cloudlanguagetools 11.4.0__tar.gz → 11.5.0__tar.gz

{cloudlanguagetools-11.4.0 → cloudlanguagetools-11.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cloudlanguagetools
-Version: 11.4.0
+Version: 11.5.0
 Summary: Interface with various cloud APIs for language processing such as translation, text to speech
 Home-page: https://github.com/Language-Tools/cloud-language-tools-core
 Author: Luc

{cloudlanguagetools-11.4.0 → cloudlanguagetools-11.5.0}/cloudlanguagetools/azure.py

@@ -186,7 +186,7 @@ class AzureTranslationLanguage(cloudlanguagetools.translationlanguage.Translatio
         return self.language_id
 
 class AzureTransliterationLanguage(cloudlanguagetools.transliterationlanguage.TransliterationLanguage):
-    def __init__(self, language_id, from_script, to_script, from_script_name, to_script_name):
+    def __init__(self, language_id, from_script, to_script, from_script_name, to_script_name, from_native_name, to_native_name):
         self.service = cloudlanguagetools.constants.Service.Azure
         self.service_fee = cloudlanguagetools.constants.ServiceFee.paid
         self.language_id = language_id
@@ -195,13 +195,15 @@ class AzureTransliterationLanguage(cloudlanguagetools.transliterationlanguage.Tr
         self.to_script = to_script
         self.from_script_name = from_script_name
         self.to_script_name = to_script_name
+        self.from_native_name = from_native_name
+        self.to_native_name = to_native_name
 
     def get_transliteration_name(self):
-        result = f'{self.language.lang_name} ({self.from_script_name} to {self.to_script_name}), {self.service.name}'
+        result = f'{self.language.lang_name} ({self.from_script_name}/{self.from_native_name} to {self.to_script_name}/{self.to_native_name}), {self.service.name}'
         return result
 
     def get_transliteration_shortname(self):
-        result = f'{self.from_script_name} to {self.to_script_name}, {self.service.name}'
+        result = f'{self.from_script_name}/{self.from_native_name} to {self.to_script_name}/{self.to_native_name}, {self.service.name}'
         return result        
 
     def get_transliteration_key(self):
@@ -402,16 +404,24 @@ class AzureService(cloudlanguagetools.service.Service):
         result = []
         azure_data = self.get_supported_languages()
         for language_id, data in azure_data['transliteration'].items():
-            # get the first script
-            first_script = data['scripts'][0]
-            from_script =  first_script['code']
-            to_script = first_script['toScripts'][0]['code']
-            from_script_name = first_script['name']
-            to_script_name = first_script['toScripts'][0]['name']
-            # print(language_id, from_script, to_script)
-            # assert(to_script == 'Latn')
             try:
-                result.append(AzureTransliterationLanguage(language_id, from_script, to_script, from_script_name, to_script_name))
+                # get the first script
+                for from_script_data in data['scripts']:
+                    from_script =  from_script_data['code']
+                    from_native_name = from_script_data['nativeName']
+                    for to_script_data in from_script_data['toScripts']:
+                        to_script = to_script_data['code']
+                        from_script_name = from_script_data['name']
+                        to_script_name = to_script_data['name']
+                        to_native_name = to_script_data['nativeName']
+                    result.append(AzureTransliterationLanguage(
+                        language_id, 
+                        from_script, 
+                        to_script, 
+                        from_script_name, 
+                        to_script_name,
+                        from_native_name,
+                        to_native_name))
             except KeyError:
                 logging.error(f'could not process transliteration language for {language_id}, {data}', exc_info=True)
         return result

{cloudlanguagetools-11.4.0 → cloudlanguagetools-11.5.0}/cloudlanguagetools.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cloudlanguagetools
-Version: 11.4.0
+Version: 11.5.0
 Summary: Interface with various cloud APIs for language processing such as translation, text to speech
 Home-page: https://github.com/Language-Tools/cloud-language-tools-core
 Author: Luc

{cloudlanguagetools-11.4.0 → cloudlanguagetools-11.5.0}/setup.py

@@ -6,7 +6,7 @@ from setuptools.command.install import install
 # twine upload dist/*
 
 setup(name='cloudlanguagetools',
-      version='11.4.0',
+      version='11.5.0',
       description='Interface with various cloud APIs for language processing such as translation, text to speech',
       long_description=open('README.rst', encoding='utf-8').read(),
       url='https://github.com/Language-Tools/cloud-language-tools-core',

