From 5583f3a5dcc94ae3e11064ec66a0ad1ae6e95b72 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Tue, 30 Mar 2021 23:37:35 +0530
Subject: [PATCH 01/26] Fixing Issue  #894

It seems removing this line resolves the issue #894 : search_dates() wrong result for 12:xx am
---
 dateparser/search/search.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index aa71c7299..22d47af04 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -88,8 +88,6 @@ def split_if_not_parsed(self, item, original):
 
     def parse_item(self, parser, item, translated_item, parsed, need_relative_base):
         relative_base = None
-        item = item.replace('ngày', '')
-        item = item.replace('am', '')
         parsed_item = parser.get_date_data(item)
         is_relative = date_is_relative(translated_item)
 

From fff511249b3e13a332ecd87b5e4e4e5fa66ef5d2 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Wed, 31 Mar 2021 01:33:16 +0530
Subject: [PATCH 02/26] Tests are corrected for recognising am

The test files were corrected to recognize am previously the am was ignored because of issue #894

After merging the test file  issue #894 is resolved
---
 tests/test_search.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 067601569..876b77636 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -461,15 +461,16 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                     January UTC
                     June 5 am utc
                     June 23th 5 pm EST
-                    May 31, 8am UTC""",
+                    May 31
+                    8am UTC""",
               [('May 2020', datetime.datetime(2020, 5, datetime.datetime.utcnow().day, 0, 0)),
                ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('2023', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
-               ('June 5 am utc', datetime.datetime(2023, 6, 5, 0, 0, tzinfo=pytz.utc)),
+               ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
                ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
-               ('8am UTC', datetime.datetime(2023, 8, 31, 0, 0, tzinfo=pytz.utc))]),
+               ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From a77cb4ca4fc3066a40459ea3236906b7404e9d79 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Wed, 31 Mar 2021 01:35:10 +0530
Subject: [PATCH 03/26] Fixing the issue #894

---
 dateparser/search/search.py | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index 22d47af04..c04e55e00 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -86,8 +86,15 @@ def split_if_not_parsed(self, item, original):
                 possible_splits.extend(self.split_by(item, original, splitter))
         return possible_splits
 
-    def parse_item(self, parser, item, translated_item, parsed, need_relative_base):
+    def parse_item(self, parser, item, translated_item, parsed, need_relative_base, language):
         relative_base = None
+
+        if language != "en":
+            item = item.replace('ngày', '')
+            item = item.replace('am', '')
+
+
+        #CHANGE HERE
         parsed_item = parser.get_date_data(item)
         is_relative = date_is_relative(translated_item)
 
@@ -99,7 +106,7 @@ def parse_item(self, parser, item, translated_item, parsed, need_relative_base):
             parsed_item = parser.get_date_data(item)
         return parsed_item, is_relative
 
-    def parse_found_objects(self, parser, to_parse, original, translated, settings):
+    def parse_found_objects(self, parser, to_parse, original, translated, settings, language):
         parsed = []
         substrings = []
         need_relative_base = True
@@ -109,7 +116,7 @@ def parse_found_objects(self, parser, to_parse, original, translated, settings):
             if len(item) <= 2:
                 continue
 
-            parsed_item, is_relative = self.parse_item(parser, item, translated[i], parsed, need_relative_base)
+            parsed_item, is_relative = self.parse_item(parser, item, translated[i], parsed, need_relative_base, language)
             if parsed_item['date_obj']:
                 parsed.append((parsed_item, is_relative))
                 substrings.append(original[i].strip(" .,:()[]-'"))
@@ -129,7 +136,7 @@ def parse_found_objects(self, parser, to_parse, original, translated, settings):
                         if len(jtem) <= 2:
                             continue
                         parsed_jtem, is_relative_jtem = self.parse_item(
-                            parser, jtem, split_translated[j], current_parsed, need_relative_base)
+                            parser, jtem, split_translated[j], current_parsed, need_relative_base, language)
                         current_parsed.append((parsed_jtem, is_relative_jtem))
                         current_substrings.append(split_original[j].strip(' .,:()[]-'))
                 possible_parsed.append(current_parsed)
@@ -151,9 +158,12 @@ def search_parse(self, shortname, text, settings):
             languages = [shortname]
             to_parse = original
 
+        
+
+
         parser = DateDataParser(languages=languages, settings=settings)
         parsed, substrings = self.parse_found_objects(parser=parser, to_parse=to_parse,
-                                                      original=original, translated=translated, settings=settings)
+                                                      original=original, translated=translated, settings=settings, language=shortname)
         parser._settings = Settings()
         return list(zip(substrings, [i[0]['date_obj'] for i in parsed]))
 

From 3a93e8ae2a9eb5d8cbfab1a6a3db90feaeb9b989 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Wed, 31 Mar 2021 15:20:22 +0530
Subject: [PATCH 04/26] Removing unnecessary  comments

---
 dateparser/search/search.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index c04e55e00..58f4ebc10 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -93,8 +93,6 @@ def parse_item(self, parser, item, translated_item, parsed, need_relative_base,
             item = item.replace('ngày', '')
             item = item.replace('am', '')
 
-
-        #CHANGE HERE
         parsed_item = parser.get_date_data(item)
         is_relative = date_is_relative(translated_item)
 

From 0ea06bcc4c7b2c3572dc37fc9c45fd65785b3ffe Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Wed, 31 Mar 2021 15:58:29 +0530
Subject: [PATCH 05/26] Improving test_search.py

---
 tests/test_search.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 876b77636..259a736bf 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -464,10 +464,10 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                     May 31
                     8am UTC""",
               [('May 2020', datetime.datetime(2020, 5, datetime.datetime.utcnow().day, 0, 0)),
-               ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
-               ('2023', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 0, 0)),
-               ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
-               ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
+               ('June 2020', datetime.datetime(2020, 6, 30, 0, 0)),
+               ('2023', datetime.datetime(2023, 6, 30, 0, 0)),
+               ('January UTC', datetime.datetime(2023, 1, 30, 0, 0, tzinfo=pytz.utc)),
+               ('June 5 am utc', datetime.datetime(2023, 6, 30, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
                ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
                ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),

From 4e729e8ce5108b3bfdd2026e2cef4fc14d86bbf5 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Wed, 31 Mar 2021 16:04:35 +0530
Subject: [PATCH 06/26] Correcting 'am' for German Language

