[fix] make flask_babel.gettext() work in engine modules (L10n & threads)

incident:
  flask_babel.gettext() does not work in the engine modules.

cause:
  the request() and response() functions of the engine modules run in the
  processor, whose search() method runs in a thread and in the threads the
  context of the Flask app does not exist. The context of the Flask app is
  needed by the gettext() function for the L10n.

Solution:
  copy context of the Flask app into the threads. [1]

special case:
  We cannot equip the search() method of the processors with the decorator [1],
  because the decorator requires a context (Flask app) that does not yet exist
  at the time of the initialization of the processors (the initialization of the
  processors is part of the initialization of the Flask app).

[1] https://flask.palletsprojects.com/en/2.3.x/api/#flask.copy_current_request_context

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

searx/search/__init__.py

@@ -8,6 +8,7 @@ from timeit import default_timer
 from uuid import uuid4
 
 import flask
+from flask import copy_current_request_context
 import babel
 
 from searx import settings
@@ -140,8 +141,9 @@ class Search:
         search_id = str(uuid4())
 
         for engine_name, query, request_params in requests:
+            _search = copy_current_request_context(PROCESSORS[engine_name].search)
             th = threading.Thread(  # pylint: disable=invalid-name
-                target=PROCESSORS[engine_name].search,
+                target=_search,
                 args=(query, request_params, self.result_container, self.start_time, self.actual_timeout),
                 name=search_id,
             )

searxng_extra/standalone_searx.py

@@ -1,12 +1,26 @@
 #!/usr/bin/env python
 # lint: pylint
-
 # SPDX-License-Identifier: AGPL-3.0-or-later
 # (C) Copyright Contributors to the SearXNG project.
-# (C) Copyright Contributors to the searx project (2014 - 2021)
 
 """Script to run SearXNG from terminal.
 
+  DON'T USE THIS SCRIPT!!
+
+.. danger::
+
+   Be warned, using the ``standalone_searx.py`` won't give you privacy!
+
+   On the contrary, this script behaves like a SearXNG server: your IP is
+   exposed and tracked by all active engines (google, bing, qwant, ... ), with
+   every query!
+
+.. note::
+
+   This is an old and grumpy hack / SearXNG is a Flask application with
+   client/server structure, which can't be turned into a command line tool the
+   way it was done here.
+
 Getting categories without initiate the engine will only return `['general']`
 
 >>> import searx.engines
@@ -23,54 +37,6 @@ Example to use this script:
 
     $ python3 searxng_extra/standalone_searx.py rain
 
-.. danger::
-
-   Be warned, using the ``standalone_searx.py`` won't give you privacy!
-
-   On the contrary, this script behaves like a SearXNG server: your IP is
-   exposed and tracked by all active engines (google, bing, qwant, ... ), with
-   every query!
-
-Example to run it from python:
-
->>> import importlib
-... import json
-... import sys
-... import searx.engines
-... import searx.search
-... search_query = 'rain'
-... # initialize engines
-... searx.search.initialize()
-... # load engines categories once instead of each time the function called
-... engine_cs = list(searx.engines.categories.keys())
-... # load module
-... spec = importlib.util.spec_from_file_location(
-...     'utils.standalone_searx', 'searxng_extra/standalone_searx.py')
-... sas = importlib.util.module_from_spec(spec)
-... spec.loader.exec_module(sas)
-... # use function from module
-... prog_args = sas.parse_argument([search_query], category_choices=engine_cs)
-... search_q = sas.get_search_query(prog_args, engine_categories=engine_cs)
-... res_dict = sas.to_dict(search_q)
-... sys.stdout.write(json.dumps(
-...     res_dict, sort_keys=True, indent=4, ensure_ascii=False,
-...     default=sas.json_serial))
-{
-    "answers": [],
-    "infoboxes": [ {...} ],
-    "paging": true,
-    "results": [... ],
-    "number_of_results": 820000000.0,
-    "search": {
-        "lang": "all",
-        "pageno": 1,
-        "q": "rain",
-        "safesearch": 0,
-        "timerange": null
-    },
-    "suggestions": [...]
-}
-
 """  # pylint: disable=line-too-long
 
