DISTINCT ?item ?name ?article ?lang
 WHERE {
   hint:Query hint:optimizer "None".
@@ -48,7 +49,7 @@ WHERE {
 ORDER BY ?item ?lang
 """
 
-SPARQL_DESCRIPTION = """
+SPARQL_DESCRIPTION: str = """
 SELECT DISTINCT ?item ?itemDescription
 WHERE {
   VALUES ?item { %IDS% }
@@ -58,43 +59,43 @@ WHERE {
 ORDER BY ?itemLang
 """
 
-NOT_A_DESCRIPTION = [
+NOT_A_DESCRIPTION: list[str] = [
-    'web site',
+    "web site",
-    'site web',
+    "site web",
-    'komputa serĉilo',
+    "komputa serĉilo",
-    'interreta serĉilo',
+    "interreta serĉilo",
-    'bilaketa motor',
+    "bilaketa motor",
-    'web search engine',
+    "web search engine",
-    'wikimedia täpsustuslehekülg',
+    "wikimedia täpsustuslehekülg",
 ]
 
-SKIP_ENGINE_SOURCE = [
+SKIP_ENGINE_SOURCE: list[tuple[str, str]] = [
     # fmt: off
-    ('gitlab', 'wikidata')
+    ("gitlab", "wikidata")
     # descriptions are about wikipedia disambiguation pages
     # fmt: on
 ]
 
-WIKIPEDIA_LANGUAGES = {}
+WIKIPEDIA_LANGUAGES: dict[str, str] = {}
-LANGUAGES_SPARQL = ''
+LANGUAGES_SPARQL: str = ""
-IDS = None
+IDS: str = ""
-WIKIPEDIA_LANGUAGE_VARIANTS = {'zh_Hant': 'zh-tw'}
+WIKIPEDIA_LANGUAGE_VARIANTS: dict[str, str] = {"zh_Hant": "zh-tw"}
+
+# descriptions[engine][language] = [description, source]
+descriptions: dict[str, dict[str, list[str]]] = {}
+wd_to_engine_name: dict[str, set[str]] = {}
 
 
-descriptions = {}
+def normalize_description(description: str):
-wd_to_engine_name = {}
-
-
-def normalize_description(description):
     for c in [chr(c) for c in range(0, 31)]:
-        description = description.replace(c, ' ')
+        description = description.replace(c, " ")
-    description = ' '.join(description.strip().split())
+    description = " ".join(description.strip().split())
     return description
 
 
-def update_description(engine_name, lang, description, source, replace=True):
+def update_description(engine_name: str, lang: str, description: str, source: str, replace: bool = True) -> None:
     if not isinstance(description, str):
-        return
+        return  # pyright: ignore[reportUnreachable]
     description = normalize_description(description)
     if description.lower() == engine_name.lower():
         return
@@ -102,54 +103,71 @@ def update_description(engine_name, lang, description, source, replace=True):
         return
     if (engine_name, source) in SKIP_ENGINE_SOURCE:
         return
-    if ' ' not in description:
+    if " " not in description:
         # skip unique word description (like "website")
         return
     if replace or lang not in descriptions[engine_name]:
         descriptions[engine_name][lang] = [description, source]
 
 
-def get_wikipedia_summary(wikipedia_url, searxng_locale):
+def get_wikipedia_summary(wikipedia_url: str, searxng_locale: str):
     # get the REST API URL from the HTML URL
 
     # Headers
-    headers = {'User-Agent': searxng_useragent()}
+    headers = {
+        "User-Agent": gen_useragent(),
+        "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
+        "Accept-Encoding": "gzip, deflate",
+        "DNT": "1",
+        "Upgrade-Insecure-Requests": "1",
+        "Sec-GPC": "1",
+        "Cache-Control": "max-age=0",
+        "Sec-Fetch-Dest": "document",
+        "Sec-Fetch-Mode": "navigate",
+        "Sec-Fetch-Site": "same-origin",
+        "Sec-Fetch-User": "?1",
+    }
 
     if searxng_locale in WIKIPEDIA_LANGUAGE_VARIANTS:
-        headers['Accept-Language'] = WIKIPEDIA_LANGUAGE_VARIANTS.get(searxng_locale)
+        headers["Accept-Language"] = WIKIPEDIA_LANGUAGE_VARIANTS[searxng_locale]
 
