Skip to content

armgit / wger

Public Imported from GitHub
0

Self hosted FLOSS fitness/workout, nutrition and weight tracker

Add endpoint to check the language of inputs

Roland Geider2d72522ac455
diff --git a/wger/core/api/serializers.py b/wger/core/api/serializers.py
index c37ddcb6f..68c9825d9 100644
--- a/wger/core/api/serializers.py
+++ b/wger/core/api/serializers.py
@@ -21,7 +21,7 @@ from django.contrib.auth import authenticate
from django.contrib.auth.models import User
from django.contrib.auth.password_validation import validate_password
from django.http import HttpRequest
-
+from lingua import LanguageDetectorBuilder
# Third Party
from rest_framework import serializers
from rest_framework.fields import empty
@@ -36,7 +36,6 @@ from wger.core.models import (
WeightUnit,
)
-
logger = logging.getLogger(__name__)
@@ -197,3 +196,60 @@ class RoutineWeightUnitSerializer(serializers.ModelSerializer):
class Meta:
model = WeightUnit
fields = ['id', 'name']
+
+
+class LanguageCheckSerializer(serializers.Serializer):
+ """
+ Serializer for language check
+ """
+ language = serializers.PrimaryKeyRelatedField(queryset=Language.objects.all(), required=False)
+ language_code = serializers.CharField(required=False, min_length=2, max_length=2)
+ input = serializers.CharField(min_length=10)
+
+ def validate(self, data):
+ """
+ Check that the detected language of the description corresponds with the
+ provided language.
+ """
+ language = data.get('language')
+ language_code = data.get('language_code')
+
+ if not language and not language_code:
+ raise serializers.ValidationError({
+ "language": "Either a language ID or a language code must be provided."
+ })
+
+ if not language:
+ try:
+ language = Language.objects.get(short_name=language_code)
+ except Language.DoesNotExist:
+ raise serializers.ValidationError({
+ "language": f'Language with code "{language_code}" does not exist.'
+ })
+
+ # Try to detect the language
+ detector = LanguageDetectorBuilder.from_all_languages().with_preloaded_language_models().build()
+ input_str = data.get('input')
+
+ detected_language = detector.detect_language_of(input_str)
+ detected_language_code = detected_language.iso_code_639_1.name.lower()
+ confidence_values = detector.compute_language_confidence_values(input_str)
+ logger.debug(
+ f"Detected language: {detected_language_code}, "
+ f"confidence values: {confidence_values}, "
+ f"input: {input_str}"
+ )
+
+ if detected_language_code != language.short_name.lower():
+ raise serializers.ValidationError({'check': {
+ "result": False,
+ "detected_language": detected_language_code,
+ "message":
+ f'The detected language is "{detected_language.name.capitalize()}" ({detected_language_code}), '
+ f'which does not match your selected language "{language.full_name.capitalize()}" '
+ f'({language.short_name}). If you believe this is incorrect, try adding more content '
+ f'or rephrasing your text, as language detection works better with longer or more '
+ f'complete sentences.'
+ }})
+
+ return super().validate(data)
diff --git a/wger/core/api/views.py b/wger/core/api/views.py
index 597c3da6b..c694554a1 100644
--- a/wger/core/api/views.py
+++ b/wger/core/api/views.py
@@ -23,7 +23,6 @@ from django.conf import settings
from django.contrib.auth.models import User
from django.utils.decorators import method_decorator
from django.views.decorators.cache import cache_page
-
# Third Party
from django_email_verification import send_email
from drf_spectacular.types import OpenApiTypes
@@ -37,7 +36,7 @@ from rest_framework import (
status,
viewsets,
)
-from rest_framework.decorators import action
+from rest_framework.decorators import action, api_view
from rest_framework.fields import (
BooleanField,
CharField,
@@ -56,7 +55,7 @@ from wger.core.api.serializers import (
RoutineWeightUnitSerializer,
UserLoginSerializer,
UserprofileSerializer,
- UserRegistrationSerializer,
+ UserRegistrationSerializer, LanguageCheckSerializer,
)
from wger.core.forms import UserLoginForm
from wger.core.models import (