---
 tests/test_search.py | 1005 ++++++++++--------------------------------
 1 file changed, 226 insertions(+), 779 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 259a736bf..848d65fed 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -1,780 +1,227 @@
-from parameterized import parameterized, param
-from tests import BaseTestCase
-from dateparser.timezone_parser import StaticTzInfo
-from dateparser.search.search import DateSearchWithDetection
-from dateparser.search import search_dates
-from dateparser.conf import Settings, apply_settings
-from dateparser_data.settings import default_parsers
-import datetime
-import pytz
-
-
-class TestTranslateSearch(BaseTestCase):
-    def setUp(self):
-        super().setUp()
-        self.search_with_detection = DateSearchWithDetection()
-        self.exact_language_search = self.search_with_detection.search
-
-    def run_search_dates_function_invalid_languages(self, text, languages, error_type):
-        try:
-            search_dates(text=text, languages=languages)
-        except Exception as error:
-            self.error = error
-            self.assertIsInstance(self.error, error_type)
-
-    def check_error_message(self, message):
-        self.assertEqual(str(self.error), message)
-
-    @parameterized.expand([
-        # English
-        param('en', "Sep 03 2014"),
-        param('en', "friday, 03 september 2014"),
-        param('en', 'Aug 06, 2018 05:05 PM CDT'),
-
-        # Chinese
-        param('zh', "1年11个月"),
-        param('zh', "1年11個月"),
-        param('zh', "2015年04月08日10点05"),
-        param('zh', "2015年04月08日10:05"),
-        param('zh', "2013年04月08日"),
-        param('zh', "周一"),
-        param('zh', "礼拜一"),
-        param('zh', "周二"),
-        param('zh', "礼拜二"),
-        param('zh', "周三"),
-        param('zh', "礼拜三"),
-        param('zh', "星期日 2015年04月08日10:05"),
-        param('zh', "周六 2013年04月08日"),
-        param('zh', "下午3:30"),
-        param('zh', "凌晨3:30"),
-        param('zh', "中午"),
-
-        # French
-        param('fr', "20 Février 2012"),
-        param('fr', "Mercredi 19 Novembre 2013"),
-        param('fr', "18 octobre 2012 à 19 h 21 min"),
-
-        # German
-        param('de', "29. Juni 2007"),
-        param('de', "Montag 5 Januar, 2015"),
-
-        # Hungarian
-        param('hu', '2016 augusztus 11'),
-        param('hu', '2016-08-13 szombat 10:21'),
-        param('hu', '2016. augusztus 14. vasárnap 10:21'),
-        param('hu', 'hétfő'),
-        param('hu', 'tegnapelőtt'),
-        param('hu', 'ma'),
-        param('hu', '2 hónappal ezelőtt'),
-        param('hu', '2016-08-13 szombat 10:21 GMT'),
-
-        # Spanish
-        param('es', "Miércoles 31 Diciembre 2014"),
-
-        # Italian
-        param('it', "Giovedi Maggio 29 2013"),
-        param('it', "19 Luglio 2013"),
-
-        # Portuguese
-        param('pt', "22 de dezembro de 2014 às 02:38"),
-
-        # Russian
-        param('ru', "5 августа 2014 г в 12:00"),
-        # Real: param('ru', "5 августа 2014 г. в 12:00"),
-
-        # Turkish
-        param('tr', "2 Ocak 2015 Cuma, 16:49"),
-
-        # Czech
-        param('cs', "22. prosinec 2014 v 2:38"),
-
-        # Dutch
-        param('nl', "maandag 22 december 2014 om 2:38"),
-
-        # Romanian
-        param('ro', "22 Decembrie 2014 la 02:38"),
-
-        # Polish
-        param('pl', "4 stycznia o 13:50"),
-        param('pl', "29 listopada 2014 o 08:40"),
-
-        # Ukrainian
-        param('uk', "30 листопада 2013 о 04:27"),
-
-        # Belarusian
-        param('be', "5 снежня 2015 г у 12:00"),
-        # Real: param('be', "5 снежня 2015 г. у 12:00"), Issue: Abbreviation segmentation.
-        param('be', "11 верасня 2015 г у 12:11"),
-        # Real: param('be', "11 верасня 2015 г. у 12:11"),
-        param('be', "3 стд 2015 г у 10:33"),
-        # Real: param('be', "3 стд 2015 г. у 10:33"),
-
-        # Arabic
-        param('ar', "6 يناير، 2015، الساعة 05:16 مساءً"),
-        param('ar', "7 يناير، 2015، الساعة 11:00 صباحاً"),
-
-        # Vietnamese
-        # Disabled - wrong segmentation at "Thứ Năm"
-        # param('vi', "Thứ Năm, ngày 8 tháng 1 năm 2015"),
-        # Disabled - wrong segmentation at "Thứ Tư"
-        # param('vi', "Thứ Tư, 07/01/2015 | 22:34"),
-        param('vi', "9 Tháng 1 2015 lúc 15:08"),
-
-        # Thai
-        # Disabled - spacing differences
-        # param('th', "เมื่อ กุมภาพันธ์ 09, 2015, 09:27:57 AM"),
-        # param('th', "เมื่อ กรกฎาคม 05, 2012, 01:18:06 AM"),
-
-        # Tagalog
-        param('tl', "Biyernes Hulyo 3, 2015"),
-        param('tl', "Pebrero 5, 2015 7:00 pm"),
-        # Indonesian
-        param('id', "06 Sep 2015"),
-        param('id', "07 Feb 2015 20:15"),
-
-        # Miscellaneous
-        param('en', "2014-12-12T12:33:39-08:00"),
-        param('en', "2014-10-15T16:12:20+00:00"),
-        param('en', "28 Oct 2014 16:39:01 +0000"),
-        # Disabled - wrong split at "a las".
-        # param('es', "13 Febrero 2015 a las 23:00"),
-
-        # Danish
-        param('da', "Sep 03 2014"),
-        param('da', "fredag, 03 september 2014"),
-        param('da', "fredag d. 3 september 2014"),
-
-        # Finnish
-        param('fi', "maanantai tammikuu 16, 2015"),
-        param('fi', "ma tammi 16, 2015"),
-        param('fi', "tiistai helmikuu 16, 2015"),
-        param('fi', "ti helmi 16, 2015"),
-        param('fi', "keskiviikko maaliskuu 16, 2015"),
-        param('fi', "ke maalis 16, 2015"),
-        param('fi', "torstai huhtikuu 16, 2015"),
-        param('fi', "to huhti 16, 2015"),
-        param('fi', "perjantai toukokuu 16, 2015"),
-        param('fi', "pe touko 16, 2015"),
-        param('fi', "lauantai kesäkuu 16, 2015"),
-        param('fi', "la kesä 16, 2015"),
-        param('fi', "sunnuntai heinäkuu 16, 2015"),
-        param('fi', "su heinä 16, 2015"),
-        param('fi', "su elokuu 16, 2015"),
-        param('fi', "su elo 16, 2015"),
-        param('fi', "su syyskuu 16, 2015"),
-        param('fi', "su syys 16, 2015"),
-        param('fi', "su lokakuu 16, 2015"),
-        param('fi', "su loka 16, 2015"),
-        param('fi', "su marraskuu 16, 2015"),
-        param('fi', "su marras 16, 2015"),
-        param('fi', "su joulukuu 16, 2015"),
-        param('fi', "su joulu 16, 2015"),
-        param('fi', "1. tammikuuta, 2016"),
-        param('fi', "tiistaina, 27. lokakuuta 2015"),
-
-        # Japanese
-        param('ja', "午後3時"),
-        param('ja', "2時"),
-        param('ja', "11時42分"),
-        param('ja', "3ヶ月"),
-        param('ja', "約53か月前"),
-        param('ja', "3月"),
-        param('ja', "十二月"),
-        param('ja', "2月10日"),
-        param('ja', "2013年2月"),
-        param('ja', "2013年04月08日"),
-        param('ja', "2016年03月24日 木曜日 10時05分"),
-        param('ja', "2016年3月20日 21時40分"),
-        param('ja', "2016年03月21日 23時05分11秒"),
-        param('ja', "2016年3月21日(月) 14時48分"),
-        param('ja', "2016年3月20日(日) 21時40分"),
-        param('ja', "2016年3月20日 (日) 21時40分"),
-
-        # Hebrew
-        param('he', "20 לאפריל 2012"),
-        param('he', "יום רביעי ה-19 בנובמבר 2013"),
-        param('he', "18 לאוקטובר 2012 בשעה 19:21"),
-        # Disabled - wrong split at "יום ה'".
-        # param('he', "יום ה' 6/10/2016"),
-        param('he', "חצות"),
-        param('he', "1 אחר חצות"),
-        param('he', "3 לפנות בוקר"),
-        param('he', "3 בבוקר"),
-        param('he', "3 בצהריים"),
-        param('he', "6 לפנות ערב"),
-        param('he', "6 אחרי הצהריים"),
-        param('he', "6 אחרי הצהרים"),
-
-        # Bangla
-        param('bn', "সেপ্টেম্বর 03 2014"),
-        param('bn', "শুক্রবার, 03 সেপ্টেম্বর 2014"),
-
-        # Hindi
-        param('hi', 'सोमवार 13 जून 1998'),
-        param('hi', 'मंगल 16 1786 12:18'),
-        param('hi', 'शनि 11 अप्रैल 2002 03:09'),
-
-        # Swedish
-        param('sv', "Sept 03 2014"),
-        param('sv', "fredag, 03 september 2014"),
-    ])
-    def test_search_date_string(self, shortname, datetime_string):
-        result = self.exact_language_search.search(shortname, datetime_string, settings=Settings())[1][0]
-        self.assertEqual(result, datetime_string)
-
-    @parameterized.expand([
-        # Arabic
-        param('ar', 'في 29 يوليو 1938 غزت القوات اليابانية الاتحاد'
-                    ' السوفييتي ووقعت أولى المعارك والتي انتصر فيها السوفييت، وعلى الرغم من ذلك رفضت'
-                    ' اليابان الاعتراف بذلك وقررت في 11 مايو 1939 تحريك الحدود المنغولية حتى نهر غول،'
-                    ' حيث وقعت معركة خالخين غول والتي انتصر فيها الجيش الأحمر على جيش كوانتونغ',
-              [('في 29 يوليو 1938', datetime.datetime(1938, 7, 29, 0, 0)),
-               ('في 11 مايو 1939', datetime.datetime(1939, 5, 11, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Belarusian
-        param('be', 'Пасля апублікавання Патсдамскай дэкларацыі 26 ліпеня 1945 года і адмовы Японіі капітуляваць '
-                    'на яе ўмовах ЗША скінулі атамныя бомбы.',
-              [('26 ліпеня 1945 года і', datetime.datetime(1945, 7, 26, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Bulgarian
-        param('bg', 'На 16 юни 1944 г. започват въздушни '
-                    'бомбардировки срещу Япония, използувайки новозавладените острови като бази.',
-              [('На 16 юни 1944 г', datetime.datetime(1944, 6, 16, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Chinese
-        param('zh', '不過大多數人仍多把第二次世界大戰的爆發定為1939年9月1日德國入侵波蘭開始，這次入侵行動隨即導致英國與法國向德國宣戰。',
-              [('1939年9月1', datetime.datetime(1939, 9, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Czech
-        param('cs', 'V roce 1920 byla proto vytvořena Společnost národů, jež měla fungovat jako fórum, '
-                    'na němž měly národy mírovým způsobem urovnávat svoje spory.',
-              [('1920', datetime.datetime(1920, 1, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Danish
-        param('da', 'Krigen i Europa begyndte den 1. september 1939, da Nazi-Tyskland invaderede Polen, '
-                    'og endte med Nazi-Tysklands betingelsesløse overgivelse den 8. maj 1945.',
-              [('1. september 1939', datetime.datetime(1939, 9, 1, 0, 0)),
-               ('8. maj 1945', datetime.datetime(1945, 5, 8, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Dutch
-        param('nl', ' De meest dramatische uitbreiding van het conflict vond plaats op 22 juni 1941 met de '
-                    'Duitse aanval op de Sovjet-Unie.',
-              [('22 juni 1941', datetime.datetime(1941, 6, 22, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # English
-        param('en', 'I will meet you tomorrow at noon',
-              [('tomorrow at noon', datetime.datetime(2000, 1, 2, 12, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        param('en', 'in a minute',
-              [('in a minute', datetime.datetime(2000, 1, 1, 0, 1))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-        param('en', 'July 13th.\r\n July 14th',
-              [('July 13th', datetime.datetime(2000, 7, 13, 0, 0)),
-               ('July 14th', datetime.datetime(2000, 7, 14, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-        param('en', 'last updated Aug 06, 2018 05:05 PM CDT',
-              [(
-                  'Aug 06, 2018 05:05 PM CDT',
-                  datetime.datetime(
-                      2018, 8, 6, 17, 5, tzinfo=StaticTzInfo(
-                          'CDT', datetime.timedelta(seconds=-18000)
-                      ))
-              )],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-        param('en', '25th march 2015 , i need this report today.',
-              [('25th march 2015', datetime.datetime(2015, 3, 25))],
-              settings={'PARSERS': [parser for parser in default_parsers
-                                    if parser != 'relative-time']}),
-        param('en', '25th march 2015 , i need this report today.',
-              [('25th march 2015', datetime.datetime(2015, 3, 25)),
-               ('today', datetime.datetime(2000, 1, 1))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Filipino / Tagalog
-        param('tl', 'Maraming namatay sa mga Hapon hanggang sila\'y sumuko noong Agosto 15, 1945.',
-              [('noong Agosto 15, 1945', datetime.datetime(1945, 8, 15, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Finnish
-        param('fi', 'Iso-Britannia ja Ranska julistivat sodan Saksalle 3. syyskuuta 1939.',
-              [('3. syyskuuta 1939', datetime.datetime(1939, 9, 3, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # French
-        param('fr', 'La 2e Guerre mondiale, ou Deuxième Guerre mondiale4, est un conflit armé à '
-                    'l\'échelle planétaire qui dura du 1 septembre 1939 au 2 septembre 1945.',
-              [('1 septembre 1939', datetime.datetime(1939, 9, 1, 0, 0)),
-               ('2 septembre 1945', datetime.datetime(1945, 9, 2, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Hebrew
-        param('he', 'במרץ 1938 "אוחדה" אוסטריה עם גרמניה (אנשלוס). ',
-              [('במרץ 1938', datetime.datetime(1938, 3, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Hindi
-        param('hi',
-              'जुलाई 1937 में, मार्को-पोलो ब्रिज हादसे का बहाना लेकर जापान ने चीन पर हमला कर दिया और चीनी साम्राज्य '
-              'की राजधानी बीजिंग पर कब्जा कर लिया,',
-              [('जुलाई 1937 में', datetime.datetime(1937, 7, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Hungarian
-        param('hu', 'A háború Európában 1945. május 8-án Németország feltétel nélküli megadásával, '
-                    'míg Ázsiában szeptember 2-án, Japán kapitulációjával fejeződött be.',
-              [('1945. május 8-án', datetime.datetime(1945, 5, 8, 0, 0)),
-               ('szeptember 2-án', datetime.datetime(2000, 9, 2, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Georgian
-        param('ka', '1937 წელს დაიწყო იაპონია-ჩინეთის მეორე ომი.',
-              [('1937', datetime.datetime(1937, 1, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # German
-        param('de', 'Die UdSSR blieb gemäß dem Neutralitätspakt '
-                    'vom 13. April 1941 gegenüber Japan vorerst neutral.',
-              [('Die', datetime.datetime(1999, 12, 28, 0, 0)),
-               ('13. April 1941', datetime.datetime(1941, 4, 13, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Indonesian
-        param('id', 'Kekaisaran Jepang menyerah pada tanggal 15 Agustus 1945, sehingga mengakhiri perang '
-                    'di Asia dan memperkuat kemenangan total Sekutu atas Poros.',
-              [('tanggal 15 Agustus 1945', datetime.datetime(1945, 8, 15, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Italian
-        param('it', ' Con questo il 2 ottobre 1935 prese il via la campagna '
-                    'd\'Etiopia. Il 9 maggio 1936 venne proclamato l\'Impero. ',
-              [('2 ottobre 1935', datetime.datetime(1935, 10, 2, 0, 0)),
-               ('9 maggio 1936', datetime.datetime(1936, 5, 9, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Japanese
-        param('ja', '1939年9月1日、ドイツ軍がポーランドへ侵攻したことが第二次世界大戦の始まりとされている。',
-              [('1939年9月1', datetime.datetime(1939, 9, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Persian
-        param('fa', 'نگ جهانی دوم جنگ جدی بین سپتامبر 1939 و 2 سپتامبر 1945 بود.',
-              [('سپتامبر 1939', datetime.datetime(1939, 9, 1, 0, 0)),
-               ('2 سپتامبر 1945', datetime.datetime(1945, 9, 2, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Polish
-        param('pl', 'II wojna światowa – największa wojna światowa w historii, '
-                    'trwająca od 1 września 1939 do 2 września 1945 (w Europie do 8 maja 1945)',
-              [('1 września 1939', datetime.datetime(1939, 9, 1, 0, 0)),
-               ('2 września 1945 (w', datetime.datetime(1945, 9, 2, 0, 0)),
-               ('8 maja 1945', datetime.datetime(1945, 5, 8, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Portuguese
-        param('pt', 'Em outubro de 1936, Alemanha e Itália formaram o Eixo Roma-Berlim.',
-              [('Em outubro de 1936', datetime.datetime(1936, 10, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Romanian
-        param('ro', 'Pe 17 septembrie 1939, după semnarea unui acord de încetare a focului cu Japonia, '
-                    'sovieticii au invadat Polonia dinspre est.',
-              [('17 septembrie 1939', datetime.datetime(1939, 9, 17, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Russian
-        param('ru', 'Втора́я мирова́я война́ (1 сентября 1939 — 2 сентября 1945) — '
-                    'война двух мировых военно-политических коалиций, ставшая крупнейшим вооружённым '
-                    'конфликтом в истории человечества.',
-              [('1 сентября 1939', datetime.datetime(1939, 9, 1, 0, 0)),
-               ('2 сентября 1945', datetime.datetime(1945, 9, 2, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Spanish
-        param('es', 'Desde finales de 1939 hasta inicios de 1941 Alemania conquistó o sometió '
-                    'gran parte de la Europa continental.',
-              [('de 1939', datetime.datetime(1939, 1, 1, 0, 0)),
-               ('de 1941', datetime.datetime(1941, 1, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Swedish
-        param('sv', 'Efter kommunisternas seger 1922 drog de allierade och Japan bort sina trupper.',
-              [('1922', datetime.datetime(1922, 1, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Thai
-        param('th',
-              'และเมื่อวันที่ 11 พฤษภาคม 1939 '
-              'ญี่ปุ่นตัดสินใจขยายพรมแดนญี่ปุ่น-มองโกเลียขึ้นไปถึงแม่น้ำคัลคินกอลด้วยกำลัง',
-              [('11 พฤษภาคม 1939', datetime.datetime(1939, 5, 11, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Turkish
