Merge pull request #95 from kuspia/master

deep_translator/constants.py

@@ -29,9 +29,8 @@ GOOGLE_CODES_TO_LANGUAGES = {
     'ca': 'catalan',
     'ceb': 'cebuano',
     'ny': 'chichewa',
-    'zh': 'chinese',
-    'zh-cn': 'chinese (simplified)',
-    'zh-tw': 'chinese (traditional)',
+    'zh-CN': 'chinese (simplified)',
+    'zh-TW': 'chinese (traditional)',
     'co': 'corsican',
     'hr': 'croatian',
     'cs': 'czech',
@@ -66,8 +65,9 @@ GOOGLE_CODES_TO_LANGUAGES = {
     'kn': 'kannada',
     'kk': 'kazakh',
     'km': 'khmer',
+    'rw': 'kinyarwanda',
     'ko': 'korean',
-    'ku': 'kurdish (kurmanji)',
+    'ku': 'kurdish',
     'ky': 'kyrgyz',
     'lo': 'lao',
     'la': 'latin',
@@ -82,9 +82,10 @@ GOOGLE_CODES_TO_LANGUAGES = {
     'mi': 'maori',
     'mr': 'marathi',
     'mn': 'mongolian',
-    'my': 'myanmar (burmese)',
+    'my': 'myanmar',
     'ne': 'nepali',
     'no': 'norwegian',
+    'or': 'odia',
     'ps': 'pashto',
     'fa': 'persian',
     'pl': 'polish',
@@ -108,11 +109,14 @@ GOOGLE_CODES_TO_LANGUAGES = {
     'sv': 'swedish',
     'tg': 'tajik',
     'ta': 'tamil',
+    'tt': 'tatar',
     'te': 'telugu',
     'th': 'thai',
     'tr': 'turkish',
+    'tk': 'turkmen',
     'uk': 'ukrainian',
     'ur': 'urdu',
+    'ug': 'uyghur',
     'uz': 'uzbek',
     'vi': 'vietnamese',
     'cy': 'welsh',
@@ -120,12 +124,18 @@ GOOGLE_CODES_TO_LANGUAGES = {
     'yi': 'yiddish',
     'yo': 'yoruba',
     'zu': 'zulu',
-    'fil': 'Filipino',
-    'he': 'Hebrew'
 }
 
 GOOGLE_LANGUAGES_TO_CODES = {v: k for k, v in GOOGLE_CODES_TO_LANGUAGES.items()}
 
+# This dictionary maps the primary name of language to its secondary names in list manner (if any)
+GOOGLE_LANGUAGES_SECONDARY_NAMES = {
+    'myanmar': ['burmese'],
+    'odia': ['oriya'],
+    'kurdish':  ['kurmanji']
+}
+
+
 PONS_CODES_TO_LANGUAGES = {
     'ar': 'arabic',
     'bg': 'bulgarian',
@@ -247,4 +257,4 @@ QCRI_CODE_TO_LANGUAGE = {
 
 QCRI_LANGUAGE_TO_CODE = {
     v: k for k, v in QCRI_CODE_TO_LANGUAGE.items()
-}
+}

deep_translator/google_trans.py

@@ -2,7 +2,7 @@
 google translator API
 """
 
-from .constants import BASE_URLS, GOOGLE_LANGUAGES_TO_CODES
+from .constants import BASE_URLS, GOOGLE_LANGUAGES_TO_CODES, GOOGLE_LANGUAGES_SECONDARY_NAMES
 from .exceptions import TooManyRequests, LanguageNotSupportedException, TranslationNotFound, NotValidPayload, RequestError
 from .parent import BaseTranslator
 from bs4 import BeautifulSoup
@@ -27,8 +27,19 @@ class GoogleTranslator(BaseTranslator):
         self.__base_url = BASE_URLS.get("GOOGLE_TRANSLATE")
         self.proxies = proxies
 
-        if self.is_language_supported(source, target):
-            self._source, self._target = self._map_language_to_code(source.lower(), target.lower())
+        # code snipppet that converts the language into lower-case and skip lower-case conversion for abbreviations
+        # since abbreviations like zh-CN if converted to lower-case will result into error
+        #######################################
+        source_lower = source
+        target_lower = target
+        if not source in self._languages.values():
+            source_lower=source.lower()
+        if not target in self._languages.values():
+            target_lower=target.lower()
+        #######################################
+
+        if self.is_language_supported(source_lower, target_lower):
+            self._source, self._target = self._map_language_to_code(source_lower, target_lower)
 
         super(GoogleTranslator, self).__init__(base_url=self.__base_url,
                                                source=self._source,
@@ -51,6 +62,17 @@ class GoogleTranslator(BaseTranslator):
         """
         return GoogleTranslator.supported_languages if not as_dict else GoogleTranslator._languages
 