     # URL path : from HTML URL to REST API URL
     parsed_url = urlparse(wikipedia_url)
     # remove the /wiki/ prefix
-    article_name = parsed_url.path.split('/wiki/')[1]
+    article_name = parsed_url.path.split("/wiki/")[1]
     # article_name is already encoded but not the / which is required for the REST API call
-    encoded_article_name = article_name.replace('/', '%2F')
+    encoded_article_name = article_name.replace("/", "%2F")
-    path = '/api/rest_v1/page/summary/' + encoded_article_name
+    path = "/api/rest_v1/page/summary/" + encoded_article_name
     wikipedia_rest_url = parsed_url._replace(path=path).geturl()
     try:
         response = searx.network.get(wikipedia_rest_url, headers=headers, timeout=10)
         response.raise_for_status()
     except Exception as e:  # pylint: disable=broad-except
-        print("     ", wikipedia_url, e)
+        print("     get_wikipedia_summary: ", wikipedia_rest_url, e)
         return None
     api_result = json.loads(response.text)
-    return api_result.get('extract')
+    return api_result.get("extract")
 
 
-def get_website_description(url, lang1, lang2=None):
+def get_website_description(url: str, lang1: str | None, lang2: str | None = None):
     headers = {
-        'User-Agent': gen_useragent(),
+        "User-Agent": gen_useragent(),
-        'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
+        "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
-        'DNT': '1',
+        "Accept-Encoding": "gzip, deflate",
-        'Upgrade-Insecure-Requests': '1',
+        "DNT": "1",
-        'Sec-GPC': '1',
+        "Upgrade-Insecure-Requests": "1",
-        'Cache-Control': 'max-age=0',
+        "Sec-GPC": "1",
+        "Cache-Control": "max-age=0",
+        "Sec-Fetch-Dest": "document",
+        "Sec-Fetch-Mode": "navigate",
+        "Sec-Fetch-Site": "same-origin",
+        "Sec-Fetch-User": "?1",
     }
     if lang1 is not None:
         lang_list = [lang1]
         if lang2 is not None:
             lang_list.append(lang2)
-        headers['Accept-Language'] = f'{",".join(lang_list)};q=0.8'
+        headers["Accept-Language"] = f'{",".join(lang_list)};q=0.8'
     try:
         response = searx.network.get(url, headers=headers, timeout=10)
         response.raise_for_status()
@@ -167,7 +185,7 @@ def get_website_description(url, lang1, lang2=None):
     if not description:
         description = extract_text(html.xpath('/html/head/title'))
     lang = extract_text(html.xpath('/html/@lang'))
-    if lang is None and len(lang1) > 0:
+    if lang is None and lang1 and len(lang1) > 0:
         lang = lang1
     lang = lang or 'en'
     lang = lang.split('_')[0]
@@ -178,15 +196,15 @@ def get_website_description(url, lang1, lang2=None):
 def initialize():
     global IDS, LANGUAGES_SPARQL
     searx.search.initialize()
-    wikipedia_engine = searx.engines.engines['wikipedia']
+    wikipedia_engine = searx.engines.engines["wikipedia"]
 
-    locale2lang = {'nl-BE': 'nl'}
+    locale2lang = {"nl-BE": "nl"}
     for sxng_ui_lang in LOCALE_NAMES:
 
         sxng_ui_alias = locale2lang.get(sxng_ui_lang, sxng_ui_lang)
         wiki_lang = None
 
-        if sxng_ui_alias in wikipedia_engine.traits.custom['WIKIPEDIA_LANGUAGES']:
+        if sxng_ui_alias in wikipedia_engine.traits.custom["WIKIPEDIA_LANGUAGES"]:
             wiki_lang = sxng_ui_alias
         if not wiki_lang:
             wiki_lang = wikipedia_engine.traits.get_language(sxng_ui_alias)
@@ -195,27 +213,29 @@ def initialize():
             continue
         WIKIPEDIA_LANGUAGES[sxng_ui_lang] = wiki_lang
 
-    LANGUAGES_SPARQL = ', '.join(f"'{l}'" for l in set(WIKIPEDIA_LANGUAGES.values()))
+    LANGUAGES_SPARQL = ", ".join(f"'{l}'" for l in set(WIKIPEDIA_LANGUAGES.values()))
     for engine_name, engine in searx.engines.engines.items():
         descriptions[engine_name] = {}
-        wikidata_id = getattr(engine, "about", {}).get('wikidata_id')
+        if engine.about.wikidata_id:
-        if wikidata_id is not None:
+            wd_to_engine_name.setdefault(engine.about.wikidata_id, set()).add(engine_name)
-            wd_to_engine_name.setdefault(wikidata_id, set()).add(engine_name)
 