-        param('tr', 'Almanya’nın Polonya’yı işgal ettiği 1 Eylül 1939 savaşın başladığı '
-                    'tarih olarak genel kabul görür.',
-              [('1 Eylül 1939', datetime.datetime(1939, 9, 1, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Ukrainian
-        param('uk', 'Інші дати, що розглядаються деякими авторами як дати початку війни: початок японської '
-                    'інтервенції в Маньчжурію 13 вересня 1931, початок другої японсько-китайської війни 7 '
-                    'липня 1937 року та початок угорсько-української війни 14 березня 1939 року.',
-              [('13 вересня 1931', datetime.datetime(1931, 9, 13, 0, 0)),
-               ('7 липня 1937', datetime.datetime(1937, 7, 7, 0, 0)),
-               ('14 березня 1939', datetime.datetime(1939, 3, 14, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-
-        # Vietnamese
-        param('vi', 'Ý theo gương Đức, đã tiến hành xâm lược Ethiopia năm 1935 và sát '
-                    'nhập Albania vào ngày 12 tháng 4 năm 1939.',
-              [('năm 1935', datetime.datetime(1935, 1, 1, 0, 0)),
-               ('ngày 12 tháng 4 năm 1939', datetime.datetime(1939, 4, 12, 0, 0))],
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
-    ])
-    @apply_settings
-    def test_search_and_parse(self, shortname, string, expected, settings=None):
-        result = self.exact_language_search.search_parse(shortname, string, settings=settings)
-        self.assertEqual(result, expected)
-
-    @parameterized.expand([
-        # English
-        param('en', 'January 3, 2017 - February 1st',
-              [('January 3, 2017', datetime.datetime(2017, 1, 3, 0, 0)),
-               ('February 1st', datetime.datetime(2017, 2, 1, 0, 0))]),
-        param('en', '2014 was good! October was excellent!'
-                    ' Friday, 21 was especially good!',
-              [('2014', datetime.datetime(
-                  2014, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
-                ),
-               ('October', datetime.datetime(2014, 10, datetime.datetime.utcnow().day, 0, 0)),
-               ('Friday, 21', datetime.datetime(2014, 10, 21, 0, 0))]),
-        param('en', """May 2020
-                    June 2020
-                    2023
-                    January UTC
-                    June 5 am utc
-                    June 23th 5 pm EST
-                    May 31
-                    8am UTC""",
-              [('May 2020', datetime.datetime(2020, 5, datetime.datetime.utcnow().day, 0, 0)),
-               ('June 2020', datetime.datetime(2020, 6, 30, 0, 0)),
-               ('2023', datetime.datetime(2023, 6, 30, 0, 0)),
-               ('January UTC', datetime.datetime(2023, 1, 30, 0, 0, tzinfo=pytz.utc)),
-               ('June 5 am utc', datetime.datetime(2023, 6, 30, 5, 0, tzinfo=pytz.utc)),
-               ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
-               ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
-               ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
-
-        # Russian
-        param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',
-              [('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
-               ('20 марта', datetime.datetime(2001, 3, 20, 0, 0)),
-               ('21 марта', datetime.datetime(2001, 3, 21, 0, 0))]),
-        # relative dates
-        param('ru', '19 марта 2001. Сегодня был хороший день. 2 дня назад был хороший день. '
-                    'Вчера тоже был хороший день.',
-              [('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
-               ('Сегодня', datetime.datetime(2001, 3, 19, 0, 0)),
-               ('2 дня назад', datetime.datetime(2001, 3, 17, 0, 0)),
-               ('Вчера', datetime.datetime(2001, 3, 18, 0, 0))]),
-        param('ru', '19 марта 2001. Сегодня был хороший день. Два дня назад был хороший день. Хорошая была неделя. '
-                    'Думаю, через неделю будет еще лучше.',
-              [('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
-               ('Сегодня', datetime.datetime(2001, 3, 19, 0, 0)),
-               ('Два дня назад', datetime.datetime(2001, 3, 17, 0, 0)),
-               ('через неделю', datetime.datetime(2001, 3, 26, 0, 0))]),
-
-        # Hungarian
-        param('hu', '1962 augusztus 11 Föld körüli pályára bocsátották a szovjet Vosztok-3 űrhajót, '
-                    'mely páros űrrepülést hajtott végre a másnap föld körüli pályára bocsátott Vosztok-4-gyel.'
-                    '2 hónappal ezelőtt furcsa, nem forgó jellegű szédülést tapasztaltam.',
-              [('1962 augusztus 11', datetime.datetime(1962, 8, 11, 0, 0)),
-               ('2 hónappal ezelőtt', datetime.datetime(1962, 6, 11, 0, 0))]),
-
-        # Vietnamese
-        param('vi', '1/1/1940. Vào tháng 8 năm 1940, với lực lượng lớn của Pháp tại Bắc Phi chính thức trung lập '
-                    'trong cuộc chiến, Ý mở một cuộc tấn công vào thuộc địa Somalia của Anh tại Đông Phi. '
-                    'Đến tháng 9 quân Ý vào đến Ai Cập (cũng đang dưới sự kiểm soát của Anh). ',
-              [('1/1/1940', datetime.datetime(1940, 1, 1, 0, 0)),
-               ('tháng 8 năm 1940', datetime.datetime(1940, 8, 1, 0, 0)),
-               ('tháng 9', datetime.datetime(1940, 9, 1, 0, 0))])
-    ])
-    @apply_settings
-    def test_relative_base_setting(self, shortname, string, expected, settings=None):
-        result = self.exact_language_search.search_parse(shortname, string, settings=settings)
-        self.assertEqual(result, expected)
-
-    @parameterized.expand([
-        # English
-        param('en', 'July 12th, 2014. July 13th, July 14th',
-              [('July 12th, 2014', datetime.datetime(2014, 7, 12, 0, 0)),
-               ('July 13th', datetime.datetime(2014, 7, 13, 0, 0)),
-               ('July 14th', datetime.datetime(2014, 7, 14, 0, 0))]),
-        param('en', '2014. July 13th July 14th',
-              [('2014', datetime.datetime(
-                  2014, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
-                ),
-               ('July 13th', datetime.datetime(2014, 7, 13, 0, 0)),
-               ('July 14th', datetime.datetime(2014, 7, 14, 0, 0))]),
-        param('en', 'July 13th 2014 July 14th 2014',
-              [('July 13th 2014', datetime.datetime(2014, 7, 13, 0, 0)),
-               ('July 14th 2014', datetime.datetime(2014, 7, 14, 0, 0))]),
-        param('en', 'July 13th 2014 July 14th',
-              [('July 13th 2014', datetime.datetime(2014, 7, 13, 0, 0)),
-               ('July 14th', datetime.datetime(2014, 7, 14, 0, 0))]),
-        param('en', 'July 13th, 2014 July 14th, 2014',
-              [('July 13th, 2014', datetime.datetime(2014, 7, 13, 0, 0)),
-               ('July 14th, 2014', datetime.datetime(2014, 7, 14, 0, 0))]),
-        param('en', '2014. July 12th, July 13th, July 14th',
-              [('2014', datetime.datetime(
-                  2014, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
-                ),
-               ('July 12th', datetime.datetime(2014, 7, 12, 0, 0)),
-               ('July 13th', datetime.datetime(2014, 7, 13, 0, 0)),
-               ('July 14th', datetime.datetime(2014, 7, 14, 0, 0))]),
-
-        # Swedish
-        param('sv', '1938–1939 marscherade tyska soldater i Österrike samtidigt som '
-                    'österrikiska soldater marscherade i Berlin.',
-              [('1938', datetime.datetime(
-                  1938, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
-                ),
-               ('1939', datetime.datetime(
-                   1939, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
-                )]),
-
-        # German
-        param('de', 'Verteidiger der Stadt kapitulierten am 2. Mai 1945. Am 8. Mai 1945 (VE-Day) trat '
-                    'bedingungslose Kapitulation der Wehrmacht in Kraft',
-              [('am 2. Mai 1945', datetime.datetime(1945, 5, 2, 0, 0)),
-               ('Am 8. Mai 1945', datetime.datetime(1945, 5, 8, 0, 0))]),
-
-    ])
-    @apply_settings
-    def test_splitting_of_not_parsed(self, shortname, string, expected, settings=None):
-        result = self.exact_language_search.search_parse(shortname, string, settings=settings)
-        self.assertEqual(result, expected)
-
-    @parameterized.expand([
-        # Arabic
-        param('ar', 'في 29 يوليو 1938 غزت القوات اليابانية الاتحاد'
-                    ' السوفييتي ووقعت أولى المعارك والتي انتصر فيها السوفييت، وعلى الرغم من ذلك رفضت'
-                    ' اليابان الاعتراف بذلك وقررت في 11 مايو 1939 تحريك الحدود المنغولية حتى نهر غول،'),
-
-        # Belarusian
-        param('be', 'Пасля апублікавання Патсдамскай дэкларацыі 26 ліпеня 1945 года і адмовы Японіі капітуляваць '
-                    'на яе ўмовах ЗША скінулі атамныя бомбы.'),
-
-        # Bulgarian
-        param('bg', 'На 16 юни 1944 г. започват въздушни '
-                    'бомбардировки срещу Япония, използувайки новозавладените острови като бази.'),
+from collections.abc import Set
+
+from dateparser.languages.loader import LocaleDataLoader
+from dateparser.conf import apply_settings, Settings
+from dateparser.date import DateDataParser
+from dateparser.search.text_detection import FullTextLanguageDetector
+import regex as re
+
+
+RELATIVE_REG = re.compile("(ago|in|from now|tomorrow|today|yesterday)")
+
+
+def date_is_relative(translation):
+    return re.search(RELATIVE_REG, translation) is not None
+
+
+class _ExactLanguageSearch:
+    def __init__(self, loader):
+        self.loader = loader
+        self.language = None
+
+    def get_current_language(self, shortname):
+        if self.language is None or self.language.shortname != shortname:
+            self.language = self.loader.get_locale(shortname)
+
+    def search(self, shortname, text, settings):
+        self.get_current_language(shortname)
+        result = self.language.translate_search(text, settings=settings)
+        return result
+
+    @staticmethod
+    def set_relative_base(substring, already_parsed):
+        if len(already_parsed) == 0:
+            return substring, None
+
+        i = len(already_parsed) - 1
+        while already_parsed[i][1]:
+            i -= 1
+            if i == -1:
+                return substring, None
+        relative_base = already_parsed[i][0]['date_obj']
+        return substring, relative_base
+
+    def choose_best_split(self, possible_parsed_splits, possible_substrings_splits):
+        rating = []
+        for i in range(len(possible_parsed_splits)):
+            num_substrings = len(possible_substrings_splits[i])
+            num_substrings_without_digits = 0
+            not_parsed = 0
+            for j, item in enumerate(possible_parsed_splits[i]):
+                if item[0]['date_obj'] is None:
+                    not_parsed += 1
+                if not any(char.isdigit() for char in possible_substrings_splits[i][j]):
+                    num_substrings_without_digits += 1
+            rating.append([
+                num_substrings,
+                0 if not_parsed == 0 else (float(not_parsed) / float(num_substrings)),
+                0 if num_substrings_without_digits == 0 else (
+                    float(num_substrings_without_digits) / float(num_substrings))])
+            best_index, best_rating = min(enumerate(rating), key=lambda p: (p[1][1], p[1][0], p[1][2]))
+        return possible_parsed_splits[best_index], possible_substrings_splits[best_index]
+
+    def split_by(self, item, original, splitter):
+        if item.count(splitter) <= 2:
+            return [[item.split(splitter), original.split(splitter)]]
+
+        item_all_split = item.split(splitter)
+        original_all_split = original.split(splitter)
+        all_possible_splits = [[item_all_split, original_all_split]]
+        for i in range(2, 4):
+            item_partially_split = []
+            original_partially_split = []
+            for j in range(0, len(item_all_split), i):
+                item_join = splitter.join(item_all_split[j:j + i])
+                original_join = splitter.join(original_all_split[j:j + i])
+                item_partially_split.append(item_join)
+                original_partially_split.append(original_join)
+            all_possible_splits.append([item_partially_split, original_partially_split])
+        return all_possible_splits
+
+    def split_if_not_parsed(self, item, original):
+        splitters = [',', '،', '——', '—', '–', '.', ' ']
+        possible_splits = []
+        for splitter in splitters:
+            if splitter in item and item.count(splitter) == original.count(splitter):
+                possible_splits.extend(self.split_by(item, original, splitter))
+        return possible_splits
+
+    def parse_item(self, parser, item, translated_item, parsed, need_relative_base, language):
+        relative_base = None
+
+        if language == "de":
+            item = item.replace('am', '')
+
+        parsed_item = parser.get_date_data(item)
+        is_relative = date_is_relative(translated_item)
+
+        if need_relative_base:
+            item, relative_base = self.set_relative_base(item, parsed)
+
+        if relative_base:
+            parser._settings.RELATIVE_BASE = relative_base
+            parsed_item = parser.get_date_data(item)
+        return parsed_item, is_relative
+
+    def parse_found_objects(self, parser, to_parse, original, translated, settings, language):
+        parsed = []
+        substrings = []
+        need_relative_base = True
+        if settings.RELATIVE_BASE:
+            need_relative_base = False
+        for i, item in enumerate(to_parse):
+            if len(item) <= 2:
+                continue
+
+            parsed_item, is_relative = self.parse_item(parser, item, translated[i], parsed, need_relative_base, language)
+            if parsed_item['date_obj']:
+                parsed.append((parsed_item, is_relative))
+                substrings.append(original[i].strip(" .,:()[]-'"))
+                continue
+
+            possible_splits = self.split_if_not_parsed(item, original[i])
+            if not possible_splits:
+                continue
+
+            possible_parsed = []
+            possible_substrings = []
+            for split_translated, split_original in possible_splits:
+                current_parsed = []
+                current_substrings = []
+                if split_translated:
+                    for j, jtem in enumerate(split_translated):
+                        if len(jtem) <= 2:
+                            continue
+                        parsed_jtem, is_relative_jtem = self.parse_item(
+                            parser, jtem, split_translated[j], current_parsed, need_relative_base, language)
+                        current_parsed.append((parsed_jtem, is_relative_jtem))
+                        current_substrings.append(split_original[j].strip(' .,:()[]-'))
+                possible_parsed.append(current_parsed)
+                possible_substrings.append(current_substrings)
+            parsed_best, substrings_best = self.choose_best_split(possible_parsed, possible_substrings)
+            for k in range(len(parsed_best)):
+                if parsed_best[k][0]['date_obj']:
+                    parsed.append(parsed_best[k])
+                    substrings.append(substrings_best[k])
+        return parsed, substrings
+
+    def search_parse(self, shortname, text, settings):
+        translated, original = self.search(shortname, text, settings)
+        bad_translate_with_search = ['vi', 'hu']   # splitting done by spaces and some dictionary items contain spaces
+        if shortname not in bad_translate_with_search:
+            languages = ['en']
+            to_parse = translated
+        else:
+            languages = [shortname]
+            to_parse = original
+
+        
+
+
+        parser = DateDataParser(languages=languages, settings=settings)
+        parsed, substrings = self.parse_found_objects(parser=parser, to_parse=to_parse,
+                                                      original=original, translated=translated, settings=settings, language=shortname)
+        parser._settings = Settings()
+        return list(zip(substrings, [i[0]['date_obj'] for i in parsed]))
+
+
+class DateSearchWithDetection:
+    """
+    Class which executes language detection of string in a natural language, translation of a given string,
+    search of substrings which represent date and/or time and parsing of these substrings.
+
+    """
+    def __init__(self):
+        self.loader = LocaleDataLoader()
+        self.available_language_map = self.loader.get_locale_map()
+        self.search = _ExactLanguageSearch(self.loader)
+
+    def detect_language(self, text, languages):
+        if isinstance(languages, (list, tuple, Set)):
+
+            if all([language in self.available_language_map for language in languages]):
+                languages = [self.available_language_map[language] for language in languages]
+            else:
+                unsupported_languages = set(languages) - set(self.available_language_map.keys())
+                raise ValueError(
+                    "Unknown language(s): %s" % ', '.join(map(repr, unsupported_languages)))
+        elif languages is not None:
+            raise TypeError("languages argument must be a list (%r given)" % type(languages))
+
+        if languages:
+            self.language_detector = FullTextLanguageDetector(languages=languages)
+        else:
+            self.language_detector = FullTextLanguageDetector(list(self.available_language_map.values()))
+
+        return self.language_detector._best_language(text)
 
