Merge pull request #184 from edx/clintonb/language-index

Language tag updates

course_discovery/apps/api/tests/test_serializers.py

@@ -347,7 +347,7 @@ class CourseRunSearchSerializerTests(TestCase):
         return d.strftime('%Y-%m-%dT%H:%M:%S') if d else None
 
     def serialize_language(self, language):
-        return language.name
+        return language.macrolanguage
 
     def test_data(self):
         course_run = CourseRunFactory()

course_discovery/apps/course_metadata/search_indexes.py

@@ -85,7 +85,7 @@ class CourseRunIndex(BaseCourseIndex, indexes.Indexable):
     image_url = indexes.CharField(model_attr='image_url', null=True)
 
     def _prepare_language(self, language):
-        return language.name
+        return language.macrolanguage
 
     def prepare_language(self, obj):
         if obj.language:

course_discovery/apps/ietf_language_tags/migrations/0002_language_tag_data_migration.py

@@ -7,126 +7,126 @@ from django.db import migrations, models
 LANGTAGS = (
     ("Afrikaans", "af"),
     ("Albanian", "sq"),
-    ("Arabic – Algeria", "ar-dz"),
-    ("Arabic – Bahrain", "ar-bh"),
-    ("Arabic – Egypt", "ar-eg"),
-    ("Arabic – Iraq", "ar-iq"),
-    ("Arabic – Jordan", "ar-jo"),
-    ("Arabic – Kuwait", "ar-kw"),
-    ("Arabic – Lebanon", "ar-lb"),
-    ("Arabic – Libya", "ar-ly"),
-    ("Arabic – Morocco", "ar-ma"),
-    ("Arabic – Oman", "ar-om"),
-    ("Arabic – Qatar", "ar-qa"),
-    ("Arabic – Saudi Arabia", "ar-sa"),
-    ("Arabic – Syria", "ar-sy"),
-    ("Arabic – Tunisia", "ar-tn"),
-    ("Arabic – United Arab Emirates", "ar-ae"),
-    ("Arabic – Yemen", "ar-ye"),
+    ("Arabic - Algeria", "ar-dz"),
+    ("Arabic - Bahrain", "ar-bh"),
+    ("Arabic - Egypt", "ar-eg"),
+    ("Arabic - Iraq", "ar-iq"),
+    ("Arabic - Jordan", "ar-jo"),
+    ("Arabic - Kuwait", "ar-kw"),
+    ("Arabic - Lebanon", "ar-lb"),
+    ("Arabic - Libya", "ar-ly"),
+    ("Arabic - Morocco", "ar-ma"),
+    ("Arabic - Oman", "ar-om"),
+    ("Arabic - Qatar", "ar-qa"),
+    ("Arabic - Saudi Arabia", "ar-sa"),
+    ("Arabic - Syria", "ar-sy"),
+    ("Arabic - Tunisia", "ar-tn"),
+    ("Arabic - United Arab Emirates", "ar-ae"),
+    ("Arabic - Yemen", "ar-ye"),
     ("Armenian", "hy"),
-    ("Azeri – Latin", "az-az"),
+    ("Azeri - Latin", "az-az"),
     ("Basque (Basque)", "eu"),
     ("Belarusian", "be"),
     ("Bulgarian", "bg"),
     ("Catalan", "ca"),
-    ("Chinese – China", "zh-cn"),
-    ("Chinese – Hong Kong SAR", "zh-hk"),
-    ("Chinese – Macau SAR", "zh-mo"),
-    ("Chinese – Singapore", "zh-sg"),
-    ("Chinese – Taiwan", "zh-tw"),
+    ("Chinese - China", "zh-cn"),
+    ("Chinese - Hong Kong SAR", "zh-hk"),
+    ("Chinese - Macau SAR", "zh-mo"),
+    ("Chinese - Singapore", "zh-sg"),
+    ("Chinese - Taiwan", "zh-tw"),
     ("Croatian", "hr"),
     ("Czech", "cs"),
     ("Danish", "da"),
-    ("Dutch – Belgium", "nl-be"),
-    ("Dutch – Netherlands", "nl-nl"),
-    ("English – Australia", "en-au"),
-    ("English – Belize", "en-bz"),
-    ("English – Canada", "en-ca"),
-    ("English – Caribbean", "en-cb"),
-    ("English – India", "en-in"),
-    ("English – Ireland", "en-ie"),
-    ("English – Jamaica", "en-jm"),
-    ("English – Malaysia", "en-my"),
-    ("English – New Zealand", "en-nz"),
-    ("English – Phillippines", "en-ph"),
-    ("English – Singapore", "en-sg"),
-    ("English – Southern Africa", "en-za"),
-    ("English – Trinidad", "en-tt"),
-    ("English – Great Britain", "en-gb"),
-    ("English – United States", "en-us"),
-    ("English – Zimbabwe", "en-zw"),
+    ("Dutch - Belgium", "nl-be"),
+    ("Dutch - Netherlands", "nl-nl"),
+    ("English - Australia", "en-au"),
+    ("English - Belize", "en-bz"),
+    ("English - Canada", "en-ca"),
+    ("English - Caribbean", "en-cb"),
+    ("English - India", "en-in"),
+    ("English - Ireland", "en-ie"),
+    ("English - Jamaica", "en-jm"),
+    ("English - Malaysia", "en-my"),
+    ("English - New Zealand", "en-nz"),
+    ("English - Phillippines", "en-ph"),
+    ("English - Singapore", "en-sg"),
+    ("English - Southern Africa", "en-za"),
+    ("English - Trinidad", "en-tt"),
+    ("English - Great Britain", "en-gb"),
+    ("English - United States", "en-us"),
+    ("English - Zimbabwe", "en-zw"),
     ("Estonian", "et"),
     ("Farsi", "fa"),
     ("Finnish", "fi"),
     ("Faroese", "fo"),