-    IDS = ' '.join(list(map(lambda wd_id: 'wd:' + wd_id, wd_to_engine_name.keys())))
+    IDS = " ".join(list(map(lambda wd_id: "wd:" + wd_id, wd_to_engine_name.keys())))
 
 
 def fetch_wikidata_descriptions():
-    print('Fetching wikidata descriptions')
+    print("Fetching wikidata descriptions")
     searx.network.set_timeout_for_thread(60)
     result = wikidata.send_wikidata_query(
-        SPARQL_DESCRIPTION.replace('%IDS%', IDS).replace('%LANGUAGES_SPARQL%', LANGUAGES_SPARQL)
+        SPARQL_DESCRIPTION.replace("%IDS%", IDS).replace("%LANGUAGES_SPARQL%", LANGUAGES_SPARQL)
     )
-    if result is not None:
+    if not result:
-        for binding in result['results']['bindings']:
+        print("ERROR: fetching wikiDATA descriptions - SPARQL_DESCRIPTION query without results.")
-            wikidata_id = binding['item']['value'].replace('http://www.wikidata.org/entity/', '')
+        return
-            wikidata_lang = binding['itemDescription']['xml:lang']
+
-            desc = binding['itemDescription']['value']
+    for binding in result["results"]["bindings"]:
+        wikidata_id = binding["item"]["value"].replace("http://www.wikidata.org/entity/", "")
+        wikidata_lang = binding["itemDescription"]["xml:lang"]
+        desc = binding["itemDescription"]["value"]
         for engine_name in wd_to_engine_name[wikidata_id]:
             for searxng_locale in LOCALE_NAMES:
                 if WIKIPEDIA_LANGUAGES[searxng_locale] != wikidata_lang:
@@ -224,41 +244,53 @@ def fetch_wikidata_descriptions():
                     f"    engine: {engine_name:20} / wikidata_lang: {wikidata_lang:5}",
                     f"/ len(wikidata_desc): {len(desc)}",
                 )
-                    update_description(engine_name, searxng_locale, desc, 'wikidata')
+                update_description(engine_name, searxng_locale, desc, "wikidata")
 
 
 def fetch_wikipedia_descriptions():
-    print('Fetching wikipedia descriptions')
+    print("Fetching wikipedia descriptions")
     result = wikidata.send_wikidata_query(
-        SPARQL_WIKIPEDIA_ARTICLE.replace('%IDS%', IDS).replace('%LANGUAGES_SPARQL%', LANGUAGES_SPARQL)
+        SPARQL_WIKIPEDIA_ARTICLE.replace("%IDS%", IDS).replace("%LANGUAGES_SPARQL%", LANGUAGES_SPARQL)
     )
-    if result is not None:
+    if not result:
-        for binding in result['results']['bindings']:
+        print("ERROR: fetching wikiPEDIA descriptions - SPARQL_WIKIPEDIA_ARTICLE query without results.")
-            wikidata_id = binding['item']['value'].replace('http://www.wikidata.org/entity/', '')
+        return
-            wikidata_lang = binding['name']['xml:lang']
+
-            wikipedia_url = binding['article']['value']  # for example the URL https://de.wikipedia.org/wiki/PubMed
+    # pylint: disable=too-many-nested-blocks
+    for binding in result["results"]["bindings"]:
+        wikidata_id = binding["item"]["value"].replace("http://www.wikidata.org/entity/", "")
+        wikidata_lang = binding["name"]["xml:lang"]
+        wikipedia_url = binding["article"]["value"]  # for example the URL https://de.wikipedia.org/wiki/PubMed
         for engine_name in wd_to_engine_name[wikidata_id]:
             for searxng_locale in LOCALE_NAMES:
                 if WIKIPEDIA_LANGUAGES[searxng_locale] != wikidata_lang:
                     continue
                 desc = get_wikipedia_summary(wikipedia_url, searxng_locale)
                 if not desc:
+                    if descriptions.get(searxng_locale, {}).get(engine_name) is None:
+                        _descr = ENGINE_DESCRIPTIONS.get(searxng_locale, {}).get(engine_name)
+                        if _descr is not None:
+                            if len(_descr) == 2 and _descr[1] == 'ref':
+                                ref_engine, ref_lang = _descr[0].split(':')
+                                _descr = ENGINE_DESCRIPTIONS[ref_lang][ref_engine]
+                                update_description(engine_name, searxng_locale, _descr[0], _descr[1])
+
                     continue
                 print(
                     f"    engine: {engine_name:20} / wikidata_lang: {wikidata_lang:5}",
                     f"/ len(wikipedia_desc): {len(desc)}",
                 )
-                    update_description(engine_name, searxng_locale, desc, 'wikipedia')
+                update_description(engine_name, searxng_locale, desc, "wikipedia")
 
 
-def normalize_url(url):
+def normalize_url(url: str):
-    url = url.replace('{language}', 'en')
+    url = url.replace("{language}", "en")
-    url = urlparse(url)._replace(path='/', params='', query='', fragment='').geturl()
+    url = urlparse(url)._replace(path="/", params="", query="", fragment="").geturl()
-    url = url.replace('https://api.', 'https://')
+    url = url.replace("https://api.", "https://")
     return url
 