+    def is_secondary(self, lang):
+        """
+        Function to check if lang is a secondary name of any primary language
+        @param lang: language name
+        @return: primary name of a language if found otherwise False
+        """
+        for primary_name, secondary_names in GOOGLE_LANGUAGES_SECONDARY_NAMES.items():
+            if lang in secondary_names:
+                return primary_name
+        return False
+
     def _map_language_to_code(self, *languages):
         """
         map language to its corresponding code (abbreviation) if the language was passed by its full name by the user
@@ -63,7 +85,7 @@ class GoogleTranslator(BaseTranslator):
             elif language in self._languages.keys():
                 yield self._languages[language]
             else:
-                raise LanguageNotSupportedException(language)
+                yield self._languages[self.is_secondary(language)]
 
     def is_language_supported(self, *languages):
         """
@@ -74,7 +96,8 @@ class GoogleTranslator(BaseTranslator):
         for lang in languages:
             if lang != 'auto' and lang not in self._languages.keys():
                 if lang != 'auto' and lang not in self._languages.values():
-                    raise LanguageNotSupportedException(lang)
+                    if self.is_secondary(lang) == False:
+                        raise LanguageNotSupportedException(lang)
         return True
 
     def translate(self, text, **kwargs):
@@ -181,8 +204,6 @@ class GoogleTranslator(BaseTranslator):
 
         return arr
 
-
-
 if __name__ == '__main__':
     translator = GoogleTranslator(source='ru', target='uk')
     t = translator.translate("Я разработчик") # => "I am a developer"