-    ("French – France", "fr-fr"),
-    ("French – Belgium", "fr-be"),
-    ("French – Canada", "fr-ca"),
-    ("French – Luxembourg", "fr-lu"),
-    ("French – Switzerland", "fr-ch"),
-    ("Irish – Ireland", "gd-ie"),
-    ("Scottish Gaelic – United Kingdom", "gd"),
-    ("German – Germany", "de-de"),
-    ("German – Austria", "de-at"),
-    ("German – Liechtenstein", "de-li"),
-    ("German – Luxembourg", "de-lu"),
-    ("German – Switzerland", "de-ch"),
+    ("French - France", "fr-fr"),
+    ("French - Belgium", "fr-be"),
+    ("French - Canada", "fr-ca"),
+    ("French - Luxembourg", "fr-lu"),
+    ("French - Switzerland", "fr-ch"),
+    ("Irish - Ireland", "gd-ie"),
+    ("Scottish Gaelic - United Kingdom", "gd"),
+    ("German - Germany", "de-de"),
+    ("German - Austria", "de-at"),
+    ("German - Liechtenstein", "de-li"),
+    ("German - Luxembourg", "de-lu"),
+    ("German - Switzerland", "de-ch"),
     ("Greek", "el"),
     ("Hebrew", "he"),
     ("Hindi", "hi"),
     ("Hungarian", "hu"),
     ("Icelandic", "is"),
     ("Indonesian", "id"),
-    ("Italian – Italy", "it-it"),
-    ("Italian – Switzerland", "it-ch"),
+    ("Italian - Italy", "it-it"),
+    ("Italian - Switzerland", "it-ch"),
     ("Japanese", "ja"),
     ("Korean", "ko"),
     ("Latvian", "lv"),
     ("Lithuanian", "lt"),
     ("F.Y.R.O. Macedonia", "mk"),
-    ("Malay – Malaysia", "ms-my"),
-    ("Malay – Brunei", "ms-bn"),
+    ("Malay - Malaysia", "ms-my"),
+    ("Malay - Brunei", "ms-bn"),
     ("Maltese", "mt"),
     ("Marathi", "mr"),
-    ("Norwegian – Bokmål", "nb-no"),
-    ("Norwegian – Nynorsk", "nn-no"),
+    ("Norwegian - Bokmål", "nb-no"),
+    ("Norwegian - Nynorsk", "nn-no"),
     ("Polish", "pl"),
-    ("Portuguese – Portugal", "pt-pt"),
-    ("Portuguese – Brazil", "pt-br"),
+    ("Portuguese - Portugal", "pt-pt"),
+    ("Portuguese - Brazil", "pt-br"),
     ("Raeto-Romance", "rm"),
-    ("Romanian – Romania", "ro"),
-    ("Romanian – Republic of Moldova", "ro-mo"),
+    ("Romanian - Romania", "ro"),
+    ("Romanian - Republic of Moldova", "ro-mo"),
     ("Russian", "ru"),
-    ("Russian – Republic of Moldova", "ru-mo"),
+    ("Russian - Republic of Moldova", "ru-mo"),
     ("Sanskrit", "sa"),
-    ("Serbian – Latin", "sr-sp"),
+    ("Serbian - Latin", "sr-sp"),
     ("Setsuana", "tn"),
     ("Slovenian", "sl"),
     ("Slovak", "sk"),
     ("Sorbian", "sb"),
-    ("Spanish – Spain (Modern)", "es-es"),
-    ("Spanish – Argentina", "es-ar"),
-    ("Spanish – Bolivia", "es-bo"),
-    ("Spanish – Chile", "es-cl"),
-    ("Spanish – Colombia", "es-co"),
-    ("Spanish – Costa Rica", "es-cr"),
-    ("Spanish – Dominican Republic", "es-do"),
-    ("Spanish – Ecuador", "es-ec"),
-    ("Spanish – Guatemala", "es-gt"),
-    ("Spanish – Honduras", "es-hn"),
-    ("Spanish – Mexico", "es-mx"),
-    ("Spanish – Nicaragua", "es-ni"),
-    ("Spanish – Panama", "es-pa"),
-    ("Spanish – Peru", "es-pe"),
-    ("Spanish – Puerto Rico", "es-pr"),
-    ("Spanish – Paraguay", "es-py"),
-    ("Spanish – El Salvador", "es-sv"),
-    ("Spanish – Uruguay", "es-uy"),
-    ("Spanish – Venezuela", "es-ve"),
+    ("Spanish - Spain (Modern)", "es-es"),
+    ("Spanish - Argentina", "es-ar"),
+    ("Spanish - Bolivia", "es-bo"),
+    ("Spanish - Chile", "es-cl"),
+    ("Spanish - Colombia", "es-co"),
+    ("Spanish - Costa Rica", "es-cr"),
+    ("Spanish - Dominican Republic", "es-do"),
+    ("Spanish - Ecuador", "es-ec"),
+    ("Spanish - Guatemala", "es-gt"),
+    ("Spanish - Honduras", "es-hn"),
+    ("Spanish - Mexico", "es-mx"),
+    ("Spanish - Nicaragua", "es-ni"),
+    ("Spanish - Panama", "es-pa"),
+    ("Spanish - Peru", "es-pe"),
+    ("Spanish - Puerto Rico", "es-pr"),
+    ("Spanish - Paraguay", "es-py"),
+    ("Spanish - El Salvador", "es-sv"),
+    ("Spanish - Uruguay", "es-uy"),
+    ("Spanish - Venezuela", "es-ve"),
     ("Southern Sotho", "st"),
     ("Swahili", "sw"),
-    ("Swedish – Sweden", "sv-se"),
-    ("Swedish – Finland", "sv-fi"),
+    ("Swedish - Sweden", "sv-se"),
+    ("Swedish - Finland", "sv-fi"),
     ("Tamil", "ta"),
     ("Tatar", "tt"),
     ("Thai", "th"),
@@ -134,7 +134,7 @@ LANGTAGS = (
     ("Tsonga", "ts"),
     ("Ukrainian", "uk"),
     ("Urdu", "ur"),
-    ("Uzbek – Latin", "uz-uz"),
+    ("Uzbek - Latin", "uz-uz"),
     ("Vietnamese", "vi"),
     ("Xhosa", "xh"),
     ("Yiddish", "yi"),

