[fix] set language_support for engines with languages in traits (#6258)

In the past, the engine option ``language_support`` was not consistently
maintained; with this patch, a ValueError is now thrown if an engine has
languages in its traits but language_support is not set to True.

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

searx/engines/__init__.py

@@ -21,7 +21,7 @@ import msgspec
 
 from searx import logger, settings
 from searx.utils import load_module
-
+from searx.data import ENGINE_TRAITS
 from searx.enginelib import Engine, EngineAbout
 
 logger = logger.getChild('engines')
@@ -196,7 +196,7 @@ def update_engine_attributes(engine: "Engine | types.ModuleType", engine_data: d
         engine.about = EngineAbout(**kvargs)
     except TypeError as exc:
         raise TypeError(
-            f"engine {engine_data['name']} ({engine_data['engine']}) - in the about section --> {exc}"
+            f"engine '{engine_data['name']}' ({engine_data['engine']}) - in the about section --> {exc}"
         ) from exc
 
     for param_name, param_value in engine_data.items():
@@ -214,6 +214,9 @@ def update_engine_attributes(engine: "Engine | types.ModuleType", engine_data: d
         if not hasattr(engine, arg_name):
             setattr(engine, arg_name, copy.deepcopy(arg_value))
 
+    if ENGINE_TRAITS.get(engine.name, {}).get("languages") and not engine.language_support:
+        raise ValueError(f"engine '{engine.name}' ({engine_data['engine']}) language_support should be set to True")
+
 
 def update_attributes_for_tor(engine: "Engine | types.ModuleType"):
     if using_tor_proxy(engine) and hasattr(engine, 'onion_url'):

searx/engines/annas_archive.py

@@ -64,6 +64,7 @@ about: dict[str, t.Any] = {
 # engine dependent config
 categories = ["files", "books"]
 paging: bool = True
+language_support = True
 
 # search-url
 base_url: list[str] | str = []

searx/engines/archlinux.py

@@ -35,6 +35,7 @@ about = {
 categories = ["it", "software wikis"]
 paging = True
 main_wiki = "wiki.archlinux.org"
+language_support = True
 
 
 def request(query, params):

searx/engines/boardreader.py

@@ -34,6 +34,7 @@ about = {
 categories = ["general", "social media"]
 paging = True
 time_range_support = True
+language_support = True
 
 base_url = "https://boardreader.com"
 time_range_map = {"day": "1", "week": "7", "month": "30", "year": "365"}

searx/engines/dailymotion.py

@@ -40,6 +40,7 @@ categories = ["videos"]
 paging = True
 page_size = 10
 
+language_support = True
 time_range_support = True
 time_delta_dict = {
     "day": timedelta(days=1),

searx/engines/duckduckgo.py

@@ -203,6 +203,7 @@ about: dict[str, str | bool] = {
 categories: list[str] = ["general", "web"]
 paging: bool = True
 time_range_support: bool = True
+language_support = True
 safesearch: bool = True
 """DDG-lite: user can't select but the results are filtered."""
 

searx/engines/duckduckgo_extra.py

@@ -28,6 +28,7 @@ about = {
     "require_api_key": False,
     "results": "JSON (site requires js to get images)",
 }
+language_support = True
 
 # engine dependent config
 categories = []

searx/engines/duckduckgo_weather.py

@@ -26,6 +26,7 @@ about = {
     "require_api_key": False,
     "results": "JSON",
 }
+language_support = True
 
 # engine dependent config
 categories = ["weather"]

searx/engines/google.py

@@ -57,6 +57,7 @@ max_page = 50
 .. _Google max 50 pages: https://github.com/searxng/searxng/issues/2982
 """
 time_range_support = True
+language_support = True
 safesearch = True
 
 time_range_dict = {"day": "d", "week": "w", "month": "m", "year": "y"}

searx/engines/google_images.py

@@ -43,6 +43,7 @@ max_page = 50
 """
 
 time_range_support = True
+language_support = True
 safesearch = True
 
 filter_mapping = {0: 'images', 1: 'active', 2: 'active'}

searx/engines/google_news.py

@@ -66,6 +66,7 @@ about = {
 categories = ["news"]
 paging = False
 time_range_support = False
+language_support = True
 
 # Google-News results are always *SafeSearch*. Option 'safesearch' is set to
 # False here.

searx/engines/mojeek.py

@@ -20,6 +20,7 @@ about = {
 }
 paging = True  # paging is only supported for general search
 safesearch = True
+language_support = True
 time_range_support = True  # time range search is supported for general and news
 max_page = 10
 

searx/engines/odysee.py

@@ -26,6 +26,7 @@ about = {
 # Engine configuration
 paging = True
 time_range_support = True
+language_support = True
 results_per_page = 20
 categories = ["videos"]
 

searx/engines/peertube.py

@@ -25,6 +25,7 @@ about = {
     "require_api_key": False,
     "results": "JSON",
 }
+language_support = True
 
 # engine dependent config
 categories = ["videos"]

searx/engines/radio_browser.py

@@ -26,6 +26,7 @@ about = {
     "require_api_key": False,
     "results": "JSON",
 }
+language_support = True
 paging = True
 categories = ["music", "radio"]
 

searx/engines/sepiasearch.py

@@ -25,6 +25,7 @@ about = {
     "require_api_key": False,
     "results": 'JSON',
 }
+language_support = True
 
 # engine dependent config
 categories = ['videos']

searx/engines/startpage.py

@@ -131,6 +131,7 @@ max_page = 18
 """Tested 18 pages maximum (argument ``page``), to be save max is set to 20."""
 
 time_range_support = True
+language_support = True
 safesearch = True
 
 time_range_dict = {"day": "d", "week": "w", "month": "m", "year": "y"}

searx/engines/wikidata.py

@@ -40,6 +40,7 @@ about = {
     "require_api_key": False,
     "results": 'JSON',
 }
+language_support = True
 
 display_type = ["infobox"]
 """A list of display types composed from ``infobox`` and ``list``.  The latter

searx/engines/wikipedia.py

@@ -72,6 +72,7 @@ about = {
     "require_api_key": False,
     "results": "JSON",
 }
+language_support = True
 
 display_type = ["infobox"]
 """A list of display types composed from ``infobox`` and ``list``.  The latter

searx/engines/yep.py

@@ -22,6 +22,7 @@ about = {
     "require_api_key": False,
     "results": "JSON",
 }
+language_support = True
 
 base_url = "https://api.yep.com"
 web_base_url = "https://yep.com"

searx/engines/zlibrary.py

@@ -61,6 +61,7 @@ about: dict[str, t.Any] = {
 
 categories: list[str] = ["files", "books"]
 paging: bool = True
+language_support = True
 base_url: str = "https://zlibrary-global.se"
 
 zlib_year_from: str = ""