-        # Chinese
-        param('zh', '不過大多數人仍多把第二次世界大戰的爆發定為1939年9月1日德國入侵波蘭開始，2015年04月08日10点05。'),
-
-        # Czech
-        param('cs', 'V rok 1920 byla proto vytvořena Společnost národů, jež měla fungovat jako fórum, '
-                    'na němž měly národy mírovým způsobem urovnávat svoje spory.'),
-
-        # Danish
-        param('da', 'Krigen i Europa begyndte den 1. september 1939, da Nazi-Tyskland invaderede Polen, '
-                    'og endte med Nazi-Tysklands betingelsesløse overgivelse den 8. marts 1945.'),
-
-        # Dutch
-        param('nl', ' De meest dramatische uitbreiding van het conflict vond plaats op Maandag 22 juni 1941  met de '
-                    'Duitse aanval op de Sovjet-Unie.'),
-
-        # English
-        param('en', 'I will meet you tomorrow at noon'),
-
-        # Filipino / Tagalog
-        param('tl', 'Maraming namatay sa mga Hapon hanggang sila\'y sumuko noong Agosto 15, 1945.'),
-
-        # Finnish
-        param('fi', 'Iso-Britannia ja Ranska julistivat sodan Saksalle 3. syyskuuta 1939.'),
-
-        # French
-        param('fr', 'La Seconde Guerre mondiale, ou Deuxième Guerre mondiale4, est un conflit armé à '
-                    'l\'échelle planétaire qui dura du 1 septembre 1939 au 2 septembre 1945.'),
-
-        # Hebrew
-        param('he', 'במרץ 1938 "אוחדה" אוסטריה עם גרמניה (אנשלוס). '),
-
-        # Hindi
-        param('hi',
-              'जुलाई 1937 में, मार्को-पोलो ब्रिज हादसे का बहाना लेकर जापान ने चीन पर हमला कर दिया और चीनी साम्राज्य '
-              'की राजधानी बीजिंग पर कब्जा कर लिया,'),
-
-        # Hungarian
-        param('hu', 'A háború Európában 1945. május 8-án Németország feltétel nélküli megadásával, '
-                    'míg Ázsiában szeptember 2-án, Japán kapitulációjával fejeződött be.'),
-
-        # Georgian
-        param('ka', '1937 წელს დაიწყო იაპონია-ჩინეთის მეორე ომი.'),
-
-        # German
-        param('de', 'Die UdSSR blieb dem Neutralitätspakt '
-                    'vom 13. April 1941 gegenüber Japan vorerst neutral.'),
-
-        # Indonesian
-        param('id', 'Kekaisaran Jepang menyerah pada tanggal 15 Agustus 1945, sehingga mengakhiri perang '
-                    'di Asia dan memperkuat kemenangan total Sekutu atas Poros.'),
-
-        # Italian
-        param('it', ' Con questo il 2 ottobre 1935 prese il via la campagna '
-                    'd\'Etiopia. Il 9 maggio 1936 venne proclamato l\'Impero. '),
-
-        # Japanese
-        param('ja', '1933年（昭和8年）12月23日午前6時39分、宮城（現：皇居）内の産殿にて誕生。'),
-
-        # Persian
-        param('fa', 'نگ جهانی دوم جنگ جدی بین سپتامبر 1939 و 2 سپتامبر 1945 بود.'),
-
-        # Polish
-        param('pl', 'II wojna światowa – największa wojna światowa w historii, '
-                    'trwająca od 1 września 1939 do 2 września 1945 (w Europie do 8 maja 1945)'),
-
-        # Portuguese
-        param('pt', 'Em outubro de 1936, Alemanha e Itália formaram o Eixo Roma-Berlim.'),
-
-        # Romanian
-        param('ro', 'Pe 17 septembrie 1939, după semnarea unui acord de încetare a focului cu Japonia, '
-                    'sovieticii au invadat Polonia dinspre est.'),
-
-        # Russian
-        param('ru', 'Втора́я мирова́я война́ (1 сентября 1939 — 2 сентября 1945) — '
-                    'война двух мировых военно-политических коалиций, ставшая крупнейшим вооружённым '
-                    'конфликтом в истории человечества.'),
-
-        # Spanish
-        param('es', '11 junio 2010'),
-
-        # Swedish
-        param('sv', ' den 15 augusti 1945 då Kejsardömet'),
-
-        # Thai
-        param('th',
-              'และเมื่อวันที่ 11 พฤษภาคม 1939 '
-              'ญี่ปุ่นตัดสินใจขยายพรมแดนญี่ปุ่น-มองโกเลียขึ้นไปถึงแม่น้ำคัลคินกอลด้วยกำลัง'),
-
-        # Turkish
-        param('tr', 'Almanya’nın Polonya’yı işgal ettiği 1 Eylül 1939 savaşın başladığı '
-                    'tarih olarak genel kabul görür.'),
-
-        # Ukrainian
-        param('uk', 'Інші дати, що розглядаються деякими авторами як дати початку війни: початок японської '
-                    'інтервенції в Маньчжурію 13 вересня 1931, початок другої японсько-китайської війни 7 '
-                    'липня 1937 року та початок угорсько-української війни 14 березня 1939 року.'),
-
-        # Vietnamese
-        param('vi', 'Ý theo gương Đức, đã tiến hành xâm lược Ethiopia năm 1935 và sát '
-                    'nhập Albania vào ngày 12 tháng 4 năm 1939.'),
-
-        # Only digits
-        param('en', '2007'),
-    ])
-    def test_detection(self, shortname, text):
-        result = self.search_with_detection.detect_language(text, languages=None)
-        self.assertEqual(result, shortname)
-
-    @parameterized.expand([
-        param(text='19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',
-              languages=['en', 'ru'],
-              settings=None,
-              expected=[('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
-                        ('20 марта', datetime.datetime(2001, 3, 20, 0, 0)),
-                        ('21 марта', datetime.datetime(2001, 3, 21, 0, 0))]),
-
-        param(text='Em outubro de 1936, Alemanha e Itália formaram o Eixo Roma-Berlim.',
-              languages=None,
-              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)},
-              expected=[('Em outubro de 1936', datetime.datetime(1936, 10, 1, 0, 0))]),
-
-        param(text='19 марта 2001, 20 марта, 21 марта был отличный день.',
-              languages=['en', 'ru'],
-              settings=None,
-              expected=[('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
-                        ('20 марта', datetime.datetime(2001, 3, 20, 0, 0)),
-                        ('21 марта', datetime.datetime(2001, 3, 21, 0, 0))]),
-
-        # Dates not found
-        param(text='',
-              languages=None,
-              settings=None,
-              expected=None),
-
-        # Language not detected
-        param(text='Привет',
-              languages=['en'],
-              settings=None,
-              expected=None),
-
-        # ZeroDivisionError
-        param(text="DECEMBER 21 19.87 87",
-              languages=None,
-              settings=None,
-              expected=[('DECEMBER 21 19', datetime.datetime(2019, 12, 21, 0, 0))]
-              ),
-        param(text='bonjour, pouvez vous me joindre svp par telephone 08 11 58 54 41',
-              languages=None,
-              settings={'STRICT_PARSING': True},
-              expected=None),
-        param(text="a Americ",
-              languages=None,
-              settings=None,
-              expected=None),
-
-        # Date with comma and apostrophe
-        param(text="9/3/2017  , ",
-              languages=['en'],
-              settings=None,
-              expected=[('9/3/2017', datetime.datetime(2017, 9, 3, 0, 0))]),
-        param(text="9/3/2017  ' ",
-              languages=['en'],
-              settings=None,
-              expected=[('9/3/2017', datetime.datetime(2017, 9, 3, 0, 0))]),
-    ])
-    def test_date_search_function(self, text, languages, settings, expected):
-        result = search_dates(text, languages=languages, settings=settings)
-        self.assertEqual(result, expected)
-
-    @parameterized.expand([
-        param(text="15 de outubro de 1936",
-              add_detected_language=True,
-              expected=[
-                  ("15 de outubro de 1936", datetime.datetime(1936, 10, 15, 0, 0), "pt")
-              ]),
-        param(text="15 de outubro de 1936",
-              add_detected_language=False,
-              expected=[
-                  ("15 de outubro de 1936", datetime.datetime(1936, 10, 15, 0, 0))
-              ]),
-    ])
-    def test_search_dates_returning_detected_languages_if_requested(
-        self, text, add_detected_language, expected
-    ):
-        result = search_dates(text, add_detected_language=add_detected_language)
-        self.assertEqual(result, expected)
-
-    @parameterized.expand([
-        param(text='19 марта 2001',
-              languages='wrong type: str instead of list'),
-    ])
-    def test_date_search_function_invalid_languages_type(self, text, languages):
-        self.run_search_dates_function_invalid_languages(text=text, languages=languages, error_type=TypeError)
-        self.check_error_message("languages argument must be a list (<class \'str\'> given)")
-
-    @parameterized.expand([
-        param(text='19 марта 2001',
-              languages=['unknown language code']),
-    ])
-    def test_date_search_function_invalid_language_code(self, text, languages):
-        self.run_search_dates_function_invalid_languages(text=text, languages=languages, error_type=ValueError)
-        self.check_error_message("Unknown language(s): 'unknown language code'")
+    @apply_settings
+    def search_dates(self, text, languages=None, settings=None):
+        """
+        Find all substrings of the given string which represent date and/or time and parse them.
+
+        :param text:
+            A string in a natural language which may contain date and/or time expressions.
+        :type text: str
+        :param languages:
+            A list of two letters language codes.e.g. ['en', 'es']. If languages are given, it will not attempt
+            to detect the language.
+        :type languages: list
+        :param settings:
+               Configure customized behavior using settings defined in :mod:`dateparser.conf.Settings`.
+        :type settings: dict
+
+        :return: a dict mapping keys to two letter language code and a list of tuples of pairs:
+                substring representing date expressions and corresponding :mod:`datetime.datetime` object.
+            For example:
+            {'Language': 'en', 'Dates': [('on 4 October 1957', datetime.datetime(1957, 10, 4, 0, 0))]}
+            If language of the string isn't recognised returns:
+            {'Language': None, 'Dates': None}
+        :raises: ValueError - Unknown Language
+        """
+
+        language_shortname = self.detect_language(text=text, languages=languages)
+        if not language_shortname:
+            return {'Language': None, 'Dates': None}
+        return {'Language': language_shortname, 'Dates': self.search.search_parse(language_shortname, text,
+                                                                                  settings=settings)}

