Merge pull request #187 from mozilla/feature/url-case

Add custom middleware and URL resolvers for ISO 3166 language prefixes.

donate/core/middleware.py

@@ -0,0 +1,63 @@
+from django.conf import settings
+from django.conf.urls.i18n import is_language_prefix_patterns_used
+from django.middleware.locale import LocaleMiddleware as BaseLocaleMiddleware
+from django.urls import get_script_prefix, is_valid_path
+from django.utils import translation
+from django.utils.cache import patch_vary_headers
+
+from .utils import get_language_from_request, language_code_to_iso_3166
+
+
+class LocaleMiddleware(BaseLocaleMiddleware):
+    """
+    A replacement for django.middleware.locale.LocaleMiddleware that
+    redirects to an ISO 3166 format language prefix (en-US) instead of
+    the form used by Django (en-us)
+    """
+
+    def process_request(self, request):
+        urlconf = getattr(request, 'urlconf', settings.ROOT_URLCONF)
+        i18n_patterns_used, prefixed_default_language = is_language_prefix_patterns_used(urlconf)
+        # We're using our version of get_language_from_request here.
+        language = get_language_from_request(request, check_path=i18n_patterns_used)
+        language_from_path = translation.get_language_from_path(request.path_info)
+        if not language_from_path and i18n_patterns_used and not prefixed_default_language:
+            language = settings.LANGUAGE_CODE
+        translation.activate(language)
+        request.LANGUAGE_CODE = translation.get_language()
+
+    def process_response(self, request, response):
+        language = translation.get_language()
+        # Rewrite language to the format we want it in.
+        language = language_code_to_iso_3166(language)
+        language_from_path = translation.get_language_from_path(request.path_info)
+        urlconf = getattr(request, 'urlconf', settings.ROOT_URLCONF)
+        i18n_patterns_used, prefixed_default_language = is_language_prefix_patterns_used(urlconf)
+        if (response.status_code == 404 and not language_from_path and
+                i18n_patterns_used and prefixed_default_language):
+            # Maybe the language code is missing in the URL? Try adding the
+            # language prefix and redirecting to that URL.
+            language_path = '/%s%s' % (language, request.path_info)
+            path_valid = is_valid_path(language_path, urlconf)
+            path_needs_slash = (
+                not path_valid and (
+                    settings.APPEND_SLASH and not language_path.endswith('/') and
+                    is_valid_path('%s/' % language_path, urlconf)
+                )
+            )
+
+            if path_valid or path_needs_slash:
+                script_prefix = get_script_prefix()
+                # Insert language after the script prefix and before the
+                # rest of the URL
+                language_url = request.get_full_path(force_append_slash=path_needs_slash).replace(
+                    script_prefix,
+                    '%s%s/' % (script_prefix, language),
+                    1
+                )
+                return self.response_redirect_class(language_url)
+
+        if not (i18n_patterns_used and language_from_path):
+            patch_vary_headers(response, ('Accept-Language',))
+        response.setdefault('Content-Language', language)
+        return response

donate/core/tests/test_middleware.py

@@ -0,0 +1,21 @@
+from django.http import HttpResponse
+from django.test import RequestFactory, TestCase
+from django.utils import translation
+
+from ..middleware import LocaleMiddleware
+
+
+class LocaleMiddlewareTestCase(TestCase):
+
+    def test_process_request_sets_iso_3166_language_code(self):
+        request = RequestFactory().get('/en-GB/')
+        LocaleMiddleware().process_request(request)
+        self.assertEqual(request.LANGUAGE_CODE, 'en-gb')
+
+    def test_process_response_redirects_to_iso_3166_language_code(self):
+        request = RequestFactory().get('/')
+        response = HttpResponse(status=404)
+        translation.activate('en-gb')
+        new_response = LocaleMiddleware().process_response(request, response)
+        self.assertEqual(new_response.status_code, 302)
+        self.assertEqual(new_response['Location'], '/en-GB/')