 import argparse

tests/unit/settings/test_settings.yml

@@ -3,3 +3,8 @@
 use_default_settings: true
 search:
   formats: [html, csv, json, rss]
+
+engines:
+  - name: general dummy
+    engine: demo_offline
+    timeout: 3

tests/unit/test_search.py

@@ -43,6 +43,12 @@ class SearchQueryTestCase(SearxTestCase):
 
 
 class SearchTestCase(SearxTestCase):
+    def setUp(self):
+
+        from searx import webapp  # pylint disable=import-outside-toplevel
+
+        self.app = webapp.app
+
     @classmethod
     def setUpClass(cls):
         searx.search.initialize(TEST_ENGINES)
@@ -53,7 +59,8 @@ class SearchTestCase(SearxTestCase):
             'test', [EngineRef(PUBLIC_ENGINE_NAME, 'general')], 'en-US', SAFESEARCH, PAGENO, None, None
         )
         search = searx.search.Search(search_query)
-        search.search()
+        with self.app.test_request_context('/search'):
+            search.search()
         self.assertEqual(search.actual_timeout, 3.0)
 
     def test_timeout_query_above_default_nomax(self):
@@ -62,7 +69,8 @@ class SearchTestCase(SearxTestCase):
             'test', [EngineRef(PUBLIC_ENGINE_NAME, 'general')], 'en-US', SAFESEARCH, PAGENO, None, 5.0
         )
         search = searx.search.Search(search_query)
-        search.search()
+        with self.app.test_request_context('/search'):
+            search.search()
         self.assertEqual(search.actual_timeout, 3.0)
 
     def test_timeout_query_below_default_nomax(self):
@@ -71,7 +79,8 @@ class SearchTestCase(SearxTestCase):
             'test', [EngineRef(PUBLIC_ENGINE_NAME, 'general')], 'en-US', SAFESEARCH, PAGENO, None, 1.0
         )
         search = searx.search.Search(search_query)
-        search.search()
+        with self.app.test_request_context('/search'):
+            search.search()
         self.assertEqual(search.actual_timeout, 1.0)
 
     def test_timeout_query_below_max(self):
@@ -80,7 +89,8 @@ class SearchTestCase(SearxTestCase):
             'test', [EngineRef(PUBLIC_ENGINE_NAME, 'general')], 'en-US', SAFESEARCH, PAGENO, None, 5.0
         )
         search = searx.search.Search(search_query)
-        search.search()
+        with self.app.test_request_context('/search'):
+            search.search()
         self.assertEqual(search.actual_timeout, 5.0)
 
     def test_timeout_query_above_max(self):
@@ -89,7 +99,8 @@ class SearchTestCase(SearxTestCase):
             'test', [EngineRef(PUBLIC_ENGINE_NAME, 'general')], 'en-US', SAFESEARCH, PAGENO, None, 15.0
         )
         search = searx.search.Search(search_query)
-        search.search()
+        with self.app.test_request_context('/search'):
+            search.search()
         self.assertEqual(search.actual_timeout, 10.0)
 
     def test_external_bang(self):
@@ -119,6 +130,7 @@ class SearchTestCase(SearxTestCase):
         )
 
         search = searx.search.Search(search_query)
-        results = search.search()
+        with self.app.test_request_context('/search'):
+            results = search.search()
         # This should not redirect
         self.assertTrue(results.redirect_url is None)