From 5773f76b3eb97051c05536cd712c34109c60f7e3 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Wed, 31 Mar 2021 16:09:11 +0530
Subject: [PATCH 07/26] Update test_search.py

---
 tests/test_search.py | 1005 ++++++++++++++++++++++++++++++++----------
 1 file changed, 779 insertions(+), 226 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 848d65fed..259a736bf 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -1,227 +1,780 @@
-from collections.abc import Set
-
-from dateparser.languages.loader import LocaleDataLoader
-from dateparser.conf import apply_settings, Settings
-from dateparser.date import DateDataParser
-from dateparser.search.text_detection import FullTextLanguageDetector
-import regex as re
-
-
-RELATIVE_REG = re.compile("(ago|in|from now|tomorrow|today|yesterday)")
-
-
-def date_is_relative(translation):
-    return re.search(RELATIVE_REG, translation) is not None
-
-
-class _ExactLanguageSearch:
-    def __init__(self, loader):
-        self.loader = loader
-        self.language = None
-
-    def get_current_language(self, shortname):
-        if self.language is None or self.language.shortname != shortname:
-            self.language = self.loader.get_locale(shortname)
-
-    def search(self, shortname, text, settings):
-        self.get_current_language(shortname)
-        result = self.language.translate_search(text, settings=settings)
-        return result
-
-    @staticmethod
-    def set_relative_base(substring, already_parsed):
-        if len(already_parsed) == 0:
-            return substring, None
-
-        i = len(already_parsed) - 1
-        while already_parsed[i][1]:
-            i -= 1
-            if i == -1:
-                return substring, None
-        relative_base = already_parsed[i][0]['date_obj']
-        return substring, relative_base
-
-    def choose_best_split(self, possible_parsed_splits, possible_substrings_splits):
-        rating = []
-        for i in range(len(possible_parsed_splits)):
-            num_substrings = len(possible_substrings_splits[i])
-            num_substrings_without_digits = 0
-            not_parsed = 0
-            for j, item in enumerate(possible_parsed_splits[i]):
-                if item[0]['date_obj'] is None:
-                    not_parsed += 1
-                if not any(char.isdigit() for char in possible_substrings_splits[i][j]):
-                    num_substrings_without_digits += 1
-            rating.append([
-                num_substrings,
-                0 if not_parsed == 0 else (float(not_parsed) / float(num_substrings)),
-                0 if num_substrings_without_digits == 0 else (
-                    float(num_substrings_without_digits) / float(num_substrings))])
-            best_index, best_rating = min(enumerate(rating), key=lambda p: (p[1][1], p[1][0], p[1][2]))
-        return possible_parsed_splits[best_index], possible_substrings_splits[best_index]
-
-    def split_by(self, item, original, splitter):
-        if item.count(splitter) <= 2:
-            return [[item.split(splitter), original.split(splitter)]]
-
-        item_all_split = item.split(splitter)
-        original_all_split = original.split(splitter)
-        all_possible_splits = [[item_all_split, original_all_split]]
-        for i in range(2, 4):
-            item_partially_split = []
-            original_partially_split = []
-            for j in range(0, len(item_all_split), i):
-                item_join = splitter.join(item_all_split[j:j + i])
-                original_join = splitter.join(original_all_split[j:j + i])
-                item_partially_split.append(item_join)
-                original_partially_split.append(original_join)
-            all_possible_splits.append([item_partially_split, original_partially_split])
-        return all_possible_splits
-
-    def split_if_not_parsed(self, item, original):
-        splitters = [',', '،', '——', '—', '–', '.', ' ']
-        possible_splits = []
-        for splitter in splitters:
-            if splitter in item and item.count(splitter) == original.count(splitter):
-                possible_splits.extend(self.split_by(item, original, splitter))
-        return possible_splits
-
-    def parse_item(self, parser, item, translated_item, parsed, need_relative_base, language):
-        relative_base = None
-
-        if language == "de":
-            item = item.replace('am', '')
-
-        parsed_item = parser.get_date_data(item)
-        is_relative = date_is_relative(translated_item)
-
-        if need_relative_base:
-            item, relative_base = self.set_relative_base(item, parsed)
-
-        if relative_base:
-            parser._settings.RELATIVE_BASE = relative_base
-            parsed_item = parser.get_date_data(item)
-        return parsed_item, is_relative
-
-    def parse_found_objects(self, parser, to_parse, original, translated, settings, language):
-        parsed = []
-        substrings = []
-        need_relative_base = True
-        if settings.RELATIVE_BASE:
-            need_relative_base = False
-        for i, item in enumerate(to_parse):
-            if len(item) <= 2:
-                continue
-
-            parsed_item, is_relative = self.parse_item(parser, item, translated[i], parsed, need_relative_base, language)
-            if parsed_item['date_obj']:
-                parsed.append((parsed_item, is_relative))
-                substrings.append(original[i].strip(" .,:()[]-'"))
-                continue
-
-            possible_splits = self.split_if_not_parsed(item, original[i])
-            if not possible_splits:
-                continue
-
-            possible_parsed = []
-            possible_substrings = []
-            for split_translated, split_original in possible_splits:
-                current_parsed = []
-                current_substrings = []
-                if split_translated:
-                    for j, jtem in enumerate(split_translated):
-                        if len(jtem) <= 2:
-                            continue
-                        parsed_jtem, is_relative_jtem = self.parse_item(
-                            parser, jtem, split_translated[j], current_parsed, need_relative_base, language)
-                        current_parsed.append((parsed_jtem, is_relative_jtem))
-                        current_substrings.append(split_original[j].strip(' .,:()[]-'))
-                possible_parsed.append(current_parsed)
-                possible_substrings.append(current_substrings)
-            parsed_best, substrings_best = self.choose_best_split(possible_parsed, possible_substrings)
-            for k in range(len(parsed_best)):
-                if parsed_best[k][0]['date_obj']:
-                    parsed.append(parsed_best[k])
-                    substrings.append(substrings_best[k])
-        return parsed, substrings
-
-    def search_parse(self, shortname, text, settings):
-        translated, original = self.search(shortname, text, settings)
-        bad_translate_with_search = ['vi', 'hu']   # splitting done by spaces and some dictionary items contain spaces
-        if shortname not in bad_translate_with_search:
-            languages = ['en']
-            to_parse = translated
-        else:
-            languages = [shortname]
-            to_parse = original
-
-        
-
-
-        parser = DateDataParser(languages=languages, settings=settings)
-        parsed, substrings = self.parse_found_objects(parser=parser, to_parse=to_parse,
-                                                      original=original, translated=translated, settings=settings, language=shortname)
-        parser._settings = Settings()
-        return list(zip(substrings, [i[0]['date_obj'] for i in parsed]))
-
-
-class DateSearchWithDetection:
-    """
-    Class which executes language detection of string in a natural language, translation of a given string,
-    search of substrings which represent date and/or time and parsing of these substrings.
-
-    """
-    def __init__(self):
-        self.loader = LocaleDataLoader()
-        self.available_language_map = self.loader.get_locale_map()
-        self.search = _ExactLanguageSearch(self.loader)
-
-    def detect_language(self, text, languages):
-        if isinstance(languages, (list, tuple, Set)):
-
-            if all([language in self.available_language_map for language in languages]):
-                languages = [self.available_language_map[language] for language in languages]
-            else:
-                unsupported_languages = set(languages) - set(self.available_language_map.keys())
-                raise ValueError(
-                    "Unknown language(s): %s" % ', '.join(map(repr, unsupported_languages)))
-        elif languages is not None:
-            raise TypeError("languages argument must be a list (%r given)" % type(languages))
-
-        if languages:
-            self.language_detector = FullTextLanguageDetector(languages=languages)
-        else:
-            self.language_detector = FullTextLanguageDetector(list(self.available_language_map.values()))
-
-        return self.language_detector._best_language(text)
-
+from parameterized import parameterized, param
+from tests import BaseTestCase
+from dateparser.timezone_parser import StaticTzInfo
+from dateparser.search.search import DateSearchWithDetection
+from dateparser.search import search_dates
+from dateparser.conf import Settings, apply_settings
+from dateparser_data.settings import default_parsers
+import datetime
+import pytz
+
+
+class TestTranslateSearch(BaseTestCase):
+    def setUp(self):
+        super().setUp()
+        self.search_with_detection = DateSearchWithDetection()
+        self.exact_language_search = self.search_with_detection.search
+
+    def run_search_dates_function_invalid_languages(self, text, languages, error_type):
+        try:
+            search_dates(text=text, languages=languages)
+        except Exception as error:
+            self.error = error
+            self.assertIsInstance(self.error, error_type)
+
+    def check_error_message(self, message):
+        self.assertEqual(str(self.error), message)
+
+    @parameterized.expand([
+        # English
+        param('en', "Sep 03 2014"),
+        param('en', "friday, 03 september 2014"),
+        param('en', 'Aug 06, 2018 05:05 PM CDT'),
+
+        # Chinese
+        param('zh', "1年11个月"),
+        param('zh', "1年11個月"),
+        param('zh', "2015年04月08日10点05"),
+        param('zh', "2015年04月08日10:05"),
+        param('zh', "2013年04月08日"),
+        param('zh', "周一"),
+        param('zh', "礼拜一"),
+        param('zh', "周二"),
+        param('zh', "礼拜二"),
+        param('zh', "周三"),
+        param('zh', "礼拜三"),
+        param('zh', "星期日 2015年04月08日10:05"),
+        param('zh', "周六 2013年04月08日"),
+        param('zh', "下午3:30"),
+        param('zh', "凌晨3:30"),
+        param('zh', "中午"),
+
+        # French
+        param('fr', "20 Février 2012"),
+        param('fr', "Mercredi 19 Novembre 2013"),
+        param('fr', "18 octobre 2012 à 19 h 21 min"),
+
+        # German
+        param('de', "29. Juni 2007"),
+        param('de', "Montag 5 Januar, 2015"),
+
+        # Hungarian
+        param('hu', '2016 augusztus 11'),
+        param('hu', '2016-08-13 szombat 10:21'),
+        param('hu', '2016. augusztus 14. vasárnap 10:21'),
+        param('hu', 'hétfő'),
+        param('hu', 'tegnapelőtt'),
+        param('hu', 'ma'),
+        param('hu', '2 hónappal ezelőtt'),
+        param('hu', '2016-08-13 szombat 10:21 GMT'),
+
+        # Spanish
+        param('es', "Miércoles 31 Diciembre 2014"),
+
+        # Italian
+        param('it', "Giovedi Maggio 29 2013"),
+        param('it', "19 Luglio 2013"),
+
+        # Portuguese
+        param('pt', "22 de dezembro de 2014 às 02:38"),
+
+        # Russian
+        param('ru', "5 августа 2014 г в 12:00"),
+        # Real: param('ru', "5 августа 2014 г. в 12:00"),
+
+        # Turkish
+        param('tr', "2 Ocak 2015 Cuma, 16:49"),
+
+        # Czech
+        param('cs', "22. prosinec 2014 v 2:38"),
+
+        # Dutch
+        param('nl', "maandag 22 december 2014 om 2:38"),
+
+        # Romanian
+        param('ro', "22 Decembrie 2014 la 02:38"),
+
+        # Polish
+        param('pl', "4 stycznia o 13:50"),
+        param('pl', "29 listopada 2014 o 08:40"),
+
+        # Ukrainian
+        param('uk', "30 листопада 2013 о 04:27"),
+
+        # Belarusian
+        param('be', "5 снежня 2015 г у 12:00"),
+        # Real: param('be', "5 снежня 2015 г. у 12:00"), Issue: Abbreviation segmentation.
+        param('be', "11 верасня 2015 г у 12:11"),
+        # Real: param('be', "11 верасня 2015 г. у 12:11"),
+        param('be', "3 стд 2015 г у 10:33"),
+        # Real: param('be', "3 стд 2015 г. у 10:33"),
+
+        # Arabic
+        param('ar', "6 يناير، 2015، الساعة 05:16 مساءً"),
+        param('ar', "7 يناير، 2015، الساعة 11:00 صباحاً"),
+
+        # Vietnamese
+        # Disabled - wrong segmentation at "Thứ Năm"
+        # param('vi', "Thứ Năm, ngày 8 tháng 1 năm 2015"),
+        # Disabled - wrong segmentation at "Thứ Tư"
+        # param('vi', "Thứ Tư, 07/01/2015 | 22:34"),
+        param('vi', "9 Tháng 1 2015 lúc 15:08"),
+
+        # Thai
+        # Disabled - spacing differences
+        # param('th', "เมื่อ กุมภาพันธ์ 09, 2015, 09:27:57 AM"),
+        # param('th', "เมื่อ กรกฎาคม 05, 2012, 01:18:06 AM"),
+
+        # Tagalog
+        param('tl', "Biyernes Hulyo 3, 2015"),
+        param('tl', "Pebrero 5, 2015 7:00 pm"),
+        # Indonesian
+        param('id', "06 Sep 2015"),
+        param('id', "07 Feb 2015 20:15"),
+
+        # Miscellaneous
+        param('en', "2014-12-12T12:33:39-08:00"),
+        param('en', "2014-10-15T16:12:20+00:00"),
+        param('en', "28 Oct 2014 16:39:01 +0000"),
+        # Disabled - wrong split at "a las".
+        # param('es', "13 Febrero 2015 a las 23:00"),
+
+        # Danish
+        param('da', "Sep 03 2014"),
+        param('da', "fredag, 03 september 2014"),
+        param('da', "fredag d. 3 september 2014"),
+
+        # Finnish
+        param('fi', "maanantai tammikuu 16, 2015"),
+        param('fi', "ma tammi 16, 2015"),
+        param('fi', "tiistai helmikuu 16, 2015"),
+        param('fi', "ti helmi 16, 2015"),
+        param('fi', "keskiviikko maaliskuu 16, 2015"),
+        param('fi', "ke maalis 16, 2015"),
+        param('fi', "torstai huhtikuu 16, 2015"),
+        param('fi', "to huhti 16, 2015"),
+        param('fi', "perjantai toukokuu 16, 2015"),
+        param('fi', "pe touko 16, 2015"),
+        param('fi', "lauantai kesäkuu 16, 2015"),
+        param('fi', "la kesä 16, 2015"),
+        param('fi', "sunnuntai heinäkuu 16, 2015"),
+        param('fi', "su heinä 16, 2015"),
+        param('fi', "su elokuu 16, 2015"),
+        param('fi', "su elo 16, 2015"),
+        param('fi', "su syyskuu 16, 2015"),
+        param('fi', "su syys 16, 2015"),
+        param('fi', "su lokakuu 16, 2015"),
+        param('fi', "su loka 16, 2015"),
+        param('fi', "su marraskuu 16, 2015"),
+        param('fi', "su marras 16, 2015"),
+        param('fi', "su joulukuu 16, 2015"),
+        param('fi', "su joulu 16, 2015"),
+        param('fi', "1. tammikuuta, 2016"),
+        param('fi', "tiistaina, 27. lokakuuta 2015"),
+
+        # Japanese
+        param('ja', "午後3時"),
+        param('ja', "2時"),
+        param('ja', "11時42分"),