@@ -74,7 +73,6 @@ from wger.version import (
get_version,
)
-
logger = logging.getLogger(__name__)
@@ -406,3 +404,14 @@ class RoutineWeightUnitViewSet(viewsets.ReadOnlyModelViewSet):
serializer_class = RoutineWeightUnitSerializer
ordering_fields = '__all__'
filterset_fields = ('name',)
+
+
+@api_view(['POST'])
+def check_language(request):
+ """
+ Checks the language of a string
+ """
+ serializer = LanguageCheckSerializer(data=request.data)
+ serializer.is_valid(raise_exception=True)
+
+ return Response({'result': True})
diff --git a/wger/core/tests/api/test_language_check.py b/wger/core/tests/api/test_language_check.py
new file mode 100644
index 000000000..0c11b3c5d
--- /dev/null
+++ b/wger/core/tests/api/test_language_check.py
@@ -0,0 +1,74 @@
+# This file is part of wger Workout Manager.
+#
+# wger Workout Manager is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# wger Workout Manager is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with Workout Manager. If not, see <http://www.gnu.org/licenses/>.
+
+# wger
+from wger.core.tests.api_base_test import ApiBaseTestCase
+from wger.core.tests.base_testcase import BaseTestCase
+
+
+class LanguageCheckApiTestCase(BaseTestCase, ApiBaseTestCase):
+ url = '/api/v2/check-language/'
+
+ def test_check_correct_language_with_code(self):
+ """
+ Test that the language is correctly detected - using language code
+ """
+ response = self.client.post(
+ self.url,
+ data={'language_code': 'de', 'input': 'Hallo Welt - das ist ein Test'}
+ ).json()
+ self.assertEqual(response, {'result': True})
+ self.assertFalse(response.get('check'))
+
+ def test_check_correct_language_with_id(self):
+ """
+ Test that the language is correctly detected - using language id
+ """
+ response = self.client.post(
+ self.url,
+ data={'language': 1, 'input': 'Hallo Welt - das ist ein Test'}
+ ).json()
+ self.assertEqual(response, {'result': True})
+ self.assertFalse(response.get('check'))
+
+ def test_check_wrong_language(self):
+ """
+ Test that the wrong language results in an error
+ """
+ response = self.client.post(
+ self.url,
+ data={'language_code': 'fr', 'input': 'a breed of hunting dog from Japan.'}
+ ).json()
+ self.assertTrue(response['check'].get('detected_language'))
+
+ def test_unknown_language(self):
+ """
+ Test that an unknown language returns an error
+ """
+ response = self.client.post(
+ self.url,
+ data={'language_code': 'zz', 'input': 'a breed of hunting dog from Japan.'}
+ ).json()
+ self.assertTrue(response.get('language'))
+
+ def test_no_languages(self):
+ """
+ Test that either a language ID or a language code must be provided
+ """
+ response = self.client.post(
+ self.url,
+ data={'input': 'a breed of hunting dog from Japan.'}
+ ).json()
+ self.assertTrue(response.get('language'))
diff --git a/wger/core/tests/test_permission_api.py b/wger/core/tests/api/test_permission_api.py
similarity index 97%
rename from wger/core/tests/test_permission_api.py
rename to wger/core/tests/api/test_permission_api.py
index acc114074..d5d3ff2a5 100644
--- a/wger/core/tests/test_permission_api.py
+++ b/wger/core/tests/api/test_permission_api.py
@@ -25,8 +25,8 @@ class CheckPermissionApiTestCase(BaseTestCase, ApiBaseTestCase):
url = '/api/v2/check-permission/'
error_message = "Please pass a permission name in the 'permission' parameter"
- def get_resource_name(self):
- return 'check-permission'
+ # def get_resource_name(self):
+ # return 'check-permission'
def test_check_permission_anonymous_no_parameters(self):
"""
diff --git a/wger/exercises/api/serializers.py b/wger/exercises/api/serializers.py
index 8b60d00b9..e7ba2f553 100644
--- a/wger/exercises/api/serializers.py
+++ b/wger/exercises/api/serializers.py
@@ -37,7 +37,6 @@ from wger.exercises.models import (
Variation,
)
from wger.utils.cache import CacheKeyMapper
-from wger.utils.constants import CC_BY_SA_4_LICENSE_ID
class ExerciseSerializer(serializers.ModelSerializer):
@@ -212,12 +211,31 @@ class ExerciseCommentSubmissionSerializer(serializers.ModelSerializer):
