[fix] google: use Nokia UA (#6546)

searx/engines/google_images.py

@@ -1,122 +1,75 @@
 # SPDX-License-Identifier: AGPL-3.0-or-later
-"""This is the implementation of the Google Images engine using the internal
-Google API used by the Google Go Android app.
+"""Google Images: see :py:obj:`searx.engines.google`."""
 
-This internal API offer results in
-
-- JSON (``_fmt:json``)
-- Protobuf_ (``_fmt:pb``)
-- Protobuf_ compressed? (``_fmt:pc``)
-- HTML (``_fmt:html``)
-- Protobuf_ encoded in JSON (``_fmt:jspb``).
-
-.. _Protobuf: https://en.wikipedia.org/wiki/Protocol_Buffers
-"""
-
-from urllib.parse import urlencode
-from json import loads
+import typing as t
+from urllib.parse import parse_qs, unquote, urlparse
 
 from searx.engines.google import fetch_traits  # pylint: disable=unused-import
-from searx.engines.google import (
-    get_google_info,
-    time_range_dict,
-    detect_google_sorry,
-)
+from searx.engines.google import google_request, wml_dom
+from searx.result_types import EngineResults
+from searx.utils import eval_xpath_list
+
+if t.TYPE_CHECKING:
+    from searx.extended_types import SXNG_Response
+    from searx.search.processors import OnlineParams
 
 # about
 about = {
-    "website": 'https://images.google.com',
-    "wikidata_id": 'Q521550',
-    "official_api_documentation": 'https://developers.google.com/custom-search',
+    "website": "https://images.google.com",
+    "wikidata_id": "Q521550",
+    "official_api_documentation": "https://developers.google.com/custom-search",
     "use_official_api": False,
     "require_api_key": False,
-    "results": 'JSON',
+    "results": "XML",
 }
 
 # engine dependent config
-categories = ['images', 'web']
+categories = ["images", "web"]
 paging = True
 max_page = 50
-"""`Google max 50 pages`_
+"""Google supports up to 50 pages of results, see the `Google max_page discussion`_.
 
-.. _Google max 50 pages: https://github.com/searxng/searxng/issues/2982
+.. _Google max_page discussion: https://github.com/searxng/searxng/issues/2982
 """
 
 time_range_support = True
 language_support = True
 safesearch = True
 
-filter_mapping = {0: 'images', 1: 'active', 2: 'active'}
+filter_mapping = {0: "images", 1: "active", 2: "active"}
 
 
-def request(query, params):
-    """Google-Image search request"""
-
-    google_info = get_google_info(params, traits)
-
-    query_url = (
-        'https://'
-        + google_info['subdomain']
-        + '/search'
-        + '?'
-        + urlencode({'q': query, 'tbm': "isch", **google_info['params'], 'asearch': 'isch'})
-        # don't urlencode this because wildly different AND bad results
-        # pagination uses Zero-based numbering
-        + f'&async=_fmt:json,p:1,ijn:{params["pageno"] - 1}'
+def request(query: str, params: "OnlineParams") -> None:
+    google_request(
+        query,
+        params,
+        {"tbm": "isch"},
+        eng_traits=traits,
+        safesearch_map=filter_mapping,
+        use_locales=False,
     )
 
-    if params['time_range'] in time_range_dict:
-        query_url += '&' + urlencode({'tbs': 'qdr:' + time_range_dict[params['time_range']]})
-    if params['safesearch']:
-        query_url += '&' + urlencode({'safe': filter_mapping[params['safesearch']]})
-    params['url'] = query_url
-    params['cookies'] = google_info['cookies']
-    params['headers'].update(google_info['headers'])
-    # this ua will allow getting ~50 results instead of 10. #1641
-    params['headers']['User-Agent'] = (
-        'NSTN/3.60.474802233.release Dalvik/2.1.0 (Linux; U; Android 12;' f' {google_info.get("country", "US")}) gzip'
-    )
 
-    return params
+def response(resp: "SXNG_Response") -> EngineResults:
+    results = EngineResults()
+    dom = wml_dom(resp)
 
-
-def response(resp):
-    """Get response from google's search request"""
-    results = []
-
-    detect_google_sorry(resp)
-
-    json_start = resp.text.find('{"ischj":')
-    json_data = loads(resp.text[json_start:])
-
-    for item in json_data["ischj"].get("metadata", []):
-        result_item = {
-            'url': item["result"]["referrer_url"],
-            'title': item["result"]["page_title"],
-            'content': item["text_in_grid"]["snippet"],
-            'source': item["result"]["site_title"],
-            'resolution': f'{item["original_image"]["width"]} x {item["original_image"]["height"]}',
-            'img_src': item["original_image"]["url"],
-            'thumbnail_src': item["thumbnail"]["url"],
-            'template': 'images.html',
-        }
-
-        author = item["result"].get('iptc', {}).get('creator')
-        if author:
-            result_item['author'] = ', '.join(author)
-
-        copyright_notice = item["result"].get('iptc', {}).get('copyright_notice')
-        if copyright_notice:
-            result_item['source'] += ' | ' + copyright_notice
-
-        freshness_date = item["result"].get("freshness_date")
-        if freshness_date:
-            result_item['source'] += ' | ' + freshness_date
-
-        file_size = item.get('gsa', {}).get('file_size')
-        if file_size:
-            result_item['source'] += ' (%s)' % file_size
-
-        results.append(result_item)
+    for link in eval_xpath_list(dom, '//a[contains(@href, "/imgres?")]'):
+        qs = parse_qs(urlparse(link.get("href", "")).query)
+        img_src = qs.get("imgurl", [""])[0]
+        url = qs.get("imgrefurl", [""])[0]
+        if not img_src or not url:
+            continue
+        width, height = qs.get("w", [""])[0], qs.get("h", [""])[0]
+        tbnid = qs.get("tbnid", [""])[0]
+        results.add(
+            results.types.Image(
+                url=url,
+                title=unquote(urlparse(img_src).path.rsplit("/", 1)[-1]) or urlparse(url).netloc,
+                img_src=img_src,
+                thumbnail_src=f"https://encrypted-tbn0.gstatic.com/images?q=tbn:{tbnid}",
+                resolution=f"{width} x {height}" if width and height else "",
+            )
+        )
 
     return results