Sync upstream

searx/engines/tineye.py

@@ -14,10 +14,16 @@ billion images `[tineye.com] <https://tineye.com/how>`_.
 
 """
 
+from typing import TYPE_CHECKING
 from urllib.parse import urlencode
 from datetime import datetime
 from flask_babel import gettext
 
+if TYPE_CHECKING:
+    import logging
+
+    logger = logging.getLogger()
+
 about = {
     "website": 'https://tineye.com',
     "wikidata_id": 'Q2382535',
@@ -34,7 +40,7 @@ categories = ['general']
 paging = True
 safesearch = False
 base_url = 'https://tineye.com'
-search_string = '/result_json/?page={page}&{query}'
+search_string = '/api/v1/result_json/?page={page}&{query}'
 
 FORMAT_NOT_SUPPORTED = gettext(
     "Could not read that image url. This may be due to an unsupported file"
@@ -120,7 +126,7 @@ def parse_tineye_match(match_json):
 
             crawl_date = backlink_json.get("crawl_date")
             if crawl_date:
-                crawl_date = datetime.fromisoformat(crawl_date[:-3])
+                crawl_date = datetime.strptime(crawl_date, '%Y-%m-%d')
             else:
                 crawl_date = datetime.min
 
@@ -150,29 +156,15 @@ def parse_tineye_match(match_json):
 
 def response(resp):
     """Parse HTTP response from TinEye."""
-    results = []
 
-    try:
+    # handle the 422 client side errors, and the possible 400 status code error
+    if resp.status_code in (400, 422):
         json_data = resp.json()
-    except Exception as exc:  # pylint: disable=broad-except
-        msg = "can't parse JSON response // %s" % exc
-        logger.error(msg)
-        json_data = {'error': msg}
-
-    # handle error codes from Tineye
-
-    if resp.is_error:
-        if resp.status_code in (400, 422):
-
-            message = 'HTTP status: %s' % resp.status_code
-            error = json_data.get('error')
-            s_key = json_data.get('suggestions', {}).get('key', '')
-
-            if error and s_key:
-                message = "%s (%s)" % (error, s_key)
-            elif error:
-                message = error
+        suggestions = json_data.get('suggestions', {})
+        message = f'HTTP Status Code: {resp.status_code}'
 
+        if resp.status_code == 422:
+            s_key = suggestions.get('key', '')
             if s_key == "Invalid image URL":
                 # test https://docs.searxng.org/_static/searxng-wordmark.svg
                 message = FORMAT_NOT_SUPPORTED
@@ -182,16 +174,23 @@ def response(resp):
             elif s_key == 'Download Error':
                 # test https://notexists
                 message = DOWNLOAD_ERROR
+            else:
+                logger.warning("Unknown suggestion key encountered: %s", s_key)
+        else:  # 400
+            description = suggestions.get('description')
+            if isinstance(description, list):
+                message = ','.join(description)
 
-            # see https://github.com/searxng/searxng/pull/1456#issuecomment-1193105023
-            # results.append({'answer': message})
-            logger.error(message)
+        # see https://github.com/searxng/searxng/pull/1456#issuecomment-1193105023
+        # results.append({'answer': message})
+        logger.error(message)
+        return []
 
-            return results
+    # Raise for all other responses
+    resp.raise_for_status()
 
-        resp.raise_for_status()
-
-    # append results from matches
+    results = []
+    json_data = resp.json()
 
     for match_json in json_data['matches']:
 