+        param('ja', "3ヶ月"),
+        param('ja', "約53か月前"),
+        param('ja', "3月"),
+        param('ja', "十二月"),
+        param('ja', "2月10日"),
+        param('ja', "2013年2月"),
+        param('ja', "2013年04月08日"),
+        param('ja', "2016年03月24日 木曜日 10時05分"),
+        param('ja', "2016年3月20日 21時40分"),
+        param('ja', "2016年03月21日 23時05分11秒"),
+        param('ja', "2016年3月21日(月) 14時48分"),
+        param('ja', "2016年3月20日(日) 21時40分"),
+        param('ja', "2016年3月20日 (日) 21時40分"),
+
+        # Hebrew
+        param('he', "20 לאפריל 2012"),
+        param('he', "יום רביעי ה-19 בנובמבר 2013"),
+        param('he', "18 לאוקטובר 2012 בשעה 19:21"),
+        # Disabled - wrong split at "יום ה'".
+        # param('he', "יום ה' 6/10/2016"),
+        param('he', "חצות"),
+        param('he', "1 אחר חצות"),
+        param('he', "3 לפנות בוקר"),
+        param('he', "3 בבוקר"),
+        param('he', "3 בצהריים"),
+        param('he', "6 לפנות ערב"),
+        param('he', "6 אחרי הצהריים"),
+        param('he', "6 אחרי הצהרים"),
+
+        # Bangla
+        param('bn', "সেপ্টেম্বর 03 2014"),
+        param('bn', "শুক্রবার, 03 সেপ্টেম্বর 2014"),
+
+        # Hindi
+        param('hi', 'सोमवार 13 जून 1998'),
+        param('hi', 'मंगल 16 1786 12:18'),
+        param('hi', 'शनि 11 अप्रैल 2002 03:09'),
+
+        # Swedish
+        param('sv', "Sept 03 2014"),
+        param('sv', "fredag, 03 september 2014"),
+    ])
+    def test_search_date_string(self, shortname, datetime_string):
+        result = self.exact_language_search.search(shortname, datetime_string, settings=Settings())[1][0]
+        self.assertEqual(result, datetime_string)
+
+    @parameterized.expand([
+        # Arabic
+        param('ar', 'في 29 يوليو 1938 غزت القوات اليابانية الاتحاد'
+                    ' السوفييتي ووقعت أولى المعارك والتي انتصر فيها السوفييت، وعلى الرغم من ذلك رفضت'
+                    ' اليابان الاعتراف بذلك وقررت في 11 مايو 1939 تحريك الحدود المنغولية حتى نهر غول،'
+                    ' حيث وقعت معركة خالخين غول والتي انتصر فيها الجيش الأحمر على جيش كوانتونغ',
+              [('في 29 يوليو 1938', datetime.datetime(1938, 7, 29, 0, 0)),
+               ('في 11 مايو 1939', datetime.datetime(1939, 5, 11, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Belarusian
+        param('be', 'Пасля апублікавання Патсдамскай дэкларацыі 26 ліпеня 1945 года і адмовы Японіі капітуляваць '
+                    'на яе ўмовах ЗША скінулі атамныя бомбы.',
+              [('26 ліпеня 1945 года і', datetime.datetime(1945, 7, 26, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Bulgarian
+        param('bg', 'На 16 юни 1944 г. започват въздушни '
+                    'бомбардировки срещу Япония, използувайки новозавладените острови като бази.',
+              [('На 16 юни 1944 г', datetime.datetime(1944, 6, 16, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Chinese
+        param('zh', '不過大多數人仍多把第二次世界大戰的爆發定為1939年9月1日德國入侵波蘭開始，這次入侵行動隨即導致英國與法國向德國宣戰。',
+              [('1939年9月1', datetime.datetime(1939, 9, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Czech
+        param('cs', 'V roce 1920 byla proto vytvořena Společnost národů, jež měla fungovat jako fórum, '
+                    'na němž měly národy mírovým způsobem urovnávat svoje spory.',
+              [('1920', datetime.datetime(1920, 1, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Danish
+        param('da', 'Krigen i Europa begyndte den 1. september 1939, da Nazi-Tyskland invaderede Polen, '
+                    'og endte med Nazi-Tysklands betingelsesløse overgivelse den 8. maj 1945.',
+              [('1. september 1939', datetime.datetime(1939, 9, 1, 0, 0)),
+               ('8. maj 1945', datetime.datetime(1945, 5, 8, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Dutch
+        param('nl', ' De meest dramatische uitbreiding van het conflict vond plaats op 22 juni 1941 met de '
+                    'Duitse aanval op de Sovjet-Unie.',
+              [('22 juni 1941', datetime.datetime(1941, 6, 22, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # English
+        param('en', 'I will meet you tomorrow at noon',
+              [('tomorrow at noon', datetime.datetime(2000, 1, 2, 12, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        param('en', 'in a minute',
+              [('in a minute', datetime.datetime(2000, 1, 1, 0, 1))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+        param('en', 'July 13th.\r\n July 14th',
+              [('July 13th', datetime.datetime(2000, 7, 13, 0, 0)),
+               ('July 14th', datetime.datetime(2000, 7, 14, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+        param('en', 'last updated Aug 06, 2018 05:05 PM CDT',
+              [(
+                  'Aug 06, 2018 05:05 PM CDT',
+                  datetime.datetime(
+                      2018, 8, 6, 17, 5, tzinfo=StaticTzInfo(
+                          'CDT', datetime.timedelta(seconds=-18000)
+                      ))
+              )],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+        param('en', '25th march 2015 , i need this report today.',
+              [('25th march 2015', datetime.datetime(2015, 3, 25))],
+              settings={'PARSERS': [parser for parser in default_parsers
+                                    if parser != 'relative-time']}),
+        param('en', '25th march 2015 , i need this report today.',
+              [('25th march 2015', datetime.datetime(2015, 3, 25)),
+               ('today', datetime.datetime(2000, 1, 1))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Filipino / Tagalog
+        param('tl', 'Maraming namatay sa mga Hapon hanggang sila\'y sumuko noong Agosto 15, 1945.',
+              [('noong Agosto 15, 1945', datetime.datetime(1945, 8, 15, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Finnish
+        param('fi', 'Iso-Britannia ja Ranska julistivat sodan Saksalle 3. syyskuuta 1939.',
+              [('3. syyskuuta 1939', datetime.datetime(1939, 9, 3, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # French
+        param('fr', 'La 2e Guerre mondiale, ou Deuxième Guerre mondiale4, est un conflit armé à '
+                    'l\'échelle planétaire qui dura du 1 septembre 1939 au 2 septembre 1945.',
+              [('1 septembre 1939', datetime.datetime(1939, 9, 1, 0, 0)),
+               ('2 septembre 1945', datetime.datetime(1945, 9, 2, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Hebrew
+        param('he', 'במרץ 1938 "אוחדה" אוסטריה עם גרמניה (אנשלוס). ',
+              [('במרץ 1938', datetime.datetime(1938, 3, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Hindi
+        param('hi',
+              'जुलाई 1937 में, मार्को-पोलो ब्रिज हादसे का बहाना लेकर जापान ने चीन पर हमला कर दिया और चीनी साम्राज्य '
+              'की राजधानी बीजिंग पर कब्जा कर लिया,',
+              [('जुलाई 1937 में', datetime.datetime(1937, 7, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Hungarian
+        param('hu', 'A háború Európában 1945. május 8-án Németország feltétel nélküli megadásával, '
+                    'míg Ázsiában szeptember 2-án, Japán kapitulációjával fejeződött be.',
+              [('1945. május 8-án', datetime.datetime(1945, 5, 8, 0, 0)),
+               ('szeptember 2-án', datetime.datetime(2000, 9, 2, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Georgian
+        param('ka', '1937 წელს დაიწყო იაპონია-ჩინეთის მეორე ომი.',
+              [('1937', datetime.datetime(1937, 1, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # German
+        param('de', 'Die UdSSR blieb gemäß dem Neutralitätspakt '
+                    'vom 13. April 1941 gegenüber Japan vorerst neutral.',
+              [('Die', datetime.datetime(1999, 12, 28, 0, 0)),
+               ('13. April 1941', datetime.datetime(1941, 4, 13, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Indonesian
+        param('id', 'Kekaisaran Jepang menyerah pada tanggal 15 Agustus 1945, sehingga mengakhiri perang '
+                    'di Asia dan memperkuat kemenangan total Sekutu atas Poros.',
+              [('tanggal 15 Agustus 1945', datetime.datetime(1945, 8, 15, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Italian
+        param('it', ' Con questo il 2 ottobre 1935 prese il via la campagna '
+                    'd\'Etiopia. Il 9 maggio 1936 venne proclamato l\'Impero. ',
+              [('2 ottobre 1935', datetime.datetime(1935, 10, 2, 0, 0)),
+               ('9 maggio 1936', datetime.datetime(1936, 5, 9, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Japanese
+        param('ja', '1939年9月1日、ドイツ軍がポーランドへ侵攻したことが第二次世界大戦の始まりとされている。',
+              [('1939年9月1', datetime.datetime(1939, 9, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Persian
+        param('fa', 'نگ جهانی دوم جنگ جدی بین سپتامبر 1939 و 2 سپتامبر 1945 بود.',
+              [('سپتامبر 1939', datetime.datetime(1939, 9, 1, 0, 0)),
+               ('2 سپتامبر 1945', datetime.datetime(1945, 9, 2, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Polish
+        param('pl', 'II wojna światowa – największa wojna światowa w historii, '
+                    'trwająca od 1 września 1939 do 2 września 1945 (w Europie do 8 maja 1945)',
+              [('1 września 1939', datetime.datetime(1939, 9, 1, 0, 0)),
+               ('2 września 1945 (w', datetime.datetime(1945, 9, 2, 0, 0)),
+               ('8 maja 1945', datetime.datetime(1945, 5, 8, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Portuguese
+        param('pt', 'Em outubro de 1936, Alemanha e Itália formaram o Eixo Roma-Berlim.',
+              [('Em outubro de 1936', datetime.datetime(1936, 10, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Romanian
+        param('ro', 'Pe 17 septembrie 1939, după semnarea unui acord de încetare a focului cu Japonia, '
+                    'sovieticii au invadat Polonia dinspre est.',
+              [('17 septembrie 1939', datetime.datetime(1939, 9, 17, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Russian
+        param('ru', 'Втора́я мирова́я война́ (1 сентября 1939 — 2 сентября 1945) — '
+                    'война двух мировых военно-политических коалиций, ставшая крупнейшим вооружённым '
+                    'конфликтом в истории человечества.',
+              [('1 сентября 1939', datetime.datetime(1939, 9, 1, 0, 0)),
+               ('2 сентября 1945', datetime.datetime(1945, 9, 2, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Spanish
+        param('es', 'Desde finales de 1939 hasta inicios de 1941 Alemania conquistó o sometió '
+                    'gran parte de la Europa continental.',
+              [('de 1939', datetime.datetime(1939, 1, 1, 0, 0)),
+               ('de 1941', datetime.datetime(1941, 1, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Swedish
+        param('sv', 'Efter kommunisternas seger 1922 drog de allierade och Japan bort sina trupper.',
+              [('1922', datetime.datetime(1922, 1, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Thai
+        param('th',
+              'และเมื่อวันที่ 11 พฤษภาคม 1939 '
+              'ญี่ปุ่นตัดสินใจขยายพรมแดนญี่ปุ่น-มองโกเลียขึ้นไปถึงแม่น้ำคัลคินกอลด้วยกำลัง',
+              [('11 พฤษภาคม 1939', datetime.datetime(1939, 5, 11, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Turkish
+        param('tr', 'Almanya’nın Polonya’yı işgal ettiği 1 Eylül 1939 savaşın başladığı '
+                    'tarih olarak genel kabul görür.',
+              [('1 Eylül 1939', datetime.datetime(1939, 9, 1, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Ukrainian
+        param('uk', 'Інші дати, що розглядаються деякими авторами як дати початку війни: початок японської '
+                    'інтервенції в Маньчжурію 13 вересня 1931, початок другої японсько-китайської війни 7 '
+                    'липня 1937 року та початок угорсько-української війни 14 березня 1939 року.',
+              [('13 вересня 1931', datetime.datetime(1931, 9, 13, 0, 0)),
+               ('7 липня 1937', datetime.datetime(1937, 7, 7, 0, 0)),
+               ('14 березня 1939', datetime.datetime(1939, 3, 14, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+
+        # Vietnamese
+        param('vi', 'Ý theo gương Đức, đã tiến hành xâm lược Ethiopia năm 1935 và sát '
+                    'nhập Albania vào ngày 12 tháng 4 năm 1939.',
+              [('năm 1935', datetime.datetime(1935, 1, 1, 0, 0)),
+               ('ngày 12 tháng 4 năm 1939', datetime.datetime(1939, 4, 12, 0, 0))],
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)}),
+    ])
+    @apply_settings
+    def test_search_and_parse(self, shortname, string, expected, settings=None):
+        result = self.exact_language_search.search_parse(shortname, string, settings=settings)
+        self.assertEqual(result, expected)
+
+    @parameterized.expand([
+        # English
+        param('en', 'January 3, 2017 - February 1st',
+              [('January 3, 2017', datetime.datetime(2017, 1, 3, 0, 0)),
+               ('February 1st', datetime.datetime(2017, 2, 1, 0, 0))]),
+        param('en', '2014 was good! October was excellent!'
+                    ' Friday, 21 was especially good!',
+              [('2014', datetime.datetime(
+                  2014, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
+                ),
+               ('October', datetime.datetime(2014, 10, datetime.datetime.utcnow().day, 0, 0)),
+               ('Friday, 21', datetime.datetime(2014, 10, 21, 0, 0))]),
+        param('en', """May 2020
+                    June 2020
+                    2023
+                    January UTC
+                    June 5 am utc
+                    June 23th 5 pm EST
+                    May 31
+                    8am UTC""",
+              [('May 2020', datetime.datetime(2020, 5, datetime.datetime.utcnow().day, 0, 0)),
+               ('June 2020', datetime.datetime(2020, 6, 30, 0, 0)),
+               ('2023', datetime.datetime(2023, 6, 30, 0, 0)),
+               ('January UTC', datetime.datetime(2023, 1, 30, 0, 0, tzinfo=pytz.utc)),
+               ('June 5 am utc', datetime.datetime(2023, 6, 30, 5, 0, tzinfo=pytz.utc)),
+               ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
+               ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
+               ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
+
+        # Russian
+        param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',
+              [('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
+               ('20 марта', datetime.datetime(2001, 3, 20, 0, 0)),
+               ('21 марта', datetime.datetime(2001, 3, 21, 0, 0))]),
+        # relative dates
+        param('ru', '19 марта 2001. Сегодня был хороший день. 2 дня назад был хороший день. '
+                    'Вчера тоже был хороший день.',
+              [('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
+               ('Сегодня', datetime.datetime(2001, 3, 19, 0, 0)),
+               ('2 дня назад', datetime.datetime(2001, 3, 17, 0, 0)),
+               ('Вчера', datetime.datetime(2001, 3, 18, 0, 0))]),
+        param('ru', '19 марта 2001. Сегодня был хороший день. Два дня назад был хороший день. Хорошая была неделя. '
+                    'Думаю, через неделю будет еще лучше.',
+              [('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