tests/unit/test_standalone_searx.py

@@ -1,113 +0,0 @@
-# -*- coding: utf-8 -*-
-"""Test utils/standalone_searx.py"""
-import datetime
-import io
-import sys
-
-from mock import Mock, patch
-from nose2.tools import params
-
-from searx.search import SearchQuery, EngineRef, initialize
-from searxng_extra import standalone_searx as sas
-from tests import SearxTestCase
-
-
-class StandaloneSearx(SearxTestCase):
-    """Unit test for standalone_searx."""
-
-    @classmethod
-    def setUpClass(cls):
-        engine_list = [{'engine': 'dummy', 'name': 'engine1', 'shortcut': 'e1'}]
-
-        initialize(engine_list)
-
-    def test_parse_argument_no_args(self):
-        """Test parse argument without args."""
-        with patch.object(sys, 'argv', ['standalone_searx']), self.assertRaises(SystemExit):
-            sys.stderr = io.StringIO()
-            sas.parse_argument()
-            sys.stdout = sys.__stderr__
-
-    def test_parse_argument_basic_args(self):
-        """Test parse argument with basic args."""
-        query = 'red box'
-        exp_dict = {
-            'query': query,
-            'category': 'general',
-            'lang': 'all',
-            'pageno': 1,
-            'safesearch': '0',
-            'timerange': None,
-        }
-        args = ['standalone_searx', query]
-        with patch.object(sys, 'argv', args):
-            res = sas.parse_argument()
-            self.assertEqual(exp_dict, vars(res))
-        res2 = sas.parse_argument(args[1:])
-        self.assertEqual(exp_dict, vars(res2))
-
-    def test_to_dict(self):
-        """test to_dict."""
-        self.assertEqual(
-            sas.to_dict(sas.get_search_query(sas.parse_argument(['red box']))),
-            {
-                'search': {'q': 'red box', 'pageno': 1, 'lang': 'all', 'safesearch': 0, 'timerange': None},
-                'results': [],
-                'infoboxes': [],
-                'suggestions': [],
-                'answers': [],
-                'paging': False,
-                'number_of_results': 0,
-            },
-        )
-
-    def test_to_dict_with_mock(self):
-        """test to dict."""
-        with patch.object(sas.searx.search, 'Search') as mock_s:
-            m_search = mock_s().search()
-            m_sq = Mock()
-            self.assertEqual(
-                sas.to_dict(m_sq),
-                {
-                    'answers': [],
-                    'infoboxes': m_search.infoboxes,
-                    'paging': m_search.paging,
-                    'results': m_search.get_ordered_results(),
-                    'number_of_results': m_search.number_of_results,
-                    'search': {
-                        'lang': m_sq.lang,
-                        'pageno': m_sq.pageno,
-                        'q': m_sq.query,
-                        'safesearch': m_sq.safesearch,
-                        'timerange': m_sq.time_range,
-                    },
-                    'suggestions': [],
-                },
-            )
-
-    def test_get_search_query(self):
-        """test get_search_query."""
-        args = sas.parse_argument(
-            [
-                'rain',
-            ]
-        )
-        search_q = sas.get_search_query(args)
-        self.assertTrue(search_q)
-        self.assertEqual(
-            search_q, SearchQuery('rain', [EngineRef('engine1', 'general')], 'all', 0, 1, None, None, None)
-        )
-
-    def test_no_parsed_url(self):
-        """test no_parsed_url func"""
-        self.assertEqual(sas.no_parsed_url([{'parsed_url': 'http://example.com'}]), [{}])
-
-    @params((datetime.datetime(2020, 1, 1), '2020-01-01T00:00:00'), ('a'.encode('utf8'), 'a'), (set([1]), [1]))
-    def test_json_serial(self, arg, exp_res):
-        """test json_serial func"""
-        self.assertEqual(sas.json_serial(arg), exp_res)
-
-    def test_json_serial_error(self):
-        """test error on json_serial."""
-        with self.assertRaises(TypeError):
-            sas.json_serial('a')

tests/unit/test_webapp.py

@@ -198,9 +198,7 @@ class ViewsTestCase(SearxTestCase):
         result = self.app.get('/preferences')
         self.assertEqual(result.status_code, 200)
         self.assertIn(b'<form id="search_form" method="post" action="/preferences"', result.data)
-        self.assertIn(
+        self.assertIn(b'<div id="categories_container">', result.data)
-            b'<input type="checkbox" id="checkbox_general" name="category_general" checked="checked"/>', result.data
-        )
         self.assertIn(b'<legend id="pref_ui_locale">Interface language</legend>', result.data)
 
     def test_browser_locale(self):