tests/test_data.py

@@ -0,0 +1,112 @@
+test_text_standard = 'Hello world.'
+
+TRANSLATED_RESULTS = {
+    "afrikaans": "Hello Wêreld.",
+    "albanian": "Përshendetje Botë.",
+    "amharic": "ሰላም ልዑል.",
+    "arabic": "مرحبا بالعالم.",
+    "armenian": "Բարեւ աշխարհ.",
+    "azerbaijani": "Salam dünya.",
+    "basque": "Kaixo Mundua.",
+    "belarusian": "Прывітанне Сусвет.",
+    "bengali": "ওহে বিশ্ব.",
+    "bosnian": "Zdravo svijete.",
+    "bulgarian": "Здравей свят.",
+    "catalan": "Hola món.",
+    "cebuano": "Kumusta kalibutan.",
+    "chichewa": "Moni Dziko Lapansi.",
+    "chinese (simplified)": "你好，世界。",
+    "chinese (traditional)": "你好，世界。",
+    "corsican": "Bonghjornu mondu.",
+    "croatian": "Pozdrav svijete.",
+    "czech": "Ahoj světe.",
+    "danish": "Hej Verden.",
+    "dutch": "Hallo Wereld.",
+    "esperanto": "Saluton mondo.",
+    "estonian": "Tere, Maailm.",
+    "filipino": "Kamusta mundo",
+    "finnish": "Hei maailma.",
+    "french": "Bonjour le monde.",
+    "frisian": "Hallo wrâld.",
+    "galician": "Ola mundo.",
+    "georgian": "Გამარჯობა მსოფლიო.",
+    "german": "Hallo Welt.",
+    "greek": "Γειά σου Κόσμε.",
+    "gujarati": "હેલો વર્લ્ડ.",
+    "haitian creole": "Bonjou mond.",
+    "hausa": "Sannu Duniya.",
+    "hawaiian": "Aloha honua.",
+    "hebrew": "שלום עולם.",
+    "hindi": "नमस्ते दुनिया।",
+    "hmong": "Nyob zoo ntiaj teb.",
+    "hungarian": "Helló Világ.",
+    "icelandic": "Halló heimur.",
+    "igbo": "Ndewo Ụwa.",
+    "indonesian": "Halo Dunia.",
+    "irish": "Dia duit ar domhan.",
+    "italian": "Ciao mondo.",
+    "japanese": "こんにちは世界。",
+    "javanese": "Halo jagad.",
+    "kannada": "ಹಲೋ ವಿಶ್ವ.",
+    "kazakh": "Сәлем Әлем.",
+    "khmer": "សួស្តី​ពិភពលោក។",
+    "kinyarwanda": "Mwaramutse isi.",
+    "korean": "안녕하세요 세계입니다.",
+    "kurdish": "Hello cîhanê.",
+    "kyrgyz": "Салам дүйнө.",
+    "lao": "ສະ​ບາຍ​ດີ​ຊາວ​ໂລກ.",
+    "latin": "Salve mundi.",
+    "latvian": "Sveika pasaule.",
+    "lithuanian": "Labas pasauli.",
+    "luxembourgish": "Moien Welt.",
+    "macedonian": "Здраво свету.",
+    "malagasy": "Hello World.",
+    "malay": "Hai dunia.",
+    "malayalam": "ഹലോ വേൾഡ്.",
+    "maltese": "Hello dinja.",
+    "maori": "Kia ora te ao.",
+    "marathi": "नमस्कार जग.",
+    "mongolian": "Сайн уу ертөнц.",
+    "myanmar": "မင်္ဂလာပါကမ္ဘာလောက။",
+    "nepali": "नमस्कार संसार।",
+    "norwegian": "Hei Verden.",
+    "odia": "ନମସ୍କାର ବିଶ୍ୱବାସି।",
+    "pashto": "سلام نړی.",
+    "persian": "سلام دنیا.",
+    "polish": "Witaj świecie.",
+    "portuguese": "Olá Mundo.",
+    "punjabi": "ਸਤਿ ਸ੍ਰੀ ਅਕਾਲ ਦੁਨਿਆ.",
+    "romanian": "Salut Lume.",
+    "russian": "Привет, мир.",
+    "samoan": "Talofa lalolagi.",
+    "scots gaelic": "Hàlo a Shaoghail.",
+    "serbian": "Здраво Свете.",
+    "sesotho": "Lefatše Lumela.",
+    "shona": "Mhoro nyika.",
+    "sindhi": "هيلو دنيا.",
+    "sinhala": "හෙලෝ වර්ල්ඩ්.",
+    "slovak": "Ahoj svet.",
+    "slovenian": "Pozdravljen, svet.",
+    "somali": "Salaamu calaykum.",
+    "spanish": "Hola Mundo.",
+    "sundanese": "Halo Dunya.",
+    "swahili": "Salamu, Dunia.",
+    "swedish": "Hej världen.",
+    "tajik": "Салом Ҷаҳон.",
+    "tamil": "வணக்கம் உலகம்.",
+    "tatar": "Сәлам, Дөнья.",
+    "telugu": "హలో వరల్డ్.",
+    "thai": "สวัสดีชาวโลก.",
+    "turkish": "Selam Dünya.",
+    "turkmen": "Salam dünýä.",
+    "ukrainian": "Привіт Світ.",
+    "urdu": "سلام دنیا۔",
+    "uyghur": "ياخشىمۇسىز دۇنيا.",
+    "uzbek": "Salom Dunyo.",
+    "vietnamese": "Chào thế giới.",
+    "welsh": "Helo Byd.",
+    "xhosa": "Molo Lizwe.",
+    "yiddish": "העלא וועלט.",
+    "yoruba": "Mo ki O Ile Aiye.",
+    "zulu": "Sawubona Mhlaba."
+}