+               ('Сегодня', datetime.datetime(2001, 3, 19, 0, 0)),
+               ('Два дня назад', datetime.datetime(2001, 3, 17, 0, 0)),
+               ('через неделю', datetime.datetime(2001, 3, 26, 0, 0))]),
+
+        # Hungarian
+        param('hu', '1962 augusztus 11 Föld körüli pályára bocsátották a szovjet Vosztok-3 űrhajót, '
+                    'mely páros űrrepülést hajtott végre a másnap föld körüli pályára bocsátott Vosztok-4-gyel.'
+                    '2 hónappal ezelőtt furcsa, nem forgó jellegű szédülést tapasztaltam.',
+              [('1962 augusztus 11', datetime.datetime(1962, 8, 11, 0, 0)),
+               ('2 hónappal ezelőtt', datetime.datetime(1962, 6, 11, 0, 0))]),
+
+        # Vietnamese
+        param('vi', '1/1/1940. Vào tháng 8 năm 1940, với lực lượng lớn của Pháp tại Bắc Phi chính thức trung lập '
+                    'trong cuộc chiến, Ý mở một cuộc tấn công vào thuộc địa Somalia của Anh tại Đông Phi. '
+                    'Đến tháng 9 quân Ý vào đến Ai Cập (cũng đang dưới sự kiểm soát của Anh). ',
+              [('1/1/1940', datetime.datetime(1940, 1, 1, 0, 0)),
+               ('tháng 8 năm 1940', datetime.datetime(1940, 8, 1, 0, 0)),
+               ('tháng 9', datetime.datetime(1940, 9, 1, 0, 0))])
+    ])
+    @apply_settings
+    def test_relative_base_setting(self, shortname, string, expected, settings=None):
+        result = self.exact_language_search.search_parse(shortname, string, settings=settings)
+        self.assertEqual(result, expected)
+
+    @parameterized.expand([
+        # English
+        param('en', 'July 12th, 2014. July 13th, July 14th',
+              [('July 12th, 2014', datetime.datetime(2014, 7, 12, 0, 0)),
+               ('July 13th', datetime.datetime(2014, 7, 13, 0, 0)),
+               ('July 14th', datetime.datetime(2014, 7, 14, 0, 0))]),
+        param('en', '2014. July 13th July 14th',
+              [('2014', datetime.datetime(
+                  2014, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
+                ),
+               ('July 13th', datetime.datetime(2014, 7, 13, 0, 0)),
+               ('July 14th', datetime.datetime(2014, 7, 14, 0, 0))]),
+        param('en', 'July 13th 2014 July 14th 2014',
+              [('July 13th 2014', datetime.datetime(2014, 7, 13, 0, 0)),
+               ('July 14th 2014', datetime.datetime(2014, 7, 14, 0, 0))]),
+        param('en', 'July 13th 2014 July 14th',
+              [('July 13th 2014', datetime.datetime(2014, 7, 13, 0, 0)),
+               ('July 14th', datetime.datetime(2014, 7, 14, 0, 0))]),
+        param('en', 'July 13th, 2014 July 14th, 2014',
+              [('July 13th, 2014', datetime.datetime(2014, 7, 13, 0, 0)),
+               ('July 14th, 2014', datetime.datetime(2014, 7, 14, 0, 0))]),
+        param('en', '2014. July 12th, July 13th, July 14th',
+              [('2014', datetime.datetime(
+                  2014, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
+                ),
+               ('July 12th', datetime.datetime(2014, 7, 12, 0, 0)),
+               ('July 13th', datetime.datetime(2014, 7, 13, 0, 0)),
+               ('July 14th', datetime.datetime(2014, 7, 14, 0, 0))]),
+
+        # Swedish
+        param('sv', '1938–1939 marscherade tyska soldater i Österrike samtidigt som '
+                    'österrikiska soldater marscherade i Berlin.',
+              [('1938', datetime.datetime(
+                  1938, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
+                ),
+               ('1939', datetime.datetime(
+                   1939, datetime.datetime.utcnow().month, datetime.datetime.utcnow().day, 0, 0)
+                )]),
+
+        # German
+        param('de', 'Verteidiger der Stadt kapitulierten am 2. Mai 1945. Am 8. Mai 1945 (VE-Day) trat '
+                    'bedingungslose Kapitulation der Wehrmacht in Kraft',
+              [('am 2. Mai 1945', datetime.datetime(1945, 5, 2, 0, 0)),
+               ('Am 8. Mai 1945', datetime.datetime(1945, 5, 8, 0, 0))]),
+
+    ])
     @apply_settings
-    def search_dates(self, text, languages=None, settings=None):
-        """
-        Find all substrings of the given string which represent date and/or time and parse them.
-
-        :param text:
-            A string in a natural language which may contain date and/or time expressions.
-        :type text: str
-        :param languages:
-            A list of two letters language codes.e.g. ['en', 'es']. If languages are given, it will not attempt
-            to detect the language.
-        :type languages: list
-        :param settings:
-               Configure customized behavior using settings defined in :mod:`dateparser.conf.Settings`.
-        :type settings: dict
-
-        :return: a dict mapping keys to two letter language code and a list of tuples of pairs:
-                substring representing date expressions and corresponding :mod:`datetime.datetime` object.
-            For example:
-            {'Language': 'en', 'Dates': [('on 4 October 1957', datetime.datetime(1957, 10, 4, 0, 0))]}
-            If language of the string isn't recognised returns:
-            {'Language': None, 'Dates': None}
-        :raises: ValueError - Unknown Language
-        """
-
-        language_shortname = self.detect_language(text=text, languages=languages)
-        if not language_shortname:
-            return {'Language': None, 'Dates': None}
-        return {'Language': language_shortname, 'Dates': self.search.search_parse(language_shortname, text,
-                                                                                  settings=settings)}
+    def test_splitting_of_not_parsed(self, shortname, string, expected, settings=None):
+        result = self.exact_language_search.search_parse(shortname, string, settings=settings)
+        self.assertEqual(result, expected)
+
+    @parameterized.expand([
+        # Arabic
+        param('ar', 'في 29 يوليو 1938 غزت القوات اليابانية الاتحاد'
+                    ' السوفييتي ووقعت أولى المعارك والتي انتصر فيها السوفييت، وعلى الرغم من ذلك رفضت'
+                    ' اليابان الاعتراف بذلك وقررت في 11 مايو 1939 تحريك الحدود المنغولية حتى نهر غول،'),
+
+        # Belarusian
+        param('be', 'Пасля апублікавання Патсдамскай дэкларацыі 26 ліпеня 1945 года і адмовы Японіі капітуляваць '
+                    'на яе ўмовах ЗША скінулі атамныя бомбы.'),
+
+        # Bulgarian
+        param('bg', 'На 16 юни 1944 г. започват въздушни '
+                    'бомбардировки срещу Япония, използувайки новозавладените острови като бази.'),
+
+        # Chinese
+        param('zh', '不過大多數人仍多把第二次世界大戰的爆發定為1939年9月1日德國入侵波蘭開始，2015年04月08日10点05。'),
+
+        # Czech
+        param('cs', 'V rok 1920 byla proto vytvořena Společnost národů, jež měla fungovat jako fórum, '
+                    'na němž měly národy mírovým způsobem urovnávat svoje spory.'),
+
+        # Danish
+        param('da', 'Krigen i Europa begyndte den 1. september 1939, da Nazi-Tyskland invaderede Polen, '
+                    'og endte med Nazi-Tysklands betingelsesløse overgivelse den 8. marts 1945.'),
+
+        # Dutch
+        param('nl', ' De meest dramatische uitbreiding van het conflict vond plaats op Maandag 22 juni 1941  met de '
+                    'Duitse aanval op de Sovjet-Unie.'),
+
+        # English
+        param('en', 'I will meet you tomorrow at noon'),
+
+        # Filipino / Tagalog
+        param('tl', 'Maraming namatay sa mga Hapon hanggang sila\'y sumuko noong Agosto 15, 1945.'),
+
+        # Finnish
+        param('fi', 'Iso-Britannia ja Ranska julistivat sodan Saksalle 3. syyskuuta 1939.'),
+
+        # French
+        param('fr', 'La Seconde Guerre mondiale, ou Deuxième Guerre mondiale4, est un conflit armé à '
+                    'l\'échelle planétaire qui dura du 1 septembre 1939 au 2 septembre 1945.'),
+
+        # Hebrew
+        param('he', 'במרץ 1938 "אוחדה" אוסטריה עם גרמניה (אנשלוס). '),
+
+        # Hindi
+        param('hi',
+              'जुलाई 1937 में, मार्को-पोलो ब्रिज हादसे का बहाना लेकर जापान ने चीन पर हमला कर दिया और चीनी साम्राज्य '
+              'की राजधानी बीजिंग पर कब्जा कर लिया,'),
+
+        # Hungarian
+        param('hu', 'A háború Európában 1945. május 8-án Németország feltétel nélküli megadásával, '
+                    'míg Ázsiában szeptember 2-án, Japán kapitulációjával fejeződött be.'),
+
+        # Georgian
+        param('ka', '1937 წელს დაიწყო იაპონია-ჩინეთის მეორე ომი.'),
+
+        # German
+        param('de', 'Die UdSSR blieb dem Neutralitätspakt '
+                    'vom 13. April 1941 gegenüber Japan vorerst neutral.'),
+
+        # Indonesian
+        param('id', 'Kekaisaran Jepang menyerah pada tanggal 15 Agustus 1945, sehingga mengakhiri perang '
+                    'di Asia dan memperkuat kemenangan total Sekutu atas Poros.'),
+
+        # Italian
+        param('it', ' Con questo il 2 ottobre 1935 prese il via la campagna '
+                    'd\'Etiopia. Il 9 maggio 1936 venne proclamato l\'Impero. '),
+
+        # Japanese
+        param('ja', '1933年（昭和8年）12月23日午前6時39分、宮城（現：皇居）内の産殿にて誕生。'),
+
+        # Persian
+        param('fa', 'نگ جهانی دوم جنگ جدی بین سپتامبر 1939 و 2 سپتامبر 1945 بود.'),
+
+        # Polish
+        param('pl', 'II wojna światowa – największa wojna światowa w historii, '
+                    'trwająca od 1 września 1939 do 2 września 1945 (w Europie do 8 maja 1945)'),
+
+        # Portuguese
+        param('pt', 'Em outubro de 1936, Alemanha e Itália formaram o Eixo Roma-Berlim.'),
+
+        # Romanian
+        param('ro', 'Pe 17 septembrie 1939, după semnarea unui acord de încetare a focului cu Japonia, '
+                    'sovieticii au invadat Polonia dinspre est.'),
+
+        # Russian
+        param('ru', 'Втора́я мирова́я война́ (1 сентября 1939 — 2 сентября 1945) — '
+                    'война двух мировых военно-политических коалиций, ставшая крупнейшим вооружённым '
+                    'конфликтом в истории человечества.'),
+
+        # Spanish
+        param('es', '11 junio 2010'),
+
+        # Swedish
+        param('sv', ' den 15 augusti 1945 då Kejsardömet'),
+
+        # Thai
+        param('th',
+              'และเมื่อวันที่ 11 พฤษภาคม 1939 '
+              'ญี่ปุ่นตัดสินใจขยายพรมแดนญี่ปุ่น-มองโกเลียขึ้นไปถึงแม่น้ำคัลคินกอลด้วยกำลัง'),
+
+        # Turkish
+        param('tr', 'Almanya’nın Polonya’yı işgal ettiği 1 Eylül 1939 savaşın başladığı '
+                    'tarih olarak genel kabul görür.'),
+
+        # Ukrainian
+        param('uk', 'Інші дати, що розглядаються деякими авторами як дати початку війни: початок японської '
+                    'інтервенції в Маньчжурію 13 вересня 1931, початок другої японсько-китайської війни 7 '
+                    'липня 1937 року та початок угорсько-української війни 14 березня 1939 року.'),
+
+        # Vietnamese
+        param('vi', 'Ý theo gương Đức, đã tiến hành xâm lược Ethiopia năm 1935 và sát '
+                    'nhập Albania vào ngày 12 tháng 4 năm 1939.'),
+
+        # Only digits
+        param('en', '2007'),
+    ])
+    def test_detection(self, shortname, text):
+        result = self.search_with_detection.detect_language(text, languages=None)
+        self.assertEqual(result, shortname)
+
+    @parameterized.expand([
+        param(text='19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',
+              languages=['en', 'ru'],
+              settings=None,
+              expected=[('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
+                        ('20 марта', datetime.datetime(2001, 3, 20, 0, 0)),
+                        ('21 марта', datetime.datetime(2001, 3, 21, 0, 0))]),
+
+        param(text='Em outubro de 1936, Alemanha e Itália formaram o Eixo Roma-Berlim.',
+              languages=None,
+              settings={'RELATIVE_BASE': datetime.datetime(2000, 1, 1)},
+              expected=[('Em outubro de 1936', datetime.datetime(1936, 10, 1, 0, 0))]),
+
+        param(text='19 марта 2001, 20 марта, 21 марта был отличный день.',
+              languages=['en', 'ru'],
+              settings=None,
+              expected=[('19 марта 2001', datetime.datetime(2001, 3, 19, 0, 0)),
+                        ('20 марта', datetime.datetime(2001, 3, 20, 0, 0)),
+                        ('21 марта', datetime.datetime(2001, 3, 21, 0, 0))]),
+
+        # Dates not found
+        param(text='',
+              languages=None,
+              settings=None,
+              expected=None),
+
+        # Language not detected
+        param(text='Привет',
+              languages=['en'],
+              settings=None,
+              expected=None),
+
+        # ZeroDivisionError
+        param(text="DECEMBER 21 19.87 87",
+              languages=None,
+              settings=None,
+              expected=[('DECEMBER 21 19', datetime.datetime(2019, 12, 21, 0, 0))]
+              ),
+        param(text='bonjour, pouvez vous me joindre svp par telephone 08 11 58 54 41',
+              languages=None,
+              settings={'STRICT_PARSING': True},
+              expected=None),
+        param(text="a Americ",
+              languages=None,
+              settings=None,
+              expected=None),
+
+        # Date with comma and apostrophe
+        param(text="9/3/2017  , ",
+              languages=['en'],
+              settings=None,
+              expected=[('9/3/2017', datetime.datetime(2017, 9, 3, 0, 0))]),
+        param(text="9/3/2017  ' ",
+              languages=['en'],
+              settings=None,
+              expected=[('9/3/2017', datetime.datetime(2017, 9, 3, 0, 0))]),
+    ])
+    def test_date_search_function(self, text, languages, settings, expected):
+        result = search_dates(text, languages=languages, settings=settings)
+        self.assertEqual(result, expected)
+
+    @parameterized.expand([
+        param(text="15 de outubro de 1936",
+              add_detected_language=True,
+              expected=[
+                  ("15 de outubro de 1936", datetime.datetime(1936, 10, 15, 0, 0), "pt")
+              ]),
+        param(text="15 de outubro de 1936",
+              add_detected_language=False,
+              expected=[
+                  ("15 de outubro de 1936", datetime.datetime(1936, 10, 15, 0, 0))
+              ]),
+    ])
+    def test_search_dates_returning_detected_languages_if_requested(
+        self, text, add_detected_language, expected
+    ):
+        result = search_dates(text, add_detected_language=add_detected_language)
+        self.assertEqual(result, expected)
+
+    @parameterized.expand([
+        param(text='19 марта 2001',
+              languages='wrong type: str instead of list'),
+    ])
+    def test_date_search_function_invalid_languages_type(self, text, languages):
+        self.run_search_dates_function_invalid_languages(text=text, languages=languages, error_type=TypeError)
+        self.check_error_message("languages argument must be a list (<class \'str\'> given)")
+
+    @parameterized.expand([
+        param(text='19 марта 2001',
+              languages=['unknown language code']),
+    ])
+    def test_date_search_function_invalid_language_code(self, text, languages):
+        self.run_search_dates_function_invalid_languages(text=text, languages=languages, error_type=ValueError)
+        self.check_error_message("Unknown language(s): 'unknown language code'")