searx/engines/yandex.py

@@ -0,0 +1,133 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""Yandex (Web, images)"""
+
+from json import loads
+from urllib.parse import urlencode
+from html import unescape
+from lxml import html
+from searx.exceptions import SearxEngineCaptchaException
+from searx.utils import humanize_bytes, eval_xpath, eval_xpath_list, extract_text, extr
+
+
+# Engine metadata
+about = {
+    "website": 'https://yandex.com/',
+    "wikidata_id": 'Q5281',
+    "official_api_documentation": "?",
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": 'HTML',
+}
+
+# Engine configuration
+categories = []
+paging = True
+search_type = ""
+
+# Search URL
+base_url_web = 'https://yandex.com/search/site/'
+base_url_images = 'https://yandex.com/images/search'
+
+results_xpath = '//li[contains(@class, "serp-item")]'
+url_xpath = './/a[@class="b-serp-item__title-link"]/@href'
+title_xpath = './/h3[@class="b-serp-item__title"]/a[@class="b-serp-item__title-link"]/span'
+content_xpath = './/div[@class="b-serp-item__content"]//div[@class="b-serp-item__text"]'
+
+
+def catch_bad_response(resp):
+    if resp.url.path.startswith('/showcaptcha'):
+        raise SearxEngineCaptchaException()
+
+
+def request(query, params):
+    query_params_web = {
+        "tmpl_version": "releases",
+        "text": query,
+        "web": "1",
+        "frame": "1",
+        "searchid": "3131712",
+    }
+
+    query_params_images = {
+        "text": query,
+        "uinfo": "sw-1920-sh-1080-ww-1125-wh-999",
+    }
+
+    if params['pageno'] > 1:
+        query_params_web.update({"p": params["pageno"] - 1})
+        query_params_images.update({"p": params["pageno"] - 1})
+
+    params["cookies"] = {'cookie': "yp=1716337604.sp.family%3A0#1685406411.szm.1:1920x1080:1920x999"}
+
+    if search_type == 'web':
+        params['url'] = f"{base_url_web}?{urlencode(query_params_web)}"
+    elif search_type == 'images':
+        params['url'] = f"{base_url_images}?{urlencode(query_params_images)}"
+
+    return params
+
+
+def response(resp):
+    if search_type == 'web':
+
+        catch_bad_response(resp)
+
+        dom = html.fromstring(resp.text)
+
+        results = []
+
+        for result in eval_xpath_list(dom, results_xpath):
+            results.append(
+                {
+                    'url': extract_text(eval_xpath(result, url_xpath)),
+                    'title': extract_text(eval_xpath(result, title_xpath)),
+                    'content': extract_text(eval_xpath(result, content_xpath)),
+                }
+            )
+
+        return results
+
+    if search_type == 'images':
+
+        catch_bad_response(resp)
+
+        html_data = html.fromstring(resp.text)
+        html_sample = unescape(html.tostring(html_data, encoding='unicode'))
+
+        content_between_tags = extr(
+            html_sample, '{"location":"/images/search/', 'advRsyaSearchColumn":null}}', default="fail"
+        )
+        json_data = '{"location":"/images/search/' + content_between_tags + 'advRsyaSearchColumn":null}}'
+
+        if content_between_tags == "fail":
+            content_between_tags = extr(html_sample, '{"location":"/images/search/', 'false}}}')
+            json_data = '{"location":"/images/search/' + content_between_tags + 'false}}}'
+
+        json_resp = loads(json_data)
+
+        results = []
+        for _, item_data in json_resp['initialState']['serpList']['items']['entities'].items():
+            title = item_data['snippet']['title']
+            source = item_data['snippet']['url']
+            thumb = item_data['image']
+            fullsize_image = item_data['viewerData']['dups'][0]['url']
+            height = item_data['viewerData']['dups'][0]['h']
+            width = item_data['viewerData']['dups'][0]['w']
+            filesize = item_data['viewerData']['dups'][0]['fileSizeInBytes']
+            humanized_filesize = humanize_bytes(filesize)
+
+            results.append(
+                {
+                    'title': title,
+                    'url': source,
+                    'img_src': fullsize_image,
+                    'filesize': humanized_filesize,
+                    'thumbnail_src': thumb,
+                    'template': 'images.html',
+                    'resolution': f'{width} x {height}',
+                }
+            )
+
+        return results
+
+    return []

searx/plugins/hostnames.py

@@ -1,7 +1,19 @@
 # SPDX-License-Identifier: AGPL-3.0-or-later
 # pylint: disable=too-many-branches
-"""In addition to rewriting/replace reslut URLs, the *hoostnames* plugin offers
-other features.
+"""
+.. attention::
+
+    The **"Hostname replace"** plugin has been replace by **"Hostnames
+    plugin"**, see :pull:`3463` & :pull:`3552`.
+
+The **Hostnames plugin** can be enabled by adding it to the
+``enabled_plugins`` **list** in the ``setting.yml`` like so.
+
+  .. code:: yaml
+
+     enabled_plugins:
+       - 'Hostnames plugin'
+       ...
 
 - ``hostnames.replace``: A **mapping** of regular expressions to hostnames to be
   replaced by other hostnames.

searx/settings.yml

@@ -1814,6 +1814,22 @@ engines:
     engine: unsplash
     shortcut: us
 
+  - name: yandex
+    engine: yandex
+    categories: general
+    search_type: web
+    shortcut: yd
+    disabled: true
+    inactive: true
+
+  - name: yandex images
+    engine: yandex
+    categories: images
+    search_type: images
+    shortcut: ydi
+    disabled: true
+    inactive: true
+
   - name: yandex music
     engine: yandex_music
     shortcut: ydm