🐛(back) allow ASCII characters in user sub field

All ASCII characters are allowed in a sub, we change the sub validator
to reflect this.

CHANGELOG.md

@@ -25,6 +25,7 @@ and this project adheres to
   - #1270
   - #1282
 - ♻️(backend) fallback to email identifier when no name #1298
+- 🐛(backend) allow ASCII characters in user sub field #1295 
 
 ### Fixed
 

src/backend/core/api/serializers.py

@@ -13,7 +13,7 @@ from django.utils.translation import gettext_lazy as _
 import magic
 from rest_framework import serializers
 
-from core import choices, enums, models, utils
+from core import choices, enums, models, utils, validators
 from core.services.ai_services import AI_ACTIONS
 from core.services.converter_services import (
     ConversionError,
@@ -422,7 +422,7 @@ class ServerCreateDocumentSerializer(serializers.Serializer):
     content = serializers.CharField(required=True)
     # User
     sub = serializers.CharField(
-        required=True, validators=[models.User.sub_validator], max_length=255
+        required=True, validators=[validators.sub_validator], max_length=255
     )
     email = serializers.EmailField(required=True)
     language = serializers.ChoiceField(

src/backend/core/migrations/0024_add_is_masked_field_to_link_trace.py

@@ -2,6 +2,8 @@
 
 from django.db import migrations, models
 
+import core.validators
+
 
 class Migration(migrations.Migration):
     dependencies = [
@@ -33,4 +35,17 @@ class Migration(migrations.Migration):
                 verbose_name="language",
             ),
         ),
+        migrations.AlterField(
+            model_name="user",
+            name="sub",
+            field=models.CharField(
+                blank=True,
+                help_text="Required. 255 characters or fewer. ASCII characters only.",
+                max_length=255,
+                null=True,
+                unique=True,
+                validators=[core.validators.sub_validator],
+                verbose_name="sub",
+            ),
+        ),
     ]

src/backend/core/models.py

@@ -14,7 +14,7 @@ from django.contrib.auth import models as auth_models
 from django.contrib.auth.base_user import AbstractBaseUser
 from django.contrib.postgres.fields import ArrayField
 from django.contrib.sites.models import Site
-from django.core import mail, validators
+from django.core import mail
 from django.core.cache import cache
 from django.core.files.base import ContentFile
 from django.core.files.storage import default_storage
@@ -39,6 +39,7 @@ from .choices import (
     RoleChoices,
     get_equivalent_link_definition,
 )
+from .validators import sub_validator
 
 logger = getLogger(__name__)
 
@@ -136,22 +137,12 @@ class UserManager(auth_models.UserManager):
 class User(AbstractBaseUser, BaseModel, auth_models.PermissionsMixin):
     """User model to work with OIDC only authentication."""
 
-    sub_validator = validators.RegexValidator(
-        regex=r"^[\w.@+-:]+\Z",
-        message=_(
-            "Enter a valid sub. This value may contain only letters, "
-            "numbers, and @/./+/-/_/: characters."
-        ),
-    )
-
     sub = models.CharField(
         _("sub"),
-        help_text=_(
-            "Required. 255 characters or fewer. Letters, numbers, and @/./+/-/_/: characters only."
-        ),
+        help_text=_("Required. 255 characters or fewer. ASCII characters only."),
         max_length=255,
-        unique=True,
         validators=[sub_validator],
+        unique=True,
         blank=True,
         null=True,
     )

src/backend/core/tests/documents/test_api_documents_create_for_owner.py

@@ -148,7 +148,7 @@ def test_api_documents_create_for_owner_invalid_sub():
     data = {
         "title": "My Document",
         "content": "Document content",
-        "sub": "123!!",
+        "sub": "invalid süb",
         "email": "john.doe@example.com",
     }
 
@@ -163,10 +163,7 @@ def test_api_documents_create_for_owner_invalid_sub():
     assert not Document.objects.exists()
 
     assert response.json() == {
-        "sub": [
-            "Enter a valid sub. This value may contain only letters, "
-            "numbers, and @/./+/-/_/: characters."
-        ]
+        "sub": ["Enter a valid sub. This value should be ASCII only."]
     }
 
 

src/backend/core/tests/test_models_users.py

@@ -44,3 +44,25 @@ def test_models_users_send_mail_main_missing():
         user.email_user("my subject", "my message")
 
     assert str(excinfo.value) == "User has no email address."
+
+
+@pytest.mark.parametrize(
+    "sub,is_valid",
+    [
+        ("valid_sub.@+-:=/", True),
+        ("invalid süb", False),
+        (12345, True),
+    ],
+)
+def test_models_users_sub_validator(sub, is_valid):
+    """The "sub" field should be validated."""
+    user = factories.UserFactory()
+    user.sub = sub
+    if is_valid:
+        user.full_clean()
+    else:
+        with pytest.raises(
+            ValidationError,
+            match=("Enter a valid sub. This value should be ASCII only."),
+        ):
+            user.full_clean()

src/backend/core/validators.py

@@ -0,0 +1,9 @@
+"""Custom validators for the core app."""
+
+from django.core.exceptions import ValidationError
+
+
+def sub_validator(value):
+    """Validate that the sub is ASCII only."""
+    if not value.isascii():
+        raise ValidationError("Enter a valid sub. This value should be ASCII only.")