From b16930169e2cdcecbf74ee0372572f6b229d9299 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Wed, 31 Mar 2021 16:10:03 +0530
Subject: [PATCH 08/26] Update search.py

---
 dateparser/search/search.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index 58f4ebc10..848d65fed 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -89,8 +89,7 @@ def split_if_not_parsed(self, item, original):
     def parse_item(self, parser, item, translated_item, parsed, need_relative_base, language):
         relative_base = None
 
-        if language != "en":
-            item = item.replace('ngày', '')
+        if language == "de":
             item = item.replace('am', '')
 
         parsed_item = parser.get_date_data(item)

From fba58dd66f3d39b397715540d32617b061af5152 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 5 Apr 2021 20:22:37 +0530
Subject: [PATCH 09/26] Fixing the static date to
 datetime.datetime.utcnow().day

---
 tests/test_search.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 259a736bf..9be2d4c0d 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -464,7 +464,7 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                     May 31
                     8am UTC""",
               [('May 2020', datetime.datetime(2020, 5, datetime.datetime.utcnow().day, 0, 0)),
-               ('June 2020', datetime.datetime(2020, 6, 30, 0, 0)),
+               ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('2023', datetime.datetime(2023, 6, 30, 0, 0)),
                ('January UTC', datetime.datetime(2023, 1, 30, 0, 0, tzinfo=pytz.utc)),
                ('June 5 am utc', datetime.datetime(2023, 6, 30, 5, 0, tzinfo=pytz.utc)),

From 2aff2958ccc5d49f3738ebc4dff291847aacb575 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 5 Apr 2021 20:26:43 +0530
Subject: [PATCH 10/26] Update test_search.py

---
 tests/test_search.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 9be2d4c0d..75632a449 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -465,12 +465,12 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                     8am UTC""",
               [('May 2020', datetime.datetime(2020, 5, datetime.datetime.utcnow().day, 0, 0)),
                ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
-               ('2023', datetime.datetime(2023, 6, 30, 0, 0)),
-               ('January UTC', datetime.datetime(2023, 1, 30, 0, 0, tzinfo=pytz.utc)),
-               ('June 5 am utc', datetime.datetime(2023, 6, 30, 5, 0, tzinfo=pytz.utc)),
+               ('2023', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 0, 0)),
+               ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
+               ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
                ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
-               ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
+               ('8am UTC', datetime.datetime(2023, 5, datetime.datetime.utcnow().day, 8, 0, tzinfo=pytz.utc))]),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From 5d8e3d53c9acdedfbfebfb96c5c6945eec140e42 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 5 Apr 2021 20:32:37 +0530
Subject: [PATCH 11/26] Update test_search.py

---
 tests/test_search.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 75632a449..e3f1bf20f 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -467,10 +467,10 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('2023', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
-               ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
+               ('June 5 am utc', datetime.datetime(2023, 6, 5, 0, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
                ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
-               ('8am UTC', datetime.datetime(2023, 5, datetime.datetime.utcnow().day, 8, 0, tzinfo=pytz.utc))]),
+               ('8am UTC', datetime.datetime(2023, 8, 31, 0, 0, tzinfo=pytz.utc))]),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From c21e4283841b7e1830c9edb046f8919b7884b41d Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 5 Apr 2021 21:01:30 +0530
Subject: [PATCH 12/26] Update test_search.py

---
 tests/test_search.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index e3f1bf20f..e91e718fa 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -467,10 +467,10 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('2023', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
-               ('June 5 am utc', datetime.datetime(2023, 6, 5, 0, 0, tzinfo=pytz.utc)),
+               ('June 5 am utc', datetime.datetime(2023, 6, 5, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
                ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
-               ('8am UTC', datetime.datetime(2023, 8, 31, 0, 0, tzinfo=pytz.utc))]),
+               ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From 34f49b9432f0ebc027670d4b69ba6f71cfe1bfd2 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 12 Apr 2021 20:50:50 +0530
Subject: [PATCH 13/26] Update dateparser/search/search.py
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Co-authored-by: Marc Hernández <noviluni@gmail.com>
---
 dateparser/search/search.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index 848d65fed..495935e81 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -103,7 +103,7 @@ def parse_item(self, parser, item, translated_item, parsed, need_relative_base,
             parsed_item = parser.get_date_data(item)
         return parsed_item, is_relative
 
-    def parse_found_objects(self, parser, to_parse, original, translated, settings, language):
+    def parse_found_objects(self, parser, to_parse, original, translated, settings, language=None):
         parsed = []
         substrings = []
         need_relative_base = True

From 50c1d30c11187d164fecf6e5db5d03301f009ecd Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 12 Apr 2021 20:55:35 +0530
Subject: [PATCH 14/26] Update search.py

---
 dateparser/search/search.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index 495935e81..8cb7bb0a0 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -86,7 +86,7 @@ def split_if_not_parsed(self, item, original):
                 possible_splits.extend(self.split_by(item, original, splitter))
         return possible_splits
 
-    def parse_item(self, parser, item, translated_item, parsed, need_relative_base, language):
+    def parse_item(self, parser, item, translated_item, parsed, need_relative_base, language=None):
         relative_base = None
 
         if language == "de":

From 0563be42091e80dec364f294b90370bbd62edc91 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 12 Apr 2021 20:57:09 +0530
Subject: [PATCH 15/26] newlines removed

---
 dateparser/search/search.py | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index 8cb7bb0a0..5fa708572 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -154,10 +154,6 @@ def search_parse(self, shortname, text, settings):
         else:
             languages = [shortname]
             to_parse = original
-
-        
-
-
         parser = DateDataParser(languages=languages, settings=settings)
         parsed, substrings = self.parse_found_objects(parser=parser, to_parse=to_parse,
                                                       original=original, translated=translated, settings=settings, language=shortname)

From c0899074abe15831637047b16adef275071c1a1e Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 12 Apr 2021 21:07:28 +0530
Subject: [PATCH 16/26] Update test_search.py

---
 tests/test_search.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index e91e718fa..876b77636 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -467,7 +467,7 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('2023', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
-               ('June 5 am utc', datetime.datetime(2023, 6, 5, 5, 0, tzinfo=pytz.utc)),
+               ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
                ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
                ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),

From dbd12ec4d9405ee65627e1fdf40ba46bfd6b0707 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Mon, 12 Apr 2021 21:28:19 +0530
Subject: [PATCH 17/26] Adding comment

---
 dateparser/search/search.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index 5fa708572..bac50c0cf 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -89,8 +89,11 @@ def split_if_not_parsed(self, item, original):
     def parse_item(self, parser, item, translated_item, parsed, need_relative_base, language=None):
         relative_base = None
 
+        # Replacing language specific words with special meaning
         if language == "de":
             item = item.replace('am', '')
+        if language == "vi":
+            item = item.replace('ngày', '')
 
         parsed_item = parser.get_date_data(item)
         is_relative = date_is_relative(translated_item)

From 4343cd4c40201bca592d58cc007e8a3954838092 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 15:48:09 +0530
Subject: [PATCH 18/26] Fixing test from line break to comma

---
 tests/test_search.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 876b77636..5d188476d 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -461,8 +461,7 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                     January UTC
                     June 5 am utc
                     June 23th 5 pm EST
-                    May 31
-                    8am UTC""",
+                    May 31, 8am UTC""",
               [('May 2020', datetime.datetime(2020, 5, datetime.datetime.utcnow().day, 0, 0)),
                ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('2023', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 0, 0)),

From 2ff140e4f3ccf61ec6d6ce3cf2fbb5d5813457ab Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 18:08:39 +0530
Subject: [PATCH 19/26] Fixing tests

---
 tests/test_search.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 5d188476d..a1e4d4b2c 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -468,8 +468,7 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
                ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
-               ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
-               ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
+               ('May 31, 8am UTC', datetime.datetime(2023, 5, 31, 0, 0, tzinfo=pytz.utc)),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From 14bb4431431d54e3493d79dd0052f2fc67cd667d Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 18:14:12 +0530
Subject: [PATCH 20/26] Update test_search.py

---
 tests/test_search.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index a1e4d4b2c..338e07cdd 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -468,7 +468,7 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
                ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
-               ('May 31, 8am UTC', datetime.datetime(2023, 5, 31, 0, 0, tzinfo=pytz.utc)),
+               ('May 31, 8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc)),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From 840e2a31417a546df93d26949dd3efd7dd721fde Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 18:22:27 +0530
Subject: [PATCH 21/26] Update test_search.py

---
 tests/test_search.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 338e07cdd..8b404acec 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -468,7 +468,7 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
                ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
-               ('May 31, 8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc)),
+               ('May 31, 8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From 2f617765b0bd27499df5941a7d594719fbc6c1d9 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 19:24:53 +0530
Subject: [PATCH 22/26] Test's fixed

---
 tests/test_search.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 8b404acec..896ff72db 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -468,7 +468,8 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('January UTC', datetime.datetime(2023, 1, datetime.datetime.utcnow().day, 0, 0, tzinfo=pytz.utc)),
                ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
-               ('May 31, 8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
+               ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
+               ('8am UTC', datetime.datetime(2023, 8, 31, 0, 0, tzinfo=pytz.utc))]),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From 4810bc6ecec5828370003c93ff4d7c21b6167743 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 19:28:16 +0530
Subject: [PATCH 23/26] Using re.sub to replace

---
 dateparser/search/search.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/dateparser/search/search.py b/dateparser/search/search.py
index bac50c0cf..1a6ea79c8 100644
--- a/dateparser/search/search.py
+++ b/dateparser/search/search.py
@@ -91,9 +91,9 @@ def parse_item(self, parser, item, translated_item, parsed, need_relative_base,
 
         # Replacing language specific words with special meaning
         if language == "de":
-            item = item.replace('am', '')
+            item = re.sub(r'am (?=\d)', '', item)
         if language == "vi":
-            item = item.replace('ngày', '')
+            item = re.sub(r'ngày (?=\d)', '', item)
 
         parsed_item = parser.get_date_data(item)
         is_relative = date_is_relative(translated_item)

From 95fbe1d8da24eee7ff31fa343252de343a715f79 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 19:29:12 +0530
Subject: [PATCH 24/26] Fixing tests

---
 tests/test_search.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 896ff72db..601020620 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -469,7 +469,7 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
                ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
-               ('8am UTC', datetime.datetime(2023, 8, 31, 0, 0, tzinfo=pytz.utc))]),
+               ('8am UTC', datetime.datetime(2023, 8, 31, 8, 0, tzinfo=pytz.utc))]),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',

From 61bfe13dfa1cd696e6068ec73abb26836dcdaa78 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 19:32:44 +0530
Subject: [PATCH 25/26] Update test_search.py

---
 tests/test_search.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index 601020620..a227178af 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -461,7 +461,8 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                     January UTC
                     June 5 am utc
                     June 23th 5 pm EST
-                    May 31, 8am UTC""",
+                    May 31
+                    8am UTC""",
               [('May 2020', datetime.datetime(2020, 5, datetime.datetime.utcnow().day, 0, 0)),
                ('June 2020', datetime.datetime(2020, 6, datetime.datetime.utcnow().day, 0, 0)),
                ('2023', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 0, 0)),

From 98b6ba6306f6b427a55ba7d6f780edef4da47a30 Mon Sep 17 00:00:00 2001
From: Gavish <gavishpoddar@hotmail.com>
Date: Fri, 16 Apr 2021 20:30:43 +0530
Subject: [PATCH 26/26] Update test_search.py

---
 tests/test_search.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/test_search.py b/tests/test_search.py
index a227178af..876b77636 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -470,7 +470,7 @@ def test_search_and_parse(self, shortname, string, expected, settings=None):
                ('June 5 am utc', datetime.datetime(2023, 6, datetime.datetime.utcnow().day, 5, 0, tzinfo=pytz.utc)),
                ('June 23th 5 pm EST', datetime.datetime(2023, 6, 23, 17, 0, tzinfo=pytz.timezone("EST"))),
                ('May 31', datetime.datetime(2023, 5, 31, 0, 0)),
-               ('8am UTC', datetime.datetime(2023, 8, 31, 8, 0, tzinfo=pytz.utc))]),
+               ('8am UTC', datetime.datetime(2023, 5, 31, 8, 0, tzinfo=pytz.utc))]),
 
         # Russian
         param('ru', '19 марта 2001 был хороший день. 20 марта тоже был хороший день. 21 марта был отличный день.',