Custom create-method to handle the 'translation' keyword argument
and set the foreign key relationship.
"""
- translation = kwargs.get('translation')
+ translation: Translation = kwargs.get('translation')
if not translation:
raise serializers.ValidationError(
"A translation object is required for creating a comment."
)
+ # Validate the language of the description
+ # -> This is done here instead of in the serializer's validate method
+ # because the language is not available in the serializer's initial_data
+ detector = LanguageDetectorBuilder.from_all_languages().with_preloaded_language_models().build()
+ language = translation.language
+
+ # Try to detect the language
+ detected_language = detector.detect_language_of(validated_data['comment'])
+ detected_language_code = detected_language.iso_code_639_1.name.lower()
+ if detected_language_code != language.short_name.lower():
+ raise serializers.ValidationError({
+ "language":
+ f'The detected language of the comment is "{detected_language.name.capitalize()}"'
+ f'({detected_language_code}), which does not match your selected language: '
+ f'"{language.full_name.capitalize()}" ({language.short_name}). If you believe '
+ f'this is incorrect, try adding more content or rephrasing your text, as '
+ f'language detection works better with longer or more complete sentences.'
+ })
+
# Create the comment with the parent translation
comment = ExerciseComment.objects.create(
translation=translation,
@@ -399,11 +417,11 @@ class ExerciseTranslationSubmissionSerializer(serializers.ModelSerializer):
if detected_language_code != language.short_name.lower():
raise serializers.ValidationError({
"language":
- f'The language of the description was detected as "{detected_language.name.lower()}" '
- f'({detected_language_code}) and does not match the chosen language '
- f'{language.full_name} ({language.short_name}). If you are sure about the '
- f'language, try rewording it or adding more content since the detection '
- f'works best with longer texts.'
+ f'The detected language of the description is "{detected_language.name.capitalize()}"'
+ f'({detected_language_code}), which does not match your selected language: '
+ f'"{language.full_name.capitalize()}" ({language.short_name}). If you believe '
+ f'this is incorrect, try adding more content or rephrasing your text, as '
+ f'language detection works better with longer or more complete sentences.'
})
return super().validate(data)
@@ -568,7 +586,7 @@ class ExerciseSubmissionSerializer(serializers.ModelSerializer):
license = serializers.PrimaryKeyRelatedField(
queryset=License.objects.all(),
required=False,
- default=License.objects.get(pk=CC_BY_SA_4_LICENSE_ID),
+ # default=License.objects.get(pk=CC_BY_SA_4_LICENSE_ID),
)
translations = ExerciseTranslationSubmissionSerializer(many=True)
@@ -606,7 +624,7 @@ class ExerciseSubmissionSerializer(serializers.ModelSerializer):
# Create the individual translations
for translation_data in translations_data:
translation_serializer = self.fields['translations'].child
- translation = translation_serializer.create(
+ translation_serializer.create(
validated_data=translation_data,
exercise=exercise
)
diff --git a/wger/urls.py b/wger/urls.py
index 5e0275b4d..55c46dd5f 100644
--- a/wger/urls.py
+++ b/wger/urls.py
@@ -287,11 +287,13 @@ urlpatterns = i18n_patterns(
urlpatterns += [
path('robots.txt', TextTemplateView.as_view(template_name='robots.txt'), name='robots'),
# API
+ path('api/v2/', include(router.urls)),
path('api/v2/exercise/search/', exercises_api_views.search, name='exercise-search'),
path('api/v2/exercise/submission/', exercises_api_views.ExerciseSubmissionViewSet.as_view(),
name='exercise-submission', ),
path('api/v2/ingredient/search/', nutrition_api_views.search, name='ingredient-search'),
- path('api/v2/', include(router.urls)),
+ path('api/v2/check-language/', core_api_views.check_language, name='check-language'),
+
# The api user login
path(
'api/v2/login/', core_api_views.UserAPILoginView.as_view({'post': 'post'}), name='api_user'