 
-def fetch_website_description(engine_name, website):
+def fetch_website_description(engine_name: str, website: str):
     print(f"- fetch website descr: {engine_name} / {website}")
     default_lang, default_description = get_website_description(website, None, None)
 
@@ -268,11 +300,11 @@ def fetch_website_description(engine_name, website):
 
     # to specify an order in where the most common languages are in front of the
     # language list ..
-    languages = ['en', 'es', 'pt', 'ru', 'tr', 'fr']
+    languages = ["en", "es", "pt", "ru", "tr", "fr"]
     languages = languages + [l for l in LOCALE_NAMES if l not in languages]
 
-    previous_matched_lang = None
+    previous_matched_lang: str | None = None
-    previous_count = 0
+    previous_count: int = 0
 
     for lang in languages:
 
@@ -307,19 +339,15 @@ def fetch_website_description(engine_name, website):
             f" / fetched lang: {fetched_lang:7} / len(desc): {len(desc)}"
         )
 
-        matched_lang = match_locale(fetched_lang, LOCALE_NAMES.keys(), fallback=lang)
+        matched_lang = match_locale(fetched_lang, list(LOCALE_NAMES.keys())) or lang
         update_description(engine_name, matched_lang, desc, website, replace=False)
 
 
 def fetch_website_descriptions():
-    print('Fetching website descriptions')
+    print("Fetching website descriptions")
     for engine_name, engine in searx.engines.engines.items():
-        website = getattr(engine, "about", {}).get('website')
+        website = engine.about.website or getattr(engine, "search_url", "") or getattr(engine, "base_url", "")
-        if website is None and hasattr(engine, "search_url"):
+        if website:
-            website = normalize_url(getattr(engine, "search_url"))
-        if website is None and hasattr(engine, "base_url"):
-            website = normalize_url(getattr(engine, "base_url"))
-        if website is not None:
             fetch_website_description(engine_name, website)
 
 
@@ -328,30 +356,32 @@ def get_engine_descriptions_filename():
 
 
 def get_output():
+    """Summary of the results, once known descriptions are not duplicated,
+    instead a reference is provided.
+
+    - from: ``descriptions[engine][language] = [description, source]``
+    - to: ``output[language][engine] = description_and_source``
+
+    ``description_and_source`` can be:
+
+    - ``[description, source]``
+    - ``description`` (if source = "wikipedia")
+    - ``[f"engine:lang", "ref"]`` reference to another existing description
+
     """
-    From descriptions[engine][language] = [description, source]
+    output: dict[str, dict[str, list[str] | str]] = {locale: {} for locale in LOCALE_NAMES}
-    To
+    seen_descriptions: dict[str, tuple[str, str]] = {}
 
-    * output[language][engine] = description_and_source
+    for engine_name, lang_descriptions in sorted(descriptions.items()):
-    * description_and_source can be:
+        for language, descr in sorted(lang_descriptions.items()):
-       * [description, source]
+            if descr[0] in seen_descriptions:
-       * description (if source = "wikipedia")
+                ref = seen_descriptions[descr[0]]
-       * [f"engine:lang", "ref"] (reference to another existing description)
+                descr = [f"{ref[0]}:{ref[1]}", "ref"]
-    """
-    output = {locale: {} for locale in LOCALE_NAMES}
-
-    seen_descriptions = {}
-
-    for engine_name, lang_descriptions in descriptions.items():
-        for language, description in lang_descriptions.items():
-            if description[0] in seen_descriptions:
-                ref = seen_descriptions[description[0]]
-                description = [f'{ref[0]}:{ref[1]}', 'ref']
             else:
-                seen_descriptions[description[0]] = (engine_name, language)
+                seen_descriptions[descr[0]] = (engine_name, language)
-                if description[1] == 'wikipedia':
+                if descr[1] == "wikipedia":
-                    description = description[0]
+                    descr = descr[0]
-            output.setdefault(language, {}).setdefault(engine_name, description)
+            output.setdefault(language, {}).setdefault(engine_name, descr)
 
     return output
 
@@ -363,8 +393,8 @@ def main():
     fetch_website_descriptions()
 
     output = get_output()
-    with DATA_FILE.open('w', encoding='utf8') as f:
+    with DATA_FILE.open("w", encoding="utf8") as f:
-        f.write(json.dumps(output, indent=1, separators=(',', ':'), sort_keys=True, ensure_ascii=False))
+        f.write(json.dumps(output, indent=1, separators=(",", ":"), sort_keys=True, ensure_ascii=False))
 
 
 if __name__ == "__main__":