donate/core/tests/test_utils.py

@@ -0,0 +1,32 @@
+from django.test import RequestFactory, TestCase
+from django.utils import translation
+
+from ..utils import ISO3166LocalePrefixPattern, get_language_from_request, language_code_to_iso_3166
+
+
+class UtilsTestCase(TestCase):
+
+    def test_get_language_code_to_iso_3166(self):
+        self.assertEqual(language_code_to_iso_3166('en-gb'), 'en-GB')
+        self.assertEqual(language_code_to_iso_3166('en-us'), 'en-US')
+        self.assertEqual(language_code_to_iso_3166('fr'), 'fr')
+
+    def test_ISO3166LocalePrefixPattern(self):
+        translation.deactivate()
+        pattern = ISO3166LocalePrefixPattern(prefix_default_language=True)
+        # Pattern should not match lowercase URLs
+        self.assertIsNone(pattern.match('en-us/'))
+        # It should match this URL
+        self.assertTrue(pattern.match('en-US/'))
+
+    def test_get_language_from_request_returns_iso_3166_language(self):
+        request = RequestFactory().get('/')
+        request.META['HTTP_ACCEPT_LANGUAGE'] = 'en-GB,en;q=0.5'
+        language = get_language_from_request(request)
+        self.assertEqual(language, 'en-GB')
+
+    def test_get_language_from_request_fallback_language(self):
+        request = RequestFactory().get('/')
+        request.META['HTTP_ACCEPT_LANGUAGE'] = 'en-FO,en;q=0.5'
+        language = get_language_from_request(request)
+        self.assertEqual(language, 'en-US')

donate/core/utils.py

@@ -0,0 +1,81 @@
+from django.conf import settings
+from django.urls import LocalePrefixPattern, URLResolver
+from django.utils.translation.trans_real import (
+    check_for_language, get_languages, get_language_from_path,
+    get_supported_language_variant, parse_accept_lang_header, language_code_re
+)
+
+
+class ISO3166LocalePrefixPattern(LocalePrefixPattern):
+    """LocalePrefixPattern subclass that enforces URL prefixes in the form en-US"""
+
+    def match(self, path):
+        language_prefix = language_code_to_iso_3166(self.language_prefix)
+        if path.startswith(language_prefix):
+            return path[len(language_prefix):], (), {}
+        return None
+
+
+def i18n_patterns(*urls, prefix_default_language=True):
+    """
+    Replacement for django.conf.urls.i18_patterns that uses ISO3166LocalePrefixPattern
+    instead of django.urls.resolvers.LocalePrefixPattern.
+    """
+    if not settings.USE_I18N:
+        return list(urls)
+    return [
+        URLResolver(
+            ISO3166LocalePrefixPattern(prefix_default_language=prefix_default_language),
+            list(urls),
+        )
+    ]
+
+
+def language_code_to_iso_3166(language):
+    """Turn a language name (en-us) into an ISO 3166 format (en-US)."""
+    language, _, country = language.lower().partition('-')
+    if country:
+        return language + '-' + country.upper()
+    return language
+
+
+def get_language_from_request(request, check_path=False):
+    """
+    Replacement for django.utils.translation.get_language_from_request.
+    The portion of code that is modified is identified below with a comment.
+    """
+    if check_path:
+        lang_code = get_language_from_path(request.path_info)
+        if lang_code is not None:
+            return lang_code
+
+    lang_code = request.COOKIES.get(settings.LANGUAGE_COOKIE_NAME)
+    if lang_code is not None and lang_code in get_languages() and check_for_language(lang_code):
+        return lang_code
+
+    try:
+        return get_supported_language_variant(lang_code)
+    except LookupError:
+        pass
+
+    accept = request.META.get('HTTP_ACCEPT_LANGUAGE', '')
+    for accept_lang, unused in parse_accept_lang_header(accept):
+        if accept_lang == '*':
+            break
+
+        # Convert lowercase region to uppercase before attempting to find a variant.
+        # This is the only portion of code that is modified from the core function.
+        accept_lang = language_code_to_iso_3166(accept_lang)
+
+        if not language_code_re.search(accept_lang):
+            continue
+
+        try:
+            return get_supported_language_variant(accept_lang)
+        except LookupError:
+            continue
+
+    try:
+        return get_supported_language_variant(settings.LANGUAGE_CODE)
+    except LookupError:
+        return settings.LANGUAGE_CODE

donate/settings.py

@@ -104,7 +104,7 @@
     'django.middleware.security.SecurityMiddleware',
     'whitenoise.middleware.WhiteNoiseMiddleware',
     'django.contrib.sessions.middleware.SessionMiddleware',
-    'django.middleware.locale.LocaleMiddleware',
+    'donate.core.middleware.LocaleMiddleware',
     'django.middleware.common.CommonMiddleware',
     'django.middleware.csrf.CsrfViewMiddleware',
     'django.contrib.auth.middleware.AuthenticationMiddleware',
@@ -190,12 +190,16 @@
     },
 ]
 
