Merge 7afa72708c into 3535377c9a

searx/engines/brave.py

@@ -30,6 +30,13 @@ Configured ``brave`` engines:
     ...
     brave_category: news
 
+  - name: brave.goggles
+    brave_category: goggles
+    time_range_support: true
+    paging: true
+    ...
+    brave_category: goggles
+
 
 .. _brave regions:
 
@@ -58,6 +65,23 @@ region are mapped to regions in SearXNG (see :py:obj:`babel
    low quality.
 
 
+.. _brave googles:
+
+Brave Goggles
+=============
+
+.. _list of Goggles: https://search.brave.com/goggles/discover
+.. _Goggles Whitepaper: https://brave.com/static-assets/files/goggles.pdf
+.. _Goggles Quickstart: https://github.com/brave/goggles-quickstart
+
+Goggles allow you to choose, alter, or extend the ranking of Brave Search
+results (`Goggles Whitepaper`_).  Goggles are openly developed by the community
+of Brave Search users.
+
+Select from the `list of Goggles`_ people have published, or create your own
+(`Goggles Quickstart`_).
+
+
 .. _brave languages:
 
 Brave languages
@@ -95,7 +119,7 @@ Implementations
 
 """
 
-from typing import TYPE_CHECKING
+from typing import Any, TYPE_CHECKING
 
 from urllib.parse import (
     urlencode,
@@ -133,12 +157,14 @@ about = {
 base_url = "https://search.brave.com/"
 categories = []
 brave_category = 'search'
-"""Brave supports common web-search, video search, image and video search.
+Goggles = Any
+"""Brave supports common web-search, videos, images, news, and goggles search.
 
 - ``search``: Common WEB search
 - ``videos``: search for videos
 - ``images``: search for images
 - ``news``: search for news
+- ``goggles``: Common WEB search with custom rules
 """
 
 brave_spellcheck = False
@@ -151,7 +177,7 @@ in SearXNG, the spellchecking is disabled by default.
 send_accept_language_header = True
 paging = False
 """Brave only supports paging in :py:obj:`brave_category` ``search`` (UI
-category All)."""
+category All) and in the goggles category."""
 max_page = 10
 """Tested 9 pages maximum (``&offset=8``), to be save max is set to 10.  Trying
 to do more won't return any result and you will most likely be flagged as a bot.
@@ -183,12 +209,15 @@ def request(query, params):
     if brave_spellcheck:
         args['spellcheck'] = '1'
 
-    if brave_category == 'search':
+    if brave_category in ('search', 'goggles'):
         if params.get('pageno', 1) - 1:
             args['offset'] = params.get('pageno', 1) - 1
         if time_range_map.get(params['time_range']):
             args['tf'] = time_range_map.get(params['time_range'])
 
+    if brave_category == 'goggles':
+        args['goggles_id'] = Goggles
+
     params["url"] = f"{base_url}{brave_category}?{urlencode(args)}"
 
     # set properties in the cookies
@@ -209,7 +238,7 @@ def request(query, params):
 
 def response(resp):
 
-    if brave_category == 'search':
+    if brave_category in ('search', 'goggles'):
         return _parse_search(resp)
 
     datastr = ""
@@ -252,7 +281,9 @@ def _parse_search(resp):
         if url is None or title_tag is None or not urlparse(url).netloc:  # partial url likely means it's an ad
             continue
 
-        content_tag = eval_xpath_getindex(result, './/div[@class="snippet-description"]', 0, default='')
+        content_tag = eval_xpath_getindex(
+            result, './/div[@class="snippet-description desktop-default-regular"]', 0, default=''
+        )
         img_src = eval_xpath_getindex(result, './/img[contains(@class, "thumb")]/@src', 0, default='')
 
         item = {

searx/settings.yml

@@ -2154,6 +2154,16 @@ engines:
     categories: news
     brave_category: news
 
+  # - name: brave.goggles
+  #   engine: brave
+  #   network: brave
+  #   shortcut: brgog
+  #   time_range_support: true
+  #   paging: true
+  #   categories: [general, web]
+  #   brave_category: goggles
+  #   Goggles: # required! This should be a URL ending in .goggle
+
   - name: lib.rs
     shortcut: lrs
     engine: xpath