tests/test_google_trans.py

@@ -4,23 +4,182 @@
 
 import pytest
 from deep_translator import exceptions, GoogleTranslator
+from deep_translator.constants import GOOGLE_CODES_TO_LANGUAGES
+#from test_data import test_text_standard, TRANSLATED_RESULTS
+import random
+
+test_text_standard = 'Hello world.'
+
+TRANSLATED_RESULTS = {
+    "afrikaans": "Hello Wêreld.",
+    "albanian": "Përshendetje Botë.",
+    "amharic": "ሰላም ልዑል.",
+    "arabic": "مرحبا بالعالم.",
+    "armenian": "Բարեւ աշխարհ.",
+    "azerbaijani": "Salam dünya.",
+    "basque": "Kaixo Mundua.",
+    "belarusian": "Прывітанне Сусвет.",
+    "bengali": "ওহে বিশ্ব.",
+    "bosnian": "Zdravo svijete.",
+    "bulgarian": "Здравей свят.",
+    "catalan": "Hola món.",
+    "cebuano": "Kumusta kalibutan.",
+    "chichewa": "Moni Dziko Lapansi.",
+    "chinese (simplified)": "你好，世界。",
+    "chinese (traditional)": "你好，世界。",
+    "corsican": "Bonghjornu mondu.",
+    "croatian": "Pozdrav svijete.",
+    "czech": "Ahoj světe.",
+    "danish": "Hej Verden.",
+    "dutch": "Hallo Wereld.",
+    "esperanto": "Saluton mondo.",
+    "estonian": "Tere, Maailm.",
+    "filipino": "Kamusta mundo",
+    "finnish": "Hei maailma.",
+    "french": "Bonjour le monde.",
+    "frisian": "Hallo wrâld.",
+    "galician": "Ola mundo.",
+    "georgian": "Გამარჯობა მსოფლიო.",
+    "german": "Hallo Welt.",
+    "greek": "Γειά σου Κόσμε.",
+    "gujarati": "હેલો વર્લ્ડ.",
+    "haitian creole": "Bonjou mond.",
+    "hausa": "Sannu Duniya.",
+    "hawaiian": "Aloha honua.",
+    "hebrew": "שלום עולם.",
+    "hindi": "नमस्ते दुनिया।",
+    "hmong": "Nyob zoo ntiaj teb.",
+    "hungarian": "Helló Világ.",
+    "icelandic": "Halló heimur.",
+    "igbo": "Ndewo Ụwa.",
+    "indonesian": "Halo Dunia.",
+    "irish": "Dia duit ar domhan.",
+    "italian": "Ciao mondo.",
+    "japanese": "こんにちは世界。",
+    "javanese": "Halo jagad.",
+    "kannada": "ಹಲೋ ವಿಶ್ವ.",
+    "kazakh": "Сәлем Әлем.",
+    "khmer": "សួស្តី​ពិភពលោក។",
+    "kinyarwanda": "Mwaramutse isi.",
+    "korean": "안녕하세요 세계입니다.",
+    "kurdish": "Hello cîhanê.",
+    "kyrgyz": "Салам дүйнө.",
+    "lao": "ສະ​ບາຍ​ດີ​ຊາວ​ໂລກ.",
+    "latin": "Salve mundi.",
+    "latvian": "Sveika pasaule.",
+    "lithuanian": "Labas pasauli.",
+    "luxembourgish": "Moien Welt.",
+    "macedonian": "Здраво свету.",
+    "malagasy": "Hello World.",
+    "malay": "Hai dunia.",
+    "malayalam": "ഹലോ വേൾഡ്.",
+    "maltese": "Hello dinja.",
+    "maori": "Kia ora te ao.",
+    "marathi": "नमस्कार जग.",
+    "mongolian": "Сайн уу ертөнц.",
+    "myanmar": "မင်္ဂလာပါကမ္ဘာလောက။",
+    "nepali": "नमस्कार संसार।",
+    "norwegian": "Hei Verden.",
+    "odia": "ନମସ୍କାର ବିଶ୍ୱବାସି।",
+    "pashto": "سلام نړی.",
+    "persian": "سلام دنیا.",
+    "polish": "Witaj świecie.",
+    "portuguese": "Olá Mundo.",
+    "punjabi": "ਸਤਿ ਸ੍ਰੀ ਅਕਾਲ ਦੁਨਿਆ.",
+    "romanian": "Salut Lume.",
+    "russian": "Привет, мир.",
+    "samoan": "Talofa lalolagi.",
+    "scots gaelic": "Hàlo a Shaoghail.",
+    "serbian": "Здраво Свете.",
+    "sesotho": "Lefatše Lumela.",
+    "shona": "Mhoro nyika.",
+    "sindhi": "هيلو دنيا.",
+    "sinhala": "හෙලෝ වර්ල්ඩ්.",
+    "slovak": "Ahoj svet.",
+    "slovenian": "Pozdravljen, svet.",
+    "somali": "Salaamu calaykum.",
+    "spanish": "Hola Mundo.",
+    "sundanese": "Halo Dunya.",
+    "swahili": "Salamu, Dunia.",
+    "swedish": "Hej världen.",
+    "tajik": "Салом Ҷаҳон.",
+    "tamil": "வணக்கம் உலகம்.",
+    "tatar": "Сәлам, Дөнья.",
+    "telugu": "హలో వరల్డ్.",
+    "thai": "สวัสดีชาวโลก.",
+    "turkish": "Selam Dünya.",
+    "turkmen": "Salam dünýä.",
+    "ukrainian": "Привіт Світ.",
+    "urdu": "سلام دنیا۔",
+    "uyghur": "ياخشىمۇسىز دۇنيا.",
+    "uzbek": "Salom Dunyo.",
+    "vietnamese": "Chào thế giới.",
+    "welsh": "Helo Byd.",
+    "xhosa": "Molo Lizwe.",
+    "yiddish": "העלא וועלט.",
+    "yoruba": "Mo ki O Ile Aiye.",
+    "zulu": "Sawubona Mhlaba."
+}
 