course_discovery/apps/ietf_language_tags/migrations/0003_fix_language_tag_names.py

+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+from django.db import migrations
+
+
+def fix_tag_names(apps, schema_editor):
+    """ Replace dashes (—) in tag names with hyphens (-)."""
+    LanguageTag = apps.get_model('ietf_language_tags', 'LanguageTag')
+
+    for tag in LanguageTag.objects.all():
+        tag.name = tag.name.replace('—', '-')
+        tag.save()
+
+
+class Migration(migrations.Migration):
+    dependencies = [
+        ('ietf_language_tags', '0002_language_tag_data_migration'),
+    ]
+
+    operations = [
+        migrations.RunPython(fix_tag_names),
+    ]

course_discovery/apps/ietf_language_tags/models.py

@@ -10,3 +10,7 @@ class LanguageTag(models.Model):
 
     def __str__(self):
         return '{code} - {name}'.format(code=self.code, name=self.name)
+
+    @property
+    def macrolanguage(self):
+        return self.name.split('-')[0].strip()

course_discovery/apps/ietf_language_tags/tests/test_models.py

@@ -15,3 +15,8 @@ class LanguageTagTests(TestCase):
         name = 'Test LanguageTag'
         tag = LanguageTag(code=code, name=name)
         self.assertEqual(str(tag), '{code} - {name}'.format(code=code, name=name))
+
+    def test_macrolanguage(self):
+        """ Verify the property returns the macrolanguage for a given LanguageTag. """
+        en_us = LanguageTag(code='en-us', name='English - United States')
+        self.assertEqual(en_us.macrolanguage, 'English')