-LANGUAGE_CODE = 'en-us'
+LANGUAGE_CODE = 'en-US'
 TIME_ZONE = 'UTC'
 USE_I18N = True
 USE_L10N = True
 USE_TZ = True
 
+LOCALE_PATHS = [
+    app('locale'),
+]
+
 LANGUAGES = [
     ('ach', 'Acholi'),
     ('ar', 'Arabic'),
@@ -209,16 +213,16 @@
     ('de', 'German'),
     ('dsb', 'Sorbian, Lower'),
     ('el', 'Greek'),
-    ('en-us', 'English (US)'),
-    ('en-ca', 'English (Canada)'),
-    ('en-gb', 'English (Great Britain)'),
+    ('en-US', 'English (US)'),
+    ('en-CA', 'English (Canada)'),
+    ('en-GB', 'English (Great Britain)'),
     ('es', 'Spanish'),
     ('et', 'Estonian'),
     ('fr', 'French'),
-    ('fy-nl', 'Frisian'),
-    ('gu-in', 'Gujarati'),
+    ('fy-NL', 'Frisian'),
+    ('gu-IN', 'Gujarati'),
     ('he', 'Hebrew'),
-    ('hi-in', 'Hindi'),
+    ('hi-IN', 'Hindi'),
     ('hr', 'Croatian'),
     ('hsb', 'Sorbian, Upper'),
     ('hu', 'Hungarian'),
@@ -234,26 +238,26 @@
     ('ms', 'Malay'),
     ('ml', 'Malayalam'),
     ('mr', 'Marathi'),
-    ('nb-no', 'Norwegian Bokmål'),
+    ('nb-NO', 'Norwegian Bokmål'),
     ('nl', 'Dutch'),
-    ('nn-no', 'Norwegian Nynorsk'),
+    ('nn-NO', 'Norwegian Nynorsk'),
     ('pl', 'Polish'),
-    ('pt-br', 'Portuguese (Brazil)'),
-    ('pt-pt', 'Portuguese (Portugal)'),
-    ('pa-in', 'Punjabi'),
+    ('pt-BR', 'Portuguese (Brazil)'),
+    ('pt-PT', 'Portuguese (Portugal)'),
+    ('pa-IN', 'Punjabi'),
     ('ro', 'Romanian'),
     ('ru', 'Russian'),
     ('sk', 'Slovak'),
     ('sl', 'Slovenian'),
     ('sq', 'Albanian'),
-    ('sv-se', 'Swedish'),
+    ('sv-SE', 'Swedish'),
     ('ta', 'Tamil'),
     ('te', 'Telugu'),
     ('th', 'Thai'),
     ('tr', 'Turkish'),
     ('uz', 'Uzbek'),
-    ('zh-cn', 'Chinese (China)'),
-    ('zh-tw', 'Chinese (Taiwan)'),
+    ('zh-CN', 'Chinese (China)'),
+    ('zh-TW', 'Chinese (Taiwan)'),
 ]
 
 STATICFILES_STORAGE = 'whitenoise.storage.CompressedManifestStaticFilesStorage'

donate/urls.py

@@ -1,6 +1,5 @@
 from django.apps import apps
 from django.conf import settings
-from django.conf.urls.i18n import i18n_patterns
 from django.contrib import admin
 from django.urls import include, path
 from django.views.decorators.cache import cache_page
@@ -12,6 +11,7 @@
 from wagtail.documents import urls as wagtaildocs_urls
 
 from donate.payments import urls as payments_urls
+from donate.core.utils import i18n_patterns
 
 # Patterns not subject to i18n
 urlpatterns = [