{cloudlanguagetools-11.4.0 → cloudlanguagetools-11.5.0}/tests/test_audio.py

@@ -143,6 +143,11 @@ class TestAudio(unittest.TestCase):
         # logging.info(f'verify_service_audio: service: {service} audio_language: {audio_language}')
         voices = self.get_voice_list_service_audio_language(service, audio_language)
         self.assertGreaterEqual(len(voices), 1, f'at least one voice for service {service}, language {audio_language}')
+
+        if service == Service.Google:
+            # exclude Journey voices, they don't use the standard interface
+            voices = [x for x in voices if 'Journey' not in x['voice_name']]
+
         # pick 3 random voices
         max_voices = 3
         if len(voices) > max_voices:
@@ -179,7 +184,8 @@ class TestAudio(unittest.TestCase):
         # pprint.pprint(mandarin_azure_voices)
 
         xiaochen = [x for x in mandarin_azure_voices if 'Xiaochen' in x.name]
-        self.assertTrue(len(xiaochen) == 2) # there is a regular and a multilingual
+        self.assertEqual(len(xiaochen), 3, str(xiaochen)) # there is a regular and a multilingual
+        # and also a DragonHD
 
         xiaochen_single_language = [x for x in xiaochen if len(x.audio_languages) == 1][0]
         self.assertEquals(xiaochen_single_language.audio_languages, [AudioLanguage.zh_CN])
@@ -257,7 +263,7 @@ class TestAudio(unittest.TestCase):
         mandarin_azure_voices = [x for x in azure_voices if AudioLanguage.zh_CN in x.audio_languages]
 
         xiaochen = [x for x in mandarin_azure_voices if 'Xiaochen' in x.name]
-        self.assertTrue(len(xiaochen) == 2) # there is a regular and a multilingual
+        self.assertEqual(len(xiaochen), 3) # there is a regular and a multilingual, and dragonhd
 
         xiaochen_single_language = [x for x in xiaochen if len(x.audio_languages) == 1][0]
         xiaochen_multilingual = [x for x in xiaochen if len(x.audio_languages) > 1][0]
@@ -526,7 +532,7 @@ class TestAudio(unittest.TestCase):
 
     def test_google_format_wav(self):
         fr_voice = self.get_voice_by_lambda(Service.Google, 
-            lambda x: AudioLanguage.fr_FR in x.audio_languages, assert_unique=False)
+            lambda x: AudioLanguage.fr_FR in x.audio_languages and 'Journey' not in x.name, assert_unique=False)
         self.verify_wav_voice(fr_voice, self.FRENCH_INPUT_TEXT, 'fr-FR')
 
     def test_naver_format_wav(self):

{cloudlanguagetools-11.4.0 → cloudlanguagetools-11.5.0}/tests/test_translation.py

@@ -226,14 +226,41 @@ class TestTranslation(unittest.TestCase):
         self.assertTrue
         self.assertEqual(result['Watson'], "<b><span style=\"font-weight: 400;\">What's wrong with you?</span></b>")
 
-    def test_transliteration(self):
-        # pytest test_translation.py -k test_transliteration
+    def test_transliteration_azure_custom(self):
+        # pytest tests/test_translation.py -k test_transliteration_azure_custom
+        
+        # chinese
+        source_text = '成本很低'
+        service = 'Azure'
+        # transliteration_key = transliteration_option['transliteration_key']
+        transliteration_key = {
+            'language_id': 'zh-Hans', 
+            'from_script': 'Hans', 
+            'to_script': 'Latn',
+        }
+        result = self.manager.get_transliteration(source_text, service, transliteration_key)
+        self.assertIn(result, ['chéng běn hěn dī', 'chéngběn hěndī'])
+
+        transliteration_key = {
+            'language_id': 'zh-Hant', 
+            'from_script': 'Hans', 
+            'to_script': 'Hant',
+        }
+        result = self.manager.get_transliteration('讲话', service, transliteration_key)
+        self.assertEqual(result, '講話')
+
+    def test_transliteration_azure(self):
+        # pytest tests/test_translation.py -k test_transliteration_azure
         
         # chinese
         source_text = '成本很低'
         from_language = Language.zh_cn.name
         service = 'Azure'
-        transliteration_candidates = [x for x in self.transliteration_language_list if x['language_code'] == from_language and x['service'] == service]
+        transliteration_candidates = [
+            x for x in self.transliteration_language_list 
+            if x['language_code'] == from_language 
+            and x['service'] == service
+            and 'to Latin' in x['transliteration_shortname']]
         self.assertTrue(len(transliteration_candidates) == 1)
         transliteration_option = transliteration_candidates[0]
         service = transliteration_option['service']