From 0476de443ec5a688386f84a48fcf2e8876abe286 Mon Sep 17 00:00:00 2001
From: Allen <64094914+allendema@users.noreply.github.com>
Date: Thu, 17 Oct 2024 04:57:21 +0000
Subject: [PATCH] [enh] use longest title and test get_ordered_results()

---
 searx/results.py           | 11 ++++---
 tests/unit/test_results.py | 61 ++++++++++++++++++++++++++++++--------
 2 files changed, 55 insertions(+), 17 deletions(-)

diff --git a/searx/results.py b/searx/results.py
index 7c973ca8f..2b677b105 100644
--- a/searx/results.py
+++ b/searx/results.py
@@ -12,7 +12,6 @@ from searx import logger
 from searx.engines import engines
 from searx.metrics import histogram_observe, counter_add, count_error
 
-
 CONTENT_LEN_IGNORED_CHARS_REGEX = re.compile(r'[,;:!?\./\\\\ ()-_]', re.M | re.U)
 WHITESPACE_REGEX = re.compile('( |\t|\n)+', re.M | re.U)
 
@@ -133,7 +132,7 @@ def result_score(result, priority):
     weight = 1.0
 
     for result_engine in result['engines']:
-        if hasattr(engines[result_engine], 'weight'):
+        if hasattr(engines.get(result_engine), 'weight'):
             weight *= float(engines[result_engine].weight)
 
     weight *= len(result['positions'])
@@ -332,10 +331,14 @@ class ResultContainer:
         return None
 
     def __merge_duplicated_http_result(self, duplicated, result, position):
-        # using content with more text
+        # use content with more text
         if result_content_len(result.get('content', '')) > result_content_len(duplicated.get('content', '')):
             duplicated['content'] = result['content']
 
+        # use title with more text
+        if result_content_len(result.get('title', '')) > len(duplicated.get('title', '')):
+            duplicated['title'] = result['title']
+
         # merge all result's parameters not found in duplicate
         for key in result.keys():
             if not duplicated.get(key):
@@ -347,7 +350,7 @@ class ResultContainer:
         # add engine to list of result-engines
         duplicated['engines'].add(result['engine'])
 
-        # using https if possible
+        # use https if possible
         if duplicated['parsed_url'].scheme != 'https' and result['parsed_url'].scheme == 'https':
             duplicated['url'] = result['parsed_url'].geturl()
             duplicated['parsed_url'] = result['parsed_url']
diff --git a/tests/unit/test_results.py b/tests/unit/test_results.py
index 72486bbc7..608d3c8c3 100644
--- a/tests/unit/test_results.py
+++ b/tests/unit/test_results.py
@@ -2,9 +2,26 @@
 # pylint: disable=missing-module-docstring
 
 from searx.results import ResultContainer
+from searx.engines import load_engines
 from tests import SearxTestCase
 
 
+def make_test_engine_dict(**kwargs) -> dict:
+    test_engine = {
+        # fmt: off
+        'name': None,
+        'engine': None,
+        'categories': 'general',
+        'shortcut': 'dummy',
+        'timeout': 3.0,
+        'tokens': [],
+        # fmt: on
+    }
+
+    test_engine.update(**kwargs)
+    return test_engine
+
+
 def fake_result(url='https://aa.bb/cc?dd=ee#ff', title='aaa', content='bbb', engine='wikipedia', **kwargs):
     result = {
         # fmt: off
@@ -19,23 +36,41 @@ def fake_result(url='https://aa.bb/cc?dd=ee#ff', title='aaa', content='bbb', eng
 
 
 class ResultContainerTestCase(SearxTestCase):  # pylint: disable=missing-class-docstring
+    def setUp(self) -> None:
+        stract_engine = make_test_engine_dict(name="stract", engine="stract", shortcut="stra")
+        duckduckgo_engine = make_test_engine_dict(name="duckduckgo", engine="duckduckgo", shortcut="ddg")
+        mojeek_engine = make_test_engine_dict(name="mojeek", engine="mojeek", shortcut="mjk")
+
+        load_engines([stract_engine, duckduckgo_engine, mojeek_engine])
+
+        self.container = ResultContainer()
+
+    def tearDown(self):
+        load_engines([])
+
     def test_empty(self):
-        c = ResultContainer()
-        self.assertEqual(c.get_ordered_results(), [])
+        self.assertEqual(self.container.get_ordered_results(), [])
 
     def test_one_result(self):
-        c = ResultContainer()
-        c.extend('wikipedia', [fake_result()])
-        self.assertEqual(c.results_length(), 1)
+        self.container.extend('wikipedia', [fake_result()])
+
+        self.assertEqual(self.container.results_length(), 1)
 
     def test_one_suggestion(self):
-        c = ResultContainer()
-        c.extend('wikipedia', [fake_result(suggestion=True)])
-        self.assertEqual(len(c.suggestions), 1)
-        self.assertEqual(c.results_length(), 0)
+        self.container.extend('wikipedia', [fake_result(suggestion=True)])
+
+        self.assertEqual(len(self.container.suggestions), 1)
+        self.assertEqual(self.container.results_length(), 0)
 
     def test_result_merge(self):
-        c = ResultContainer()
-        c.extend('wikipedia', [fake_result()])
-        c.extend('wikidata', [fake_result(), fake_result(url='https://example.com/')])
-        self.assertEqual(c.results_length(), 2)
+        self.container.extend('wikipedia', [fake_result()])
+        self.container.extend('wikidata', [fake_result(), fake_result(url='https://example.com/')])
+
+        self.assertEqual(self.container.results_length(), 2)
+
+    def test_result_merge_by_title(self):
+        self.container.extend('stract', [fake_result(engine='stract', title='short title')])
+        self.container.extend('duckduckgo', [fake_result(engine='duckduckgo', title='normal title')])
+        self.container.extend('mojeek', [fake_result(engine='mojeek', title='this long long title')])
+
+        self.assertEqual(self.container.get_ordered_results()[0].get('title', ''), 'this long long title')