 
 @pytest.fixture
 def google_translator():
     """Sample pytest fixture.
-
     See more at: http://doc.pytest.org/en/latest/fixture.html
     """
     return GoogleTranslator(target='en')
 
 
+def case_sensitivity_checks():
+    test_lang = 'Czech'
+    test_text = 'Hi, the sky is dark while the moon is white. Hurrah!! Denver is a city name in Colorado.'
+    translated_text = 'Ahoj, obloha je tmavá, zatímco měsíc je bílý. Hurá!! Denver je název města v Coloradu.'
+    test_cases = []
+    n = len(test_lang)
+    mx = 1 << n
+    test = test_lang.lower()
+    for i in range(mx):
+        combination = [k for k in test_lang]
+        for j in range(n):
+            if (((i >> j) & 1) == 1):
+                combination[j] = test_lang[j].upper()
+        temp = ""
+        for i in combination:
+            temp += i
+        test_cases.append(temp)
+    random_cases = 5
+    random_test_cases = random.sample(test_cases, random_cases)  # randomly choosing any five cases since list is in order of 2^n containing all cases
+    for case in random_test_cases:
+        assert GoogleTranslator(source='en', target=case).translate(test_text) == translated_text
+
+def multiple_names_lang_checks():
+    assert GoogleTranslator(source='en', target='burMeSe').translate("Hello") == 'မင်္ဂလာပါ'
+    assert GoogleTranslator(source='en', target='Oriya').translate("What's up?") == 'କଣ ଚାଲିଛି?'
+    assert GoogleTranslator(source='en', target='kurManJi').translate("Nice is dice.") == 'Xweş xweş e.'
+
+def test_random_tranlations_cases_multiple_names():
+    random_sample_size = 5
+    d = dict.fromkeys(list(TRANSLATED_RESULTS.keys()))
+    random_lang_names = random.sample(d.keys(), random_sample_size)
+    random_subset_dict = {k: TRANSLATED_RESULTS[k] for k in random_lang_names}
+    for lang, translation in random_subset_dict.items():
+        assert GoogleTranslator(source='en', target=lang).translate(test_text_standard) == translation
+
+    case_sensitivity_checks()
+    multiple_names_lang_checks()
+
 def test_content(google_translator):
     """Sample pytest test function with the pytest fixture as an argument."""
     # from bs4 import BeautifulSoup
     # assert 'GitHub' in BeautifulSoup(response.content).title.string
     assert google_translator.translate(text='좋은') == "good"
 
+def test_abbreviations_and_languages_mapping():
+    for abb, lang in GOOGLE_CODES_TO_LANGUAGES.items():
+        if(abb!= 'en'):
+            g1 = GoogleTranslator(abb)
+            g2 = GoogleTranslator(lang)
+            assert g1._source == g2._source
 
 def test_inputs():
     with pytest.raises(exceptions.LanguageNotSupportedException):
@@ -29,13 +188,6 @@ def test_inputs():
     with pytest.raises(exceptions.LanguageNotSupportedException):
         GoogleTranslator(source="auto", target="nothing")
 
-    # test abbreviations and languages
-    g1 = GoogleTranslator("en", "fr")
-    g2 = GoogleTranslator("english", "french")
-    assert g1._source == g2._source
-    assert g1._target == g2._target
-
-
 def test_payload(google_translator):
 
     with pytest.raises(exceptions.NotValidPayload):
@@ -59,6 +211,5 @@ def test_payload(google_translator):
     #for _ in range(1):
     #assert google_translator.translate(text='좋은') == "good"
 
-
 def test_one_character_words():
     assert GoogleTranslator(source='es', target='en').translate('o') == 'or'